MR. JUSTICE NEUBERGER:
A. INTRODUCTORY
The parties
- This case comprises a number of conjoined patent actions relating to the protein erythropoietin ("EPO"). Two patents are involved. They are European Patent (UK) No: 0148605B2 ("the 605 patent" or "605") and European Patent (UK) No: 0411678B ("the 678 patent" or "678").
- Over ten different parties are concerned in the actions, but they can be effectively divided into three groups. First, there are Kirin-Amgen Inc., Ortho-McNeil Pharmaceutical Inc., Ortho Biotech Inc., Ortho Biotech Products LP and Janssen-Cilag Ltd ("Amgen"). Secondly, Roche Diagnostics GmbH, Roche Diagnostics Ltd and Genetics Institute Inc. ("Roche"). Thirdly, there are Hoechst Marion Roussel Inc., Hoechst Marion Roussel Ltd and Transkaryotic Therapies Inc. ("TKT"). For the purpose of these proceedings the precise nature of the interest or involvement of each member of each group does not matter.
- The registered proprietor of the 605 patent is Amgen, and the registered proprietor of the 678 patent is Roche.
- So far as 605 is concerned, Roche and TKT each contend that it should be revoked; their respective arguments for so saying are, in the main, somewhat different, but they are not mutually inconsistent. Amgen argues that Roche and TKT infringe 605. The nature of Roche's processes is rather different from those of TKT.
- As to 678, Amgen argues that it should be revoked, and Roche contends that Amgen infringes. TKT is not involved in the arguments on 678, save, of course, to the extent that they may impinge on the arguments relating to 605.
- The 605 Patent has been considered by the Technical Board of Appeal of the European Patent Office on two occasions (T412/93, on 21st November 1994 and T636/97 on 26th March 1998). The 678 Patent was also considered by the Technical Board of Appeal on 16th April 1999, T277/95.
- The 605 Patent has also been the subject of proceedings in two other Member States of the European Patent Convention ("EPC") and elsewhere. Thus it has been considered by the District Court of the Hague on 13th March 1993, upheld by the Netherlands Appeal Court on 27th January 2000 (Kirin-Amgen Inc. -v- Boehringer Mannheim GmbH). It has also been the subject of a decision of the German Patent Court on 14th December 2000, (Hoffman La Roche A.G. -v- Kirin-Amgen Inc.). There have also been decisions of the Federal Court and Court of Appeal in Canada (15th February 1999 and 20th December 2000, Kirin-Amgen Inc. -v- Hoffman La Roche Ltd) and of the Federal Court of Australia (25th June 1998, Genetics Institute Inc. -v- Kirin-Amgen Inc.) in relation to similar patents in their jurisdiction. There have also been proceedings on the equivalent US Patent in the District Court of Massachusetts and Court of Appeal, Federal Circuit (11th December 1989 and 5th March 1991, Amgen Inc. -v- Chugai Pharmaceutical Co. Ltd).
Erythropoietin
- Natural human EPO is a hormone which circulates in the bloodstream of healthy people. It stimulates the production of red blood cells. Those cells are, of course, an essential component of blood. They contain haemoglobin, the agent which carries oxygen from the lungs to those parts of the body which require oxygen. Anaemia is a condition which involves the lack of sufficient red blood cells; it leads to loss of the oxygen-carrying capability of blood, which results in a serious, often indeed debilitating, loss in the quality of life.
- In healthy humans, EPO is produced, albeit in very small, indeed minuscule, quantities in the kidneys. Only very little EPO is needed, at least partly, because red blood cells have a relatively long life - about four months. However, when the level of oxygen in the blood drops, the kidneys respond by manufacturing more EPO. In 1983, it was known that EPO production was stimulated by hypoxia (reduced oxygen transported by blood cells).
- People suffering from kidney failure are normally treated by a blood purification process. It was discovered that they nonetheless become anaemic in the absence of additional treatment, so they required regular blood transfusions. By 1967, it was realised that EPO deficiency was the key factor explaining this anaemia, and it was hoped that administration of EPO to such patients might deal with the problem, if and when EPO became available in sufficient quantities. Because EPO was naturally produced in such very small amounts, it proved difficult to isolate it in any significant quantity. Nor was it easy to determine the structure of EPO, or the gene responsible for its production, although, at least according to Amgen's case, considerable effort was put into this exercise. By 1977 very small amounts of EPO had been reported as isolated from the urine of patients suffering from aplastic anaemia.
- On Amgen's case, Dr Fu-Kuen Lin, a scientist then employed by Amgen, made a breakthrough; he cloned the DNA responsible for the manufacture of EPO - i.e. EPO DNA. Amgen contend that his invention not only provided the DNA sequence that codes for EPO, but also new glycoprotein products, which are therapeutically useful pharmaceutical compositions, cells and processes to make EPO available in sufficient quantity and quality to treat the hundreds of thousands of patients world-wide suffering from various types of anaemia.
- Indeed, accurately, so far as I can see, Amgen also argue that EPO manufactured substantially in accordance with the teaching of Dr Lin in 605, "recombinant EPO", reverses the most debilitating effect of renal failure not solved by dialysis. They also say that patients to whom recombinant EPO is administered have received an enormous benefit and that the availability of recombinant EPO has been the most important advance in the treatment of end stage renal failure in the past 20 years.
Outline of issues
- The hearing of these proceedings lasted over 25 full days, during which I heard evidence from a number of very eminent scientists relating to a number of different issues of fact and expert opinion. I also entertained argument relating to a fair number of different points, some of which may be of some general significance, in the field of patent law, and especially in the field of genetic engineering patents.
- I propose, in Part B, to set out the technical background, without which anybody ignorant of the science of, and techniques involved in, genetic engineering (such as myself until very recently) could not understand the issues and arguments. I shall then turn, in Part C, to the state of the common general knowledge as at the relevant time for the purpose of 605 - i.e. around 1984. When dealing with the common general knowledge, I will trespass into developments which have occurred since the relevant time, hopefully making it clear when I am doing so. While, by definition, developments after 1984 could not have been known as at the relevant date, those developments are relevant when considering issues of infringement, as well, at least arguably, of construction, insufficiency and even novelty.
- I will then, in Part D, seek to summarise, and quote the centrally relevant parts of, the 605 patent itself. It is, perhaps inevitably, a long document, involving a net 24 pages (each running to some 60 lines) of description, including 12 Examples (some of which are two-pronged), a total of 31 claims, 21 tables of sequences (one of which runs to five pages) and 4 figures.
- I shall then deal, in Part E, with various issues of construction which have been raised in relation to 605. I think it is then appropriate to deal with an issue which in a sense is one of construction, albeit that it could also be said to be a question of policy, and could, indeed, be dealt with in relation to the argument that the patent, or certain claims in the patent, are invalid for want of novelty. However, because it is arguably an issue of construction, and also because it is a point of principle, I propose to deal with the construction and effect of product by process claims immediately after I have discussed the other issues of construction, in Part F.
- So far as 605 is concerned, that then leaves issues of infringement and validity. Reported cases suggest that it is more normal to deal with infringement before validity, but I propose to deal with the validity issues first. This is primarily because there is one important and difficult issue relating to insufficiency and breadth of claim, which appears to me to have a significant bearing on, or at least interrelationship with, the issue of whether TKT infringes 605. I believe that it is therefore more convenient to consider that insufficiency and breadth of claim question before turning to infringement.
- There are a number of issues relating to the validity of the 605 patent. First, there are a number of insufficiency allegations in relation to 605 and its claims. Some of these allegations are of a detailed and technical nature; others raise points of principle. Because the insufficiency issues need fully detailed discussion, I have divided up the issues. In Part G, I set out the law on this topic and summarise the issues. In Part H, I deal with the one specific insufficiency allegation which requires fairly extensive consideration. In Part I, I turn to various other insufficiency issues relating to Claim 1 of 605. Part J is concerned with the insufficiency arguments on Claims 19 and 20. In Part K, I deal with the permissible breadth of the Claims in light of the approach of the House of Lords in Biogen Inc -v- Medeva plc [1997] RPC1.
- Having dealt with insufficiency, I will deal in Part L with the argument that 605 is invalid because it constitutes a discovery rather than an invention. In Part M, I will then turn to the attack on 605 based on the contention that it suffers from lack of novelty. The final attack on 605 which I must discuss is that it is invalid on the ground that it contains added matter: that is in Part N.
- Having considered the issues relating to the alleged invalidity of 605, I will then turn to the question whether Roche infringe 605 (Part O), and then to the question whether TKT infringe 605 (Part P). I will then turn, in Part Q, to deal with 678, Roche's patent. In summary, the issue (albeit that it raises a number of sub-issues) is whether, as Amgen contends, 678 was anticipated by 605, and is therefore invalid. I will then summarise my conclusions, in Part R.
The witnesses
- The expert evidence in this case was extensive in terms of quantity (and indeed volume of paper) and impressive in terms of quality. The witnesses were experts in one of two main fields (albeit that there was, inevitably, a degree of overlap in the expertise of at least some of the witnesses). Those fields were molecular biology and glycoprotein chemistry and biochemistry. When preparing for trial, and indeed in opening, the three parties tried to ensure that there was no significant overlap between the evidence given by one expert and the evidence given by another expert for the same party. To the extent that there was any such overlap, it was agreed that any cross examination on the issue need only be directed to one of the experts (to be selected by the party calling the witnesses).
- Amgen called three expert witnesses in relation to the molecular biology issues. Professor Randolph Wall is a Professor in the Department of Microbiology and Immunology at the UCLA School of Medicine and a member of the UCLA Molecular Biology Institute. He received his PhD in microbiology in 1970, and for the last 28 years his interest has focused mostly on molecular genetics and immunology using recombinant DNA technology, and in particular cloning. His evidence was directed primarily to the state of recombinant DNA and cloning technology as at the end of 1983, the attacks on the validity of 605 on the grounds of lack of novelty and insufficiency, and the alleged infringement of 605 by Roche.
- Dr Sydney Brenner obtained a D.Phil. at Oxford, and worked in the Medical Research Council - "the MRC" - Laboratory of Molecular Biology from 1956 to 1987 (as the Director for the last 9 years). He then became a director of the MRC Unit of Molecular Genetics in Cambridge until 1992. He has retired from that post, and he is now a director of the Molecular Sciences Institute in Berkeley, California. It was as a result of research which he carried out with Dr Francis Crick that the genetic code was shown to be the triplet code, or codon, a concept I will try and explain in a little more detail below. His evidence involved explaining in a little detail the technology involved in, and disclosed by, 605, and the technology involved in TKT's work (so-called "gene activation technology") leading to TKT's allegedly infringing product, and the connection, at least as he saw it, between the disclosure of 605 and TKT's work.
- Dr Michael Gait is a senior group leader in the MRC Laboratory of Molecular Biology at the MRC Unit in Cambridge. Throughout his career since 1973, he has researched and practised innovative methods of synthesising oligonucleotides and synthetic genes, and has carried out related work. His evidence was directed towards the state of the art in synthesis of genes in 1984, and the feasibility of carrying out some of the work taught by the 605 patent, or suggested in the evidence or argument.
- Roche also called three expert witnesses on the molecular biology issues. Professor William Brammar has been Professor of Biochemistry at Leicester since 1977, having acquired his PhD in microbial physiology at University College London during the 1960s. His evidence was directed towards most of the issues which Professor Wall dealt with on behalf of Amgen, save that he was not concerned with the issue of whether Roche infringed 605, although he did consider whether human cDNA was claimed or enabled by 605, an issue which is of relevance to the question of Roche's infringement.
- Professor Hans Gassen is the Director of the Institute of Biochemistry at Darmstadt University of Technology in Germany, having obtained his doctorate in 1966. Much of his research has been in the area of synthesis, analysis and biological functions of DNA and RNA. In terms of his evidence, he was effectively the opposite number of Dr Gait.
- Dr Edward Fritsch worked for Genetics Institute Inc. ("GI", one of the Roche parties) from 1982 until 2000, having been a Vice-President from 1986. He was called in part as a witness of fact, because he was part of the GI team seeking to sequence and obtain EPO and the EPO gene and to find a way of expressing EPO in commercial quantities, at the same time as Dr Lin was carrying out his research. However, he also gave expert evidence, particularly in relation to human tissue as a possible source of EPO in 1984, and also as to certain cell lines.
- TKT called one expert witness on the molecular biology issues. He was Professor Nicholas Proudfoot, the Professor of Experimental Pathology at Oxford University. He has worked at Oxford since 1984, having obtained his PhD at Cambridge in 1975. His evidence was directed to the state of the art in 1984, how the various claims of 605 would strike the notional skilled reader at the time, and a comparison of TKT's technology with the state of the art in 1984 and the teaching of 605.
- On the glycoprotein issues, Amgen called two expert witnesses. The first was Professor Richard Cummings, Professor of Biochemistry and Molecular Biology in the University of Oklahoma Health Sciences Centre. Since 1980, he has worked in molecular biology. His research has concentrated on glycoproteins, and in particular the carbohydrate portions, including biosynthesis, structure and their biological effects. He gave evidence relating to the construction, sufficiency and novelty of 605, and also relating to the added matter issue. His evidence also extended to the issues relating to 678.
- Sir John Walker worked at the MRC Laboratory of Molecular Biology between 1974 and 1998, and is now Director of the MRC's Dunn Human Nutrition Unit, both in Cambridge. He is a Fellow of the Royal Society and a Nobel Laureate in Chemistry. His evidence related to the isolation and purification of EPO, and the carrying out of SDS-PAGE and blotting experiments (whose function is to identify the approximate molecular weight of polypeptides).
- Roche called one expert on the glycoprotein issues. He was Professor Henrik Clausen, the Senior Associate Professor (Docent) in the Department of Oral Diagnostics in the School of Dentistry at Copenhagen University; he is also a Visiting Professor of Harvard Medical School. Throughout his career, he has concentrated on research in protein and carbohydrate chemistry and immunology, and over the past 10 years his research has focused on the biology of carbohydrates and biosynthesis and genetic regulation of glycosylation of proteins. His evidence related to the purity and isolation of glycoproteins, a comparison of urinary EPO and recombinant EPO by reference to published papers, and experiments carried out for the purpose of these proceedings. His evidence was also directed to the insufficiency issues relating to 605 (so far as they were glycoprotein, as opposed to molecular biology, issues) and the validity of 678.
- TKT called two expert witnesses on the glycoprotein issues. Professor Phillips Robbins is Professor Emeritus in the Department of Biology at MIT and Professor in the Department of Molecular and Cell Biology in Boston University of Dental Medicine. He obtained a PhD in Biochemistry in 1955, and much of his research has been into topics connected with the glycosylation of proteins. His evidence, which was of a fairly general nature (which does not in any way cast doubt upon its expertise or relevance) concerned the state of knowledge as to the glycosylation of proteins in general, and of EPO in particular, both as at the relevant date for 605 and thereafter.
- Professor Paul Matsudaira is a member of the Whitehead Institute for Biomedical Research, and Professor of Biology in Bioengineering at MIT, and obtained his PhD in biology in 1981. Much of his research has involved electrophoresis and electrophoretic methods, which includes SDS-PAGE and blotting. His evidence was largely concerned with analysing, and expressing views about, relevant experiments which were reported in learned journals or had been carried out for the purpose of these proceedings.
- There were also some witnesses of fact, apart from Dr Fritsch. Amgen put in written evidence from Dr Christopher Winearls, a consultant nephrologist at the Renal Unit at the Oxford Radcliffe Hospital. He was the lead clinician in the research team which first described the successful use of Amgen's recombinant EPO to treat renal anaemia. He was also involved in trials of such EPO in transfusion-dependent patients. His evidence explained the substantial benefits of having recombinant EPO available to treat patients with chronic renal failure or anaemia. There was also written evidence from Professor Stuart Orkin, Professor of Paediatrics at Harvard Medical School, who worked on a number of different genetic technology projects. One of those projects related to EPO, and started in 1981 and lasted for about 3 years. Like Dr Fritsch, therefore, he was in a group which was effectively working in competition with Dr Lin with a view to obtaining and sequencing the EPO gene and EPO itself, and thereafter seeking to express EPO in cells. Amgen also put in a witness statement from Dr Lin, the inventor of 605.
- Evidence of fact was also given by Dr Thomas Strickland. He obtained his PhD in biochemistry in 1981, was a consultant for Amgen between 1995 and 1999, and then became employed by Amgen, for whom he now works as Quality Technical Evaluator. He carried out experiments for the purpose of these proceedings, including purification procedures with a view to isolating EPO from human urine, and then running the product on SDS-PAGE against certain recombinant EPOs. There was also evidence from Dr Jeffrey Browne, who received a PhD in molecular biology in 1982, and worked for Amgen between 1981 and 1998. Although not directly responsible for much of the work carried on in Amgen's laboratory in 1984 and 1985, investigating the performance of various urinary EPOs and various recombinant EPOs on SDS-PAGE, he was indirectly involved in some of the work, and was named in one or two papers published in that connection in learned journals. Dr Strickland's name also appeared on at least one of those papers.
- Amgen also put in a proof from Professor Julian Davies, Professor of Microbiology and Immunology at the University of British Columbia, who worked for Biogen, another well-known major commercial company carrying on research in the biotechnology field, between 1981 and 1985. He explained how, between 1980 and 1983, a team of scientists at Biogen, of which he was one, tried to clone the EPO gene. Amgen also put in one or two other witness statements which I do not think necessary specifically to describe, at any rate at this stage.
- TKT had a witness of fact. He was Dr Michael Heartlein, Vice President, Molecular Biology, at TKT, having obtained his doctorate in genetics in 1984. His evidence concerned the experiments carried out by TKT in connection with its technology, and in particular the experiments carried out in relation to seeking to identify the optimum location for the artificial promoter relative to the encoding part of the EPO gene.
- Apart from the experiments carried out in Amgen's laboratories in 1984 and 1985, to which I have already briefly referred, and the work carried out by Dr Strickland on behalf of Amgen for the purpose of these proceedings, some experiments were also put in on behalf of Roche and on behalf of TKT. I will deal with the nature of, evidence and arguments about, and the effect of, those experiments in due course in this judgment.
- The expert witnesses from whom I heard were all very distinguished. I have very briefly summarised their respective careers, and have not gone into their many distinguished posts, the very substantial number of publications in prestigious journals they have published, the books they have written, edited or contributed to, the courses they have run, or the honours which they have been awarded. Inevitably, some of the experts would be regarded, at any rate by some people, as even more distinguished than others. There is no suggestion that any of these witnesses was anything other than a highly respected expert in the field in which he professed to practice, and more than competent to give evidence on the topic or topics on which he spoke.
- Equally, there is no attack on the honesty of any of the experts, quite rightly in my view. However, it is alleged against some of the expert witnesses that they were not as impartial as they are required to be by the Civil Procedure Rules ("the CPR"). I have to say that, in respect of one expert, namely Professor Cummings, I consider that allegation to be well-founded; the reasons for this will appear later. However, it is only fair to Professor Cummings to make it clear that this in no way impugns his expertise or honesty. He certainly did not strike me as so partisan as to lead me to conclude that his evidence should be rejected whenever it conflicted with evidence the other way. On the contrary, on some of the issues on which he was in conflict with other experts, I found his evidence more convincing than theirs.
B. TECHNICAL BACKGROUND
Proteins
- EPO is a protein. Proteins are molecules which carry out many vital biochemical functions. They consist of one or more chains of polypeptides. (EPO only has one polypeptide chain, so I shall ignore proteins with more than one chain). A polypeptide is a chain of amino acids. There are twenty different amino acids, and some of them are more common than others. Each amino acid has a carboxyl group (-COOH) at one end and an amino group (-NH2) at the other end. Each amino acid can be written:
H
|
NH2—C—COOH, where R is its specific constituent, which varies from one
|
R
amino acid to another. This specific constituent is known as the side chain of the amino acid, because, when the amino acid is in a polypeptide, this constituent is, in effect, a branch off the polypeptide chain.
- In a polypeptide, amino acids are linked by a peptide bond, which is formed with the loss of an -OH from the carboxyl group of one amino acid and an -H from the amino group of the other acid, i.e. with the effective loss of a water molecule. Thus, the simplest polypeptide (i.e. one with two amino acids) has one peptide bond and may be written:
H O H
| || |
NH2—C—C—N—C—COOH. Strictly each amino acid component in a
| | |
R1 H R2
polypeptide or protein is referred to as an amino acid residue, but a residue is sometimes called simply an amino acid.
- Many side chains are subjected to modification during the process of their formation. An important and relevant example of such modification is the addition of saccharides (or sugar residues), a process known as glycosylation.
- While glycosylation does not normally occur in bacteria, it does occur in higher organisms. Glycosylation can involve addition of a single sugar residue, or a substantial number of sugar residues. Such residues can substantially vary in size, structure, and composition one from the other. There are two common types of glycosylation, N-glycosylation and O-glycosylation. The reference to N and O is a reference to the atom (nitrogen and oxygen respectively) to which the carbohydrate residue (known as the "glycan") attaches. A glycosylated protein is known as a glycoprotein. EPO is in a glycoprotein.
- In a polypeptide, the amino acid chain is treated as starting at the amino end of the first amino acid, known as the N-terminus or NH2-terminus, and ending with a carboxyl group of the last amino acid, known as the C-terminus or COOH-terminus. A polypeptide only becomes a protein once it is folded into the correct shape, often with cross linking between cysteine amino acids known as disulphide bridges. If a protein loses its correct shape, it ceases to be effective as a protein: it is denatured. A protein also has some internal sequences - known as active sites - which are more important for its biological functions than other such sequences.
- Although a given protein has a specific sequence of amino acids in its polypeptide chain, a change, or indeed a number of changes, in the identity of the amino acids in a protein may not destroy its effectiveness. A single change might reduce or even destroy a protein's effectiveness; another change or other changes might have no effect on, or actually improve, its effectiveness; sometimes a change can result in side effects. Variants of protein where amino acid residues have been changed (or even removed) without the effectiveness of the protein being lost are known as analogues of the protein.
- In general the greater the number of changes in the amino acids in a given protein the more likely it would be that its effectiveness will be diminished or will cease. Further, as I have mentioned, some of the amino acid sequences in a protein were of greater importance than others. Accordingly, changes, even a number of changes, in the amino acids in less significant locations might have little, if any effect; whereas a single change in a more sensitive area, in particular the protein's active site or (if more than one) its active sites, could destroy the effectiveness of the protein. The characteristics of amino acids vary, so a change from one amino acid to another with similar characteristics might be expected to be less significant than a change to an amino acid with different characteristics.
The genetic code
- In order to make (or "express") a protein, the amino acids have to be assembled in the correct sequence. The sequence for a particular protein is specified by a gene which is present in the genetic material of a cell - i.e. within the DNA in the cell. The "genetic code" is the relationship between a sequence of genetic material and the amino acid sequence of the corresponding protein.
- All living organisms are composed of cells. Eukaryotic cells have a nucleus containing the cell's genetic material; the nucleus is surrounded by a cytoplasm which is in turn bordered by the cell membrane. Prokaryotic cells have a single cellular compartment bounded by a cell membrane and a cell wall. Mammals (including humans) plants and fungi are made up of eukaryotic cells; bacteria are made up of prokaryotic cells.
- The genetic material in a cell is made from deoxyribonucleic acid, DNA, which consists of two strands of complexes called nucleotides. Each nucleotide includes a nitrogenous base, a phosphate and a sugar. DNA contains four different types of nucleotide, each of which is characterised by a different nitrogenous base, namely Adenine, Guanine, Cytosine and Thymine, abbreviated respectively to A, G, C and T. The nucleotide bases in DNA are linked by bonds between a carbon atom of one nucleotide, known as a 3'-carbon, and another carbon atom, known as the 5'-carbon, of the neighbouring nucleotide. By convention a strand of DNA begins with the 5'end and finishes with the 3'end. The 5'end is "upstream" (to the left of the DNA sequence as written out) and the 3'end is "downstream" (to the right).
- In its normal state, DNA consists of two strands of nucleotides in the well-known "double helix" configuration. Each of the two strands of DNA is complementary - i.e. linked to the other strand - by the invariable rule that A pairs with T, and G pairs with C. In eukaryotic cells, the double helix DNA is divided into a number of different chromosomes, which are compact structures, in which the DNA is complexed with many proteins.
- Cells divide and duplicate their genetic material. Accordingly, on division of a cell, each DNA molecule becomes two identical molecules each carrying the same information as was in the original. This process, known as replication, involves the two strands of DNA separating so that each strand acts as a template, with new nucleotides being added one by one on the basis already mentioned, namely A pairing with T, and G with C.
- The order of the nucleotides in a particular segment of DNA - called a gene - provides the blueprint for the "expression" of a particular protein. In effect, the DNA blueprint is a succession of code words, each consisting of a set of three nucleotides, known as codons, in the gene. Each codon (i.e. each set of three nucleotides) specifies a particular amino acid. With three exceptions, a particular codon instructs the cellular machinery to add a particular amino acid to a growing polypeptide chain. The three exceptions are "stop codons" which instruct that machinery where to stop the manufacture of the protein.
- As there are twenty different amino acids and four different nucleotide bases, once one excludes the three stop codons, there are 61 possible permutations of bases to encode 20 amino acids. Two amino acids are encoded by only one group of three bases, i.e. by only one codon. Some amino acids are encoded by as many as six different codons. The majority of amino acids are encoded by two or four codons.
- This feature of the genetic code, namely that (with two exceptions) each amino acid can be encoded by more than one codon, is known as degeneracy. As a result of degeneracy, once a protein's amino acid sequence is known, many (normally many millions or billions) of different possible gene sequences could theoretically encode that protein. However, only very few may in fact do so.
Expression of proteins
- The cells of a given organism almost all contain the same DNA, namely the same genetic material which, according to the evidence, can encode for tens of thousands of different proteins ("the genome"). However, each particular type of cell produces only a few of the proteins encoded by the DNA. Thus virtually all human cells contain DNA encoding for insulin, but only pancreatic cells actually produce that protein. In eukaryotic cells, as I have mentioned, DNA is divided into a number of different chromosomes, in which the DNA is combined with certain proteins. DNA sequences in a particular cell may, depending on their chromosomal regions, be "open" or "closed". If "open" (like the insulin gene in pancreatic cells), the sequence is capable of expression; if "closed", it is not. The EPO gene is included in the DNA in Chromosome 7.
- When a cell responsible for the production of a particular protein detects the need for that protein, the relevant segment of DNA that codes for that protein (i.e. the relevant gene) produces a substance related to DNA, called ribonucleic acid, RNA. This is made from a slightly different nucleic acid from DNA, and consists of a single strand rather than a double strand. Further, in place of one of DNA's bases, Thymine, it has a different base, Uracil: it contains U instead of T. The RNA manufactured in the cell is known as mRNA - i.e. Messenger RNA. mRNA is made by a process called transcription. Transcription involves an enzyme called RNA polymerase travelling along the relevant gene making the single stranded RNA with bases complementary to the bases on the DNA. Thus, A is reproduced as U (there being no T in RNA), G is reproduced as C, T is reproduced as A, and C is reproduced as G.
- In higher organisms (i.e. those with eukaryotic cells), structural genes have non-coding sequences of bases, called "introns", between the portions, called "exons", which correspond to the portions of the mRNA which encode the protein. The introns are effectively edited out before translation of the mRNA, leaving only those regions of codons in the mRNA which are protein encoding, i.e. the exons. This process, which involves the removal of the segments which do not encode for the protein (most notably the introns) and joining the encoding exons together, is known as splicing. The splicing process is determined by sequences near the exon-intron junctions and the intron-exon junctions in the RNA; these sequences are known respectively as splice donor sites and splice acceptor sites.
- Once the transcription of the mRNA is complete, its code must be "translated" to the amino acid sequence. For this purpose, the RNA is released from the cell cytoplasm (in the case of a eukaryotic cell). Translation takes place in a complex known as the ribosome, a microscopic structure in the cell. The ribosome effectively reads off the base sequence of the whole of the mRNA (starting at the 5'end with a start codon) and translates it into polypeptide by use of another type of RNA called transfer RNA, tRNA. This translation ends when a stop codon is reached. The tRNA contains the appropriate codons for the amino acids of the particular protein, and is joined by specific enzymes to the amino acid which corresponds to the particular codon. These amino acids are expressed and then joined by peptide bonds, with the assistance of an enzyme called peptidyl transferase.
- Genes from higher organisms also have a non-coding regulatory region before the encoding regions (i.e. upstream of the first encoding region) and another non-encoding region downstream of the last exon. The start of the gene, i.e. where the RNA polymerase starts transcribing, is known as the cap site. The regulatory region controls the transcriptional activity of the gene. The promoter is that portion of the regulatory region which initiates the transcription and is upstream of the cap site. The promoter is thus upstream of the encoding region of the gene, and it provides a binding site for RNA polymerase, the enzyme which transcribes the gene into mRNA. In fact, there are sometimes two promoter sites in a particular gene. Immediately before the start of the protein encoding region is a series of nucleotides known as the Kozak sequence, which, in effect, helps the ribosome to recognise when translation should start. Dr Kozak found that certain sequences worked better than others for this purpose.
- The encoding region of the gene which actually transcribes is also called the structural gene or the reading frame. Failure to translate the mRNA precisely from the correct reading frame (i.e. the correct triplet of nucleotides) results in failure to produce the desired protein. The first codon, known as the initiation codon, of the first encoding exon in a gene is almost always ATG (or AUG in the corresponding RNA). Immediately after the last encoding exon of the gene is the terminator sequence which instructs the RNA polymerase to stop copying the DNA. There then follows the downstream non-coding region, which almost always ends with a number of consecutive adenines - the so called "poly A tail".
- Chemical modifications may occur to proteins immediately after, or even during, their expression. Thus, after expression, mammalian proteins are normally glycosylated - i.e. sugar or saccharide sub-units are chemically attached to some side chains of the protein. Glycosylation can enable a protein to persist longer in the bloodstream. Further, after (or during) translation, certain amino acids may be cleaved from the protein. In many cases, the first several amino acids at the N-terminus of the protein serve as a chemical signal directing the protein into the first of a series of membrane-enclosed compartments within the cell. These initial amino acids are often referred to as a leader peptide. The leader peptide, which marks the protein for secretion from the cell, causes the protein to enter the secretory pathway, which is a series of membrane-enclosed compartments within the cytoplasm of the cell.
- The first of the compartments within the secretory pathway is called the rough endoplasmic reticulum ("RER"). As a protein produced with a leader peptide emerges from the ribosome, it directs the ribosome to bind to the RER so that the growing protein is inserted through the membrane of the RER into its internal space. The leader peptide sequence may then be cleaved from the protein (which therefore becomes "mature"), and the protein remains in the central cavity of the RER. It is also within the RER that initial glycosylation occurs. Then, the protein moves to another array of membrane-enclosed compartments, called the Golgi apparatus ("the Golgi"), where certain modifications to the sugar chains take place. Finally, the protein is secreted by the cell and travels through the body to act on other cells.
C. COMMON GENERAL KNOWLEDGE IN 1984 AND RECENT SUBSEQUENT DEVELOPMENTS
Common general knowledge: the law
- A patent must be construed as at the relevant date through the eyes of a person (or, as in the present case, a team of persons) suitably skilled, albeit not inventive, in the relevant art or arts. Such an addressee is "likely to have a practical interest in the subject matter of the invention (i.e. "skilled in the art")" and is deemed to have "practical knowledge and experience of the kind of work in which the invention was intended to be used" - see per Lord Diplock in Catnic Components Limited -v- Hill & Smith Limited [1982] RPC 183 at 242 to 243.
- The skilled addressee is treated as informed with "common general knowledge", and the extent of this common general knowledge has been discussed in a number of cases. In particular, I would refer to Beloit Technologies Inc. -v- Valmet Paper Machinery Inc. (No. 2) [1997] RPC 489 at 494 line 21 to 495 line 28 and Raychem Corporation's Patents [1999] RPC 497 at 503 line 39 to 504 line 19. In the latter case, Laddie J said:
"The common general knowledge... is not limited to material [the notional addressee] has memorised and has at the front of his mind. It includes all that material in the field he is working in which he knows exists, which he would refer to as a matter of course if he cannot remember it and which he understands is generally regarded as sufficiently reliable to use as a foundation for further work or to help understand the pleaded prior art. This does not mean that everything on the shelf which is capable of being referred to without difficulty is common general knowledge nor does it mean that every word in a common text book is either. In the case of standard text books it is likely that all or most of the main text will be common general knowledge."
- In the former case, Aldous LJ said:
"It has never been easy to differentiate between common general knowledge and that which is known by some. It has become particularly difficult with the modern ability to circulate and retrieve information. ...The notional skilled addressee is the ordinary man who may not have the advantages that some employees of large companies may have. The information in a patent specification is addressed to such a man and must contain sufficient details for him to understand and apply the invention.
...It follows that evidence that a fact is known or even well-known to a witness does not establish that that fact forms part of the common general knowledge. Neither does it follow that it will form part of the common general knowledge if it is recorded in a document."
While this emphasises that the notional skilled man is not the highest common denominator in the field, it seems to me equally clear that he is not the lowest either.
- Aldous LJ went on to quote with approval what Luxmoore J said in British Acoustic Films 53 RPC 221 at 250:
"[I]t is not sufficient to prove common general knowledge that a particular disclosure is made... in a scientific journal, no matter how wide the circulation of that journal may be, in the absence of any evidence that the disclosure is accepted generally by those who are engaged in the art to which the disclosure relates. The piece of particular knowledge as disclosed in the scientific paper does not become common general knowledge merely because it is widely read, and still less because it is widely circulated. Such a piece of knowledge only becomes general knowledge when it is generally known and accepted without question by the bulk of those who are engaged in the particular art; in other words, when it becomes part of their common stock of knowledge relating to the art."
- Accordingly, particularly in a field where the technology is unfamiliar, save perhaps in the most general terms, even to members of the public who may consider themselves to be generally well educated, the court will rely very much on expert evidence before it in deciding whether a particular fact was or was not within the stock of common general knowledge of the skilled addressee at the relevant date. As with expert evidence on questions of construction, the court will normally have to bear in mind that the experts from whom it is hearing will normally suffer from two problems. The first is that they are giving evidence about the state of knowledge as at a date significantly earlier than that on which they are considering the matter; particularly in an area where the science and technology are advancing very fast, as in the present case, that is a particular difficulty. Secondly, the court will normally be hearing from experts who are considerably more skilled, considerably more informed, and indeed inventive, than the notional addressee.
- However, this should not cause the court to take too restrictive a view of what constitutes common general knowledge, not least because to some extent these two points cut both ways. In a fast growing area, the amount of available information in the form of publications and other communications, is likely to be substantially greater at the hearing date than at the publication date, particularly where the gap, as here, is more than 15 years. To that extent, it would be easier for a particular fact to have been common general knowledge at an earlier date, as that fact would have had, as it were, less competition. Secondly, there is no intrinsic reason why a person who was inventive or particularly skilled in the field should have more knowledge than an uninventive person of average skill; indeed, it is not unknown for the averagely skilled and uninventive in any field to be rather better informed than their more inventive and skilful colleagues. On the other hand, where, as here, the expert witnesses are people of outstanding experience, as well as outstanding skill and inventiveness, one would expect them to have more knowledge than the notional addressee, albeit that one must remember that one is removing, as it were, more than 15 years experience from each of the witnesses when judging common general knowledge as at 1984.
- In the following sections of this part of the judgment, I propose to deal, in the main, with the common general knowledge of the notional addressee as at the priority date, to the extent that it was either undisputed or, in my view, indisputable. However, because I think it helpful when considering issues of construction and sufficiency, I also propose to explain some of the developments which occurred in the field after 1984, while trying to make it clear when I am dealing with that aspect.
Recombinant production of proteins
II. RECOMBINANT PRODUCTION OF PROTEINS
- From the 1920s, some proteins were obtained by purifying them from human or animal tissues or from bodily fluids. During the 1970s, recombinant DNA technology was developed. This is a technique which involves isolating or synthesising the gene which codes for the desired protein, combining the gene with other ("vector") DNA, inserting this "recombinant" DNA into a "host cell", which then expresses the protein.
- One reason for preparing recombinant DNA is to produce - or to "express" - large quantities of the protein encoded by a particular gene. In order to express a particular protein using recombinant DNA, one needs enough information regarding the gene encoding the protein, so that additional vector DNA segments can be recombined with the gene to effect protein expression. This requires the relevant gene to be isolated, identified and characterised. The difficulty of this exercise will be appreciated from the fact that the amount of DNA present in each cell of a human (and most higher animals) amounts to around 6,000,000,000 nucleotides in two complementary strands of DNA (i.e. 3,000,000,000 "base pairs"). This is known as the genome, and the evidence in this case suggested that it comprises over 50,000 genes (although very recently reported research appears to suggest that this may be somewhat on the high side).
- A frequent starting point is the selection of a suitable library of genetic information and a set of "probes" with which to screen a "library" consisting of strands of DNA. There are two normal types of libraries that can be used for these purposes: genomic DNA libraries and complementary DNA ("cDNA") libraries. A library consists of fragments of DNA resulting from cleaving the DNA at certain points. The identity of the fragments of DNA will depend of course on the source of the DNA.
- Ideally, a genomic library should contain a complete set of all the DNA sequences present in the genome, i.e. it will contain all the genetic information present in the cells from which the library was made. However, there is no guarantee that every part of the genome will find its way into the library. A genomic library is stored as pieces of the DNA in a suitable set of carriers. Complementary DNA is so called because it is the complement of the mRNA. Unlike genomic DNA, cDNA does not contain introns or promoters and rarely contains repetitive sequences. Further, only a sub-set of the genomic sequences is expressed in any one cell-type or tissue, and a cDNA library can only contain DNA sequences derived from reverse transcription of mRNA of these expressed sequences.
- As at the priority date, a genomic DNA library of repute was the Lawn library disclosed in 1978 by Lawn et al in Cell 15:1174. Its use was described in the standard work in the field, the so-called Maniatis Manual (Molecular Cloning, A Laboratory Manual, by Maniatis, Fritsch and Sambrook 1982, especially in Chapter 7). While it was not clear that the Lawn library contained all the genes, there was a high probability that the gene one was searching for would be present. By 1983 the technique of Okayama and Berg (Mol.Cell Biol 2(2):161) was also widely known and accepted as a method for constructing cDNA libraries.
- Researchers construct probes in order to screen libraries based on information concerning the actual or putative amino acid sequence of the protein of interest. Probes are typically fairly short sequences of DNA nucleotides, called oligonucleotides, made synthetically. They are constructed with bases which are complementary to the bases of the targeted gene according to the base pairing rules (A with T, C with G) so that the probes will bind - or hybridise - to the gene in the same way that the two complementary strands of the DNA molecule bind to each other in the cell. "Screening" is the process of finding a desired clone from within a population in a library. Such a screening exercise is likely to be futile if the probes are based on unreliable amino acid sequencing of the relevant part of the protein of interest.
- A significant problem with genomic libraries in 1983 was that there was no index or map leading researchers to a desired clone or gene: as explained above, each gene comprises a tiny fraction of the overall DNA, or genome. Another problem that compounded the difficulty of using such libraries is that the DNA was fragmented in a random manner. Thus, while part of the sequence of a gene of interest may be contained in a single fragment of DNA, the complete sequence may very well be split between two or even more fragments, and those fragments might well contain sequences which were not part of the gene concerned.
- If the targeted gene (or a significant part of it) has previously been isolated, identified and sequenced, the probing exercise is normally relatively easy. One designs a probe which is an exact complement to that gene (or the part). Otherwise knowledge of the amino acid sequence of at least a fragment of the targeted protein is essential for determining the constitution of the probe. Armed with that knowledge, probes may be designed and used to screen the library in an attempt to find the gene which codes for that sequence. Probes are designed by determining which codons code for the amino acid sequence which is being targeted. However, this process is greatly complicated by the existence of degeneracy.
- The targeted sequence will inevitably include amino acids which are coded for by more than one codon so, in order to be sure of having a sequence which is an exact match to the gene of interest, one has to screen the library with a set of probes containing multiple nucleotide sequences each of which will be different. A "fully degenerate" set of probes covers all possible sequences coding for the targeted amino acid sequence. As a matter of logic, the longer the amino acid sequence selected for targeting, the greater the number of probes required. The probes are normally "labelled" with a radioactive atom so they can be traced.
- The library, which is to be screened, is first "plated out" so that one can distinguish individual clones among the potentially millions of clones in the library. The DNA in the individual clones is denatured, i.e., the double-stranded helix structure is disrupted, making it single-stranded. The probes are then reacted with the denatured colonies or plaques in the two types of libraries. The plates are washed to remove the surplus probe material. They are then inspected to see whether any of the probes have stuck ("hybridised") to any parts of the library - i.e. whether there are any "positives". Positives may be missed if the background "noise" is too high. If a probe hybridises to one of the clones in the library, then the clone is sequenced to determine whether it contains part or all of the gene of interest. The ability to confirm the sequence is limited by the information known concerning the gene and/or the corresponding protein.
- Screening a genomic DNA library is complicated by the fact that the probe for the selected amino acid sequence may target parts of the exon portions of the gene that are interrupted by an intron. Such a probe, described as "spanning an intron", might not hybridise to the clones in the library; thus it would not enable identification of the desired gene. Until the gene has been cloned (i.e. isolated, identified and sequenced), the existence, number and location of any introns are usually unknown. Screening a cDNA library also has problems. For the library to be useful, the cell or tissue from which it has been obtained needs to have been expressing a sufficient amount of the mRNA of interest at the time when the cell or tissue was harvested. If the cell or tissue was not expressing sufficient amounts of the mRNA of interest, or was expressing no such mRNA, there will not have been enough or any "message", hence there will be insufficient or nothing appropriate from which the complementary DNA can be derived.
- "False positives" can result where a probe matches or hybridises to a sequence outside of the region of interest. Given the size of the DNA libraries (especially genomic DNA libraries) and the fact that there are only four bases in DNA, it is possible that a selected sequence may be repeated several times in the genome. The longer the probe the less likely that will be. However, longer probes may show certain sequences which would also lead to false positives.
- The conditions in which the probing takes place can vary. The more stringent the conditions the more difficult it is for the probes to hybridise with the DNA strands. Accordingly, high stringency conditions have the advantage of reducing the potential for false positives: however, they have the disadvantage of increasing the risk of there being no positives. The stringency of the hybridisation conditions is normally related to the temperature and the salt concentration of the solution in which the hybridisation exercise is being carried out. A technique often used is to increase the stringency gradually in a particular experiment, so that a number of probes hybridise initially, and thereafter, as the stringency increases, the number of hybridising probes decreases, ending ideally with only one probe hybridising, namely the gene one is looking for.
- After hybridising, the resultant material is often subjected to "washing" which involves subjecting it to somewhat higher stringency conditions in a solution. This is with a view to washing off all or some false positives which still remain. Once a particular gene has been isolated, identified and characterised, there are various ways in which that information can be utilised to form recombinant DNA, that will enable a cell that contains the gene to produce greater quantities of the desired protein.
- While the basic processes of transcription and translation are common to all cells, not all types of cell are appropriate as host cells. Bacteria cells cannot deal with introns and therefore cannot deal with genomic DNA (as opposed to cDNA). Further, bacteria do not glycosylate proteins; accordingly, proteins, such as EPO, which require glycosylation in order to be fully effective, need to be produced in eukaryotic cells. Even in different eukaryotic cells, the same protein may be glycosylated differently.
- In 1984 the types of mammalian cells which were thought to be suitable host cells was limited. Those in conventional use included the Chinese Hamster Ovary ("CHO") cell, the baby hamster kidney ("BHK") cell, and the COS monkey ("COS") cell, as well as certain types of human cells.
Recombinant DNA techniques
- There are a number of techniques for introducing DNA into a cell (which, in 1983, would normally have been a bacterium or a yeast), with a view to expressing a gene which it would not ordinarily express. The process of introducing DNA into a cell is called transformation or transfection. If a transformed cell is to express quantities of a desired gene, the relevant DNA which has been introduced must replicate as the cell replicates. Isolated fragments of DNA do not, in general, replicate in a bacterium; this is why one inserts the relevant DNA into another piece of DNA (the vector) which replicates in the cell. Suitable vectors are the DNAs of bacterial viruses or bacteriophages ("phages") which naturally infect bacteria and replicate within them. Alternatively, bacterial plasmids (circular pieces of DNA capable of self-replication) can be used.
- In order to manufacture this recombinant DNA, the vector DNA and the DNA of interest are first both cleaved. The two sets of fragments are then incubated to join the fragments to form recombinant DNA. These recombinant DNA techniques were and are often carried out, in E.coli cells. A single transformed phage cell gives rise to a plaque. Within a particular plaque each phage particle is genetically identical. Such a group of genetically identical organisms is called a clone, and the process of creating a colony of cells which all contain the same inserted DNA is called gene cloning.
- Two principle techniques in gene cloning are (a) cutting (cleavage) and joining (ligation) of nucleic acids, and (b) synthesising oligo- and polynucleotides.
- The discovery of restriction enzymes rendered possible the cleavage of a double-stranded DNA molecule into discrete gene fragments, thereby enabling gene manipulation. Restriction enzymes catalyse the breaking of a specific type of bond between adjacent nucleotides at particular sites ("restriction sites") within a DNA molecule. They are sometimes called restriction endonucleases, distinguishing them from exonucleases, which catalyse the breaking of the bond between the last nucleotide in a nucleic acid chain and the remainder of the nucleic acid.
- Restriction endonucleases are highly specific. Most of them are of so called Type II; they cleave DNA molecules only in regions where particular sequences (usually of from 4 to 6 nucleotides) are present. The resultant DNA fragments can associate by hydrogen bonding with other DNA fragments. If produced by some enzymes, they can have overlapping 5' and/or 3'ends; for this reason the fragments are said to have "sticky ends". DNA fragments from diverse sources can be joined by means of such cohesive ends. Other enzymes make even cuts giving rise to so-called "blunt-ended" fragments with no cohesive ends at all. Some enzymes recognise tetranucleotide sequences, while others recognise longer sequences, and this has an effect on the average fragment length produced. HindII, apparently the first Type II restriction endonuclease to be discovered, is an example of an enzyme which recognises more than one sequence.
- If a restriction endonuclease is used to produce fragments with sticky ends, these fragments will be capable of associating with fragments produced by the action of the same restriction endonuclease on a different DNA molecule. They will then be able to ligate. When the sticky ends associate, the join has "nicks" in the backbone of the DNA chains a few base pairs apart in opposite strands. The E.coli bacterium has an enzyme, DNA ligase, which can be used to repair these nicks in the sugar-phosphate backbone of DNA chains to form an intact double-strand.
- A problem in the development of cloning techniques was the dependence on the availability of restriction endonuclease cleavage sites in suitable positions. Although vector molecules were developed with restriction sites at specific and useful locations, the position of sites in the DNA to be cloned was effectively random. In 1976, a method was developed to enable any double stranded DNA molecule generated by a particular restriction enzyme to be inserted into a cloning vector at a non-matching restriction endonuclease site. This involved the chemical synthesis of a short oligonucleotide containing the desired specific restriction endonuclease recognition site, known as a linker. Such linkers could be joined by DNA ligase onto the DNA to be inserted and the resulting molecule cut with a restriction enzyme to generate sticky ends. The molecule could then be inserted into the matching restriction site in the cloning vector.
- As I have mentioned, in the regulatory region upstream of the coding sequence of a gene there is a promoter. In gene cloning, an artificial promoter may be inserted into the regulatory region (or even upstream of that region). In 1983, specific viral promoters which were generally effective for production of recombinant DNA in E.coli were well known; indeed, they were commercially available. It was generally considered that insertion of the new promoter should occur as near the initiation codon (ATG) as possible, and that it should be accompanied by removal of the natural promoter.
- The resultant "recombinant" DNA is inserted - or "transfected" - into the host cell where it integrates with the native DNA essentially at random. In a eukaryotic cell, the chromosomal structure around the recombinant gene will influence whether it expresses, and, if so to what extent, just as with a naturally occurring gene. By 1983, it was known that the level of protein expression depended on the number of copies of the relevant gene present in the cell and the expression level of those genes. There were also techniques available to select cells which contained multiple - or "amplified" - copies of a particular gene.
- Indeed, DNA amplification can occur spontaneously in many lines of mammalian cells. When it occurs as a desirable event, those cells that contain amplified copies of DNA can be selected by various means from those that do not. One means of selection of cells that have amplified DNA involves the use of the gene that directs the synthesis of dihydrofolate reductase ("DHFR"), a protein normally produced in cells because it is required by the cell to synthesise the nucleotides for DNA. DHFR can be blocked by a cancer drug called methotrexate ("MTX"); this inhibition will kill the cell unless the cell can avoid the inhibition. One way the cell can avoid the inhibition by MTX is to produce more DHFR, so that it escapes the MTX inhibition.
- Amplification normally confers no advantage to the cell, and it is not stable. However, in the presence of MTX, amplification of the DHFR gene means that the amplified cells can produce more DHFR, and so are better able to survive. The cells in which no DHFR gene amplification occurs ("DHFR-cells") die. Therefore, in the presence of MTX, the cells that have spontaneously amplified the DHFR gene ("DHFR+ cells") are selected for survival. The more DHFR made by the cell, the higher the concentration of MTX in which the cells can grow.
- During amplification, the amplified DNA segment usually includes more than the DHFR gene. If another gene is within the amplified region of DNA, it will therefore be amplified with the DHFR gene. Such gene segments that are amplified along with a selectable gene, such as DHFR, are called passenger genes. To take advantage of the ability of cells to amplify passenger genes, a DHFR gene is linked recombinantly to the gene which is to be amplified. Cells containing this recombinant DNA are then treated with MTX, resulting in some cells surviving due to the presence of amplified DHFR genes. In addition to increased numbers of the DHFR gene, these cells also contain proportionately increased numbers of the passenger gene. As a result, these amplified cells will make increased amounts of the protein encoded by the passenger gene.
- Synthesising DNA developed during the late 1970's, when several independent laboratories pioneered oligonucleotide synthesis. Various scientists reported the synthesis of specific genes in learned journals. At that time, chemical synthesis of oligonucleotides was carried out by solution-phase chemistry using a phosphodiester or phosphotriester method The solid-phase method then took over using phosphotriester or phosphoramidite chemistries, which although still a manual process, was a labour-reduced process. On the basis of the evidence, synthesis of a polynucleotide with up to 400 or 500 bases was possible in 1984, but only a limited number of groups of scientists could reliably achieve it.
Homologous recombination
- As I have mentioned, when a DNA construct manufactured outside the host cell ("exogenous DNA") is transfected into a eukaryotic host cell, it integrates randomly with the chromosomes. However, by 1984, it was known, in relation to yeast and bacteria, that if the exogenous DNA had significant regions which were substantially identical, or "homologous", to sequences of the DNA in the nucleus of the host cell ("the chromosomal DNA") then the chromosomal DNA might integrate with the homologous exogenous DNA, effectively replacing the chromosomal DNA. This process is known as homologous recombination.
- In 1984, the skilled worker in the field could target native genes in the host cell ("endogenous" genes) by homologous recombination in lower organisms such as bacteria and yeast. This targeting was effected by, in effect, "knocking out" a specific gene in order to determine its function. There were no reports of gene targeting by homologous recombination in mammalian cells prior to 1985. It was, however, appreciated that targeting endogenous mammalian cells was highly desirable, not least because it would enable gene function of mammalian cells to be studied, leading to the correction of defects in genes which were disease-associated.
- In 1984, it was perceived that there were barriers to the development of gene targeting by homologous recombination in mammalian cells. The size of the mammalian genome is considerably larger than that of bacteria or yeast: given that the integration of exogenous DNA with chromosomal DNA was effectively random, this was seen to lead to obvious problems. Further, even if the exogenous DNA could be targeted at specific sites, it was not clear how one was to select the desired integration event from the vast number of other, random, insertions. Even after the first report of targeting an endogenous mammalian gene by homologous recombination (Smithies et al. 1985 Nature 317:230) in 1985, there were difficulties. The reported method required one to analyse thousands of cells to find one in which gene targeting had occurred. It is clear to my mind from the evidence that the technique of gene targeting by homologous recombination was impractical until the late 1980s (see for instance Nature 336:348 in 1998). Indeed, even now, gene targeting by homologous recombination has not been achieved in many eukaryotic organisms which have been well studied. For example, it was only successfully achieved in the past couple of years in cells of the fruit fly, which had been widely studied in the laboratory (see for instance Rong et al. 2000 Science 288:2013).
- As well as gene targeting, another use for homologous recombination is activation of endogenous genes. As already explained, the great majority of the cells of an organism contain the whole of the genome in respect of that organism. Thus, as I have explained, the great majority of cells of the human body contain DNA which will encode for all the proteins expressed by DNA in the human body, but, save in cells of a particular type, the sequence which encodes for a particular protein is "switched off". This switching off is effected by a negative regulatory element, or "NRE". Gene activation involves introducing into the genome of a human cell where the relevant gene is "switched off" by an NRE, a sequence just upstream of the gene concerned which overrides the NRE and effectively "switches on" the gene.
- Gene activation involves targeting a promoter upstream of an endogenous gene, with a view to the promoter operating with the endogenous regulatory elements to express the gene. Gene activation involves several steps. First, a targeting construct is produced: this is a polynucleotide containing a promoter (normally a viral promoter) flanked by sequences which allow the promoter to be targeted by homologous recombination to a specific location upstream of the desired endogenous gene. Secondly, the targeting construct is transfected into cells containing the desired gene. Thirdly, the flanking ("or targeting") sequences then undergo homologous recombination, as a result of which the promoter is positioned upstream of the endogenous encoding sequences. Fourthly, the "switched on" gene then expresses the desired protein. The art in this field includes the choice of flanking sequences, which determine the position of the targeted promoter with respect to the endogenous gene.
- Gene activation in mammalian cells was reported for the first time around 1991, and was obviously not being carried out, even by particularly skilled workers in this field in 1984.
Glycosylation of proteins
- As I have mentioned, many proteins, including EPO, are glycosylated - i.e. they have sugar - or saccharide - units attached to them. Glycosylation can determine the conformation of a protein, and this in turn may affect its functionality; it can also prolong the life of a protein.
- The glycosylation of any protein is and was known in 1984 to be heterogeneous. In other words, the glycans were known to consist of a combination or combinations of monosaccharides, attached to the polypeptide chain of the protein ("the back bone"), which vary from molecule to molecule. A number of different monosaccharide units are found in glycoproteins. The most common are the hexoses galactose ("Gal") and mannose ("Man"), the hexosamines N-acetylglucosamine ("GlcNAc") and N-acetygalactosamine ("GalNAc"), fucose ("Fuc") and sialic acids (most commonly N-acetylneuraminic acid, abbreviated to "NeuNAc"). The saccharide units are linked together to form oligosaccharides, by enzyme-catalysed reactions between hydroxy (OH) groups of each monosaccharide. There are many hydroxy groups on monosaccharides, and so the units can be linked together in a number of different ways. Because each monosaccharide can link to more than one other monosaccharide, oligosaccharides can be branched.
- The formation of oligosaccharides on a protein is controlled by enzymes, collectively known as glycosyltranferases. Formation of a specific linkage between two sugars is catalysed by a specific enzyme. Thus, there is at least one enzyme for each known combination of any two saccharides in any linkage in a given configuration. The final glycosylation of a protein therefore depends upon the presence or absence of particular glycosyltransferases in a cell and the extent to which they succeed, under particular conditions, in carrying out their functions. A glycosylation unit, or "glycan", can, as I have mentioned, either be N-linked or O-linked.
- In mammalian cells, it was known in 1984 that N-glycosylation starts by the attachment of a precursor oligosaccharide unit to the nitrogen atom in the side chain of the amino acid asparagine. This takes place in the RER of the cell. The precursor unit consists of two GlcNAc residues, nine mannoses and three glucoses. This precursor is then trimmed by enzymes to remove the three glucose residues and up to four of the mannose residues. If four mannoses are removed, a GlcNAc is added to one branch of the oligosaccharide. GlcNAc and fucose units can then be further added to the oligosaccharide unit.
- Further modification can occur with addition of different saccharides in different linkages (NeuNAc, Fuc, Gal, GalNAc, GlcNAc). Many of these modifications will "cap" the oligosaccharide and terminate further elongation/branching. These capping structures (and therefore the overall structures of the glycans), in general, differ among species and among individuals within the species. The blood group system is based on differences in capping structures, which is why immunogenicity to carbohydrates is one of the major barriers in blood transfusion and organ transplantation between individuals as well as between species.
- The attachment of oligosaccharides to a protein is not like the translation of mRNA into protein by ribosomes and tRNAs: there is no unique result (i.e. no particular glycan) prescribed by a given template. The process is carried out by a large number of glycosyltransferase enzymes; there are many processing stages, and most of them are governed by specific pathways or rules. Accordingly, there can be a very large number of different outcomes. In the case of complex N-linked glycoforms the branching can lead to the glycan being bi-, tri- or tetra-antennary. Each branch or antenna may include repeating structures. In an antennary structure, some antennae may be capped by sialic acid whereas others may be uncapped. Some N-linked structures may be fucosylated (i.e. have a fucose unit linked) at the innermost GlcNAc and others not. The result is that a glycoprotein produced even by one specific cell type will consist of a heterogeneous mixture of different species, termed glycoforms.
- The particular glycosyltransferase enzymes produced by a cell will depend on the genome. A mutation in the genome may knock out the gene for a particular glycosyltransferase or reduce, or indeed increase, its activity. However, there is and was in 1984, no reason to think that the introduction of a mutation in CHO cells to produce cells which are DHFR - would suffer any other mutation.
- By 1983 the general biosynthetic pathway for N-glycosylation in mammalian cells was well understood. That for O-glycosylation was less well understood. However, a number of things were known about O-glycosylation. The glycan was known to be always attached to the oxygen atom in the side chain of a serine or threonine amino acid. It was also known that there was no precursor unit attached as a starting point. O-glycosylation was understood to be initiated by attachment of a GalNAc residue in the RER and/or Golgi, and the oligosaccharide structure was built up residue by residue by individual enzymes in the Golgi. It was also known that GalNAc is not found in N-linked oligosaccharides (except in certain specific cases which are not relevant to this case). Accordingly, the presence or absence of GalNAc was regarded as diagnostic of O-linked glycosylation. What was not known in 1983/85, but is known now, is that O-glycosylation is a differentiated process that varies with cell type.
- Bacteria (including E.coli) do not glycosylate proteins. In the case of proteins such as EPO, where glycosylation is vital for in vivo activity, this means that bacteria will produce inactive material. Yeast cells (such as those of S. cerevisiae) are able to produce and attach a certain precursor and process it to remove glucose and one mannose. However, in the Golgi of yeast, only mannose is added, so yeast produce only highly mannosylated structures. Plant and insect cells also produce structures which are different from those produced by mammalian cells.
- As I have mentioned, the actual glycoproteins produced by mammalian cells are mixtures of (normally a very large number of) different glycoforms, and the glycoforms produced depend upon the glycosyltransferases which are active in the cell. Glycosylation is a highly sensitive parameter of cell biological processes. Different cell types have different glycosylation properties. Cells from the same species and of the same type can glycosylate differently.
- It is now known (but was not known in 1984) that CHO cells almost exclusively engage in a form of capping found in human cells and therefore, recombinant glycoproteins expressed in CHO cells will be appropriate for use in human therapy. In 1984, there was limited knowledge about the way in which non-human mammalian cells would glycosylate proteins produced by the expression of recombinant DNA.
- The glycosylation capacity of a cell line is dictated by the properties and functions of the glycosyltransferases expressed in the cell. However, cells normally only express a subset of the glycosyltransferases for which their genome encodes. As I have mentioned, cells from one organism normally contain the same DNA regardless of their tissue origin. Thus, a normal cell's potential for glycosylation is generally much larger than that which is dictated by the expressed enzymes.
- It was becoming clear by 1984 that the identification of a suitable cell line was important not only from the perspective of ensuring appropriate glycosylation to enable the glycoprotein to function, but also to ensure that the recombinant glycoprotein did not generate an immunogenic response. Little was known that would guide the researcher towards the best cell line for any given glycoprotein.
Purification and deglycosylation of proteins
- Any protein of interest (whether naturally occurring or recombinant) has to be separated - or "isolated" - from other material. At the present time (and in 1984) there are a number of well known techniques. Some depend on the physico-chemical differences between the protein and other material (including other proteins). These include factors such as the size, charge, hydrophobicity or solubility of the molecules. Filtration depends on the relative size of the molecules. There are various forms of chromatography which depend on differences in charge between the protein molecules and the molecules of other material from which the protein is to be isolated. Other techniques depend on the electric charge of the molecules; an example is isoelectricfocusing.
- As there will normally be a large number of different contaminatory materials, and as the proteins of interest may be present in low concentration, most purification schemes will involve a combination of steps using different properties of the protein of interest. However, as each step inevitably results in the loss of a proportion of the protein, part of the art of the protein chemist is to seek a scheme which involves the minimum number of purification steps. Glycoproteins present a particular problem because they have so many glycoforms. Their heterogeneity extends to differences in shape, size and charge (this last largely depending on the number of sialic acids).
- It was known in 1984 that various substances could remove some or all of the glycans from a glycoprotein. Treatment with neuraminidase would remove all the sialic acid residues. It was also known that endoglycosidase removed N-glycans. It was also known in 1984 that there were various ways of analysing the glycans in a glycoprotein. Normally a combination of chemical and chromatographic methods were used.
SDS-PAGE and Blotting
- SDS-PAGE is (and was in 1984) a well-established form of an experimental technique called electrophoresis; it enables the apparent molecular weights of proteins or glycoproteins to be assessed. It is inexpensive and relatively easy; its results are not precise but approximate.
- Building on the basic principle that charged proteins will move in an electrical field, the SDS-PAGE method of electrophoresis was developed to allow the mass (molecular weight) of the protein to play the most direct role on the rate of migration of the protein in the electrical field. In SDS-PAGE, an electrical field is applied across a gel, which is cast from a particular type of plastic, polyacrylamide. The gel acts like a sieve, allowing smaller proteins to pass through it faster than larger proteins.
- Most proteins are, as I have said, in a three-dimensional, "folded" structure. Although, in general, smaller molecules will move more quickly through a polyacrylamide gel than larger molecules, two molecules of the same size might appear to be of different sizes on gel electrophoresis if one is folded tightly into a compact shape and the other is not. In SDS-PAGE, the effect of the shape of the protein on its movement through the gel is minimised by disrupting the three-dimensional structure of the proteins to be tested, i.e., by unfolding or "denaturing" the protein molecules. In this manner, all the proteins have, at least roughly, the same "linear" shape.
- SDS is a detergent that carries a negative electrical charge. In SDS-PAGE, protein samples that are to be run on the gel are pre-treated with SDS. SDS has two relevant effects on proteins. First, when mixed with proteins, SDS causes the proteins to denature. Sometimes a reducing reagent, such as mercaptoethanol, is also mixed with the proteins, to disrupt certain bonds between amino acids. The result is that all the proteins being tested have nearly the same extended linear shape. Secondly, when SDS is mixed in sufficient quantities with protein molecules, it effectively "coats" the protein giving each the same charge per amino acid residue. However, SDS typically will not bind to carbohydrates. Accordingly, the shape and charge contributed by the carbohydrate side chains of a glycoprotein can affect the protein's mobility on a gel.
- SDS-PAGE does not provide the "absolute" molecular weight of the protein, but its "apparent" molecular weight. The "absolute" molecular weight of a molecule is calculated by adding together the weights of the atoms that form the molecule, and it is measured in Daltons or Kilo Daltons - KDa. The SDS-PAGE technique measures the relative mobility of molecules on a gel. A scientist may then deduce the relative apparent molecular weight of the molecules by comparing the mobility of that molecule with the mobility of other molecules and/or standard molecules having known molecular weights. As discussed above, there can be residual effects on the mobility of a molecule on a gel, particularly in the case of glycoproteins, that are not directly the result of molecular weight (e.g., charge and shape). Thus it is generally understood that SDS-PAGE provides a "relative" and "apparent" molecular weight, not an absolute molecular weight.
- A number of different lanes may be run on a given SDS-PAGE experiment. Thus, if comparing two different proteins, one may have three lanes of each of the two proteins, so that the reliability of the gel and of the samples, as well as the comparison between the two proteins can be taken into account. There may also be one or two bands of "marker", a mixture of substances of known molecular weight, to act as a guide to the apparent molecular weight of material in the other bands. Unglycosylated proteins will produce fairly narrow and sharp bands. Because their glycosylation is heterogeneous, as I have explained, most glycoproteins will have wider and more "fuzzy" bands as differences in glycosylation will normally result in different apparent molecular weights.
- The comparison of the apparent molecular weight of two materials is more reliable if they are run on adjoining lanes in the same gel. This became possible in about the mid-1970s when so-called "slab" gels became available - i.e. single gels in which separate lanes for different materials (including markers) became available. Even then, there was the possibility of distortions in the gel - e.g. due to the presence of a little air. Distortions would be less likely in relation to columns near the centre of the slab gel, as opposed to the edges.
- The result of an SDS-PAGE experiment is the separation of proteins on a gel. After the separation, it is necessary to be able to see the separated proteins. One sensitive technique for doing this is referred to as Western blotting or immunoblotting. There are three basic steps involved in a Western blot.
- First, after electrophoresis of the proteins through a polyacrylamide gel, they are transferred by blotting to a porous membrane sheet. Secondly, this membrane sheet is treated with a solution containing an antibody to the protein of interest. An antibody is a type of protein that is capable of specifically recognising and binding to part of a molecule (referred to as its epitope). After this step, the antibody will have (hopefully) bound to the separated protein of interest. Thirdly, the presence of the antibody, and thus the presence of the protein of interest, is detected. One way to detect the presence of the antibody which bound to the protein is by adding a second antibody solution to the membrane sheet, termed a two-antibody method. This second antibody solution contains the detectable antibody, which specifically binds to the first antibody. The second antibody is detected, such as by treatment with another solution that creates a visually coloured band corresponding to the location in the gel of the antibodies, and thus the protein of interest. Alternatively, the presence of the protein of interest on the gel can be detected by using a detectable label – e.g. a radioactive ion or compound to the protein. Exposure of the gel to radiographic film will produce a band identifying the presence of the labelled protein. This is called autoradiography. A method for detecting the amount of protein in a sample is radio-immunoassay ("RAI").
- A very similar exercise for DNA fragments is known as Southern blotting. A Southern blot involves separating the DNA fragments in the genomic library on an agarose gel according to size, absorbing the separated fragments from the gel onto a membrane and probing the resultant separated fragments with a DNA probe (with radioactive phosphorous).
D. THE 605 PATENT
Introduction
- The 605 patent claims a priority date of 13th December 1983 and it is entitled "Production of Erythropoietin". The application was filed on 12th December 1984, and it was published on 17th July 1985. 605 disclosed for the first time the DNA sequence for the human EPO gene, and the corresponding amino acid sequence, for human EPO. 605 disclosed novel DNA sequences, glycoprotein products, pharmaceutical compositions, and processes and cells producing EPO. The Examples in 605 teach ways to obtain these claimed inventions.
- Before the disclosure afforded by 605, the state of the art with regard to EPO was as follows. It was known to be a glycoprotein with a molecular weight of about 34,000 Daltons, or 34 Kilo Daltons or KDa. Naturally occurring human EPO had only been isolated from one source, namely urine, and then only in very limited quantities. It had therefore not been possible to obtain sufficient EPO for the purpose of determining its amino acid sequence, let alone to treat patients. A proposed sequence for the first twenty-six amino acids (from the N-terminal) had been published in 1983 (by Sue and another in PNAS USA 80:3651). However, it was incomplete and had only been put forward as "putative", and it was not believed to be reliable. Indeed, the disclosure of 605 demonstrated that two of the amino acids had been wrongly identified. Further, this reported sequence was highly degenerate which rendered its use in connection with the detection of the EPO gene extremely difficult. Nothing was known about the sequence, length, size, or location within the genome, of the EPO gene, let alone about the arrangement of exons and introns within it. Further, there was no satisfactory source for obtaining EPO DNA, because it was impossible to identify an EPO-expressing cell which could be used as an appropriately enriched source of mRNA to obtain EPO cDNA clones on the basis of the data then available. Further, probing a small genomic DNA library would not have been considered a viable option.
- In 1962, Kuratowska et al (Erythropoiesis (1962) p. 58) had reported partial production of EPO from human blood plasma and in 1973 Essers et al (Klinische Wochenschrift 51:1005) reported satisfactory treatment of anaemic patients with EPO-rich plasma. In 1977 Miyake and Goldwasser in a paper ("Miyake") in J.Biol.Chem 252: No.15 p5558, reported how they had purified "milligram quantities" of EPO "with a potency of 20,400 at 21% yield" from the urine of anaemic patients. In 1983, Ascensao et al (Blood 62(5):1132) described an EPO-producing human testicular germ cell that had been sustained in culture for two years; it had a biological activity of 100-600 milliunits/ml of cell culture. A similar level of production was reported by Sherwood et al, in Clinical Research 31(2): 323A, in 1983 for an EPO-producing human renal carcinoma cell line which was cultured for over 3 years.
- In summary, the 605 patent reveals that what Dr Lin did was as follows:
a. he located the human EPO gene within a genomic library comprising literally millions of fragments of human chromosomes, although the library contained no effective index, and, indeed, although it was not in fact known that the EPO gene was in the library;
b. he identified and extracted the EPO gene from the library;
c. he disclosed the amino acid sequence of human EPO;
d. he determined the structure of the EPO gene and disclosed its sequence, the extent and location of the introns and exons, and the up stream and down stream sequences; and
e. he used this information to effect relatively large-scale production of glycoproteins having the biological activity of naturally occurring human EPO.
- In terms of originality, what he did was "to use mixed [oligonucleotide] probes to probe a genomic [DNA] library" according to Professor Brammar, whereas prior to Dr Lin's work "genomic [DNA] libraries [had] previously been screened successfully only with cDNA probes". That evidence is consistent with that of Professor Wall who said that, in 1984, probing a genomic DNA library to seek out the EPO gene by Dr Lin's method would not have been expected to succeed.
- The 605 patent disclosed for the first time the complete nucleotide sequence and organisation of the structural region of the human EPO gene. In particular, Table VI of 605 includes the amino acid sequence of EPO, the identity and location of the nucleotide sequences of the exons, i.e. those that code for human EPO, where the initial EPO is processed to remove the leader peptide (consisting of 27 amino acids) and the sequences of the introns which precede, follow and separate the protein-coding exon sequences, including the splice sites at the intron-exon junctions.
- In brief terms, using recombinant DNA techniques then available, Dr Lin inserted a fragment of DNA containing these human EPO sequences into a plasmid, and, before transfecting this recombinant plasmid into vertebrate cells, he inserted a strong viral promoter, a DNA fragment from the simian virus known as "SV40", upstream from the EPO gene structural region of the fragment in the plasmid. This resulted in a recombinant gene in which the SV40 viral promoter was able to control the expression of human EPO from the human EPO gene.
- In order to increase the amount of RNA transcribed from this recombinant construct, and thus to increase the amount of recombinant human EPO produced by a cell transfected with the construct, he also inserted a DHFR gene at a location upstream from the viral promoter. By inserting this DHFR gene, and treating cells transfected with the construct with MTX, he was able to select cells containing amplified copies of the recombinant human EPO gene, as I have explained.
- Using recombinant techniques then available, Dr Lin inserted the genetic construct created as described above in the chromosomal DNA of a host cell line (in Example 10, a CHO cell line) by transfecting the cell with the DNA of the construct. When this genetic construct was inserted into the CHO cells, the viral promoter sequences, lying upstream from the structural region of the human EPO gene, directed the RNA polymerase to transcribe an RNA copy of the DNA sequences downstream from the promoter site, including the exons and introns that make up the structural region of the human EPO gene. The resultant primary RNA transcript was processed by other cellular enzymes to remove the intron sequences and splice the exon sequences back together to form an mRNA. The cell then used the mRNA as a template to translate the amino acid backbone of human EPO from individual amino acids.
- As described in 605, the "immature" human EPO polypeptide as it is first synthesised in the cell contains 193 amino acids, beginning with a 27-amino-acid leader peptide. As the EPO polypeptide is being synthesised in the cell, this leader peptide directs it into the RER. Once the EPO polypeptide begins to enter the RER, the leader peptide is removed. The human EPO polypeptide, now substantially "mature" and 166 amino acids in length, continues through the RER into the Golgi, where the polypeptide undergoes glycosylation before being secreted from the cell.
- After the filing of the application for 605, it was discovered that the final amino acid residue in the human EPO polypeptide (the arginine residue at position 166), is removed during or following the secretion of the polypeptide from the cell to generate the fully mature human EPO polypeptide having the 165-amino acid sequence set out in Table VI of 605.
The notional addressee
- There is a limited dispute as to the identity of the notional addressee of the Patent. It is agreed that it would consist of a team of people. The team would include three PhD's with several years experience in gene technology, molecular biology and cell biology, respectively. At least one of these three would have had experience of proteins in general and glycoproteins in particular. The team would also include two laboratory technicians well acquainted with gene technology and biochemical techniques.
- At least one member of this notional team would have skill and experience in purification and preparation of proteins and, in particular, of assessing apparent molecular weights by SDS-PAGE. There is an issue as to the extent that the team would include a member who had skill and experience in assessing the carbohydrate content of a glycoprotein which had been isolated, but that is best dealt with under the issue of Added Matter.
The disclosure and specification
- Under the heading "Background", 605 opens with these words:
"The present invention relates generally to the manipulation of genetic materials and, more particularly, to recombinant procedures making possible the production of polypeptides possessing part or all of the primary structural conformation [of EPO]" (page 2 lines 5ff).
- There then follows a summary of techniques for the manipulation of genetic materials. 605 then turns to explain why EPO is of interest, what was known about it, and the attempts to isolate it from blood and urine. At page 2 lines 41ff, there is the description of:
"The attempt [in the last decade] to manufacture industrially and pharmaceutically significant substances using organisms which either do not initially have genetically coded information concerning the desired product included in their DNA or (in the case of mammalian cells in culture) do not ordinarily express a chromosomal gene at appreciable levels. Simply put, a gene that specifies the structure of a desired polypeptide product is either isolated from a "donor" organism or chemically synthesised and then stably introduced into another organism.... Once this is done, the existing machinery for gene expression in the "transformed" or "transfected" microbial host cells operates to construct the desired product, using the exogenous DNA as a template for transcription of mRNA..."
- The patent then goes on to describe the existence of publications on such "further "recombinant DNA" methodologies for the isolation synthesis purification and amplification of genetic materials for use in the transformation of selected host organisms" (page 2 lines 50-51). At page 2 lines 55ff:
"Selected foreign ("exogenous" or "heterologous") DNA strands usually including sequences coding for desired product are prepared.... The linear viral or plasmid DNA is incubated with the foreign DNA in the presence of ligating enzymes capable of effecting a restoration process and "hybrid" vectors are formed which include the selected exogenous DNA segments "spliced" into the viral or circular DNA plasmid."
- At page 3 lines 11ff, there is this:
"At the risk of over simplification, it can be stated that three alternative principal methods can be employed: (1) the "isolation" of double-stranded DNA sequence from the genomic DNA of the donor; (2) the chemical manufacture of a DNA sequence providing a code for a polypeptide of interest; and (3) the in vitro synthesis of a double-stranded DNA sequence by enzymatic "reverse transcription" of mRNA isolated from donor cells. The last-mentioned methods which involve formation of a DNA "complement" of mRNA are generally referred to as "cDNA" methods."
- On the same page at lines 29ff, the patent continues:
"When the entire sequence of amino acid residues of the desired polypeptide is not known, direct manufacture of DNA sequences is not possible and isolation of DNA sequences coding for the polypeptide by a cDNA method becomes the method of choice despite the potential drawbacks. ...Among the standard procedures for isolating cDNA sequences of interest is the preparation of plasmid-borne cDNA "libraries" derived from reverse transcription of mRNA abundant in donor cells selected as responsible for high level expression of genes."
- On page 4 lines 6-7, there is this:
"The use of genomic DNA isolates is the least common of the three above-noted methods for developing specific DNA sequences for use in recombinant procedures."
- The patent then goes on to the Lawn library. On the same page at lines 20ff, there is a reference to a paper published in 1983 by Anderson and Kingston in PNAS (USA) 80,6838, which reports on "probing a genomic library using a mixture of labelled probes". The patent then quotes a passage from the paper which explains that "such a method [is] impractical for the isolation of mammalian protein genes when the corresponding mRNAs are unavailable".
- On the same page, at lines 34ff, the patent continues:
"There... continues to exist a need in the art for improved methods for effecting the rapid and efficient isolation of cDNA clones in instances where little is known of the amino acid sequence of the polypeptide coded for and where "enriched" tissue sources of mRNA are not readily available for use in constructing cDNA libraries."
- At page 5, lines 30ff of the 605 patent, there is this:
"Prior attempts to obtain erythropoietin in good yield from plasma or urine have proven relatively unsuccessful. Complicated and sophisticated laboratory techniques are necessary and generally result in the collection of very small amounts of impure and unstable extracts containing erythropoietin.
...
Initial attempts to isolate erythropoietin from urine yielded unstable, biologically inactive preparations of the hormone. US Letters Patent No. 3,865,801 describes a method of stabilising the biological activity of a crude substance containing erythropoietin recovered from urine. The resulting crude preparation containing erythropoietin purportedly retains 90% of erythropoietin activity, and is stable.
Another method of purifying human erythropoietin from urine of patients with aplastic anaemia is described in Miyake, et al., J. Biol. Chem., Vol. 252, No. 15 (August 10, 1977), pp. 5558-5564. This seven-step procedure includes ion exchange chromatography, ethanol precipitation, gel filtration, and adsorption chromatography, and yields a pure erythropoietin preparation with a potency of 70,400 units/mg of protein in 21% yield.
US Letters Patent No 4,397,840 to Takezawa et al describes methods of preparing "an erythropoietin product" from healthy human urine...
UK Patent Application No 2,085,887 by Sugimoto et al ... describes a process for the production of hybrid human lymphoblastoid cells, reporting production of ... erythropoietin..."
- After discussing the use of antibodies for the purpose of isolating EPO, the patent goes on to explain at page 6 lines 57ff:
"While substantial efforts appear to have been made in attempted isolation of DNA sequences coding for human and other mammalian species erythropoietin, none appear to have been successful. This is due principally to the scarcity of tissue sources, especially human tissue sources, enriched in mRNA such as would allow for construction of a cDNA library from which a DNA sequence coding for erythropoietin might be isolated by conventional techniques. Further, so little is known of the continuous sequence of amino acid residues of erythropoietin that it is not possible to construct, e.g., long polynucleotide probes readily capable of reliable use in DNA/DNA hybridisation screening of cDNA and especially genomic DNA libraries. ...It is estimated that the human gene for erythropoietin may appear as a "single copy gene" within the human genome and, in any event, the genetic material coding for human erythropoietin is likely to constitute less than 0.00005% of total human genomic DNA which would be present in a genomic library.
To date, the most successful of known reported attempts at recombinant-related methods to provide DNA sequences suitable for use in microbial expression of isolatable quantities of mammalian erythropoietin have fallen far short of the goal."
- There are then set out what are often known as "consistory clauses" which (save that the word "point" is used in place of the word "claim") follow, almost word for word, the 31 claims.
- At page 9, lines 31ff of 605, there is this:
"Vertebrate (e.g., COS-1 and CHO) cells provided by the present invention comprise the first cells ever available which can be propagated in vitro continuously and which upon growth in culture are capable of producing in the medium of their growth in excess of 100U (preferably in excess of 500U and most preferably in excess of 1,000 to 5,000U) of erythropoietin per 106 cells in 48 hours as determined by radioimmunoassay."
- The ensuing paragraphs of the patent illustrate the scope of the invention with reference to "cloned DNA sequences of monkey and human species origins and polypeptide sequences suitably deduced therefrom" (page 9, lines 42-43).
- At page 10, lines 7ff there is this:
"Novel DNA sequences of the invention include all sequences useful in securing expression in prokaryotic or eukaryotic host cells of polypeptide products having at least a part of the primary structural conformation and one or more of the biological properties of [EPO] which are comprehended by: (a) the DNA sequences set out in Tables V and VI herein or their complementary strands; (b) DNA sequences defined in (a) or fragments thereof; and (c) DNA sequences which hybridise (under hybridisation conditions such as illustrated herein or more stringent conditions) to DNA sequences."
- Thereafter, under the heading "Detailed Description", the Patent explains how DNA sequences encoding all or part of the polypeptide sequences of human and monkey species EPO have been isolated and characterised, and how each species of EPO have been expressed eukaryotically and prokaryotically, thereby producing isolatable quantities of polypeptides which display the biological properties of naturally occurring EPO. How this was achieved is then set out in rather more detail with reference to twelve Examples which take up pages 12-45 inclusive of the Patent.
- Example 1 relates to amino acid sequencing of human urinary EPO fragments and construction of mixtures of radio labelled probes based on the results of this sequencing. The Example includes Table I which sets out the sequence of 17 different polypeptide fragments, the last two of which (containing seven and twenty one amino acid residues respectfully) are said to provide a useful basis for oligonucleotide probes, in view of their comparatively low degeneracy.
- Example 2 relates to identification of monkey EPO cDNA clones and thus provides information concerning animal treatment and preliminary RIA analysis of animal sera.
- Example 3 relates to preparation of a monkey cDNA library, hybridisation procedures for screening that library a protocol for verification of positive clones, DNA sequencing of a positive cDNA clone and the generation of the monkey EPO polypeptide's amino acid sequence. Table V at pages17-18 provides the amino acid sequence for monkey EPO (including identification of the leader peptide) and the nucleotide sequence of a monkey EPO cDNA. As Table V sets out cDNA, it excludes the intervening introns, but it includes part of the non-coding sequences both upstream and downstream of the encoding regions.
- Example 4 relates to identification of positive human genomic clones and thus provides information concerning the source of the genomic library, plaque hybridisation procedures and verification of positive clones. Part A describes the use of the human foetal liver genomic library prepared in accordance with the procedures of Lawn et al Cell 18, pp: 533-543 (1979). Part B describes how the library was screened and that (at page19, lines 14ff):
"Autoradiography of the filters revealed three positive clones (reactive with both probe mixtures) among the 1,500,000 phage plaques screened. Verification of the positive clones as being EPO-encoding was obtained through DNA sequencing and electron micrographic visualization of heteroduplex formation with the monkey cDNA of Example 3. This procedure also gave evidence of multiple introns in the genomic DNA sequence."
- Example 5 relates to nucleotide analysis of one genomic clone (l hE1) and the generation of human EPO polypeptide amino acid sequence information. These are set out in Table VI on pages 20-24. This table sets out, over five pages, the sequence (consisting of the letters A, T, C and G) for the human EPO gene. It includes much of the upstream sequence (i.e. the regulatory region 5' of the first exon) - namely the 620 bases before the first exon. It includes the two endogenous transcription start sites. It then continues with the five encoding exons and the four intervening introns. It also contains the downstream (i.e. 3' of the last exon) non-coding region. Two of the 620 bases in the regulatory region are not identified, as is indicated by "XX"; some of the bases in the first intron are not identified as indicated by "[I.S.]". Under the exons it identifies the amino acids for which the codons encode: thus it gives the full amino acid sequence for human EPO.
- After describing the arrangement of introns and exons (all of which are shown in Table VI) in the sequence Example 5 continues at page 25, lines 3ff:
"Table V1 thus serves to identify the primary structural conformation (amino acid sequence) of mature human EPO as including 166 specified amino acid residues (estimated M.W. = 18,399). Also revealed in the Table is the DNA sequence coding for a 27 residue leader sequence along with 5' and 3' DNA sequences which may be significant to promoter/operator functions of the human gene operon. Sites for potential glycosylation of the mature human EPO polypeptide are designated in the Table by asterisks."
- The example then explains how the sequence of human and monkey polypeptides are compared in Table VII, in which the degree of homology is shown by reference to asterisks where the amino acids are the same. The patent observes at page 25, lines 16ff that:
"It should be noted that the deduced human and monkey EPO sequences reveal an "additional" lysine (K) residue at (human) position 116. Cross-reference to Table VI indicates that this residue is at the margin of a putative mRNA splice junction in the genomic sequence. Presence of the lysine residue in the human polypeptide sequence was further verified by sequencing of a cDNA human sequence clone prepared from mRNA isolated from COSP-1 cells transformed with the human genomic DNA in Example 7, infra."
- Example 6 describes the construction of a vector incorporating EPO-encoding DNA derived from a monkey cDNA clone, the use of the vector for transfection of COS-1 cells and the cultured growth of the transfected cells.
- Example 7 sets out construction of two vectors incorporating EPO-encoding DNA derived from a positive "human genomic DNA EPO clone", the use of the vector for transfection of COS-1 cells and the cultured growth of the transfected cells. Example 7A involves an endogenous promoter, whereas Example 7B involves the SV40 viral promoter. Example 7A refers to "initial attempts at microbial synthesis of isolatable quantities of human EPO polypeptide material coded for by the human genomic DNA EPO clone [which] also involved expression in mammalian host cells" (page 27 lines 46-47). Later, it goes on to describe how human EPO which had been introduced into a specific plasmid "was used to express human EPO polypeptide material in COS-1 cells" (page 28 line 18). A similar expression was used in relation to Example 7B as recorded on the same page at lines 41-42.
- Example 8 relates to RAI procedures performed on media supernatants obtained from the cultured growth of transfected cells according to Example 6 and 7.
- Example 9 refers to assays for the biological activity of microbially expressed EPO of Examples 6 and 7. The in vitro assay referred to is that described in Goldwasser, et al., Endocrinology 97 (2), pp: 315-323 (1975) and the in vivo activity was determined according to the general procedures of Cotes et al., Nature 191 pp: 1065-1067 (1961) and Hammond, et al., Ann. N.Y. Acad. Sci., 149 pp: 516-527 (1968).
- Example 10 exemplifies mammalian host cell expression systems for monkey EPO cDNA and human EPO genomic DNA involving CHO cells. It describes the immunological and biological activities of products of these expression systems as well as characterisation of such products. In particular, the cells described are a particular type of CHO DHFR-cell namely the DuX-B11 CHO KI cell line as described in 1980 by Urlaub et al, PNAS 77:4461. The use of MTX amplification to enhance EPO production is described at page30, lines 1ff, and the in vitro and in vivo activities are given at page30, lines 44ff. At page 31 lines 10ff the patent describes:
"A preliminary attempt was made to characterize recombinant glycoprotein products from conditioned medium of COS-1 and CHO cell expression of the human EPO gene in comparison to human urinary EPO isolates using both Western blot analysis and SDS-PAGE. These studies indicated that the CHO-produced EPO material had a somewhat higher molecular weight than the COS-1 expression product which, in turn, was slightly larger than the pooled source human urinary extract. All products were somewhat heterogeneous. Neuraminidase enzyme treatment to remove sialic acid resulted in COS-1 and CHO recombinant products of approximately equal molecular weight which were both nonetheless larger than the resulting asialo human urinary extract. Endoglycosidase F enzyme (EC 3.2.1) treatment of the recombinant CHO product and the urinary extract product to totally remove carbohydrate from both) resulted in substantially homogeneous products having essentially identical molecular weight characteristics.
Glycoprotein products provided by the present invention are thus comprehensive of products having a primary structural conformation sufficiently duplicative of that of a naturally-occurring erythropoietin to allow possession of one or more of the biological properties thereof and having an average carbohydrate composition which differs from that of naturally-occurring erythropoietin."
- Example 11 exemplifies synthetic genes encoding human EPO and EPO analogues. It begins by explaining that it "relates to the total manufacture by assembly of nucleotide bases of two structural genes encoding the human species EPO sequence of Table VI" (page 31, lines 26-27). These genes include a number of preferred codons for expression in E.coli and yeast host cells, and expression systems based on them. These genes were made without introns for expression in such cells. The gene for expression in E.coli is referred to as ECEPO (for E.coli EPO) and the gene for expression in yeast referred as SCEPO (for S.cerevisiae EPO). The structure of these constructs is given in Tables VIII to XXI. It is explained at page 38, lines 20ff that "the manufactured ECEPO gene... may be variously modified to encode erythropoietin analogues", which it then specifies.
- Example 12 relates to the immunological and biological activity profiles of "recombinant products of the manufactured ECEPO and SCEPO genes within the expression systems of Example 11". The in vitro and in vivo assay values are given at page 45 lines 29ff. Including reference to the "expression product" in line 38, which is followed in line 39ff:
"In the expression system designed for use of S. cerevisiae host cells; [the relevant] plasmid was transformed into two different strains...."
- At the conclusion of the Examples, the patent goes on to explain at page 45, lines 54ff:
"It should be readily apparent from consideration of the above illustrative examples that numerous exceptionally valuable products and processes are provided by the present invention in its many aspects.
Polypeptides provided by the invention are conspicuously useful materials, whether they are microbially expressed products or synthetic products, the primary, secondary or tertiary structural conformation of which was first made known by the present invention."
- The pharmaceutical and diagnostic values of the invention are then explained, as are the possibilities of allelic variations of EPO based on the sequences disclosed in the patent. At page 47, lines 27-28, there is this:
"In addition to naturally-occurring allelic forms of mature EPO, the present invention also embraces other "EPO products" such as polypeptide analogues of EPO and fragments of "mature" EPO."
605 goes on at lines 34ff to explain, by way of example, that such analogues can include deletions of certain amino acid residues or even part of the sequence, including "having the residues coded for by an entire exon deleted" (lines36-37) or the deletion of "one or more potential sites for glycosylation" (lines 38-39). This passage also contains reference to two or three specific amino acid sequences in EPO (as shown in Table VI) which can be removed without loss of efficacy and two or three specific substitutions which can similarly be made.
- At page 47, lines 53ff, the patent points out that:
".the cloned DNA sequences described herein which encode human and monkey EPO polypeptides are conspicuously valuable for the information which they provide concerning the amino acid sequence of mammalian erythropoietin which has heretofore been unavailable despite decades of analytical processing of isolates of naturally-occurring products."
- At page 48, lines 3ff the patent continues:
"DNA sequences of the invention are also conspicuously suitable materials for use as labelled probes in isolating EPO and related protein encoding cDNA and genomic DNA sequences of mammalian species other than human and monkey species herein specifically illustrated. The extent to which DNA sequences of the invention will have use in various alternative methods of protein synthesis (e.g., in insect cells) or in genetic therapy in humans and other mammals cannot yet be calculated. DNA sequences of the invention are expected to be useful in developing transgenic mammalian species which may serve as eukaryotic "hosts" for production of erythropoietin and erythropoietin products in quantity. See, generally, Palmiter et al., Science, 222 (4625), 809-814 (1983).
Viewed in this light, therefore, the specific disclosures of the illustrative examples are clearly not intended to be limiting upon the scope of the present invention and numerous modifications and variations are expected to occur to those skilled in the art. As one example, while DNA sequences provided by the illustrative examples include monkey cDNA and genomic DNA sequences, because this application provides amino acid sequence information essential to manufacture of DNA sequence, the invention also comprehends such manufactured DNA sequences as may be constructed based on knowledge of EPO amino acid sequences. These may code for EPO (as in Example 12) as well as for EPO fragments and EPO polypeptide analogs (i.e., "EPO Products") which may share one or more biological properties of naturally-occurring EPO but not share others (or possess others to different degrees).
In a like manner, while the above examples illustrate the invention of microbial expression of EPO products in the context of mammalian cell expression of DNA inserted in a hybrid vector of bacterial plasmid and viral genomic origins, a wide variety of expression systems are within the contemplation of the invention. Conspicuously comprehended are expression systems involving vectors of homogeneous origins applied to a variety of bacterial, yeast and mammalian cells in culture as well as to expression systems not involving vectors (such as calcium phosphate transfection of cells). In this regard, it will be understood that expression of, e.g., monkey origin DNA in monkey host cells in culture and human host cells in culture, actually constitute instances of "exogenous" DNA expression inasmuch as the EPO DNA whose high level expression is sought would not have its origins in the genome of the host. Expression systems of the invention further contemplate these practices resulting in cytoplasmic formation of EPO products and accumulation of glycosylated and non-glycosylated EPO products in host cell cytoplasm or membranes (e.g. accumulation in bacterial periplasmic spaces) or in culture medium supernatants as above illustrated, or in rather uncommon systems such as P.aeruginosa expression systems (described in Gray, et al.. Biotechnology, 2, pp. 161-165 (1984)."
The claims
- As already mentioned, there are 31 claims. For the purpose of these proceedings, it is necessary to set out over half of them. Claim 1 is in the following terms:
"1. A DNA sequence for use in securing expression in a prokaryotic or eukaryotic host cell of a polypeptide product having at least part of the primary structural confirmation [sic] of that of erythropoietin to allow possession of the biological property of causing bone marrow cells to increase production of reticulocytes and red blood cells and to increase hemoglobin [sic] synthesis or iron uptake, said DNA sequence selected from the group consisting of:
(a) the DNA sequences set out in Tables V and VI or their complementary strands;
(b) DNA sequences which hybridise under stringent conditions to the protein coding regions of the DNA sequences defined in (a) or fragments thereof; and
(c) DNA sequences which, but for the degeneracy of the genetic code, would hybridise to the DNA sequences defined in (a) and (b).
- Claims 2, 3 and 5 to 7 are in these terms:
2. A DNA sequence according to Claim 1 encoding human erythropoietin.
3. A cDNA sequence according to Claim 1 being a monkey species erythropoietin coding DNA sequence.
5. A genomic DNA sequence according to Claim 1 or 2.
6. A human species erythropoeitin coding DNA sequence according to Claim 5.
7. A DNA sequence according to Claim 6 and including the protein encoding region set forth in Table VI.
- Claim 19 is in these terms:
19. A recombinant polypeptide having part or all of the primary structural conformation of human or monkey erythropoietin as set forth in Table VI or Table V or any allelic variant or derivative thereof possessing the biological property of causing bone marrow cells to increase production of reticulocytes and red blood cells to increase hemoglobin synthesis or iron uptake and characterized by being the product of eukaryotic expression of an exogenous DNA sequence and which has higher molecular weight by SDS-PAGE from erythropoietin isolated from urinary sources.
- I turn now to Claims 20 to 23:
20. A glycoprotein polypeptide according to Claim 19 having an average carbohydrate composition which differs from that of human erythropoietin isolated from urinary sources.
21. A polypeptide according to Claim 19 or 20 wherein the exogenous sequence is a cDNA sequence.
22. A polypeptide according to Claim 19 or 20 wherein the exogenous DNA sequence is a genomic DNA sequence.
23. A polypeptide according to Claim 19 or 20 wherein the exogenous DNA sequence is carried on an autonomously replicating circular DNA plasmid or viral vector.
- Claims 26 and 27 are in these terms:
26. A polypeptide product of the expression in a eukaryotic host cell of a DNA sequence according to any of Claims 1, 2, 3, 5, 6 and 7.
27. A process for production of a polypeptide having at least part of the primary structural conformation of erythropoietin to allow possession of the biological property of causing bone marrow cells to increase production of reticulocytes and red blood cells and to increase hemoglobin synthesis or iron uptake, which process is characterized by culturing under suitable nutrient conditions a prokaryotic or eukaryotic host cell transformed or transfected with a DNA sequence according to any of Claims 1, 2, 3, 5, 6 and 7 in a manner allowing the host cell to express said polypeptide; and optionally isolating the desired polypeptide product of the expression of the DNA sequence.
- Finally, I must set out Claims 28 to 31:
28. A process according to Claim 27, characterised by culturing a host cell of any one of Claims 12 to 16 [which refer back to Claim 1].
29. A process according to Claim 27 or 28 for production of a polypeptide of any one of Claims 19 to 23 and 26.
30. A pharmaceutical composition comprising a polypeptide produced in accordance with the process of Claim 27, 28 or 29 and a pharmaceutically acceptable diluent, adjuvant or carrier.
31. A pharmaceutical composition according to Claim 30, comprising a polypeptide of any one of Claims 19 to 23 and 26.
The procedural history
- As I have mentioned, the application for the grant of 605, which I shall call 605A, was filed on 12th December 1984 at the European Patent Office. In its original form, which I shall call 605B, the patent was granted on 25th July 1990. It was then subject to opposition proceedings which came before the Technical Board of Appeal ("the Board") who gave a decision on, as I have mentioned, 21st November 1994. As is not unusual, some changes were made between filing and initial grant, and other changes were made as a result of the arguments before, and the decision of, the Board.
- There are three significant differences between (i) the application for the grant of 605, namely 605A and/or the patent granted in its original form, 605B, before the hearing before the Board, and (ii) 605 in its present form. First, there is a paragraph contained in Example 10, which was subsequently deleted. Then there are two relevant amendments to the Claims.
- The paragraph which was included in Example 10 in 605A and 605B, but was not followed through into the patent in its present form, namely 605, is to be found immediately after the first paragraph, and immediately before the second paragraph, of Example 10 in the passage I have quoted from earlier in this Judgment. The paragraph in question which was deleted (and which I shall refer to as "the deleted matter") was on page 65 of 605A (and on page 29 of 605B). It was in these terms:
"Purified human urinary EPO and a recombinant, CHO cell-purified, EPO according to the invention were subjected to carbohydrate analysis according to the procedure of Ledeen, et al. Methods in Enzymology, 83 (Part D), 139-191 (1982) as modified through use of the hydrolysis procedures of Nesser, et al., Anal.Biochem., 142, 58-67 (1984). Experimentally determined carbohydrate constitution values (expressed as molar ratios of carbohydrate in the product) for the urinary isolate were as follows: Hexoses, 1.73; N-acetylglucosamine, 1; N-acetylneuraminic acid, 0.93; Fucose, O; and N-acetylgalactosamine, O. Corresponding values for the recombinant product (derived from CHO pDSVL-gHuEPO 3-day culture media at 100 nM MTX) were as follows: Hexoses, 15.09; N-acetylglucosamine, 1; N-acetylneuraminic acid, 0.998; Fucose, O; and N-acetylgalactosamine, O. These findings are consistent with the Western blot and SDS-PAGE analysis described above."
- The second significant difference is to be found in Claim 3. In 605A, Claim 3 is in these terms:
"A polypeptide according to Claim 1 [which was in fairly similar terms to Claim 1 of 605] wherein the exogenous DNA sequence is a cDNA sequence."
In 605B Claim 3 was to:
"A cDNA sequence according to Claims 1 [which was in fairly similar terms to Claim 1 of 605] or 2 [which was a claim to a DNA sequence within Claim 1 encoding for human EPO]."
- The third significant difference is between 605B and 605. In 605B Claim 20 on page 37 was the equivalent of Claim 19 in 605. While it differed from Claim 19 in a number of respects, the only relevant feature was that, unlike Claim 19, it omitted any comparison between recombinant EPO and urinary EPO: it ended with the reference, also found in Claim 19, to "an exogenous DNA sequence".
E. CONSTRUCTION ISSUES
General
- There are a number of issues as to the proper interpretation of the Claims of the 605 patent. In order to deal with them, I need only to focus on Claim 1 and 2 and Claim 19, as the issues all relate to expressions in those three Claims, albeit that the issues also impinge on many of the other Claims, which refer directly or indirectly to Claim 1, 2 or Claim 19.
- The scope of any claim of the Patent is to be determined in accordance with Section 125 of the Patents Act 1977 ("the 1977 Act"). The effect of Section 125(1) is that an invention is taken to be that specified in the claim as interpreted by the description and any drawings contained in the specification. Section 125(3) specifically incorporates the Protocol on the Interpretation of Article 69 of the European Patent Convention ("the Protocol"). The effect of the Protocol is that claims in Patents should be construed neither too strictly nor too loosely. When interpreting a claim, the court should adopt a middle course between these two extremes, "which combines a fair protection for the patentee with a reasonable degree of certainty for third parties".
- While the patent is to be construed through the eyes of a person (or team of persons) appropriately skilled in the relevant art (or arts), construction is ultimately a matter for the court - see per Aldous LJ in Lubrizol Corporation -v- Esso Petroleum Co Ltd [1997] RPC 727 at 738. Expert evidence will often be of assistance, especially in a case such as the present, where the technology is relatively new and sophisticated. However, it is important to remember that, particularly in a fast developing field, such as that in the present case, the evidence must be judged bearing in mind the passage of time since the relevant date - in this case over 15 years ago. Where the technology has developed significantly since the priority date, the meaning of an expression may have expanded to incorporate a new development, contracted to be limited to the previous art, or even changed in a significant way. Further, as is undoubtedly the case here, the knowledge, experience and skill of the expert witnesses may be considerably greater than that of the notional addressee . In particular, the notional addressee is deemed to be non-inventive; the experts in this case were far from satisfying that criterion, albeit in their skill as opposed to their evidence.
- In Glaverbel SA -v- British Coal Corporation [1995] RPC 255 at 268 to 271, Staughton LJ set out seven propositions of law which had been agreed between the parties in that case, and which he considered to be "well founded". Those propositions were as follows:
1. "The interpretation of a patent... is a question of law", and therefore the question of construction itself is not a matter for evidence;
2. The court should have regard to the surrounding circumstances at the date of publication "(or perhaps the priority date)" but only in so far as those facts would be available to "every skilled addressee";
3. Expert evidence is admissible as to technical terms in the patent, and "it may be that expert evidence can go somewhat further";
4. Any claim must be construed in the context of the patent as a whole, but a claim "expressed in clear language... cannot be extended or cut down by reference to the rest of the specification";
5. The Court adopts "a purposive construction rather than a purely literal one derived from applying... meticulous verbal analysis" - per Lord Diplock in Catnic [1982] RPC 183 at 243;
6. Subsequent conduct cannot be relied on;
7. A claim should not be construed by reference to prior material, at least in order to avoid its effect. In Beloit -v- Valmet [1995] RPC 705 at 720, Jacob J, consistent with this observation, stated his view that Article 69 of the EPC "does not legitimately allow courts to construe claims using the prior art either to widen them or narrow them".
- As Jacob J said in Minnesota Mining & Manufacturing Co. -v- Plastus Kreativ A.B. [1997] RPC 737 at 743 (affirmed by the Court of Appeal at [1997] RPC 747):
"The words are to be construed having regard to the inventor's purpose as set out in the rest of his patent. That is why the words "purposive construction" are apt."
- The Court of Appeal has also given guidance on the correct approach to construction in Wheatley -v- Drillsafe Ltd [2000] IP&T 1067 at paragraphs 18 to 26, as applied in American Home Products Corp. -v- Novartis Pharmaceuticals UK Ltd [2000] IP&T 1308. In Wheatley, Aldous LJ (in what was a dissenting judgment, but Sedley and Mance LJJ agreed with the principles) explained at paragraph 22:
"The object of interpretation is to ascertain the intention of the author, in this case the patentee. This involves examining the words of the claim through the eyes of a person to whom the specification is directed, in the context of the specification as a whole."
- In paragraph 23 in Wheatley at [2000] IP&T 1077, Aldous LJ went on to say that the question of construction is to be judged objectively. He explained that the approach to construction embodied in Article 69 of the European Patent Convention ("the EPC") is the same as the approach adopted by Hoffmann J in Improver Corp. -v- Remington Consumer Products Limited [1990] FSR 181 at 189. As Hoffmann J made clear in that case, his approach is directly derived from that of Lord Diplock in Catnic [1982] RPC 183. The observation I have referred to of Lord Diplock in Catnic [1982] RPC 183 at 243 is entirely consistent with the approach of the English courts towards the construction of charterparties (where "detailed semantic and syntactical analysis" must "yield to business common sense" per Lord Diplock in The Antaios [1985] AC 191 at 210D) and leases (see e.g. Basingstoke and Deane B. C. -v- The Host Group Ltd [1988] 1 WLR 348). Accordingly, when construing a patent, the approach of the court should not, in my view, normally differ substantially from its approach when construing any other legal document. Thus the approach of the court to interpreting technical terms in a patent is the same as in other legal documents - see Chitty on Contracts (28th Edition) Volume I paragraph 12.041, and Part 3 of Chapter 12 generally.
- It is true that a patent, unlike most documents the court is called upon to construe, involves taking into account the principle I have mentioned of "fair protection for the patentee [and] reasonable... certainty for third parties" (see the Protocol). However, I believe that that is largely reflected in the well established rule that a patent is to be construed through the eyes of a person or persons of reasonable skill in the relevant art, who is assumed to possess the common general knowledge available in his field. In my view, this is substantially akin to the surrounding circumstances known to both parties which must be taken into account when construing a commercial contract (see for instance per Lord Hoffmann in Investors Compensation Scheme Ltd -v- West Bromwich Building Society [1998] 1 All ER 98). The fact that a patent, unlike a charter party or a lease, is a unilateral document does not seem to me of itself to be a relevant point of distinction, given that it appears that the approach to construction of notices is the same as that of contracts: see Mannai Investment Co Ltd -v- Eagle Star Life Assurance Co Ltd [1997] AC 749.
- As a matter of principle, questions of construction of a claim in a patent, and questions of infringement of that claim are separate. Indeed, there are obvious dangers in determining issues of construction of the claims of a patent while having regard to the issues of infringement. At least on the face of it, what the patent means should not be influenced by the existence or nature of an alleged infringement. Apart from anything else, the alleged infringement will normally have arisen some time after the patent has been applied for, and will therefore be something which occurred after the date by reference to which the patent is to be construed. However, some issues of construction in a patent action only arise because of the nature of the alleged infringement, and sometimes an issue of construction can hardly be understood without reference to the alleged infringement which gives rise to the issue of construction.
- Furthermore, in many cases there is something of an overlap between construction and infringement issues. In many infringement actions, where the alleged infringement is not within the literal wording of a claim, it is now well established that the court has to ask itself the three questions which Hoffmann J set out in Improver [1990] FSR 181 at 192, and which were recently cited with approval, as being in accordance with the Protocol, by Aldous LJ in Wheatley [2000] IP&T 1067. The reasoning of Hoffmann J was based on the approach of Lord Diplock in Catnic [1982] RPC 183 and that of Aldous LJ was based on the Protocol. Accordingly, at least on the face of it, there is a powerful case for saying that, ultimately, when considering the three questions, the court is really posing a composite issue of interpretation. This view is supported by Hoffmann J's statement in Improver [1990] FSR 181 at 189 that:
"In the end..... the question is always whether the alleged infringement is covered by the language of the claim."
- It nonetheless remains the case that the three Improver or Protocol questions are directed to infringement, indeed to a specific alleged infringement - note the approach of the Court in Improver. At least in this case, it seems to me that these Improver or Protocol questions, to the extent that it is necessary to deal with them, are more conveniently considered under infringement, rather than under construction. It is because of the potential interrelationship between construction, sufficiency and infringement that I will deal with these issues in that order before turning to other matters relating to 605.
- Before turning to the specific issues of construction of 605, it is right to mention that, as is perhaps inevitable, some of the issues of construction are plainly connected. Thus, the point as to the meanings of "host cell" and "securing expression" in Claim 1, and the meanings of "recombinant polypeptide" and "an exogenous DNA sequence" in Claim 19 appear to have a great deal in common. While it would obviously require an exceptional case before two connected expressions of the same claim could be given inconsistent meanings, it would be easier for connected expressions in different claims to be given inconsistent meanings. Nonetheless, given that all the claims have to be construed in the context of the patent as a whole, it seems to me that, in the absence of any good reason to the contrary, one would prefer to arrive at a conclusion whereby connected expressions in different claims were given consistent meanings. However, I do not think that is a point of much weight where the claims are independent of each other or cast very differently from each other.
Claim 1: Introductory
- There are, or at least there were, four issues of construction relating to Claim 1. The first is whether the opening words "a DNA sequence" extends to all DNA sequences, and in particular whether it could extend to a cDNA sequence coding for human EPO. The second issue is the meaning of "host cell" and "securing expression". The third issue concerns the meaning of integer (a). The fourth issue concerns the meaning of "stringent conditions" in integers (b) and (c). This latter issue is, I think, most conveniently dealt with when considering insufficiency. Accordingly, I propose to deal only with the first three issues, as I have identified them, at this stage.
Claim 1: "A DNA sequence"
- Roche initially contended that, particularly if one can take into account 605A and 605B, and what was said during the hearing before the Board in 1994, human cDNA is not included within the scope of Claim 1. In support of this, Mr Simon Thorley QC (who appears with Mr Michael Tappin and Miss Iona Berkeley for Roche) pointed out in opening to the fact that the focus of the teaching in 605 is human genomic DNA and monkey cDNA, and that any teaching relating to human cDNA is conspicuous only by its absence. Further, in light of the amendments made to 605B before the Board and in particular the amendment to Claim 3, Mr Thorley argued that, even if Claim 1 on its face, in the context of 605, appears to extend to human cDNA, it should not be so construed. However, after further consideration, Mr Thorley abandoned these arguments, albeit leaving the latter contention open in a higher court.
- For what it is worth, I consider Mr Thorley was right to abandon that point, at least at first instance. As a matter of ordinary scientific language, a "DNA sequence" is a sequence of nucleotides which can, as the patent itself explains, be derived from any source. In particular, unless there was reason to think that the draftsman of the patent intended a more limited use of the expression, reference to a DNA sequence includes any genomic DNA and any cDNA; indeed, it is perfectly apt to include synthetic or natural sequences (or, indeed, a permutation of the two). To my mind, reading Claim 1 together with Claim 2 and Claim 3 tends to support this conclusion. Claim 2 specifically refers to a DNA sequence "encoding human [EPO]" which confirms that the patent is intended to extend to human EPO (as to which there could, it is fair to say, be no real doubt anyway). By referring expressly to "a cDNA sequence" and by referring to Claim 1, it appears to me that Claim 3 serves to underline the fact that Claim 1 is indeed extending to cDNA.
- Further, once one concludes that Claim 1 extends to some cDNA, it is hard to see how it can be said, as a matter of construction, that human cDNA is not included. The fact that there is no claim specifically to a cDNA sequence for human EPO is not, to my mind, sufficient to indicate a contrary result.
- Roche sought to support their case by reference to the apparent finding of the Board that human cDNA was not enabled by the patent, and that, as a result, Claim 3 of 605A had to be amended to the form as it presently appears in 605. At least on the basis of the arguments I have heard, my view is that, as the law in this country currently stands, when construing a patent, it is not legitimate to have regard to the arguments raised by or before the Board in opposition proceedings relating to the grant or amendment of the patent in question, even though those observations may have led to the patent being amended.
- First, I do not think that Roche's argument would have been easily reconcilable with Section 125 of the 1977 Act. Secondly, Roche's approach to construction would lead to inconvenience and expense. If, when construing a patent, one could take into account what was said on behalf of the patentee at any opposition proceedings, then, it seems to me, one always would have to take into account what was said at opposition proceedings. It cannot be a matter of choice. Accordingly, no patent agent or lawyer could sensibly advise an interested third party as to the meaning and effect of a European patent without studying a transcript of the opposition proceedings (if any). Thirdly, I consider that Roche's argument would be inconsistent with the recent rejection of the proposal to amend Article 69 of the EPC to incorporate something along the lines of what is known in the United States as file wrapper estoppel.
- Fourthly, it appears that authority supports my conclusion. In Glaverbel [1995] RPC 255, the precise point was not at issue, but at 270 to 271, Staughton LJ rejected an attempt to rely upon other documents containing statements by the patentee relating to the alleged invention, although he accepted that, if the facts warranted it, statements made by the patentee could give rise to an estoppel. Of more direct relevance is Palmaz's European Patents (UK) [1999] RPC 47 at 73, where Pumfrey J referred to the submission that a particular construction of the patent was "consistent with the representations made by... the patent attorneys acting for the patentees to the European Patent Office...", and said that "this is not a legitimate approach to construction".
- It is true that the point was left open by Jacob J in Bristol Myers Squibb Limited -v- Baker Norton Limited [1999] RPC 253 at 274, and it appears that the Netherlands Supreme Court and the Stockholm City Court may well take a different view (see Ciba Geigy AG -v- Ote Optics BV (13th January 1985), NJ 1995 391 and Spanak Aktiebolag -v- Allround-Smide Aktiebolag (6th May 1997 respectively). However, as Mr Waugh points out, the Hague District Court in their decision dated 13th March 1993 (upheld by the Netherlands Appeal Court on 27th January 2000) when considering the validity and meaning of the 605 patent itself, was apparently unimpressed by this argument.
- Indeed, the Dutch decision in the present case seems to me to highlight another reason why one should, at the very least, be wary of relying on what was said in objection proceedings before the Board. Although any decision of the Board on an issue which this court has to decide are worthy of respect and consideration, it is most certainly not binding on this court, as the Dutch courts emphasised in their decisions on the 605 patent. In the Netherlands, both the court at first instance and Appeal Court concluded, contrary to the opinion of the Board, that human cDNA was within the scope of Claim 1 of 605 in its present form. Quite apart from this, I am not convinced of the factual basis for the contention advanced by Roche, even if the legal basis existed. However, as the point is not pursued, it would be inappropriate to delve into that aspect.
Claim 1: "Host cell" and "securing expression"
- The words "host cell" in Claim 1 have to be construed as part of the opening part of that Claim, which is:
"A DNA sequence for use in securing expression in a ... host cell of a polypeptide product having... the structural conformation of [EPO]."
It is common ground that the words "for use in" in the context of a patent means that the thing in question has to be suitable for the defined role, in this case securing expression of the appropriate protein in a host cell: see for instance Bristol Myers [2000] IP&T 908 at 917, where the Court of Appeal said that "what is suitable is a question of fact, not one of perception".
- There is no doubt that the teaching of the patent, involving as it does the insertion, through recombinant DNA technology, of the human EPO gene into a non-human cell (be it a yeast cell, an E.coli cell, a CHO cell or a COS cell) results in the cell concerned being a "host cell". That is because the whole of the gene encoding for human EPO does not naturally exist in the cell concerned, and therefore the gene which is introduced for the purpose of ultimately expressing EPO is a stranger to the cell, and the cell is therefore accurately described as a "host cell".
- Given that the issue of construction centres on the word "host" in the expression "host cell", the obvious question which arises is: host to what? Two possible answers are advanced to that question. The first is that of TKT, namely that the cell must be host to a "DNA sequence" which encodes EPO or an analogue of EPO. The alternative answer, supported by Amgen, is that it is a cell which is host to some "DNA sequence", which is foreign to it, but which need not have any particular characteristics, save that it is connected with the production of EPO or an analogue of EPO.
- Concentrating on Claim 1 on its own, there is linguistic force in the point that the "host cell" referred to is a cell which was "host" to that which is described in the Claim, namely "a DNA sequence", which is then further defined at the end of the Claim by reference to integers (a), (b) and (c). Accordingly, Amgen's argument that it is not limited to encoding sequences, as non-encoding sequences are included in Tables V and VI, and hence in integer (a), has considerable attraction and logic.
- However, this interpretation would appear to exclude a cell from being a "host cell" if it was host to DNA sequences upstream of the whole sequence disclosed in Table VI (which DNA sequence can be used according to molecular technology to "switch on" the endogenous encoding sequence). That is a conclusion which seems rather arbitrary, bearing in mind how the disclosure in Table VI would have rendered the sequencing of that further upstream region a relatively simple exercise even as at 1984. These inconsistencies are avoided if TKT's more limited construction is adopted.
- I am of the view that a cell is not a "host cell" unless it is host to exogenous DNA encoding for EPO or its analogue. Such a conclusion is based in part on the teaching of the 605 patent. The terms "host" and "host cell" are used consistently to describe cells which have been transfected with exogenous or foreign DNA (i.e. DNA from outside that particular cell) which encodes EPO, with a view to securing expression of EPO in those host cells. That was accepted by Dr Brenner. The Examples contained in 605 are all concerned with EPO-encoding DNA which has been isolated outside the cell and inserted into the cell to which it is foreign. Indeed, at the relevant time, the routine method of production of a recombinant protein was by cloning the gene encoding the protein and the introduction of that clone into a self-replicating organism by transfection or transformation. There was no knowledge of the technique of "switching on" an endogenous encoding sequence by transfecting the cell with exogenous DNA sequences as including an artificial promoter.
- Further, in order to fall within Claim 1, a DNA sequence must be one "for use in securing expression... of a polypeptide...". I accept that it can be said that strands of nucleotides other than the parts which actually encode a polypeptide (e.g. a promoter) are instrumental in securing expression. However, it seems to me that, particularly in the context of the teaching of 605, the notional addressee would have understood the DNA referred to as including sequences which actually encode, i.e. which directly express, rather than only consisting of sequences which indirectly cause or assist the encoding parts of the gene to express (a point I expand on when considering the meaning of "exogenous DNA" in connection with Claim 19).
- I believe that the interpretation favoured by TKT is supported by the way in which integer (b) of Claim 1 is limited to "the protein coding regions" of the sequences identified in integer (a). The reference to "fragments thereof" does not take matters any further, in the sense that that must be a reference to fragments of the encoding regions. It is true that integer (a) itself extends to the whole sequences of Tables V and VI and therefore includes non-encoding regions. However, that is because those two specific sequences have been obtained in full: the centrally relevant regions are clear from integers (b) and (c). In other words, even in Claim 1 itself the draftsman is concentrating on the encoding regions of EPO DNA.
- Further, the wider meaning is not even foreshadowed either by the state of the art at the relevant date or by the contents of the specification. As I have mentioned, the specification is purely concerned with DNA which includes the EPO-encoding regions being isolated outside a cell and then introduced into the DNA inside that cell, the host cell, with a view ultimately to expressing EPO. The discussion (for instance, at page 48, lines 6 to 10) in the 605 patent about future possible developments are no more than speculation, albeit informed speculation. More significantly, I consider that even that passage would not have been read by the notional reader as referring to the "switching on" of endogenous DNA. As Dr Brenner accepted, that had not been achieved, let alone disclosed anywhere, by 1984.
- Accordingly, in order to be a host cell for the purpose of Claim 1, I consider that the cell must have had introduced into it EPO-encoding genetic material; which had first been isolated outside the cell. However, I do not consider that it follows from this that the genetic material so introduced must, as it were, be foreign material, in the sense of coming from a different species. Thus, the fact that monkey EPO-encoding DNA is introduced into a monkey cell would not prevent that cell becoming a host cell for the purpose of Claim 1 even though there is already monkey EPO-encoding DNA present. I reach that conclusion for two reasons. First, as a matter of ordinary language and concept, a cell would be a "host" to a substance artificially introduced into it, even though there is already some of that substance naturally present. Nothing was said in the evidence which causes me to think that the notional addressee would think otherwise. Secondly, Example 6 involves introducing monkey cDNA into a COS cell (which is a monkey cell). It is true that monkey cDNA does not exist naturally, but it seems to me that it would be perverse to construe Claim 1 as not capable of extending to monkey genomic DNA in COS cells, in light of that Example. Further, if one considers the teaching of the patent more generally, it seems to me that that point is reinforced. Perhaps the most important passage for this purpose is at page 48, which refers to "mammalian cells in culture" as being "conspicuously comprehended", as well as the statement that:
"It will be understood that expression of e.g. monkey origin DNA in monkey host cells in culture and human host cells in culture actually constitute instances of "exogenous" DNA expression in as much as the EPO DNA whose high level expression is sought would not have its origins in the genome of the host."
- In connection with this latter passage, it seems to me important to bear in mind that, whatever the correct construction of Claim 1 may appear to be if one concentrates on the words used in it, the Claim must be construed in the context of the patent as a whole. If the draftsman has specifically indicated somewhere in the specification what he means by a particular expression, then that must clearly be taken into account. In patents, just as in any other documents, it is open to the parties of the draftsman "to make their own dictionary" as Romer LJ put it in Re Sassoon [1933] 1 Ch. 858 at 890-891. Of course, the mere fact that the draftsman of a patent indicates in the specification that a certain thing is "conspicuously comprehended... within the contemplation of the invention" does not automatically mean that it is in fact comprehended within a particular claim, or even any claim, in that patent. Otherwise, one would fall foul of the fourth proposition approved by Staughton LJ in Glaverbel [1995] RPC 255.
- However, I do not consider that anything in the teaching of the patent assists Amgen's contention that "host cell" in Claim 1 can extend to a cell where it is the native DNA sequence (as opposed to a DNA sequence which had been artificially introduced into the cell) unless the DNA sequence introduced into the cell is a DNA sequence within Claim 1 itself, namely a DNA sequence "selected from the group consisting of" integers (a), (b) and (c) of the Claim. Not only do I believe that this accords with the natural meaning of Claim 1, linking as it does "host cell" with "a DNA sequence", but there is nothing in the teaching of the patent to call that conclusion into question.
- I should add that this conclusion appears to me to be consistent with my conclusion as to the meaning of "human cDNA", which I discuss when considering that expression in the context of the alleged insufficiencies of Claim 1.
Claim 1: "From the group consisting of"
- On behalf of Roche, Mr Thorley contends that, in order to be within Claim 1, a DNA sequence must be one which falls within one of the three integers, (a), (b) or (c), and it is not enough if it is a sequence which is merely part of a sequence falling within an integer. Thus, as I have mentioned, in Table VI, the patent sets out the coding sequences (i.e. the exons) the intervening non-coding sequences (i.e. the introns) as well as the DNA upstream of the 5'end of the first exon to the tune of nearly six hundred bases, and the DNA downstream of the 3'end of the last exon to the tune of some five hundred bases. Accordingly, if Mr Thorley's argument is correct, Claim 1(a) would not extend to a DNA sequence which, for instance, omitted the first three bases and/or the last three bases shown in Table VI, let alone human cDNA which would omit all the introns as well as most or all of the upstream DNA and most or all of the downstream DNA.
- As a matter of pure language and precision, there is obvious force in this contention. Tables V and VI contain specific DNA sequences, and the natural language of the closing part of Claim 1 (from "said DNA sequence selected from...") can be said with force to suggest that the draftsman intended only those sequences, in whole, or, indeed, complementary strands of the whole of those sequences, to be within the integer of the Claim. It can also be said that this proposition is supported by the references to "the protein coding regions of the DNA sequences defined in (a) and the "fragments thereof" in Claim 1(b).
- On the other hand, at any rate at first sight, it seems almost absurd to construe integer (a) of Claim 1 in such a limited way. At least in the absence of integers (b) and (c), it would mean that the ambit of the Claim, while precise, was extraordinarily limited so far as the patentee was concerned. Indeed, in the absence of integers (b) and (c), it would be so limited as to call into question whether it can possibly be correct, bearing in the mind the requirement of "a purposive construction".
- In the absence of integers (b) and (c) of Claim 1, I think that it would have been necessary to strive to give a wider meaning to the Claim than it naturally bears. However, in light of the inclusion of integers (b) and (c), I consider that integer (a) should be given the limited meaning for which Mr Thorley contends. First, as already mentioned, it is the natural meaning, in light of the words of Claim 1 immediately preceding integer (a). Secondly, as I have also mentioned, to depart from the natural meaning would involve implying into integer (a) the sort of additions or modifications that are expressly contained in integer (b). Thirdly, it appears to me that integer (b) effectively solves the apparent problem thrown up by the literal construction of integer (a). Read literally, integer (a) covers the entire sequence of bases set out in Table VI and the complementary sequence (i.e. A for T and vice versa and G for C and vice versa). Integer (b) will extend not merely to any sequence which hybridises in stringent conditions to the exons of the DNA shown in Table VI (or the complement of those exons) but also to fragments of those exons. Subject to any question as to the sufficiency of integer (b), it seems to me that it was plainly intended to extend to those sequences which would be covered by the wider and non-natural reading of integer (a), urged on behalf of Amgen, through Mr Anthony Watson QC, Mr Andrew Waugh QC and Mr Colin Birss.
- Fourthly, it appears to me that if one does not give integer (a) its natural meaning, but a wider meaning as contended for by Mr Waugh, it is unclear how wide that meaning should go. Thus, it is unclear whether one should simply limit it to the exons, or whether one should permit it to extend to some of the exons. If the answer is simple, then it is hard to understand why the draftsman of the patent did not make it clear. If the answer is not simple, then it is self-evidently unsatisfactory because one is faced with uncertainty as to what to imply. Furthermore, it seems to me to involve a rather odd approach to construction to hold that one integer of a claim should be interpreted more widely than its natural meaning, where it is unclear how widely it should be interpreted, particularly where in the immediately succeeding integer the draftsman has identified specific criteria by which sequences, related to those in the first integer, but representing a much wider class than the first integer, are to be identified. If integer (b) is sufficient, and there is no other objection to Claim 1, then that would be a satisfactory "purposive" result. If it turns out that integer (b) is insufficient, then that is no reason for giving integer (a) an artificially wide meaning. In other words, although an integer of a claim can, indeed should, properly be construed bearing in mind another integer of the same claim, it does not seem to me that the construction of the first integer, should, at least normally, be altered if it transpires that the second integer is insufficient.
Claim 2
- Amgen contend that the reference in Claim 2 to "a DNA sequence according to Claim 1" effectively incorporates only the first part of Claim 1, namely:
"A DNA sequence for use in securing expression in a prokaryotic or eukaryotic host cell."
Roche argue that the whole of Claim 1 is, as it were, incorporated into Claim 2.
- Amgen's contention is that Claim 2 is limited to DNA sequences which encode human EPO and which are suitable for the stated use. Amgen also argue that this extends not merely to DNA sequences which do so encode, but also to DNA sequences which include the whole of such encoding region (e.g. as exons). In other words, to adopt Roche's language, Amgen's case is that "Claim 2 is a narrow claim which is not susceptible to the same attacks (hybridisation, degeneracy and analogues) as are mounted against Claim 1". Roche, on the other hand, argued that the whole of Claim 1 is, as it were, incorporated into Claim 2.
- Claim 2 is not conspicuously well drafted, and, accordingly, it follows that it is not clear, at least to me, precisely what the draftsman was getting at. Particularly in those circumstances, it appears to me that the appropriate approach is to seek to construe Claim 2 strictly literally, and then ask oneself whether that construction accords with common sense and the Protocol. If it does, then, as I see it, that construction should be adopted. If it does not, then one may have to cast one's eyes a little further.
- There are three components to Claim 2. The first component limits the Claim to a DNA sequence. There is no problem about that. The second component, "according to Claim 1" seems to me to incorporate, at least as a matter of simple language, the whole of Claim 1. The third component "encoding human erythropoeitin" appears to me to be introducing a factor which is intended to cut down the wider words of Claim 1. In other words, Claim 2 extends to DNA sequences which satisfy both of two requirements, namely that they fall within Claim 1 and that they encode human erythropoeitin.
- As to what is meant by "human erythropoeitin" in Claim 2, there is a difference between the parties, in that Amgen contends that it is limited to the protein human erythropoeitin with the amino acid sequence identified in Table VI of the 605 patent, whereas Roche argue that it extends to any analogue of that protein, or any allelic variant of that protein, which shows EPO-like characteristics. On this issue, I am of the view that Amgen's argument is correct. The expression "human erythropoeitin" means what it says, at least on the face of it, and is limited to the amino acid sequence identified in Table VI. Any other polypeptide which has a very similar sequence (e.g. a change of a single amino acid residue) and which has the same biological properties as EPO is not "human erythropoeitin". I draw support in reaching that conclusion from the reasoning of the Court of Appeal in American Home Products [2000] IP&T at 1308, where the Court of Appeal held that the word "rapamycin" in a claim was limited to that particular molecule, and not any variant thereof. Of course, decisions as to what one word means in one patent are of very limited assistance as to what another word means in another patent, but it appears to me that the approach of the Court of Appeal in that case does provide support for my conclusion here. Furthermore, there is the contrast between Claim 1, which refers to "a polypeptide product having at least part of the primary structural confirmation of that of erythropoeitin to allow possession of [its] biological property..." with the simple reference to "human erythropoeitin" in Claim 2.
- To support its contrary contention, Roche referred to various references to erythropoeitin or human erythropoeitin in the specification. It seems to me that only references to "human erythropoeitin" would even be capable of being of assistance on this issue. However, the more important point, as I see it, is that this is not a case where there is real assistance from the specification as to what is meant by a particular expression in a claim. The meaning of the claim in question, namely Claim 2, appears to me to be tolerably clear, particularly when compared with the immediately preceding Claim, and in light of the fourth proposition in Glaverbel [1995] RPC 255, I think it would be illegitimate to invoke anything in the specification to point to a different conclusion, save in exceptional circumstances, e.g. if the specification actually defined the term "human erythropoeitin" for the purposes of the patent.
- In these circumstances, I conclude that the qualification effected by Claim 2 to Claim 1 effectively reduces the ambit of Claim 1 to DNA which encodes the protein whose amino acid sequence is identified in Table VI, and does not extend to its analogues. I do not think it would be realistic to treat the definition as so limited that, for instance, it does not apply to the mature protein (i.e. human erythropoeitin without the leader peptide).
- I turn to the question of the applicability of integers (a) to (c) of Claim 1 into Claim 2. As a matter of language, it seems to me, as I have already said, that the whole of Claim 1 is incorporated into Claim 2, save in so far as it is effectively cut out by the closing three words of Claim 2. Accordingly, those integers would be incorporated into Claim 2, unless, as I have indicated, it leads to an odd result. So far as integer (a) is concerned, the Table VI sequence could be incorporated but it would be meaningless surplusage, but the Table V sequence would not. As to integers (b) and (c), it is conceptually possible for them to be incorporated into Claim 2. However, standing back and looking at the matter in a practical way, I have reached the conclusion that those three integers are not included by incorporation into Claim 2. It appears to me to be unreal and unnecessary to include integers (b) and (c) of Claim 1 when it comes to Claim 2, because Claim 2 has already identified perfectly satisfactorily the DNA sequences which are claimed, namely those which encode the specific protein, human EPO, i.e. the encoding sequence or a larger sequence which includes the encoding sequence in such a way that it actually can effect expression of EPO. The whole "baggage" of the integers of Claim 1 seems to me to be redundant. The notion that the integers are redundant can be said to be inconsistent with the natural reading of Claim 2 as I have identified, but that is also met by the fact that, on any view, a large portion of Claim 1 does not get translated into Claim 2, namely the reference to analogues of EPO which have the necessary biological function.
Claim 19: Introductory
- There are four issues raised in relation to Claim 19. The first issue is the meaning of "recombinant polypeptide", the second issue concerns the meaning of "being the product of eukaryotic expression of an exogenous DNA sequence". The third and fourth issues relate to the expression "higher molecular weight by SDS-PAGE" and from EPO isolated from urinary sources, namely how one assesses whether "has a higher molecular weight by SDS-PAGE", and the meaning of "[EPO] isolated from urinary sources". I shall deal with those four issues in turn.
Claim 19: "Recombinant polypeptide"
- While it is common ground that a "recombinant polypeptide" must be a polypeptide which results from recombinant DNA, TKT contend that it has a more limited meaning than that for which Amgen argue. On Amgen's case, a recombinant polypeptide is a polypeptide which has been produced using any recombinant DNA technique, including homologous recombination. In particular, it is said to extend to a case which involves using the cell's endogenous encoding DNA which has been "switched on" by an inserted exogenous construct and promoter inserted upstream of the encoding DNA. On the other hand, on TKT's case, it is limited to a polypeptide which is expressed in cells which have been transformed or transfected with exogenous DNA which encodes the polypeptide concerned (in this case EPO), and which has been isolated outside the cell, joined to a vector sequence and transfected into the cell.
- As a matter of straightforward language, one might have thought that, in order to be described as "recombinant", a polypeptide itself would have to be combined in some way, but that is obviously not what is meant by "recombinant polypeptide", and, to that extent at least, the expression can be said to be a little misleading to the uninitiated. It is clear that the word "recombinant" in the expression is referential to something other than the polypeptide: it is a reference to the means by which the polypeptide was obtained.
- In those circumstances, it appears to me that there is a powerful case for saying that the natural meaning of "recombinant" when used to describe a polypeptide is to indicate that the polypeptide has been produced by recombinant means, or, to put the same point another way, by DNA technology. As is pointed out on behalf of Amgen, the word "recombinant" derives from the concept of cleaving (normally at least two strands of) DNA and then ligating or combining two or more pieces of DNA thereby produced, resulting in sequences which would not exist in nature. As a matter of ordinary language, it might be wondered why one can describe a polypeptide as recombinant only if the DNA which encoded that polypeptide was exogenous to the cell in which the polypeptide was expressed, and had been isolated outside that cell. If the polypeptide was expressed at all, or was expressed in artificially high quantities, as a result of DNA technology involving cleaving and ligating DNA, that might be thought to be enough to enable it to be described as "recombinant".
- However, I have reached the conclusion that, in the context of this patent, viewed through the eyes of the appropriately skilled addressee in 1984, the reference to "a recombinant polypeptide" would have been understood to have a more limited meaning, namely the product of cells transformed or transfected with exogenous DNA which encodes the polypeptide. The patent throughout is concerned with recombinant polypeptides produced by exogenous encoding DNA. Dr Brenner, Amgen's witness on this topic, accepted that the term "recombinant polypeptide" was used in that sense throughout the patent.
- However, it is right to mention that Dr Brenner also stated that he thought that, even in 1983 or 1984, a polypeptide would have been regarded as recombinant if it had been expressed in "cells in which there was propagated DNA made by natural means which encoded the polypeptide". Nonetheless, he accepted that the more limited meaning of "recombinant polypeptide", for which TKT contend, was at the time "the main use of this term", at least in the patent itself.
- The witness who dealt with this issue on behalf of TKT was Professor Proudfoot. His evidence, which I found impressive on this issue, in that it was logical, careful and consistent, was based fundamentally on the proposition that recombinant DNA would have been understood as at 1984 to be DNA which had been isolated by recombinant technology and then inserted into a cell. Indeed, this was implicit in a question put to him by Mr Watson and with which he agreed. Mr Watson described "the basic invention" of 605 as "the isolation and sequencing of the EPO gene followed by the disclosure of a route to its expression".
- In addition, it appears to me that the relevant academic or technical dictionaries support the narrower construction. Thus, the 1997 edition of the Oxford Dictionary of Biochemistry and Molecular Biology describes "recombinant protein" as "a protein coded for by a gene... that has been cloned in a system that supports expression of the gene...". Glick and Pasternak in "Molecular Biotechnology" (1994) have a glossary in which "recombinant protein" is defined as "a protein whose amino acid sequence is encoded by a cloned gene". A very similar definition is to be found in the glossary of "Gene Cloning: An Introduction" by Brown (1986 and 1995). The fact that some of these books were published significantly after the priority date is a factor which should, if anything, assist Amgen's case because the use of endogenous coding sequences promoted by exogenous sequences had become pretty well known by the early 1990s.
- These dictionary definitions are consistent with the evidence of Professor Proudfoot. Certain other dictionaries were put forward as supporting Amgen's construction, but it seems to me that they suffered from the fact that, although they contained definitions of related expressions (in particular recombinant DNA), none of them defined "recombinant polypeptide" or "recombinant protein". Further, on analysis, even the dictionary definitions of recombinant DNA relied on by Amgen did not assist the contention that it extended to homologous recombination.
- Of course, the extent to which dictionaries are helpful on a question of construction is limited. Questions as to the meaning of words in documents can rarely, if ever, be determined conclusively by reference to dictionaries. Not only are dictionary definitions inevitably shorn of any relevant context, but they also represent the view of a particular person who is required to summarise a definition in a few words. Further, where, as here, the argument is whether a dictionary definition is too narrow, the conclusion that a wider meaning is correct does not contradict the dictionary meaning. Also, particularly in a new and rapidly evolving science, expressions can change their meaning from time, and can easily have slightly different meanings to different scientists in the field.
- It is right to mention some references to the allegedly infringing EPO produced by TKT, which goes under the name of GA-EPO. This is produced by introducing a targeting construct (made up of various sequences of DNA) into the genome of the cell upstream of the endogenous coding region of the EPO gene, which coding region is effectively "switched on" by the introduction of the targeting construct. The EPO thereby produced would not be a "recombinant polypeptide" within the meaning of Claim 19 on TKT's construction, albeit that it would be on Amgen's construction. In a report on behalf of HMR (one of the TKT parties) to the Food and Drug Administration ("FDA") in the United States, Professor Kagen of Boston University referred to the EPO thereby produced as "recombinant". GA-EPO is also referred to by this adjective in an internal HMR document. Indeed, Professor Matsudaira referred to GA-EPO as "recombinant" in his evidence in this case. Dr Heartlein described the targeting construct as prepared for "recombination EPO".
- Particularly when taken together, these references plainly assist the argument that a polypeptide can be described as "recombinant" even though it has been expressed by encoding regions of the DNA which are endogenous. However, while of obvious assistance to Amgen's case, they are all relatively recent descriptions (the earliest being, I think, 1996) and are therefore referable to a time when genetic engineering techniques had advanced significantly since the filing date. In particular, they were statements made at a time when techniques had enabled endogenous encoding regions which would not otherwise be active to be artificially "switched on" by inserting DNA upstream, techniques which were only dreamed of or hoped for in 1984.
- Drawing the various strands on this issue together, the position seems to me to be as follows. Standing as at 1984, the appropriately skilled person might have understood "recombinant polypeptide" to have the wider meaning for which Amgen contend, but, as at that date, the more natural and common meaning was the narrower one, as argued for by TKT. The notion that the narrower meaning is what would have been understood by the notional reader in the present case is reinforced once one considers that it is not only the more common meaning, but the meaning which accords with the teaching of the patent, and the way in which the expression is used elsewhere. On the basis of the evidence, it seems to me likely that, as the techniques of biotechnology have expanded over the past fifteen years, the wider meaning of the term "recombinant polypeptide" has gained greater currency. However, even now, it appears to me that the narrower, more limited, meaning is one which has considerable currency, particularly in light of the evidence of Professor Proudfoot and of the recent editions of the dictionaries.
- In reaching this conclusion, I have considered the meaning of "recombinant polypeptide" as much in the context of the patent generally as in the context of Claim 19 itself. If, as I consider to be correct, the meaning of the term "an exogenous DNA sequence" is limited to encoding sequences (a point dealt with in the next section of this part of the judgment) then it tends to reinforce my view that the narrower meaning of "recombinant polypeptide" is also appropriate. Having said that, it appears to me that, in the end, the meaning of "recombinant polypeptide" is not of central importance, if the meaning of "exogenous DNA sequence" is to be treated as limited to the extent I have indicated. Even if "recombinant polypeptide" on its own has the wider meaning in Claim 19, the expression can only refer to a recombinant polypeptide within the narrower meaning if it is limited to being the product of the expression of exogenous encoding DNA.
Claim 19: "Expression of an exogenous DNA sequence"
- Although there was a suggestion to the contrary at one time by Amgen, it appears clear that "expression" refers to the combined process of transcription (by which mRNA is produced from the DNA) and translation (by which the protein is produced from the mRNA transcript). In other words, it means the production of the polypeptide from DNA which must be "exogenous". It is common ground that, in order to be exogenous, DNA must be foreign to the cell: in other words, if DNA is exogenous to a cell, the cell is a "host" to the DNA. The exogenous nature of the DNA is the conceptual complement of the cell concerned being a host.
- Amgen contend that a polypeptide will be expressed by "an exogenous DNA sequence", if any part of the sequence responsible for the expression of the polypeptide (not merely the encoding part, but the promoter or any other part of the regulatory region) is foreign to the cell in which the expression takes place. TKT contends that the "exogenous DNA" referred to is effectively limited to the encoding region, and in any event it must have been constructed outside the cell, and thereafter introduced into the cell in which the expression of protein takes place.
- In this connection, it is the encoding regions which are directly responsible for "expression" in the sense that they are transcribed into mRNA, which ultimately results in the expression of the protein, whereas the non-encoding regions enable the encoding regions to effect, or assist them in effecting, such expression. As a matter of language, one can obviously describe the non-encoding but enabling regions as being part of the total DNA sequence responsible for the "expression" of the ultimate "polypeptide product". Whether one is simply referring to the encoding regions which directly express, or the whole of the DNA which enables expression to occur, when one refers to the DNA which "expresses" must depend, in my view, upon the context.
- The fact that Claim 1 appears substantially to be concerned with the encoding regions does not, to my mind, take matters any further, because, unlike many, indeed most, of the other claims, Claim 19 does not refer back, either directly or indirectly, to Claim 1 of 605.
- In my judgment, the contents of the specification, supports the view that a polypeptide which is the product of "expression of an exogenous DNA sequence" is a polypeptide which is, as it were, the direct, as opposed to the indirect, product of such expression. As with the word "host", the word "exogenous" is used on a number of occasions in the specification of 605. Thus, at page 48, it is stated that "monkey origin DNA in monkey host cells in culture and human host cells in culture actually constitute instances of "exogenous" DNA expression" because "EPO DNA whose high level expression is sought would not have its origins in the genome of the host". As I have mentioned, this indicates that "exogenous DNA" can be from the same species as the host, but, more importantly in relation to the point at issue, it does indicate that DNA that codes for the desired protein is introduced into the DNA in the host cell. I believe this passage to be of particular significance, because the draftsman was there concerned with emphasising that the teaching of the patent is intended to extend beyond the specific Examples given on the preceding pages of the 605 patent. Even though he is concerned to indicate that the patent should extend more widely than its specific teaching might suggest, he does not go so far as to suggest that it extends to endogenous encoding DNA sequences. As I have indicated, this is scarcely surprising in light of the fact that this was unachievable as at the relevant date.
- It is fair to say that there is one passage pointing the other way. On page 2 at lines 65 to 66, when describing recombinant DNA technology, there is reference to the fact that "selected foreign ("exogenous" or "heterologous") DNA strands usually including sequences coding for desired product are prepared...". This could be said to suggest that, when referring to exogenous DNA, the draftsman of the patent not merely had in mind the fact that such exogenous DNA was not limited to encoding sequences, but that it might even not include encoding sequences. However, those words are in a passage dealing with the technology very generally, and not with the meaning of "exogenous DNA sequences" in the context of "expression" in Claim 19. Further, there is no reference to any such exercise being relevant to the teaching or claims of the 605 patent itself. Quite apart from this, this passage cuts both ways: it can be said with some force that the patent should be read as involving the sort of technique that is described as being "usually" employed.
- I should also refer to another passage on the same page of the patent at lines 34 to 40, in these terms:
""Promoter" DNA sequences usually "precede" a gene in a DNA polymer and provide a site for initiation of the transcription into mRNA. "Regulator" DNA sequences, also usually "upstream" of... a gene in a given DNA polymer, bind proteins that determine the .. rate ... of transcriptional initiation. ...[T]hese sequences... cooperate to determine whether the transcription (and eventual expression) of a gene will occur."
This passage suggests that "expression" is "of a gene", and that the promoter is not part of the gene - i.e. not part of the sequence which expresses. This approach may also be seen on the same page at lines 25 to 26 where "specific DNA nucleotide sequences (genes)" are said to be "further "transcribed" into... mRNA".
- It is right to add that the draftsman does not appear to have been entirely consistent in his use of the word "gene". Thus, on page 24 at lines 45 to 47, there is this:
"In Table VI, the initial continuous DNA sequence designates a top strand of 620 bases in what is apparently an untranslated sequence immediately preceding a translated portion of the human EPO gene. More specifically, the sequence appears to comprise the 5' end of the gene which leads to a translated DNA region coding for the first four amino acids.... of a leader sequence..."
It appears to me that this under-scores the point that expressions such as "gene" can have more than one meaning, albeit within a very limited boundary. However, the important point on the present issue is that, in accordance with the language used on page 2 of the patent, it is the encoding regions which are said to "express".
- During cross examination, it became apparent that the wider construction of "exogenous DNA sequence" supported by Amgen could lead to a conceptual or linguistic difficulty. If an exogenous promoter is inserted upstream of an endogenous coding sequence, thereby activating the coding sequence, it is not entirely easy to decide whether the protein thereby expressed is the product (on Amgen's wider construction) of an exogenous sequence. The sequence immediately responsible for the expression is endogenous, but if one looks at the totality of the sequence involved in the production, it also contains exogenous material. Dr Brenner and Professor Proudfoot, both of whom were highly experienced and straightforward witnesses, took different views as to whether scientists practising in this field could have referred to such DNA as exogenous or not. No such problem arises if the DNA referred to as " exogenous" in Claim 19 is limited to the coding sequences, because, whether one judges the matter as at 1984 or today, there does not seem to be any technique, at any rate referred to in the evidence in this case, whereby part of the encoding DNA is exogenous and part endogenous.
- It is true that Professor Proudfoot accepted that if, for instance, the encoding sequences were endogenous, but there was an exogenous promoter and an exogenous leader sequence, this would involve the creation of a new gene (not limiting that word to the encoding regions). However, he took the view that, if part of the gene was exogenous and part endogenous, then it would be wrong to regard the gene as a whole either as exogenous or as endogenous: it was in part exogenous and in part endogenous. I do not see any inconsistency in those views, with which I agree so far as the language of the patent is concerned. It is true that Dr Brenner had a different opinion on the exogenous nature of such a gene, but he appeared to me to be saying that one could call such a hybrid sequence exogenous rather than that it would have been normal in 1984 to refer to it as such.
Claim 19: "Higher molecular weight by SDS-PAGE"
- The evidence and arguments on the comparison on SDS-PAGE between the performances of recombinant EPO and urinary EPO ("rEPO" and "uEPO" respectively) took up quite a substantial proportion of the factual and expert evidence in this case. It is an area where there is a particular danger of letting the evidence and argument relating to insufficiency spill over illegitimately into the area of interpretation. As I have explained, SDS-PAGE involves different substances migrating along a gel, such that the speed of migration is directly related to the apparent molecular weight of the substance (not the actual molecular weight, because the shape and charge of the molecules may also influence its speed of migration).
- If a substance loaded in a particular gel is homogenous (i.e. all its molecules have the same apparent molecular weight) then, unless the lane is over-loaded with the substance or there is some other problem, it will migrate to a narrow well-defined band. On the other hand, if the substance is heterogeneous, the band will be wider and fuzzier, because the different molecules of the same basic substance will have different apparent (and indeed actual) molecular weight. The molecules of a non-glycosylated protein are almost always homogenous, and will therefore produce a tight band. However, the molecules of a glycoprotein, even if produced from a single source (whether naturally or artificially) will nearly always be heterogeneously glycosylated, and will include a very large number of different glycoforms. As a result the band produced by a glycoprotein on SDS-PAGE will normally be fuzzy.
- Roche's contention is that, in order to give clarity to the closing part of Claim 19, it must be interpreted as meaning that all glycoforms of rEPO have a higher molecular weight than all the glycoforms of uEPO. In other words, on this argument, if one was running the two substances in adjoining columns on SDS-PAGE, the leading edge of the rEPO band (containing the molecules of rEPO with the lowest molecular weight) would be behind the trailing edge of the uEPO (containing the molecules of uEPO with the highest molecular weight).
- I do not accept that construction. To my mind, if the leading edge, the trailing edge and the centre of intensity on Western blot or RIA, of the uEPO are respectively each ahead of the leading edge, the trailing edge and the centre of intensity on Western blot or RIA, of the rEPO, that would be enough to enable the rEPO to satisfy the requirement of the closing words of Claim 19. Both as a matter of ordinary language, and looking at the patent through the eyes of the notional addressee, I consider that to be the correct meaning of the closing part of Claim 19.
- So far as ordinary language is concerned, it seems to me that one has to bear in mind the following relevant factors. First, the bare (i.e. wholly deglycosylated) rEPO is presumably (so far as the reader is concerned) identical to the bare uEPO; secondly, the nature and extent of glycosylation of the two forms of EPO is multifarious even if expressed in a single type of cell; thirdly, this results in both types of EPO having a relatively fuzzy band on SDS-PAGE. Taking these matters together, and the natural meaning of the closing part of Claim 19 of the 605 patent leads one to the conclusion that, if one can say that the start, middle and end of one band is plainly ahead of, respectively, the start, middle and end of another band, then the material in the latter band moves slower than the material in the former band, and therefore has a higher apparent molecular weight than that in the former band. Assuming that each band demonstrates a "molecular weight" in the singular rather than a band of "molecular weights" in the plural, then, even though there may be an overlap, one cannot say that one fuzzy band is equal to the other. One band is ahead of the other and hence has a lower molecular weight.
- It appears to me that that view is consistent with the evidence. When faced with neighbouring bands of different materials which clearly overlapped, but where there was no real doubt that the start, middle and end of one band was ahead respectively of the start, middle and end of the second band, Professor Matsudaira said that the former clearly had a lower apparent molecular weight than the latter. While Professor Cummings expressed himself in slightly different ways at different points in his evidence, I consider that he took the same view at least in relation to some of the reported experiments. In so far as he took a different view, it was not convincing or consistent. Although Professor Clausen appeared to take the view supported by Roche, I was not convinced by it, indeed, it was not clear to me that he adhered to it.
- I draw support from a number of the published papers to which I was referred, and in which results of SDS-PAGE exercises were reported, reproduced and analysed. Where one band, viewed as a whole, ran ahead of, but by no means completely clear of, another band, the two bands were treated as performing differentially: in other words, a detectable difference in performance was regarded as significant, indeed worthy of published reproduction and report, even though there was overlap. In such cases, the band which, viewed overall, was slower, did include some material of lower molecular weight than some of the material in the band which was overall faster, yet the latter was treated as having a lower molecular weight.
- Accordingly, both as a matter of ordinary language (particularly once one notes the technical background), and in light of the way that Claim 19 would in any event have been understood by a reader apparently skilled in the art, I do not consider that the "absolutist" construction favoured by Roche is correct.
- It is said, however, that any construction other than that favoured by Roche would lead to unacceptable uncertainty. I do not agree. As with virtually any comparative exercise, there can sometimes be doubts as to whether there is a significant difference between two specific bands in a gel, but that of itself cannot be a valid objection to Amgen's construction. Further, there are always fields where it is impossible inconvenient or unrealistic to specify a feature in precise quantitative terms. It is equally inevitable that expressing something in non-quantitative terms can lead, on occasions, to uncertainties, difficulties and differences of opinion. Those sorts of "fuzzy edges" are inevitable in science and technology, and it would be quite unrealistic to construe a patent which does not, for some reason or another, condescend to precise quantification, in an unrealistic or unnatural way simply because it would otherwise result in the odd occasion in which there could be uncertainties or differences of opinion.
- Thus, I accept that difficulties could arise in a case where the leading edge of one band runs ahead of the leading edge of another, but the centre of intensity and the trailing edge of the former runs behind those, respectively, of the latter. The large number of photocopies and photographs of different sorts of EPO suggest that this relatively rarely occurs. However, it should be noted that, even the construction advanced by Roche would not avoid uncertainty: it is clear, from the conflicting evidence of the undoubted experts on the topic in this case, that highly qualified persons who are very experienced in SDS-PAGE can disagree about where it is right to treat a particular glycoprotein band on SDS-PAGE as starting or ending.
- As I have just mentioned, the evidence included one or two cases where two types of EPO were run against each other on SDS-PAGE, which resulted in one band being substantially wider than the other, so that, for instance, although the leading edge of each band migrated effectively identically, the trailing edge of the first band was significantly behind the trailing edge of the second band. In some cases, that may well have been attributable to the fact that the first band contained substantially more material, and was overloaded. If that is the cause, then one would simply run the experiment again, without the overloading of the first column. Sometimes the cause may be the presence of impurities which are included in the material in the first band for some reason. If impurities are the cause, then the obvious answer is to improve the degree of purity of the material in the first band, and run the test again.
- However, sometimes the cause may be that the EPO in the first band is unusually heterogeneous, and in particular far more heterogeneous than the material in the second band. In such a case, where there is a marked differential in the trailing edge between two bands which have effectively identical leading edges, there is obviously more room for argument that one cannot say that the first band has a higher apparent molecular weight than the second band. However, taking the evidence as a whole on this topic, and in particular that of Professor Matsudaira, who struck me as convincing, I consider that, at least in most cases where this occurs, the material in the first band would be regarded as having a higher apparent molecular weight than that in the second band. This would primarily be because, given that EPO is always heterogeneous, and therefore will always produce a fuzzy band on SDS-PAGE, the most important indicator of "the apparent molecular weight" of the material in a particular band is the centre of the band or, as Professor Matsudaira put it, the point of greatest intensity within the band. The centre of the band is not always easy to identify, because, particularly where the trailing edge fades gradually up the column, the precise location of the centre of the band is, at least to some extent, a matter of opinion. That is why I am convinced by the formulation of Professor Matsudaira, namely the point of greatest intensity: after all, that will be the point where there is greatest concentration of EPO of a specific molecular weight.
- In these circumstances, it seems to me that, where one is running two materials against each other on SDS-PAGE, and what appears to be the normal situation (i.e. the leading edge, the centre, and the trailing edge of one band are all either substantially the same or all running ahead of their equivalents on the other band) does not obtain, it must, to some extent, be a question of judgment as to whether one band can be said to be displaying a higher apparent molecular weight than the other. In the great majority of such cases, however, I am of the view that the ordinarily skilled man in the art would have been able to say with confidence whether one band displayed a higher apparent molecular weight than the other. It would only be where one band was very diffuse indeed (not due to overloading, impurity, or some other correctable problem) or where it was impossible fairly to identify the centre of intensity of one or both of the bands, that the test would be inconclusive.
- As already mentioned, the result of a large number of SDS-PAGE experiments were in evidence in the present case. Virtually every experiment of relevance to the present case was one where, with all due respect to the experts and counsel who respectively gave evidence and advanced arguments, it appeared to me tolerably clear whether one band was or was not running ahead of, and therefore displaying a lower apparent molecular weight than, the other band. However, it is right to acknowledge that, having reviewed each of the experiments, there were one or two cases where it would not be possible to draw any safe conclusion. However, in so far as those results were the subject of oral evidence, it did appear to me that there was a probability that their unreliability was due to some correctable factor of the sort that I have mentioned. Accordingly, even taking into account the fact that there would be occasions where difficulties could arise in carrying out the exercise indicated by the closing words of Claim 19, I am quite satisfied that it is nearly always a relatively straight forward exercise which would rarely lead to any difficulties, and that one can characterise those difficulties as "puzzles set at the edge of the Claim" (to quote from General Tire & Rubber Company -v- The Firestone Tyre & Rubber Company Limited [1972] RPC 457 at 511, line 37).
Claim 19: "Erythropoeitin isolated from urinary sources"
- I turn to the question of the identification or source of the uEPO to which the draftsman of the 605 patent should be taken as referring at the end of Claim 19. It is common ground that the urinary source referred to must be human urine, but the parties differ as to what the extent and the legitimate method of purification of the uEPO may be. The description in 605 specifically refers on page 5 to EPO purified from urine by Miyake in J. Biol. Chem. 252(15):5558, but it also refers to other methods of purification of EPO from human urine immediately before and immediately after that reference. On page 6 reference is made to a paper by Yanagawa et al. in J.Biol Chem 259(5):2707, which describes another method of purifying EPO from urine.
- The first issue between the parties is whether EPO "isolated" from urinary sources carries with it the requirement that the EPO should be pure. There was some debate as to whether the notional addressee of 605 would have understood a requirement that a protein be "isolated" from certain sources would carry with it the requirement that it should be 100% pure. The evidence on this issue was not particularly satisfactory, and I suspect this is largely because words such as "isolated" and "purified" may depend very much on context for precisely what they mean. Having heard the evidence of a number of witnesses on the topic, none of whom I suspect had had to consider the point in any detail before, I incline to the view that scientists in this field tended to use the word "isolation" in a relative rather than an absolute sense. In other words, although the word "isolated" could be used to describe a substance which was 100% pure, the more normal adjective for such a substance is and was "pure", I believe that the word "isolated" was more frequently used to describe a substance which had been obtained in higher concentrations than before or in sufficient concentration for the purpose for which it was required. "Pure" without qualification would mean 100% pure; "isolated" would not: thus one might describe a substance as "isolated" to the extent of being "X% pure".
- Although this is contrary to Amgen's contention and contrary to what Sir John Walker said in oral evidence, it seems to me that the clearest support for this conclusion from the evidence is to be found in the written evidence in chief of Sir John Walker, where he described a substance "certainly not... pure and arguably... not... isolated". Further, at page 9 line 51 of the patent there is reference to "isolation and purification" of polypeptides, which tends to suggest that the draftsman had in mind that they were different concepts and that isolation comes before purification.
- To my mind, therefore, the reference in Claim 19 to EPO "isolated from urinary sources", in the normal language of scientists in this field, would not have been understood to require the EPO to be 100% pure. If one considers the context in which the word "isolated" is used in Claim 19, I believe that that conclusion is reinforced. What the skilled addressee would be led to understand from Claim 19, and indeed the teaching of the patent, is that rEPO has a higher apparent molecular weight than uEPO (see the passage in Example 10 at page 31 lines 10 to 14). Accordingly, I believe that he would understood that what the closing words of Claim 19 required of him, was to isolate EPO from urine in sufficient concentration, or to a sufficient degree of purity, to be able to carry out a comparison of that EPO with rEPO on SDS-PAGE. It is clear from the evidence that, for that purpose, the uEPO would not have to be 100% pure. I see no reason to interpret the closing part of Claim 19 as imposing a greater burden on the skilled addressees seeking to implement the teaching of the patent, or seeking to discover whether his product infringes, than that which a reasonable person in the position, and with the knowledge, of the skilled addressee, would consider necessary.
- Miyake and Yanagawa (both of whose papers are referred to in the specification of 605) obtained the urine, from which they then isolated EPO, from a number of different patients, and the urine was then "pooled". On behalf of Amgen, Mr Watson contends that the EPO isolated from urinary sources referred to in Claim 19 would be understood by the skilled addressee to be EPO isolated according to the teaching of Miyake, or alternatively EPO isolated from urine "of comparable or better purity or activity to that of Miyake". TKT, through Mr David Kitchin QC (who appears with Mr Richard Meade), argue that, provided EPO is purified from urine (whether from a single patient or pooled from many patients), it is within the closing words of Claim 19.
- I prefer the construction advanced by TKT. As Mr Kitchin says, if the draftsman had intended to refer to one type of method of purification, it would only have been too easy for him to have done so. Indeed, by referring to the different papers of Miyake and Yanagawa (and others) in the specification (at pages 5 and 6), he has demonstrated that he knew that he could refer to one or more of the various methods. In my judgment, the only sensible - indeed the natural - way of reading the closing words of Claim 19, is as a reference to all uEPO purified from human urine, at least by any of the methods described in the description, and irrespective of the source of the urine, and, in particular, whether from one person or whether pooled from several people.
- This leads me to another issue, namely whether the exercise called for by the closing words of Claim 19 should be treated as limited to EPO isolated from urinary sources in accordance with methods taught by the patent (or within the common general knowledge of the notional addressee as at the priority date), or whether methods of isolation which were only published (or within common general knowledge) after the priority date could be taken into account. As a matter of principle, of course, the patent is to be construed as at the priority date. On that basis, there is obviously a powerful argument for saying that EPO isolated from urine in accordance with methods which would not have been known of as at the priority date are simply irrelevant: they could not have been within the mind or knowledge of the skilled man as at the date he has to construe the patent. Further, it would be unsatisfactory if a particular recombinant EPO appeared to infringe the patent at the date of publication (because it had a higher molecular weight than uEPO isolated in accordance with any method known or taught as at the priority date) but it subsequently appeared not to infringe because it had the same, or a lower, molecular weight than EPO isolated from urine according to a method published subsequent to the filing date.
- However, it can be said with some force that the draftsman of the patent appears to have proceeded on the basis that EPO isolated from urine has, in effect, a consistent apparent molecular weight, which does not depend upon the method by which it is isolated. If that assumption turns out to be wrong, then it may be said that the patentee only has himself to blame. Furthermore, it can also be said that, by limiting the EPO isolated from urinary sources to EPO isolated in accordance with particular methods known as at the priority date, one is effectively rewriting the patent to put right a defect which the patentee failed to consider: after all, it would have been only too easy for the patentee to identify the method by which the urinary EPO should be isolated.
- I have concluded that, despite these points, the reference in Claim 19 to EPO isolated from urinary sources is to be read as limited to EPO isolated in accordance with methods mentioned in the 605 patent (and any other methods which were common general knowledge as at the priority date, albeit that there is no evidence of any such other methods) although I would not exclude minor variations or improvements to those methods. It appears to me that such a construction is more consistent with the principle of combining "a fair protection for the patentee with a reasonable degree of certainty for third parties". First, if different methods of isolating EPO from urine can produce different results, then one can at least minimise the uncertainty by limiting the methods of isolation to those taught in the patent. Secondly, as I have said, it appears to be very unsatisfactory that a particular recombinant EPO could initially appear to infringe, but would subsequently not infringe. I do not consider that it is helpful, or indeed permissible, in this connection to take into account the circumstances in which the reference to, and comparison with, urinary EPO was included in Claim 19 (namely the requirement of the Board in light of the position it takes on product-by-process claims, a point I deal with in greater detail below).
F. PRODUCT-BY-PROCESS CLAIMS
General
- Claim 19 and Claim 26 may be characterised as "product-by-process claims". This is because they claim a product (namely "a recombinant polypeptide..." and "a polypeptide product..." respectively) which is described (exclusively in the case of Claim 26) by reference to having been made by a specific process ("the product of eukaryotic expression..." and "[product] of the expression in a eukaryotic host cell..." respectively). They can therefore be contrasted with most of the other Claims (including Claim 1) which are product claims, or Claims 27, 28 and 29 which are process claims.
- There is an issue between the parties, which could be characterised as one of construction or of policy, in relation to Claims 19 and 26, and that issue appears to me to raise a point of principle in relation to product-by-process claims generally. Amgen's position is simple. A claim to a product "produced" or "obtained" by a particular process is limited to the identified product in so far as, and only in so far as, it has been produced or obtained by the identified process. It is not, and cannot as a matter of ordinary language, be a claim to the product concerned, if it has not been obtained or produced by the identified process. If that is right, then a recombinant polypeptide will not fall within Claim 19 unless it has actually been obtained by "eukaryotic expression of an exogenous DNA sequence" (the words "being the product of" being equivalent in this connection to "obtained" or "produced"), and any polypeptide which only satisfies the other requirements of Claim 19 would not be within its scope.
- On behalf of TKT (and indeed on behalf of Roche) Mr Kitchin (supported by Mr Thorley) argues that Amgen's contention is incorrect, and that product-by-process claims, such as Claim 19 and Claim 26, should be interpreted as extending to the product claimed, whether or not it was produced or obtained by the identified process. The importance of this point is that, if TKT and Roche are correct, then, even if the process claimed is novel, a product-by-process claim will be invalid on grounds of anticipation unless the product itself is also novel. It may be that the argument advanced by TKT and Roche as to product-by-process claims is not so much one based on construction, but on a policy as to the circumstances in which product-by-process claims should be allowed.
- Whatever the basis on which the contention on behalf of TKT and Roche rests, I consider that it is wrong. As a matter of principle, application of the normal rules of interpretation appears to me to require one to construe a product-by-process claim, just as any other claim, by giving the words of the claim, read overall in the context of the patent as a whole, their normal meaning, unless, of course, there is a good reason to the contrary. Indeed, that is what is statutorily required by Section 125(1) of the 1977 Act. As a matter of ordinary language, a claim to a product obtained or produced by a specific process cannot fairly be said to extend to any product other than one produced or obtained by the particular process. Claim 19 cannot, as a matter of ordinary language, apply to any polypeptide unless it is "recombinant" and it is "the product of eukaryotic expression of an exogenous DNA sequence". To construe such a claim as extending more widely appears to me to fly in the face of the clear meaning of the words. Further, not only does the argument advanced by Mr Kitchin, in so far as it is one of construction, involve giving the words of a claim an unnatural meaning; it involves giving the words an unnatural meaning with the result that the claim is likely to be invalid, whereas giving the words their natural meaning will be more likely to render the claim valid. As was said by Lord Brougham LC in Langston -v- Langston (1834) 2Cl&Fin 194, "you should lean towards that construction which preserves, rather than towards that which destroys", which he described as "a rule of common law and common sense".
- So far as policy is concerned, I can discern no reason in principle or in practice why a claim to a product made by a certain process could be invalid simply because the product is not novel, if the process is novel, so that a claim to the process would be valid. None of the grounds for invalidity set out in Section 72 of the 1977 Act are specifically prayed in aid on this issue by TKT or Roche. In my view rightly so; it is difficult to justify their case simply on that basis. They rely more on decisions of the Board.
- However, before turning to those decisions, it is perhaps not irrelevant to note that Article 64(2) of the EPC provides:
"If the subject-matter of the European patent is a process, the protection conferred by the patent shall extend to the products directly obtained by such process."
This is reflected in terms in the provisions of Section 60(1)(c) of the 1977 Act. As Mr Kitchin rightly points out, these provisions are concerned with infringements, not with claims. That is particularly relevant in the context of a system (such as that at the EPO and in some of the EPC signatories, such as Germany) under which issues of validity and infringement are tried separately by different courts. Nonetheless, to my mind, Article 64 and Section 60 render it more difficult to contend that a product-by-process claim should be invalid as a matter of principle, in circumstances where the process is novel and the claim only purports to extend to products obtained or produced by the process.
- I now turn to consider the arguments which point to the opposite conclusion. At least in principle, there could be three such arguments. The first is that, as a matter of construction, Claims 19 and 26 should in the context of 605, be construed as TKT contends. The argument is based, on the principle that, at least in the context of a patent which contains process claims (such as Claim 27) product-by-process claims (such as Claims 19 and 26) should be given a different meaning or effect from a process claim. This argument is based on the proposition that where a draftsman of a document has expressed himself differently in two places, there is a presumption that he did so because he intended two different effects. In my judgment, there is nothing in this point, and, to be fair, I do not understand Mr Kitchin to advance it. First, this principle of construction is far too weak to justify not giving Claim 19 or Claim 26 their natural meaning, particularly if, as a result, those Claims would be at risk of being rendered invalid. Secondly, in the context of a patent, such as 605, which also contains product claims (such as Claim 1) this argument is self-defeating, because it simply would mean that a product-by-process claim instead of being a process claim becomes a product claim, and the problem thrown up by the principle of interpretation remains, albeit in a slightly different form.
- Secondly, there is the argument that Article 64 and Section 60(1)(c) of the 1977 Act only extend protection to process claims to products "directly" obtained or produced by the process, whereas Claims 19 and 26 contain no such limitation. I do not think there is anything in that point either, and, again, I do not understand Mr Kitchin to press it. First, the Article and Section are concerned with infringement, not claims. Secondly, at least in the context of Claims 19 and 26, it appears to me that it would be difficult, if not impossible, to conceive of circumstances where the polypeptide claimed could be anything other than the direct result of the process referred to in the Claim. Thirdly, in so far as a product-by-process claim does claim a product which is not directly obtained or produced from the process, the court may have to consider whether, in the circumstances of the particular claim, it is phrased too widely so as to catch products which are not in fact novel or otherwise entitled to protection, in which case the claim might be bad. However, the mere fact that, in certain circumstances, a product-by-process claim not limited expressly or impliedly to products "directly" obtained or produced by the process could be too wide, does not, as I see it, mean that every product-by-process claim which does not contain the word "directly" must be invalid.
The decisions of the Board
- The third argument in favour of TKT's construction, which is really the way in which Mr Kitchin puts his case, is that there is a consistent line of decisions from the Board which unambiguously supports TKT's argument as to how product-by-process claims should be treated. The approach appears to originate from a decision in February 1984, T150/82, "Claim Categories/IFF" OJ EPO 1984, 309, where product-by-process claims were discussed in paragraphs 6 to 11. I must confess to finding the reasons for the decision a little difficult to follow, but it ultimately appears to me to turn on semantics rather than substantive considerations. Thus, in paragraph 10, the Board said this:
"Whilst some features of [the] end-effects may be drawn into the definition of the process for reasons of clarity and of conciseness, the product is in consequence of the invention, without being the invention itself, which is rather the novel interaction represented by the process in such cases. Any attempt to claim the in itself non-inventive product by means of product-by-process claims is claiming the mere effects instead."
- It is true that, if the product has no novel features of itself, it can be said to be non-inventive, but, at least as I see it, if it is only such products as have been made by the inventive process which is claimed, there can be no objection to the claim, as is supported (albeit only indirectly) by the provisions of Article 64(2) which, it is fair to say, the Board considered.
- I also note from the decision in T150/82 that the Board identified a difference of approach to product-by-process claims between the courts of some Member States (notably Germany) and those of the United Kingdom: see paragraphs 9 and 11. In particular, in the latter paragraph, the Board appeared to accept that English courts accept product-by-process claims as valid, provided the process is itself patentable. Although it is, of course, desirable that there is a common approach to issues such as this in the courts of all signatories to the EPC, I do not believe that it is the duty of an English court to abandon what has been the English approach to construction or policy on this issue, and to adopt what has been the German approach to construction and policy on this issue, merely because the Board has favoured the latter as opposed to the former. On the contrary: unless there is a provision in the EPC or the 1977 Act to the contrary, I believe that, at least sitting at first instance, it is my duty, in this jurisdiction, to follow what, as the Board accepts, has been the consistent approach of the English courts - albeit prior to the 1977 Act. This is particularly so if, as I believe to be right, that approach is in fact consistent with the 1977 Act.
- As I have mentioned, there have been a number of subsequent decisions of the Board which have reiterated and, indeed, somewhat developed, their approach. They are T205/83 "Vinyl Ester/Crotonic Acid Copolymers/HOECHST" OJ EPO 1985, 363 (see paragraphs 3.1 and 3.2), T219/83 "Zeolites/BASF" OJ EPO 1986 211 (see paragraph 10), T248/85 "Radiation Processing/BICC" OJ EPO 1986 261 (at paragraph 6), T130/90 "University of Texas/Recombinant Monoclonal Anti-body" [1996] EPOR 46 at 51-52 and 54-55 and T124/93 "AMOCO/Olefin Catalyst" [1996] EPOR 624 at 632 (and indeed the decision of the Board on what was Claim 19 in its original form). They are all decisions of three member Boards apart from T219/83, where the Board was composed of five members.
- At paragraph 10 of the decision in T219/83, at OJ EPO 1996, at 220, the Board said this:
"[Product-by-process] claims... have to be interpreted in an absolute sense, i.e. independently of the process."
If that is not what a product-by-process claim means as a matter of ordinary language, I cannot agree with that view. I have similar difficulty with an observation in T248/85 at OJ EPO 1986, at 266, where the Board said this of the product-by-process claims before them:
"[T]he claims do not define the extent of protection: they define the matter for which protection is sought."
So they do, but the matter in question is limited by the requirement that it must be obtained from the novel process. In T124/93 at [1996] EPOR 632, the Board referred to its previous decisions as establishing that:
"Product-by-process claims give protection for the products as such, independent from the process by which they were made...." (paragraph 3.5).
As I have indicated, as a matter of natural language, at least if worded as Claims 19 and 26 in the present case, I do not consider that they do so.
- I am reluctant not to follow the approach of the Board, particularly in light of the sheer number of consistent decisions on this point. However, I am not bound by decisions of the Board. Indeed, particularly in light of Section 125 of the 1977 Act, I am required, in any particular case, to identify the invention claimed by what is "specified in a claim". As a matter of ordinary language, I find it impossible to construe a product-by-process claim, such as Claim 19 or Claim 26 in the present case, "in an absolute sense" as the Board apparently felt able to do in T219/83 and, in effect, in the other decisions to which I have referred.
- I should make three further points before passing on from this issue. First, the guidance in paragraph in 4.7b of Chapter III, The EPO Guidelines for Substantive Examination is, as one might expect, entirely consistent with the decisions of the Board on this issue. The first sentence is in these terms:
"Claims for products defined in terms of a process of manufacture are admissible only if the products as such fulfil the requirements for patentability, i.e. inter alia that they are new and inventive."
It is not suggested that these guidelines take matters any further, in the sense that they are not binding on me, and in those circumstances I say no more about them.
- Secondly, it is right to mention that the decisions of the Board to which I have referred show that, if two conditions are satisfied, a product-by-process claim will be admitted. The first, which I have been discussing, is that the product itself must be novel. The second is that it must be impossible, or at least difficult satisfactorily, to define the product, and in particular the way in which the product is novel, other than by means of description through the process by which it is produced or obtained (see for instance paragraphs 3.3 and 4.12 in T130/90 at [1996] EPOR 46 at 52 and 55). The fact that the Board is prepared to admit product-by-process claims, albeit in the limited circumstances I have described, means that it cannot be said that product-by-process claims are inherently objectionable as a matter of principle. Otherwise, the Board would have rejected the product-by-process claim in T130/90 or have required it to be reformulated as a process claim to produce a particular product. That very possibility of such a simple reformulation suggests, at least to my mind, that the present approach of the Board is open to question.
- Thirdly, it is to be noted that the Board does not appear to have followed its own jurisprudence in this very case. Until the hearing before the Board, Claim 19, as I have mentioned, did not contain the closing words limiting the Claim to rEPO which had a higher molecular weight by SDS-PAGE than uEPO. Those words were inserted because, in their absence, Claim 19 would have fallen foul of the Board's requirement that, in a product-by-process claim, the product, as well as the process, had to be novel: without the claimed distinction between rEPO and uEPO, rEPO would not have appeared, at least on the face of the Claim, to be novel. On the other hand, the Board did not require Claim 26 to be amended in this way, or indeed in any other way. As a result, it appears to be a product-by-process claim which, on my view of the law, would not fail for want of novelty provided the process is inventive, but should have failed before the Board on grounds of lack of novelty because, on the Board's view, the product was not distinguished on the face of the Claim from naturally occurring product.
G. INSUFFICIENCY AND BREADTH OF CLAIM: THE LAW AND THE ISSUES
Introduction
- By virtue of Section 72(1)(c) of the 1977 Act a patent must be revoked if it "does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art". It is fundamental to the validity of a patent that it not merely discloses a novel product or process, but that the disclosure is "enabling": in this connection see Asahi Kasei Kogyo KK's Application [1991] RPC 485 as explained by Lord Hoffmann in Biogen Inc. -v- Medeva plc [1997] RPC 1 at 46 to 47, especially at 46 lines 45 to 50. If the disclosure of a patent falls short of being enabling, i.e. if it does not satisfy the requirements of Section 72(1)(c), then it is invalid on grounds of insufficiency. It is convenient to distinguish between two types of insufficiency, although there can obviously be a degree of overlap between them, and although they both derive from the same principle. I will call the two types "classic insufficiency" and "Biogen insufficiency". Classic insufficiency arises where the teaching of the patent does not support that which the teaching specifically purports to deliver: in this case, TKT and Roche contend that Claim 1(a) fails on the ground of classic insufficiency. A claim suffers from Biogen insufficiency if it is cast more widely than the teaching of the patent enables: in the present case, even if the teaching of 605 is enabling so far as the specific EPO sequence identified in Table VI is concerned, it is said that Claim 1 as a whole is cast too wide because it is not enabling so far as analogues of EPO are concerned.
- To adopt the expression used by Lord Hoffmann in Biogen [1997] RPC 1 at 50 line 39, a patent is insufficient on either basis if it cannot "deliver the goods". If the claim is cast more widely than the teaching justifies, the claim will be Biogen insufficient; if the claim on the face of it appears to be cast narrowly enough, it may nonetheless be classically insufficient if the teaching of the specification is not enabling.
- In the present case, as I have said, the various items of teaching in the specification, and in particular Examples 4, 7 and 10, could be said on their face to support Claim 1(a). However, in that connection, quite apart from the other arguments which they raise, Roche and TKT contend that the specification is classically insufficient on a number of grounds. Over and above that, they contend that there is no, or very little, teaching so far as analogues of EPO and of the EPO gene are concerned, as a result of which it is said that the Claims that extend to those products are Biogen insufficient.
- The issue of classic insufficiency involves consideration of a number of specific criticisms as to the inadequacy of the teaching of 605 so far as enablement of most of the Claims is concerned, and those issues turn largely on the contents of this particular patent and the evidence in relation thereto. The issues on Biogen insufficiency are potentially more far reaching in the field of biotechnology patents. I propose to deal with classic insufficiency first, and then turn to Biogen insufficiency.
- Before doing so, it is convenient briefly to consider the interrelationship of the two types of insufficiency in a little more detail in the present case. It is best taken by way of an example. Now that Roche has abandoned its argument as to the meaning of "a DNA sequence" at the opening of Claim 1, it is common ground before me that human cDNA is within the scope of the Claim. One of the points taken by TKT and Roche is that human cDNA is not enabled by the teaching of 605. Roche and TKT contend that, at least to that extent, the Claim is insufficient because, in terms of enablement, the teaching of 605 is classically insufficient so far as human cDNA is concerned. Obviously, that argument has to be resolved on its merits. However, even if that argument succeeds, the claim to human cDNA could still be valid in light of the reasoning of Lord Hoffmann in Biogen [1997] RPC 1.
- Accordingly, the interrelationship between the two insufficiency arguments is rather more complex than might first appear. It is true that the two insufficiency arguments provide alternative bases upon which TKT and Roche could in principle rely. Thus, Claim 1(a) may be classically insufficient and Claims 1(b) and (c) may be Biogen insufficient. On the other hand, at least as I see it, an argument based on Biogen [1997] RPC 1 may enable Amgen to say that, even if human cDNA or Claims 1(b) and 1(c) are not classically enabled by the teaching of the patent, they are nonetheless properly claimed by Claim 1 because of the reasoning of Lord Hoffmann in Biogen. In that sense, the description of "Biogen insufficiency" is misleading, or at least insufficient. One is as much concerned with the permissible breadth of the Claims of the patent bearing in mind its teaching and contribution.
Classic insufficiency
- In Mentor Corporation -v- Hollister Incorporated [1993] RPC 7 the Court of Appeal gave guidance as to what constitutes classic insufficiency. At [1993] RPC 12 lines 3 to 7, Lloyd LJ said this:
"On the one hand the addressee must be able to perform the invention without any further inventive step on his part. On the other hand it is not required that he should be able to perform the invention without any trial or experiment at all, in particular where the subject matter is new or especially delicate."
- On the same page at lines 50 to 52, he said this:
"In each case sufficiency will... be a question of fact and degree, depending on the nature of the invention and the other circumstances of the case."
- At [1993] RPC 13 lines 6 to 24, he cited with approval passages in the judgment of Buckley LJ in Valensi -v- British Radio Corporation [1973] RPC 337, which were to this effect:
"[T]he hypothetical addressee is not a person of exceptional skill and knowledge, ... he is not to be expected to exercise any invention nor any prolonged research, enquiry or experiment. He must, however, be prepared to display a reasonable degree of skill and common knowledge of the art in making trials and to correct obvious errors in the specification if a means of correcting them can readily be found. ...
Further... it is not only inventive steps that cannot be required of the addressee. While the addressee must be taken as a person with a will to make the instructions work, he is not to be called upon to make a prolonged study of matters which present some initial difficulty: and, in particular, if there are actual errors in the specification - if the apparatus really will not work without departing from what is described - then, unless both the existence of the error and the way to correct it can quickly be discovered by an addressee of the degree of skill and knowledge which we envisage, the description is insufficient."
- In another passage in Mentor, at [1993] RPC 14 lines 28 to 45, Lloyd LJ said this:
"It was at first argued that the skilled man should not have to carry out any research, enquiry or experiment at all, whether prolonged or otherwise. But [counsel] subsequently retreated from that extreme position. There is no support for setting so high a standard of disclosure... When, a little later, Aldous J came to apply the law to the facts of this case, he refers to "routine trials" and "normal routine matters that the skilled man would seek to do and would be able to do". [Counsel] criticises the use of the word "routine". To require the performance of routine trials is, he said, to ask too much of the addressee. I do not agree. "Routine" is just the word I would have chosen myself to describe the sort of trial and error which has always been regarded as acceptable; and "routine trials" has the further advantage that it is a positive concept, which is easily understood and applied. ....If the trials are unusually arduous or prolonged, they would hardly be described as routine."
- It is to be noted that the Board takes the same view. Thus, in T694/92 Mycogen/Modifying plant cells [1998] EPOR 114, the Board said this in paragraph 5:
"[T]he guiding principle is always that the skilled person should, after reading of the description, be able to readily perform the invention over the whole area claimed without undue burden and without needing inventive skill."
- The argument as to whether 605 is insufficient on classic grounds gives rise to three issues which it is convenient to consider at this stage. The first issue is raised by Mr Thorley who contends on behalf of Roche that, when considering whether following the teaching of the patent involves an undue burden or not, one can, indeed one should, take into account the fact that, in relation to a patent such as this, it is open to the patentee to deposit with the ATCC a cell line of the relevant cells, as indeed Roche did in connection with Example 10 of its 678 patent. His argument is that, if it is open to the patentee in a particular case to deposit a cell line, thereby enabling the reader to avoid the work which would otherwise be involved in the preparation of a similar cell line (and in particular transfecting and amplifying), then the court should, to put it at its lowest, be that much more ready to conclude that the extra work resulting from the cell line not having been deposited is more than routine.
- That argument is obviously not without its attractions, but I have come to the conclusion that it should be rejected. As Mr Thorley rightly accepts, there is no obligation on a patentee seeking to patent a claim involving cell technology to deposit a cell line: it is merely an option. Accordingly, in the absence of good reason to the contrary, it seems to me that the relevance of the ability to deposit a cell line is merely that the court should not, as it were, lean over backwards in favour of the patentee when considering an argument of classic insufficiency, if the patentee could have avoided the alleged insufficiency by depositing the cell line. However, both principle and common sense suggest to me that one should go no further than that. So far as principle is concerned, the guidance given in Mentor [1993] RPC 7 and in Valensi [1973] RPC 337 lays down the general approach to an allegation of classic insufficiency. Above all, sufficiency is to be judged by reference to the nature of the invention and the field in which it represents a contribution to the art. What constitutes "invention..., prolonged research, enquiry or experiment", judged in that context, does not seem to me to involve taking into account the fact that the relevant cell line could have been deposited.
- Quite apart from this, I find it a little difficult to see just how one would take into account the fact that the cell line could have been deposited when deciding whether the work involved in following the teaching was "routine" or not. Take a rather gross example. Suppose the court formed the view, ignoring the possibility of depositing a cell line, that six months experimentation was too much, but three months experimentation was not too much, to be routine. It is then difficult to see why three months should be no longer acceptable, merely because the patentee could have deposited a cell line. To put the same point slightly differently, it is hard to identify by how much one would reduce the three month period to take into account the right to deposit a cell line. I appreciate that it could be said that this point looks rather stronger when one puts it in abstract, rather than by reference to specific facts, but I do not think that it thereby loses its validity.
- The second issue relates to the common general knowledge which is to be treated as available to the notional addressee. I have already explained the concept of common general knowledge. The question which has been raised is whether, as Mr Thorley and Mr Kitchin contend, he is to be treated as limited to the skill and common general knowledge of the notional addressee, when deciding whether the work required of him would be more than routine from his point of view, or whether, as Mr Waugh contends, one is entitled to take into account the fact that the addressee might consult an individual or a text book in a different field, in effect to supplement the common general knowledge which he would otherwise have. I do not think that that question can be answered in abstract. As a matter of general principle, it appears to me that the court should be careful before it ascribes to the notional addressee information outside his common general knowledge, when considering whether a claim is classically insufficient or not. After all, it may be said, the whole point of deciding on the identity and qualifications of the notional addressee, is to enable the court effectively to determine the common general knowledge which is to be taken into account when determining issues of construction and sufficiency. Furthermore, if one steps outside the common general knowledge of the notional addressee, one is entering into a field of uncertainty. For instance, if he is to be assumed to consult a text book in a neighbouring field, it may depend upon which text book he consults as to what he is assumed to have picked up; if he is to consult a colleague, it may depend on where he is working or who his colleagues are.
- On the other hand, I would have thought that it would be inappropriate to conclude that, as a matter of absolute principle, once one has assessed the skill and knowledge to be attributed to the notional addressee, it is impossible in all circumstances to assume that he would not consult someone else on a certain point, when trying to implement the teaching of the patent. After all, a patent is ultimately a document which is intended to have practical, rather than theoretical, consequences, and its teaching is there for the purpose of being worked in the commercial and technical world. It seems to me, unrealistic in those circumstances to require the court to adopt a rigid approach to this sort of question. For instance, suppose the expert witnesses called by each side unequivocally agree that, although a certain fact would not be within the scope of the knowledge of the skilled addressee, it would be obvious to him (from the teaching of the patent or otherwise) that he should consult a certain book which would have the information in question (or a certain class of books all of which had that information). Even if there is no express reference to the book (or class of books) in the patent, I would have thought it wrong to proceed on the artificial, indeed erroneous, assumption that the skilled addressee would not have obtained that information.
- A third question which, to my mind, is also not capable of a generalised answer, is the extent to which the addressee should be taken to be aware of the contents of a particular paper or other document specifically referred to in the specification of the patent in suit. If the paper is referred to in the specification in connection with a particular aspect of the patent, then, even though the relevant passage in the paper is not quoted in the patent, I would have thought that, at least in general, the information identified in the paper should be treated as being within the knowledge of the addressee (even if it would not be within his common general knowledge) because he is effectively given the information, albeit indirectly, in the patent itself. On the other hand, if the paper is referred to in the patent in connection with one point, it will not by any means necessarily follow that the addressee should be treated as being aware of information contained in the paper relating to another point, which may also be relevant to the teaching of the patent. The answer must depend upon the facts of the particular case. Sometimes, the court would not be prepared to conclude that the addressee would pick up the fact that the paper contained information on one point, if it was referred to in the patent in connection with a different point, at least in the absence of cogent expert evidence to that effect. In other cases, the court may be prepared to conclude that it is obvious, once one appreciates the teaching of the patent as a whole, the purpose for which the paper is referred to, and the different information which is available in the paper, that that latter information should be treated as being available to the addressee.
Biogen insufficiency and breadth of claim: Introductory
- The issues and arguments on Biogen insufficiency in this case highlight a problem of some general significance in the field of biotechnology patents, and seem to me to represent a particularly acute problem of balancing "fair protection for the patentee" with "reasonable ... certainty for third parties". Dr Lin was the first person to succeed in sequencing the amino acid residues in EPO and most importantly the encoding regions (i.e. the exons), the splice sites and the two promoters of the EPO gene. Now that the obviousness argument has been abandoned, it is common ground that what he did was inventive, and that, subject to questions of classic insufficiency, discovery and lack of novelty, Amgen should therefore be entitled to a patent which extends to such techniques (if any) which the patent enables and which relate directly to the sequences he revealed - especially that in Table VI of 605. An important issue between the parties which remains is whether the disclosure of the patent justifies Amgen including in their claims (i) DNA sequences falling within Claim 1(b) and/or (c) whose sequences are not revealed and (ii) polypeptides which fall within Claims 1 and 19. i.e. those with similar but different sequences from that shown in Tables V and VI, and which possess the same biological properties as EPO, namely analogues of EPO. There is no significant teaching as to the sequences of the further DNA claimed in Claims 1(b) and (c) and no teaching how to get them or identify them, and very little teaching (save on page 47) as to possible analogues of EPO claimed in Claim 19 or as how to get them or identify them.
- In order to deal with the issue, it is necessary to consider in a little detail three decisions relating to the insufficiency of biochemical patents. They are Biogen [1997] RPC 1 itself, American Home Products [2000] IP&T 1308 and Chiron Corporation -v- Murex Diagnostics Limited [1996] FSR 153. I propose to consider the cases in that order, although Chiron, a decision of the Court of Appeal, was decided about a year before Biogen, a decision of the House of Lords, and the decision of the Court of Appeal in American Home Products is the most recent of the three decisions. Each of these cases provides helpful guidance as to what may constitute insufficiency in the present case. Indeed, Biogen itself goes further and lays down important principles (which have been followed in other jurisdictions as well as in England). I think it is nonetheless important, as with virtually any decisions, to have in mind the particular facts of each case, when considering the guidance given by the judgments or speeches.
Biogen insufficiency and breadth of claim: Biogen [1997] RPC1
- In Biogen [1997] RPC 1, the claim was, in effect, to any recombinant method of making antigens possessed by the hepatitis B virus; there were two types of antigen, core antigens and surface antigens. Lord Hoffmann started by considering the extent to which the priority document (known as "Biogen I") was enabling. He accepted the findings of Aldous J that core and surface antigens (HBcAg and HBsAg) could be made by the skilled man on the basis of the disclosure of that priority document, and that it was accordingly enabling to that extent. What Professor Murray, the inventor, had done, was to use a particular technique to introduce the DNA coding for the particular polypeptide into a host cell, in which the DNA had been expressed. Aldous J had held that Professor Murray's technique was not obvious at the relevant date, but that the actual result, namely the synthetic HB virus antigen, was an obvious thing to try to make.
- At [1997] RPC 43 lines 42 to 47, Lord Hoffmann said this:
"The general idea of expressing the gene for a polypeptide displaying HBV antigen specificity in a suitable host was... fairly widely entertained. The inventive concept was the notion that Professor Murray's method of achieving the goal - creating large fragments of genomic DNA, ligating them to pBR322 and introducing the hybrid molecule into E.coli - would work."
- After referring to a passage in the decision of the Board in T292/85 Genentech I/Polypeptide Expression [1989] OJ EPO 275, Lord Hoffmann went on to say at [1997] RPC 48, lines 41 to 48, that the approach in that case involved the application of:
"[A] principle of patent law which has long been established in the United Kingdom, namely that the specification must enable the invention to be performed to the full extent of the monopoly claimed. If the invention discloses a principle capable of general application, the claims may be in correspondingly general terms. The patentee need not show that he has proved its application in every individual instance. On the other hand, if the claims include a number of discrete methods or products, the patentee must enable the invention to be performed in respect of each of them" - [1997] RPC at 48 lines 41 to 49.
- In this important passage, Lord Hoffmann distinguished between two types of invention, and he then went on to expand on that aspect in these terms at [1997] RPC at 49 lines 1 to 9:
"Thus if the patentee has hit upon a new product which has a beneficial effect but cannot demonstrate that there is a common principle by which that effect will be shared by other products of the same class, he will be entitled to a patent for that product but not for the class, even though some may subsequently turn out to have the same beneficial effect..... On the other hand, if he has disclosed a beneficial property which is common to the class, he will be entitled to a patent for all products of that class (assuming them to be new) even though he has not himself made more than one or two of them."
- He then quoted with apparent approval a passage from the decision of the Board in T409/91 Exxon/Fuel Oils [1994] OJ EPO 653, which included this:
"[T]he claims must be supported by the description, in other words, it is the definition of the invention in the claims that needs support. In the Board's judgment, this requirement reflects the general legal principle that the extent of the patent monopoly, as defined by the claims, should correspond to the technical contribution to the art in order for it to be supported or justified", - [1997] RPC at 49 lines 16 to 22.
- At [1997] RPC 50 lines 36 to 45, Lord Hoffmann said this:
"But the fact that the skilled man following the teaching of Biogen I would have been able to make HBcAg and HBsAg in bacterial cells, or indeed in any cells, does not conclude the matter. I think that in concentrating up on the question of whether Professor Murray's invention could, so to speak, deliver the goods across the full width of the patent or priority document, the courts and the EPO allowed their attention to be diverted from what seems to me in this particular case the critical issue. It is not whether the claimed invention could deliver the goods, but whether the claims cover other ways in which they might be delivered: ways which owe nothing to the teaching of the patent or any principle which it disclosed."
- At [1997] RPC 51 lines 44 to 50, Lord Hoffmann discussed "the technical contribution to the art which Professor Murray made in 1978 and disclosed in Biogen 1". He said that:
"[Professor Murray's contribution] consisted in showing that despite the uncertainties which then existed over the DNA of the Dane particle - in particular, whether it included the antigen genes and whether it had introns - known recombinant techniques could nevertheless be used to make the antigens in a prokaryotic host cell."
He then went on to indicate that he was prepared to accept that "the method was shown to be capable of making both antigens and... that it would work in any otherwise suitable host cell".
- Having considered the problem which existed before the disclosure of Biogen I, the nature of Professor Murray's contribution, and the legal principles, Lord Hoffmann went on to say this at [1997] RPC 51 line 50 to 52 line 39:
"Does this contribution justify a claim to a monopoly of any recombinant method of making the antigens. In my view it does not. The claimed invention is too broad. Its excessive breadth is due, not to the inability of the teaching to produce all the promised results, but to the fact that the same results could be produced by different means. Professor Murray had won a brilliant Napoleonic victory in cutting through the uncertainties which existed in his day to achieve the desired result. But his success did not in my view establish any new principle which his successors had to follow if they were to achieve the same results. The inventive step, as I have said, was the idea of trying to express unsequenced eukaryotic DNA in a prokaryotic host. Biogen 1 discloses... the way to do it.... This, if anything, was the original element in what Professor Murray did. But once the DNA had been sequenced, no one would choose restriction enzymes on this basis. ...The metaphor use by one of the witnesses was that before the genome had been sequenced every one was working in the dark. Professor Murray invented a way of working with the genome in the dark. But he did not switch on the light and once the light was on his method was no longer needed. ...
It is said that what Professor Murray showed by his invention was that it could be done. HBV antigens could be produced by expressing Dane particle DNA in a host cell. Those who followed, even by different routes, could have greater confidence by reason of his success. I do not think that that is enough to justify a monopoly of the whole field. I suppose it could be said that Samuel Morse had shown that electric telegraphy could be done. The Wright brothers showed that heavier-than-air flight was possible, but that did not entitle them to a monopoly of heavier-than-air flying machines. It is inevitable, in a young science like electricity in the 19th century or flying at the turn of the last century or recombinant DNA technology in the 1970s, that dramatically new things will be done for the first time. The technical contribution made in such cases deserves to be recognised. But care is needed not to stifle further research and healthy competition by allowing the first person who has found a way of achieving an obviously desirable goal to monopolise every other way of doing so."
- Finally, I should quote a passage in Lord Hoffmann's speech at [1997] RPC 53 lines 7 to 14, where he was considering the approach of the Board:
"[I]n arriving at [its] conclusion, the Board directed its attention solely to the question of whether the teaching in Biogen I would enable the man skilled in the art to achieve expression of HBsAg as well as HBcAg. Nothing was said about whether the claims were too broad because expression could also be achieved without the use of the teaching which it contained, by a method of which it could not be said, in the words of the Technical Board in Genentech I, that it was "in a manner which could not have been envisaged without the invention"."
In a sense, the first sentence in that quotation can be said to be concerned with classic insufficiency, and the second sentence with Biogen insufficiency.
Biogen insufficiency and breadth of claim: American Home Products [2000] IP&T 1308
- The patent in American Home Products [2000] IP&T 1308, was concerned with the molecule rapamycin, which is produced by the bacterium Streptomyces hygroscopicus. As at the priority date, rapamycin was a known product, whose detailed molecular structure was also known. Professor Sir Roy Calne discovered that rapamycin had an immunosuppressive effect, whose mechanism was not known, but which was particularly valuable in that it inhibited organ or tissue transplant rejection. As Aldous LJ explained at [2000] IP&T 1310H:
"Because rapamycin was a known product at the priority date, it could not be patented; neither could its use as a treatment, because that would offend s. 4(2) of the Patents Act 1977. For those reasons the main claims of the patent followed the form known as "Swiss type" which is used to claim an invention for a second medical use."
- The first issue between the parties was whether, as a matter of construction, the patent was limited to the particular molecule or substance rapamycin, or whether it extended to all "derivatives thereof which exhibit the same type of inhibition to organ rejection as rapamycin and which are suitable for the preparation of a medicament inhibiting organ or tissue transplant rejection in a mammal" (see at [2000] IP&T 1319a). The second issue was, if the patent was not limited to rapamycin but extended to all such variants, it was or would be insufficient. On the issue of construction, the Court of Appeal held that the patent was limited to rapamycin alone, and in those circumstances the question of sufficiency did not arise. Nonetheless the court went on to consider that question. Aldous LJ, at [2000] IP&T 1322j to 1323a identified "two differences of substance between the parties". The first was "whether the disclosure has to be sufficient for the full ambit of the claim to be performed" and the second was the "particularity of disclosure ... necessary".
- Aldous LJ then said that the first issue had been settled by Biogen [1997] RPC 1, especially at 47 to 49. At [2000] IP&T 1325h, he said this:
"As the judge held, Professor Calne hit upon the new use for rapamycin. The specification contains an enabling disclosure of that product. Whether any particular molecule derived from rapamycin would work at all was impossible to predict with certainty nor how many would have immunosuppressant activity. Even if a rapamycin derivative were produced which had immunosuppressant activity, it would be impossible to be certain that it did not exhibit unpredictable defects. Discovering those defects would need in vivo tests which would take a long time. As the judge described the claim, it covered all the molecules which would work, but left it uncertain as to which ones do and how many of them there are. Such a claim does not reflect a class with a unifying characteristic. It is a claim to a number of compounds with the number and identity being left to the skilled person to find out."
- At [2000] IP&T 1326a-b, Aldous LJ continued:
"The invention as described was the discovery that rapamycin had [immunosuppressive] advantages. Some derivatives would be expected to have similar advantages, but the skilled person would not be able to predict which ones would have that actuality and, even if the right one was selected, it would take prolonged tests to find out whether it had the appropriate qualities. It follows that, as Lord Hoffmann pointed out in Biogen, the patent, to be sufficient, must provide an enabling disclosure across the breadth of the claim."
- I should also quote from the judgment of Aldous LJ at [2000] IP&T 1327e-g:
"There is a difference between on the one hand a specification which requires the skilled person to use his skill and application to perform the invention and, on the other, a specification which requires the skilled person to go to the expense and labour of trying to ascertain whether some product has the required properties. When carrying out the former the skilled person is trying to perform the invention, whereas the latter requires him to go further and to carry out research to ascertain how the invention is to be performed. If the latter is required the specification would appear to be insufficient.
The patentees wish to construe claim 1 to include derivatives of rapamycin which exhibit inhibition to organ rejection like rapamycin itself. Thus upon the patentees' construction, the specification must teach how to perform the invention with such derivatives of rapamycin. Upon the judge's findings of fact, the specification does not contain that teaching and therefore the patent would be insufficient, if that were the correct construction of claim 1."
- As in this case, the court was faced with an argument by the patentee that, if such a narrow conclusion were correct, that biotechnology patent, and indeed most biotechnology patents, would be effectively valueless. The force of that point can be seen when one sees the relatively tiny modification to the rapamycin molecule which resulted in the allegedly infringing product (see at [2000] IP&T 1314b-f and 1315f-j). However, Aldous LJ said this at 1329e-f:
"I do not agree that a patent limited to the second use of rapamycin is virtually valueless. The patent protects the second medical use and the long and expensive work that has been carried out to obtain regulatory approval. Thus a person who wishes to market a derivative has to make the derivative and then carry out the long and expensive work needed to get it on the market. Without the patent, other manufacturers could use the work of the patentees. In any case, I do not believe that the patent system should be used to enable a person to monopolise more than that which he has described in sufficient detail to amount to an enabling disclosure."
It should perhaps be noted that those observations were made in the context of a finding of fact made at first instance that there was "a strong probability that other molecules derived from rapamycin would exhibit similar conformation in those areas which made rapamycin efficacious and would also exhibit similar immunosuppressant activity" (at [2000] IP&T 1316e).
Biogen insufficiency and breadth of claim: Chiron [1996] FSR 153
- In Chiron [1996] FSR 153, the patent described how particles of virus had been isolated from a chimpanzee, called Rodney, suffering from a form of hepatitis known as NANBH and how strands of DNA were cloned, in familiar manner, to form a cDNA library. It went on to describe how strands of the cloned cDNA were examined for existence of codons encoding the formation of an antigenic determinant to bind antibodies already present in individuals who had been infected with NANBH. Out of a large number of samples, one proved positive in five out eight tests. The genetic sequence of that cDNA sample was determined and set out in the patent, together with the 859 amino acid sequence of the polypeptide thereby encoded. The virus was named HCV.
- Claim 1 of the patent in Chiron claimed:
"A polypeptide... comprising a contiguous sequence of at least 10 amino acids encoded by the genome of .... HCV and comprising an antigenic determinant where an HCV is characterised by:
1. A positive stranded RNA genome;
2. Said genome comprising an open reading frame [sc. the exon] encoding a polyprotein; and
3. Said polyprotein comprising an amino acid sequence having at least 40% homology to the 859 amino acid sequence in Figure 14."
- The Court of Appeal in Chiron had to consider the question of sufficiency about a year before the decision of the House of Lords in Biogen, but they did have the benefit of the reasoning of the Court of Appeal in Biogen [1995] FSR 4, to which they referred, and by which they regarded themselves bound. At [1996] FSR 181, Morritt LJ said this:
"For the appellants it was submitted that a claim to a class of products is a claim in respect of more than one invention if the claim does not specify a novel common feature. If it does not, so it was contended, the individual members of the class will only be ascertainable by empirical means so that the invention of one member is not the invention of all the others..... This was disputed by Chiron. They submit that the test is the much simpler one they deduce from the decision of this court in Biogen whether in the case of a product claim the invention of the one product is the invention of the other or others."
- After considering guidance given by the House of Lords in May & Baker Limited -v- Boots Pure Drug Co. Limited (1950) 67 RPC 23 (also relied on by Lord Hoffmann in Biogen), Morritt LJ referred to the Court of Appeal's decision in Biogen [1995] FSR 4, and then said this at [1996] FSR 182:
"The analysis of the authorities supports the argument for Chiron. Further the decision in Biogen is binding on us. However the proposition established in Biogen by the adoption of the principle expressed by Lord MacDermott in May & Baker ... provides no yardstick by which to judge whether the invention of one product is the invention of the other. In that respect it is likely that when it is found that the invention of the one is also the invention of the other that part of the test formulated by the appellants which requires a novel common feature will also be found to have been satisfied. But we do not consider that the satisfaction of that test is indispensable to the conclusion that both the products are comprised in one invention for that would be to impose an unnecessary strait jacket on the ability of the law to adapt to new circumstances."
With that principle in mind, Morritt LJ then turned to the issue of "whether individual claims relate to more than one invention".
- At [1996] FSR 183, Morritt LJ turned to the inventive concept. He said this:
"The discovery or inventive concept as found by Aldous J was "the identification and sequencing of the 5-1-1 clone as the cause of NANBH: in essence, finding out the specific sequence for HCV"... Claim 1, according to Aldous J..., is "a claim to a new class of chemicals which relate to HCV". The appellants dispute this contention by reference to the principle as formulated by them. They contend that chemicals comprised in the claim do not comprise a class and will be different both chemically and biologically from each other. They submit that the claim does not prescribe a formula by which they may be linked. Chiron claims that the discovery lay in finding and sequencing the virus and the invention is the chemical which that analysis has shown to contain the antigenic determinants of the virus to which antibodies to HCV will bind."
- Morritt LJ then stated that the Court of Appeal adopted Chiron's approach. He went on to say this at [1996] FSR 183:
"There is nothing in [Claim 1] as a matter of language or in the specification as a whole to differentiate between one polypeptide falling within the claim and any of the millions of others. At the other end of the scale there is no doubt that it was the finding and sequencing of the Rodney virus genome which was of the greatest importance for this discovery and analysis enabled the antigenic determinants to be found in the protein expressed by the sequence. In view of the fact that an antigenic determinant can only be defined by reference to the antibody which binds to it and the further fact that the immune response of an individual produces a whole range of antibodies to any given virus it is not possible to define antigenic determinants by reference to any common chemical formula. Thus the invention is the chemical comprising at least 10 amino acids in which there is an antigenic determinant to which an antibody to HCV will bind. In our view both in substance and in form this is a single invention properly defined by the common denominator of the existence of an antigenic determinant of HCV, notwithstanding that the resulting polypeptides will have divergent characteristics in other respects. The invention of one is the invention of all of them because that which is common to all is of the essence of the discovery and that which distinguishes one from another is irrelevant to that discovery."
- At [1996] FSR 184, he pointed out that:
"A different conclusion was reached in Biogen because the relevant claim required the polypeptide "to display HBV antigen specificity". That phrase covered both HBcAg and HBsAg which were different in use and composition and responded to different antibodies. We do not regard that decision as being inconsistent with our conclusion in respect of the numbers of inventions comprehended in claim 1 of the patent in this case."
This indicates that the Court of Appeal's reasons for finding the claim in Biogen too wide were a little different from those of the House of Lords. To that extent, as Roche and TKT argue, the reasoning in Chiron must be approached with circumspection.
- Importantly, because it can be said to tie in Chiron with what was said subsequently by Lord Hoffmann in Biogen [1997] RPC 1, Morritt LJ said this at [1996] FSR 185:
"[W]e are bound by Biogen to consider whether in this case on the facts the invention can be performed across its whole width. But before doing so it is useful to consider what is involved in such performance and what the invention is; the importance of the latter point having been emphasised in Biogen... First, the performance required is not the production of a commercial product but rather of a workable prototype. Secondly, the man skilled in the art is expected to display a reasonable degree of skill and common knowledge of the art in making trials and to correct obvious errors in the specification... In this connection it should be noted that it is not disputed that it can be determined by routine means whether any given polypeptide falls within claim 1 nor that out of the large number of polypeptides which do fall within claim 1 any given polypeptide falling within the claim could be made by routine means."
- Immediately thereafter, at [1996] FSR 185-6, Morritt LJ continued:
"[T]he appellants... contended that the real issue is the extent to which the patent specification has identified to by one skilled in the art which polypeptides are claim 1 polypeptides because they have at least one HCV antigenic determinant.
In our judgment the contention of the appellants is misconceived. As we have already held in relation to the point of construction claim 1 does not require the mapping or identification of all the antigenic determinants or the best of them. ....[The] claim is to a polypeptide comprising an antigenic determinant encoded by the HCV genome as identified; such a polypeptide can be defined in the sequence by exposure to an antibody and can be made by routine methods of molecular biology. The effect of the issue as formulated by the appellants is to widen the claim from what it is and then to contend that so widened it is not enabled across its width. In our judgment the issue is not as the appellants have formulated it. The consequence is that the invention as actually claimed is capable of being performed by a person skilled in the art because it is a claim to a polypeptide which can be both made and identified by routine means. Accordingly, in our judgment claim 1 is not invalid on the grounds of insufficiency."
The alleged insufficiencies in 605
- A number of different points were advanced on behalf of Roche and/or TKT as to why at least some of the Claims of 605 are insufficient. When considering those arguments, I need only consider Claims 1, 19 and 20 (although the points impinge on virtually all the other Claims). Those arguments are as follows. The first argument is that, at least without undue burden, no source of human cDNA was available to work Claim 1. The second contention is that Claim 1 is insufficient owing to the defective nature of some of the Examples. The third argument is that the 605 patent's teaching is insufficient in relation to Claim 1 claiming expression in prokaryotic cells generally (or human cells in particular). Fourthly, integer (b) and integer (c) of Claim 1 are allegedly insufficient in light of the requirement of "stringent conditions". Fifthly, there is a contention of insufficiency so far as the claims to analogues of EPO in Claim 1 and Claim19 are concerned. Sixthly, there is the alleged insufficiency of the comparison of recombinant EPO with urinary EPO alleged against Claim 19. Seventhly, there is an attack on the sufficiency of the test in Claim 20.
- It appears to me that the first, second, third and fourth allegations of insufficiency, which relate to Claim 1, the sixth allegation, which relates to Claim 19, and the seventh allegation, which relates to Claim 20, are essentially allegations of classic insufficiency. However, it may be that some of those alleged insufficiencies (particularly the third and, possibly, the first) could, as it were, be avoided when one considers the permissible breadth of the claim in light of the guidance given by the House of Lords in Biogen [1997] RPC 1, and the two decisions of the Court of Appeal to which I have referred. To the extent that it is necessary to consider that aspect of any of the six allegations of classic insufficiency, I will go on to consider them when I deal with Biogen insufficiency and breadth of claim.
- As to the fifth allegation of insufficiency, it relates to an issue upon which there is very brief and limited reference in the 605 patent (at page 47) namely analogues of EPO. It appears to me hard to argue against the proposition that the teaching of the patent in this connection is classically insufficient, in the sense that there is no real guidance as to what substitutions and deletions in relation to the various amino acid residues in EPO, as disclosed in Table VI, would or might work. Accordingly, I think it right to consider this aspect under the heading of Biogen insufficiency and breadth of claim.
- An allegation of insufficiency, which I do not need to consider at this stage at any rate, would arise if I had accepted Amgen's construction arguments as to the meaning of "host cell", "recombinant polypeptide" and "expression of exogenous DNA sequence". In that event, the claims would, as a matter of construction, have extended to expressing EPO by homologous recombination (i.e. transfecting with an exogenous DNA targeting construct sequence a cell with the consequence of "switching on" an endogenous EPO-encoding DNA sequence). TKT argues that, if any claim of the 605 patent did extend to that technique, then any such claims would be insufficient. I shall, in fact, have to consider that argument as a substantially similar argument when dealing with the issue of TKT's infringement.
- It is also right to mention that, in the light of my view as to the correct construction of Claim 2, it raises no additional problem with regard to insufficiency. The polypeptide coded for in that Claim is a single sequence (subject to the leader peptide point) namely that shown in Table VI of the 605 patent. There was no specific argument directed to the sufficiency of Claim 2, possibly because its relevance only came to light during closing speeches. In my view, there is no basis for concluding that Claim 2, as interpreted above, is insufficient.
H. CLASSIC INSUFFICIENCY: SOURCES OF HUMAN cDNA
Introduction
- As I have mentioned when considering the technical background, if one is to obtain a cDNA which encodes a specific protein, one needs a library with the appropriate DNA before one can probe for the mRNA which it encodes. Accordingly, if the skilled addressee is to obtain cDNA which encodes a specific protein, he has to make or find a DNA library which he can be tolerably confident will contain the specific mRNA for which he is searching.
- Roche's case is that if Claim 1 extends to human cDNA, then the patent is insufficient in its teaching in that connection because it does not teach or otherwise enable (e.g. through the medium of a deposited cell line) the notional reader to obtain a source of human cDNA, at least without undue effort. Amgen suggests that the reader would have appreciated that there were three different sources from which he could have obtained cDNA as at the priority date or the date of filing. The three types of source are tissues, cell expression, and gene synthesis. I shall consider those various suggested sources in turn.
- The issue has an air of unreality about it in the sense that human cDNA is not identified specifically in the 605 patent. As Mr Waugh says, the patent is ultimately concerned only with sufficient encoding and other sequences to enable EPO to be expressed by recombinant means. However, there is an argument as to what human cDNA actually is. There is no doubt that it includes the five directly encoding sequences, and that it excludes the intervening four introns. However, Roche contend that it includes the whole of the upstream (non-encoding) sequence from the cap and the downstream (non-encoding) sequence to the poly A tail. Like so many terms in this field, it seems to me that "cDNA" can have more than one meaning. It appears to me that in some contexts, Roche's definition (referred to for convenience as "full length cDNA") would be appropriate as Dr Fritsch and Professor Brammar said. On other occasions it can mean something much shorter. As Professor Brammar accepted, "A cDNA clone can indeed be very short and still be called a cDNA clone". Professor Wall had the same view.
- In my judgment, the human cDNA which the reader of the 605 patent would have in mind would be a human cDNA sequence which would "deliver the goods". One is not here concerned with abstract concepts but with an exercise with a practical commercial and/or therapeutic aim. Accordingly, what the reader would understand is that he needed "a cDNA suitable for expression", "a full length coding sequence with an ATG with some means of getting translation" (to quote Professor Brammar). If there is "no technical reason" for including the upstream or downstream sequences, as Professor Wall said, then they would not be regarded by the reader as being necessary. Professor Brammar accepted that "what one is interested with [sic] is essentially the coding sequence". As Professor Wall said, "the other elements... necessary in order to express the protein in a cell... [such as the Kozak box] could be added synthetically or by using... standard recombinant DNA techniques". This is consistent with the view of Roche at the time: they recorded that Dr Lin had "disclosed the full sequence of human cDNA". Dr Fritsch had the same view. Indeed, Roche's admission as to the nature of their allegedly infringing product identifies it as expressed through a cDNA sequence which effectively is the encoding sequence of the EPO gene. I also rely on the evidence in relation to the so-called Hu-13 clone discussed in the next section but one of this judgment.
Human tissue
- So far as tissue sources are concerned, although they relied primarily on the evidence, Roche point to what is said in the 605 patent itself at page 6 line 58 to page 7 line 2. After referring to the fact that previous efforts to isolate DNA sequences encoding for EPO have failed, the patent says this:
"This is due principally to the scarcity of tissue sources, especially human tissue sources, enriched in mRNA such as would allow for construction of a cDNA library from which a DNA sequence coding for erythropoeitin might be isolated by conventional techniques."
- On the face of it, that passage, read on its own, does support Roche's suggestion. However, that passage, like any other part of the patent, cannot be read on its own: it must be read in the context of the patent as a whole, and in particular in its immediate context. If one reads the remaining part of the paragraph in which that passage is found, one sees that reference is made to difficulties arising from the fact that, until the disclosure of 605, the genetic sequence of the EPO gene was unknown, and there was only limited assistance to be got from Sue and others (who had identified, albeit not wholly accurately or completely, the first 27 amino acids of EPO, which were rather degenerate), and that the "genetic material coding for human erythropoeitin is likely to constitute less than 0.0005% of total human genomic DNA". Professor Wall said in his evidence that the position became rather different with the disclosure of the 605 patent itself, which clearly enabled long and reliable probes, indeed probes consisting of the whole of the encoding sequence, to be prepared. Mr Thorley suggests that Professor Wall retreated from that view, but I did not understand him to have done so. What he did accept, quite correctly, was that there was no teaching in the patent as to an appropriate source for constructing a cDNA library, which is why, of course, Amgen have to rely upon common general knowledge.
- I incline to the view that, as a matter of language, even taken in its context, the passage relied on by Roche does mean that the absence of a source of human cDNA is a problem. The subsequent passage seems to me to identify additional problems rather than connected problems. However, I accept that the point is not clear. In any event, I do not think the point need be decided. The issue to be determined is ultimately not what the patent suggests, but what was revealed by the patent itself or what was common general knowledge. The only tissue source revealed in the specification of 605 is the human kidney, but it is accepted by Amgen that this would not be a practical source. Amgen's contention is that human foetal liver was a suitable source of cDNA for a person seeking to probe for the EPO gene.
- Roche contend that the foetal liver source would not have been within the common general knowledge as at the priority date. Roche also contend that, even if foetal liver as a source of a cDNA library would have been known to the notional reader, it would not have produced enough EPO mRNA for a cDNA library to contain detectable amounts of cDNA encoding EPO. I shall first consider the evidence relied on by Amgen to support its contention that the human foetal liver as an appropriate source for cDNA would have been within the common general knowledge.
- First, Amgen rely on the Lawn Library, which was of course a genomic DNA library. It was referred to in the specification of the 605 patent, and, indeed, in the Maniatis Manual (accepted to be common general knowledge). It was constructed from foetal liver cells. Indeed, it appears to me from the evidence that the foetal liver was a reasonably well established source of genetic material for various studies. The article by Lawn (cited at page 4 line 10 of the specification) expressly refers in more than one place to the fact that his library was prepared from human foetal liver sources.
- Secondly, there is the evidence of what those actually working in this field were thinking at the time. In an internal paper confidential to GI (one of the Roche parties in this litigation) and written in August 1982 in connection with GI's research to seek to clone the EPO gene, Dr Fritsch wrote:
"The cloned DNA sequences to be screened by these oligonucleotides could consist of either a cDNA library of mRNA from foetal liver cells (the liver is the major organ which produces EPO in the foetus) or the human genomic DNA library" (emphasis added).
In his evidence Dr Fritsch described the foetal liver as "the favourable tissue" and "certainly a place to look", but he did not accept that, as at 1984, it was known to be the "correct place".
- Professor Orkin of Harvard University, whose team, it may be remembered, was embarking on the same exercise at around the same time said in his unchallenged witness statement that he had "looked at human foetal liver tissue which was believed to be an enriched source of EPO mRNA". However, he also said that he and his team "did not know whether or not the human foetal liver was an appropriate source" for mRNA encoding for EPO.
- Thirdly, by the priority date, there were a number of different publications which disclosed the same information. Two text books, one published in the United Kingdom and the other in the United States, support the proposition, namely Hardisty & Weatherall on "Blood and its Disorders" (Second Edition 1982), and Wintrobe & others "Clinical Hematology" (1981) respectively. In the former text book, at page 121, various papers are referred to and are said to "indicate that the major source of erythropoeitin in foetal life is the liver and not the kidney". The latter text book states that "the liver... appears to be the major site of erythropoeitin production during fetal and neo natal development". The same view was being expressed in review articles at the time. Thus in a "Mini Review" on "Control of Erythropoeitin Production" in (1983) Proc Soc Exp. Biol and Med at 173, 289, Dr James Fisher set out to "summarise the present state of knowledge of the many... factors which may play a role in the control of erythropoeitin production". At 296, he said that "the liver has been demonstrated... to be the primary site of production of erythropoeitin in the foetus ...". To the same effect, are articles by Thomas & others, in Brit J of Obs and Gyn, 90 at 795 in the same year, and by Testa and others, in Haematologica 67(1):64 in 1981.
- I was also shown a number of papers which had been published by the priority date. Thus, Congote et al. in 1977 had published a paper entitled "Regulation of Fetal Liver Erythropoiesis" in J. Steroid. Biochem. 8:423, whose abstract began with the words "The liver is the main erythropoietic tissue of the human fetus at mid term". Two papers by Zanjani et al. had described the liver as the "primary" or "major" site of EPO "formation" or "production" in the foetus: see J. Lab. Clin. Med. 89:640 (1977) and J. Clin. Invest. 67:1183 (1981).
- In a paper published as one of the "Letters to Nature", in Nature, and received in December 1984, thirteen scientists working for GI including Dr Jacobs, Dr Miyake (who had isolated very small amounts of EPO from urine, as mentioned in the patent) and Dr Fritsch (who gave evidence) reported on "Isolation and Characterisation of Genomic and cDNA Clones of Human Erythropoeitin". At the beginning of the paper ("Jacobs"), the authors mentioned that EPO was "produced in... the liver of fetal or neo-natal mammals", citing in support two papers, one of which was one of the Zanjani papers. Jacobs also referred to Congote to support its statement that EPO was "released from... human fetal liver".
- I have reached the conclusion, despite this evidence, that, as at 1984, human foetal liver was a source of human cDNA was not within common general knowledge. I accept that a number of people who would have been interested in the disclosure and teaching of the 605 patent would have known that the foetal liver was probably the most promising place to look for such DNA. However, in the first place, in accordance with the oral evidence of Dr Fritsch and the written evidence of Professor Orkin (whose evidence was not challenged in cross examination) it appears to me that it was more a question of the foetal liver being a promising place to look, as opposed to being a source in which one could have confidence. Secondly, I am not satisfied on the evidence that the notional addressee with the appropriate level of knowledge and skill (not the highest or the lowest common denominator, but somewhere around the mean) would have been aware of the fact that a foetal liver was even a promising place to look. There was no evidence that either of the two text books was known to the notional reader of the 605 patent. Indeed, the only evidence about them was that of Dr Fritsch who had not heard of Wintrobe. None of the papers to which I have been referred were, if I may put it this way, self-evidently part of the common general knowledge. None of the articles (other than Jacobs) were in journals read by ordinary molecular biologists; indeed, Professor Wall said that he had no knowledge about whether they would have been known to the skilled person in 1984. It appears that they were put before the court as a result of one of Amgen's advisers reading Jacobs, in which those papers were cited, directly or indirectly.
- Further, none of these papers could be said to be conclusive in showing that the EPO gene was expressed in the foetal liver. As Dr Fritsch said in cross-examination, even if the papers showed or strongly suggested that EPO was present in the foetal liver, that did not mean that it was being synthesised from EPO mRNA in the liver: it could have been produced elsewhere, and activated in the foetal liver. Indeed, there were papers published in or before 1984 which specifically suggested this possibility was correct (for instance Fyhrquist et al. in a Letter to Nature published in April 1984 in Nature 308:649).
- The fact that there was reference in certain text books to the foetal liver as being a source of EPO does not cause me to change my view. The mere fact that the two text books I have mentioned contained such references does not of itself necessarily establish that the contents were common general knowledge even to haematologists. Indeed, for what it is worth, Dr Winearls, a nephrologist who gave written evidence about the management of severely anaemic patients before and after the ability to treat such patients with recombinant EPO, did not suggest that the foetal liver was a source of EPO DNA known to him.
- I also draw assistance from an article published in 1986 by eight scientists working for Amgen in the Cold Spring Harbor Symposia on Quantitative Biology ("Cold Spring Harbor") Volume 51, page 693, in which they said this:
"Isolation of a gene for human EPO proved to be particularly problematical because there was no known source of mRNA. No cell lines had been characterised that produced significant amounts of EPO that could provide enriched sources of mRNA."
These scientists then went on to refer to the article by Fyhrquist (op. cit.) as indicating that the mere fact that EPO was present in the human kidney did not mean that the encoding DNA or mRNA was present in the human kidney. This appears to be consistent with Dr Fritsch's evidence that the mere fact that EPO was or might have been known to be found in the foetal liver does not mean that the EPO mRNA was in that organ.
- I am less impressed by Roche's alternative argument that, even if the notional addressee of 605 should be treated as having known of foetal liver as a source of EPO mRNA for a cDNA library, it would not have contained detectable amounts of EPO cDNA, as Dr Fritsch suggested in his evidence. It is true that EPO is, and was known to be, a rare protein with correspondingly small, even tiny, quantities of mRNA. However, I think that, on this aspect of human tissue as a source of coding mRNA, Roche's case, and indeed, at least to an extent, Dr Fritsch's evidence, involved a degree of magnifying what might have been seen as a small potential problem into something which appeared to be, but was not really, substantial.
- Although it was accepted by Dr Fritsch that foetal liver did not require as low an oxygen level as an adult kidney to express EPO, he suggested that there would be concerns about the degree of hypoxicity necessary, and that this would depend upon the precise physiological conditions in the particular foetus. Secondly, he suggested that the foetal liver might only be producing EPO at certain stages of foetal growth. Thirdly, he suggested it was possible that the foetal EPO gene was different from the human EPO gene.
- There was little, if anything, other than what was said by Dr Fritsch, in the way of evidence to support the contention that any of these matters would have been of concern to the notional addressee of the 605 patent, or, indeed, that any such concerns would have been justified. For instance, there is nothing to suggest that EPO was expressed in the liver of foetuses only for certain periods of development, and no reason was put forward as to why that might be so. The reason that the evidence concentrated on livers of foetuses which were approximately 20 weeks old is obvious: the larger the liver, the better the prospect of getting many cells in which EPO was expressed, but foetuses more than 20 weeks old are rare, for moral and legal reasons.
- In relation to both aspects of alleged insufficiency in connection with human tissue, Mr Thorley placed some reliance on what was done by the research team at GI, who were engaged on the same exercise as Dr Lin. I am not convinced that that is of a great deal of assistance on either aspect of alleged insufficiency. The fact that one particular person did something, did not do something, found something easy, or did not find it easy, can, of course, be of some potential value when the court is called on to consider what the notional skilled man might have done, and how easy he might have found it. However, the problem with this sort of evidence is that it simply shows that a particular person took a certain amount of time and a certain amount of effort, as a result of which he got or did not get a particular result. If, as in the present case, the evidence is put forward to suggest that what was involved was a long and uncertain task, one can either take that at face value, or one can analyse how, if one omits the wrong turnings and dead-ends, the ultimate result could have been attained (if it was not achieved) or how it could have been attained much more quickly (if it was eventually attained). That would then lead to arguments as to whether the person who actually carried out the work was unlucky, incompetent, or exaggerating his difficulties, and as to whether the analysis which shows that he could have achieved his end much more quickly is merely based on wisdom of hindsight. That sort of argument then brings one straight back to the issue which the court would have to determine anyway. Over and above this, difficulties encountered in the past may be reduced or even avoided by the very disclosure of the patent in suit.
- It seems to me that the evidence as to what work was actually carried out by the GI team in relation to obtaining EPO cDNA runs into just this sort of problem. A number of steps taken by the team did not involve using a foetal liver cell line, and that could therefore be said to be irrelevant and distracting work. Some of their attempts to obtain encoding mRNA may have failed because they did not have the information revealed by the 605 patent, and so they were unable to use a long and reliable probe based on Table VI of the patent.
- In my judgment, therefore, if (contrary to my view) it had been within the common general knowledge that EPO-encoding mRNA would be found in a DNA library made up from foetal liver, then I am not persuaded that it would have involved undue effort on the part of the appropriately skilled man to screen that library with appropriate probes. While it would not have been an elementary exercise, it certainly would not have been a "research project" as suggested by Mr Thorley. In the context of this sort of technology, I do not consider that that exercise would have involved undue effort.
Cell expression
- Amgen's case is the notional reader would have appreciated that EPO cDNA could and would be cloned by taking the following steps. First, the EPO gene would be isolated from a genomic DNA library by a probe based on the information contained in Table VI of 605. Secondly, the EPO gene would be inserted into a host cell in accordance with the Examples in the patent (or indeed any other method known to a person skilled in the art): in particular, in light of the description in the patent, the cell would almost certainly be a COS or CHO cell. Thirdly, the host cell transfected with the EPO gene would express EPO mRNA, thereby providing an enriched source of this material. Fourthly, reverse transcriptase would be used to make cDNA, from which the EPO cDNA clone can be obtained.
- There is no specific teaching in connection with such a method in the patent, although Amgen is able to point to a reference to this sort of exercise in Example 5 (at page 25, lines 18 to 20 of 605), which was concerned with confirming the presence of a lysine amino acid residues in human EPO (not found in monkey EPO). The passage is in the following terms:
"Presence of the lysine residue in the human polypeptide sequence was further verified by the sequencing of a cDNA human sequence clone prepared from mRNA isolated from COS P-1 cells transformed with the human genomic DNA in Example 7 infra."
- This reference is said by Amgen to assist the contention that cell expression would have been in the mind of the notional addressee. However, in my judgment, one cannot build very much on that "passing reference" to the use of COS cells (as Professor Brammar put it). It does not purport to be a teaching specifically directed to the obtaining of human cDNA encoding for EPO. Furthermore, as Professor Wall accepted and Professor Brammar said in terms, all that the notional addressee would have understood from the passage was that the patentee had managed to obtain a clone with some human sequence in it, the sequence in question being sufficient to identify the lysine referred to in the passage I have quoted. In other words, in so far as the sentence implies that the patentee had obtained anything, it was not complete EPO cDNA, but a portion which enabled him to sequence across the lysine. I agree with Professor Brammar when he said that "there is no encouragement whatsoever in the patent how to make a human cDNA", and, ultimately, it does not seem to me that anything in the evidence of Professor Wall called that view into question. His evidence in relation to this sentence in the patent was that the notional addressee would know of the technique of obtaining cDNA from COS cells transfected with genomic DNA encoding for EPO, and would recognise the description of this procedure. This, therefore, brings one to the question of whether the notion of obtaining EPO encoding cDNA through the medium of a host cell transfected with encoding genomic DNA would have been within the common general knowledge. That, as I see it, is the central issue.
- Because attention had been concentrating on the sentence in Example 5 of the patent which I have quoted above, and because some of the papers referred to in this context involved the use of COS cells, much evidence was given over to whether, and if so to what extent, the skilled addressee would have a concern about using a COS cell for the purpose of producing the encoding mRNA. Thus, Professor Brammar said that the skilled reader would have regarded the use of COS cells as very risky, because of the inability of their splicing machinery to cope with a high degree of copying, and their general unreliability. On this issue, there is much force in Mr Waugh's argument that any perceived problem with COS cells is not really in point, because both Professor Brammar and Professor Wall agreed that there would be no or very little such concern if a CHO cell were used. Professor Wall specifically made that point in his written evidence in reply, where he said that if the skilled reader was concerned about COS cells:
"The obvious routes would be either to use a human tissue source... or... a different host such as a transformed CHO cell which could be utilised as a source of message."
Professor Brammar said that "I would have been certainly much happier trying [the CHO cell] route than going via COS".
- There is no doubt that the possibility of using transfected host cells for the purpose ultimately of expressing encoding cDNA for a particular protein had been raised by the priority date. Thus, in a paper published by Anderson and Kingston in 1983 in PNAS 80:6838 ("Anderson") ended in these terms:
"[A] genomic clone for the gene of interest could be isolated and sub-cloned into [an] expression vector; subsequent transfection of an appropriate host cell should then produce correctly spliced transcripts of the cloned gene from which cDNA could be prepared."
- While considerable reliance was placed on this observation in Anderson, by Mr Waugh on behalf of Amgen, I think it is of limited value. It seems to be a suggestion as to what might be done, and it is by no means clear that the authors had actually carried out that exercise. It may therefore have been nothing more than a bright idea, and there is certainly no reason to think that, if it was not already common general knowledge, this suggestion thereby caused it to become such. Indeed, I have no reason to think that Anderson should be treated as part of the common general knowledge of the notional addressee in the present case. It is true that Anderson is referred to in three places in 605, but the relevant passage is neither quoted nor relied on, even inferentially.
- Perhaps more importantly, in 1979, two important papers had been consecutively published in Nature. Hamer & Leder in Nature 281:35 reported that they had expressed the mouse globin gene in a monkey cell which was not the COS cell, but was a predecessor cell in the art, in the sense that it was an earlier, and less satisfactory, cell so far as transfection and expression was concerned. The complete chromosomal beta-globin gene was inserted into a vector, and this resulted in mRNA transcription, and the translation of the globin protein. Mantai et al. reported in Nature 281:40 that they had carried out a similar exercise in relation to the rabbit beta-globin gene in transformed mouse cells, resulting in beta-globin mRNA. As "land mark" publications, I believe that these two papers were part of the common general knowledge. Partly as a result, I think that the basic idea of obtaining cDNA through an appropriate host cell transfected with genomic DNA is something which would, in my judgment, have been within the contemplation of the notional addressee of the patent who wished to obtain cDNA which encoded for EPO.
- In this connection, it seems to me that answers given by Professor Brammar, a fair and honest witness called on behalf of Roche, in cross examination were revealing. He accepted that there were a number of "methods to get a cDNA" in 1984. He accepted that the sentence I have quoted from Example 5 of the 605 patent was a "short (but for the average person skilled in the art comprehensible)" way, but "not the only way" of achieving that end. He also accepted that "a person would be able to reproducibly isolate and produce a human EPO-encoding cDNA according to the methods... just described", which at least included transfecting a host cell with the appropriate genomic DNA. He also agreed that, based on the DNA sequence revealed in Table VI of 605, "the possibility existed to successfully use routinely practised techniques as they are e.g. described in Maniatis for the isolation of... cDNA probes" (my emphasis), albeit that his agreement was invited and obtained for this as "a technical proposition". He also accepted that the fact that "defective clones can be obtained by all isolations of... cDNA clones" was "known to the average person skilled in the art and he knew that he therefore had to repeat his trials once or several times". He also said that, although deriving cDNA from the mRNA expressed in a host cell transfected with genomic DNA "could be done", "it would be likely to be a difficult route because of the caveat... fishing for the authentic cDNA, with its proper splicing etc. amongst a complex pool of other transcripts".
- In my judgment, taking this evidence together with that of Professor Wall, which was to much the same effect, albeit that he was less concerned about problems, the notion of using an appropriately transfected CHO host cell for the expression of EPO-encoding mRNA, and thereby producing EPO cDNA was something which would have been within the common general knowledge of the notional addressee of the 605 patent. In a paradoxical way, it might be argued that this conclusion is called into question by the sentence I have quoted from Example 5, in the sense that it might have put off the skilled reader, because he would not have been happy about using a COS cell (for the reasons given by Professor Brammar) and it might not have occurred to him to use a CHO cell in light of that sentence. That point was not raised in evidence or argument, and I am not persuaded that it is a good one in any event.
- Indeed, I think that the use of a COS cell would also have been in the contemplation and common general knowledge of the skilled addressee. I consider that the notional addressee would have had in mind the possibility of using a COS cell line for this purpose. The concern expressed by Professor Brammar at aberrant splicing in COS cells was based on a number of papers, including one by Wojchowski et al in Biochim. Biophys. Acta 910:224 ("Wojchowski"), in 1987. Professor Wall described aberrant splicing as a comparatively rare event, and a paper by Wise et al in 1989 in Nucleic Acid Research 17(16): 6591 states that aberrant splicing would either result in poor expression of the protein or synthesis of other protein products with different molecular weights. Both of these, as I see it on the evidence, could be pretty easily demonstrated and the consequences excluded. Further, other papers (such as one by Nordstrom et al in 1986 in Biochim. Biophys. Acta 867: 152) indicate that expression in a transfected COS cell (with an exogenous construct including the genomic DNA and an SV40 promoter) resulted in more than 50% of the mRNA expressed being correctly spliced, and 78% of the transcripts could express the protein.
- Wojchowski had three contributors, one of whom was Professor Orkin. His evidence was that the work was ultimately successful and the results were achieved "within a few months". In cross examination, Professor Brammar accepted that the steps involved in dealing with problems in COS cells were "relatively routine and short". In re-examination, he described the work which Professor Orkin had to carry out as involving "steps [which] are not complex. They are fairly predictable but they impose a time lag. ... My guess would be at least a few weeks to correct the problem". Further, I draw comfort from the fact that there is no evidence of anyone using the transfected COS cell method with a view to obtaining the entire encoding region of the EPO gene and failing.
- In reaching my conclusion that the transfected host cell route to cDNA generally, and in COS cells in particular, would have been within the common general knowledge of the skilled man, I derive some comfort from the fact that patents in the biotechnology field applied for by, and granted in the United States to, GI (one of the Roche parties) seemed to support the view. Thus, in a patent applied for in 1987 (PCT/US87/02005) reference is made at page 15 to "the expression vector containing the gene [being] transfected into a mammalian cell, e.g. monkey COS cells." The application goes on to state that the "human gene is transcribed and the RNA correctly spliced ... mRNA is obtained from these cells and cDNA synthesised from the mRNA." The same sort of indication was contained in another application, PCT/US87/01537 at page 50. Indeed, there is an application for a European patent by GI, made in 1987, EP 314,705, which, at pages 11 and 12, refers to the construct being "amplified in bacteria and transfected into monkey COS cells ... . .. cDNA is synthesised from the mRNA by standard procedures." The lack of detail in these passages tends to support the view that the sort of addressee who would be reading patents of this type would be familiar with the transfected host cell route to cDNA. It is fair to say that these applications were made some three years after the application in the present case, but, at least on the evidence I have seen, including the various papers relied on by the parties, I am not convinced that that is a significant factor. In this connection, it is to be noted that Roche's case in relation to COS cells was based on a number of papers, including Wojchowski, which were published around the same time as, or later than, the filing of these patent applications by GI.
- Roche contends that the fact that the patent does not disclose a specific means of obtaining human cDNA emphasises the difficulty in 1984 of obtaining such DNA. Mr Thorley says that the obvious prize which was sought was human EPO ("HuEPO") and cDNA is a more convenient source of expression than genomic DNA, at least according to Roche's case. Accordingly, the fact that there is no teaching in relation to human cDNA in the patent Mr Thorley suggests that such DNA could not in fact be obtained easily.
- Although this point has some attraction, I have come to the conclusion that it should be rejected. At the relevant date, Amgen had a CHO cell transfected with a genomic EPO gene, which was giving good levels of expression, and which, indeed, is still giving good levels of expression, according to the evidence of Dr Browne. It is true that cDNA has advantages over genomic cDNA in terms of size and reliability. Indeed, I accept that cDNA would be preferable to genomic DNA. However, the arguments are not all one way. Professor Proudfoot explained that "having the whole gene present, or at least the gene with some of the introns, greatly enhances the final expression of that gene". With cDNA, of course, there would be no introns, and there would not be any splicing, and as Professor Proudfoot went on to explain:
"[I]n a number of cases... the actual process of splicing seems to enhance the production of messenger RNA on its transfer from the nucleus to the cytoplasm."
- In his written statement, Dr Lin said that obtaining HuEPO cDNA for the purpose of expressing EPO was not regarded as an urgent matter or an issue of priority by Amgen, and this was confirmed by Dr Browne in evidence. I see no reason to doubt that evidence. After all, Amgen's experiments as taught in the 605 patent were "delivering the goods" with genomic DNA.
- Furthermore, when he turned his attention to the matter, Dr Lin obtained a cDNA sequence which at least Amgen contend is HuEPO cDNA, known as Hu13, through the transfected COS cell route, in a matter of a few weeks. Roche contend that the Hu13 clone is not authentic cDNA in this context, because of the arrangement of certain nucleotide sequences within the clone. However, it appears to me that the essential point is that the Hu13 clone contains the whole of the natural mRNA primary transcript from the cap site at the 5'end down to the poly-A tail at the 3'end. It is only the arrangement of the sequences significantly upstream of the transcription sites which differ from the sequences in naturally occurring HuEPO cDNA.
- It is fair to say that it is not merely that the Hu13 clone is a little different from HuEPO cDNA which is "naturally occurring" (to use a word slightly inappropriate to cDNA); the different nucleotide sequences upstream of the transcription site actually gave rise to confusion in the minds of the Amgen scientists, and that confusion was only sorted out after consultation with an outside expert, Dr Suggs in early 1985.
- This aspect of the evidence relating to Hu13, upon which Roche rely, does underscore their point that, transfected COS cells are not always a reliable or predictable source of cDNA. However, it appears to me that the essential point is that the transfected COS cell route produced, after work which was routine and not lengthy, a gene which contained all the relevant necessary parts of human cDNA for the purposes of the teaching of the 605 patent. At the time, on 6th April 1984, Dr Lin specifically recorded that "Hu13 is a true human EPO clone". Of course, that is not conclusive, and he could be said to have been parti pris. Nonetheless, it is consistent with what Professor Brammar accepted as being "correct" in cross examination, namely that with the Hu13 clone "you have the whole thing at your disposal" which enables one to "engineer it to put stronger signals up front of the coding sequence" and to "conduct experiments on that to adjust expression".
- Professor Brammar also accepted that "if... your desire was to get a message that goes right the way back to the first initiation site, [the Hu13 clone] gives it to you" and that "you have the whole thing there". He also accepted that the Hu13 clone was not "incomplete" and that it would be a "false impression" if one thought that the Hu13 clone "was not complete". Professor Wall's evidence was to much the same effect. He said that his view that the Hu13 clone was a type of human cDNA was correct on the basis of his own experience and use of language, and, above all, in light of the fact that it was cDNA and it performed that function which was required of it under the patent, namely to effect human EPO expression when used in accordance with Examples 7 and 10.
- I agree with that view. If one was to encapsulate the purpose of the patent, it is to obtain the expression of human EPO in relatively substantial quantities by inserting exogenous DNA into a host cell. The identification of the DNA appropriate for this task is not in the context of an academic exercise. The Hu13 clone is cDNA, it contains all the coding for human EPO and in that sense it is human cDNA. Further and above all, it "delivers the goods", to quote Lord Hoffmann in Biogen [1997] RPC 1.
- Dr Fritsch, it is fair to record, had a different view. However, it appears to me that that view is somewhat at variance with Jacobs, a paper to which Dr Fritsch himself was party. In that paper a gene sequence including all the exons, but none of the introns, of human EPO, and only a relatively small part of the upstream sequence, and the whole of the downstream sequence, was referred to as "the isolated cDNA [which] encodes human erythropoeitin" and "erythropoeitin cDNA" and "erythropoeitin fetal liver cDNA". While none of those expressions are "human cDNA", I think the distinction is one without any relevant difference, not least because, as I have mentioned, the expression "human cDNA" is not in the patent.
Gene synthesis
- I now turn to the third method by which Amgen contend that the notional addressee would appreciate, through common general knowledge, that he could obtain human cDNA, namely gene synthesis. There was what I suspect was a somewhat arid argument between the parties as to whether a synthetic DNA sequence which was identical to the sequence of human cDNA obtained from tissue sources or through a transfected host cell could be called human cDNA. As with some of the expressions used in the claims of the 605 patent (such as "host cell" or "recombinant polypeptide"), I suspect that cDNA could be and was used to describe such synthetic DNA, at least by some people in some circumstances. The reason the point seems a little arid is that the question of what is meant by "human cDNA" in this context does not arise as a matter of the proper construction of that expression in the patent in suit. The issue really arises in relation to infringement, because, as I will explain when dealing with infringement, Roche's allegedly infringing product results from the expression of what is undoubtedly human cDNA.
- In my view, in so far as the point is relevant, the draftsman of 605 did not use the expression "cDNA" to extend to synthetic DNA. Thus, on page 3, the draftsman specifically defines "further "cDNA" methods" as applying to cDNA as the complement of mRNA in transfected host cells, and not to "the mechanical manufacture of a DNA sequence" (see at lines 12 to 16). This is confirmed by passages later on the same page. Further, as one goes through the first ten Examples in the patent, one sees references to cDNA which plainly do not extend to synthesised DNA (albeit that this is merely consistent with what is common ground, and does not exclude the possibility of "cDNA" also referring to synthesised DNA). However, Examples 11 and 12 deal with synthetic DNA, and it is not insignificant, to my mind, on this issue, that there is no reference to such DNA being thought of as cDNA. I will nonetheless proceed to refer to "synthetic cDNA", for the sake of simplicity.
- The fact that the reader of the patent, if he bothered to think about the point at all, would conclude that the draftsman did not treat cDNA as including synthetic DNA has little bearing on the question of whether a person wishing to use the teaching of the patent to make HuEPO would know that he could use synthetic cDNA for that purpose, and that this could be done without undue effort. The fact that the patent nowhere specifically suggests this course, even though no reference to cDNA in the patent extends to synthetic DNA, is of some relevance. However, the statement that "the chemical manufacture of a DNA sequence provides a code for a polypeptide of interest" is made early in the specification, as I have mentioned. Further, the notion of synthetic DNA sequences encoding for EPO, albeit not specifically human cDNA, is to be found in Example 11. However, Example 11 is concerned with DNA synthesis as a means for making sequences suitable for expression in E.coli cells and yeast cells.
- On the evidence, it seems clear that, as at the relevant time, even a leading expert in the field could not have synthesised human cDNA, in the sense of each nucleotide in the synthetic product being identical to its equivalent in the natural product (in so far as one can have "natural" human cDNA). Professor Gassen said that the manufacture of a coding region of a gene with more than 500 base pairs (which is what would be involved in the EPO cDNA gene) required changes to be made from natural sequences. The changes would be required partly to introduce or remove restriction sites to enable the sequence to be divided into sections, partly to avoid self hybridisation of oligonucleotides, and partly to minimise the risk of the oligonucleotides ligating together other than in the desired order to form the final overall cDNA sequence.
- Dr Gait, a leading figure in this field, identified two changes which would have to be made, although, if he had actually carried out the synthesis, it seems possible that a number of other changes either would have had to have been made or at least would have been desirable. In that sense, again for what it is worth, it can be said that, even if synthetic human cDNA is human cDNA for the purposes of 605, the synthetic cDNA which could have been made would not have been human cDNA precisely, although it would have been very close thereto. In terms of the nucleotide sequence, if human cDNA could have been synthesised, it would, as I see it, have about 99% homology with naturally occurring human cDNA.
- In his evidence, Dr Gait expressed confidence that it would have been possible in 1984 to synthesise human cDNA, albeit with a few changes to the bases for the reasons I have mentioned. He was an impressive witness; he was clearly very experienced in this field, and he spoke with authority. However, at least as at the relevant date, I am not satisfied that the notional reader of the patent, clothed with the requisite common general knowledge, would, or could without undue effort, have embarked on the synthetic route in order to obtain human cDNA.
- While, as I have mentioned, Dr Gait expressed confidence about the ability of experts in this particular field to synthesise human cDNA encoding for EPO at the relevant date, I cannot, in light of the evidence given, hold that it was more than a real possibility that this could have been achieved. Perhaps more importantly, had the appropriately skilled reader considered the possibility of expressing EPO through the indirect medium of synthesised human cDNA, he would not, in my view, have considered that as being more than a possible route, whose success was uncertain.
- When embarking on the exercise of synthesising genetic material, the first stage is to design the process, and the second stage is to try and implement it. The first stage involves a number of steps, the first of which is to divide the desired gene into appropriate sections. Dr Gait undertook that step. However, he did not undertake the next step, which involves considering whether, and if so how, the sequence could be made in practice, and in particular whether, and if so how, the sequence can be divided up into oligonucleotides which would not self hybridise, and which could be uniquely ligated together. As Dr Gait accepted, and as indeed published papers confirm, a so-called "gene construction map" is prepared before the second, practical, stage is undertaken (albeit that there will no doubt normally be amendments which have to be made to the design process which only come to light during the practical process).
- Professor Gassen, who was significantly less sanguine about the likelihood of success, and the relative ease, of synthesising human cDNA than Dr Gait, effectively challenged Dr Gait to produce a practicable gene construction map, and no such map was produced. Dr Gait had apparently undertaken an attempt to effect that process for the first of the three sections into which he had divided the EPO gene for the purpose of the notional synthesis, but there was no suggestion he had done this for the other two sections. As Mr Thorley points out, the evidence suggested that the design process should normally take less than a week. Although Professor Gassen did not condescend to particular problems which might be encountered in relation to the synthesis of human cDNA, it seems to me that it can fairly be said that, unless and until a specific proposal was put up to him, it would not have been easy for him to identify specific problems. Further, while this may be another way of putting the same point, the questions he was asked in cross-examination on this were not themselves particularly specific.
- Dr Gait estimated that it would take around twelve man months to make the sequence of more than 500 base pairs encoding for EPO using the techniques employed in his laboratory in 1984. First, there is no doubt that Dr Gait was one of the top experts in the world in this comparatively small field in 1984. Secondly, his estimate of twelve man months was mainly based on very limited specific experience; in fact, it was substantially on the basis of the synthesis of one protein, somatomedin C, and this succeeded first time. Not even Dr Gait, and certainly not the averagely skilled addressee, could have been remotely confident that the synthesis of human cDNA, if it was possible, would be successful first time round.
- I think there is something in Mr Waugh's point that Professor Gassen may have been somewhat more pessimistic than the notional addressee of the patent (assuming he considered the synthetic human cDNA route) and more pessimistic than the facts justified in 1984. Nonetheless, his experience in 1984 was that "the vast majority of the initial attempts" to "express [a synthetic] sequence in a host cell and hoped to produce a biologically active protein" would "be met with failure". He suggested that "the creation of the synthetic sequence" could "in some cases [be] up to a year's work". To my mind, although a highly experienced and reputable scientist, Professor Gassen was, on this issue, closer to the ordinarily skilled person than was Dr Gait.
- It is right to mention that there is contemporary evidence that Dr Fritsch of GI prepared a note in 1984 suggesting that he thought that synthesising human cDNA would not be too arduous a task. However, he was not particularly knowledgeable about DNA synthesis, and I think it would be wrong in any event to make much of his note in this connection.
Conclusion
- In these circumstances, I conclude that, while the obtaining of human cDNA through human tissues or by means of synthesis would not have been within the common general knowledge of the notional addressee of 605, the obtaining of such DNA through the route of a transfected host cell (and in particular a CHO cell, but also a COS cell) would have been within the common general knowledge.
- I draw some comfort in reaching this conclusion from what Dr Fritsch of GI (a Roche party) wrote in January 1984, when he heard what Dr Lin had achieved: "[W]ith this information they can certainly predict the human cDNA sequence and synthesise it easily".
I. CLASSIC INSUFFICIENCY: OTHER ISSUES BEARING ON CLAIM 1
The Examples: Introductory
- As Mr Thorley points out, as there was no deposit of cells, the skilled person wishing to put the teaching of 605 into effect had to carry out Example 4, then Example 7, and then Example 10. He helpfully summarises the five steps which would be involved, as follows:
1. Obtain the Lawn library;
2. Screen it with probes and isolate and clone the EPO gene (Example4);
3. Construct the expression vector and insert the EPO genomic fragment (Example 7);
4. Obtain the DuX-B11 CHO cells and transfect them (Example 10);
5. Amplify those cells using MTX (Example 10).
- It is not suggested on behalf of Roche that step 2 suffers from any insufficiency; Mr Thorley fairly describes the exercises in Example 4 as "straight forward in the sense that they all involved well known procedures". As to steps 4 and 5, relating to Example 10, Roche do not contend for any classic insufficiency there either.
- I shall deal with the alleged insufficiencies in relation to steps 1 and 3.
Example 4: The Lawn library
- So far as step 1 is concerned, Roche's case is that the Lawn library was not available as at the priority date, that it would therefore have to have been made by the notional addressee, but this was technically demanding and time consuming, and that therefore the first step alone involved undue effort.
- In my judgment, this ground of alleged insufficiency is not made out. The way in which to prepare a Lawn library was set out in detail in Chapter 9 of the Maniatis Manual, which extended to over 25 pages, and contained clear directions. Professor Wall estimated that it would take no more than a couple of months to carry out this work, and Professor Brammar agreed. Further, an advantage for the person following the teaching of the patent making his own Lawn library would be that, with the benefit of the disclosures of the 605 patent, and with knowledge of where the EPO gene is to be found in the genome, he could significantly enhance the frequency with which the EPO gene was included. Although I think Mr Thorley is right to say that the construction of a Lawn library would not have been "trivial or quick", it seems to me that it would have involved routine work which would not have taken an undue amount of time, even bearing in mind that it was the first of five steps.
- It is also argued by Mr Waugh on behalf of Amgen that the notional addressee would have been able to obtain a sample of the Lawn library from someone who himself had obtained a sample of the original library from Dr Lawn. In this connection, I believe that the evidence establishes that the original Lawn library had been distributed fairly widely, and indeed had been deposited with the ATCC. Professor Wall was not challenged on his view that he "would expect that a sample would still be available from one of the recipients of the [original] library" and that he thought it would be "a simple matter for an initial recipient to grow up some of their material to provide it to others".
- Quite apart from this, it was common ground between Professor Wall and Professor Brammar that many laboratories had, in 1984 "constructed their own genomic libraries". This means that there was, at the very least, a good prospect of the notional addressee either having such a library in his laboratory, or having indirect access to such a library through a colleague in another laboratory. Accordingly, I see no problem so far as the sufficiency of step 1 is concerned.
Example 7: Screening, isolating and probing
- A number of insufficiencies are alleged by Roche, primarily on the basis of Professor Brammar's evidence, so far as step 3 is concerned. Of course, given that Example 7 includes two Examples, namely 7A and 7B, it may well be that if one of those was insufficient, the patent would still not fail for insufficiency on this ground, provided the other was sufficient. However, I do not need to decide that point, because I think that, although Professor Brammar's criticisms were justified, in the sense that it would have been better if the 605 patent had been drafted so as to avoid them, I do not consider that any of them, even if they are taken together, render Example 7 or Example 7A non-enabling. I turn first to the specific criticisms of Example 7A by Professor Brammar.
- The first criticism of Example 7A arises out of Example 6. It is the absence of a restriction site immediately adjacent to the particular nucleotide, numbered 2448, in the gene. In 1983, it was known that there were certain "poison sequences" which might exist upstream of the encoding area of the gene, and which it would be desirable to remove before transfecting the gene into COS cells. The 605 patent teaches that fragments from 2448 to 4362 are poison sequences, and the cleavage site at 4362 was provided by a particular restriction enzyme commercially available and widely used in 1983. However, there was no restriction site at 2448. Professor Wall said that:
"The skilled addressee would appreciate that the specific terminus of this fragment did not have to be exactly 2448; any fragment within 50-100 bases would most likely have provided equivalent results."
- He then identified possible restriction sites which could have been utilised for this purpose, and also mentioned restriction enzymes which would have cleaved at those sites. I agree with his conclusion that "once the patent taught the key elements of the fragment, and its length, it was a routine matter in 1983 to generate either an identical or nearly identical fragment, or a functional equivalent." Indeed, as I understood him, Professor Brammar accepted that view in cross-examination.
- The second criticism of Example 7A is that the sequence of a linker to be attached to the end of the sequence at 2448 (or, in light of the criticism I have just been considering, a slightly different nucleotide) is not provided. Connected with this criticism is the further criticism that directions were not given as to how to attach a linker. However, Professor Brammar accepted that one "could readily go and get [the appropriate] linker from a catalogue" and that this would be "standard routine". As Mr Waugh points out, it would appear that the only information required for a linker is a restriction site which it provides, and this is readily available from the type of catalogue referred to. Furthermore, I was referred to publications in journals including Nature and the Proceedings of the National Academy of Sciences which report expression experiments involving linkers, without giving the sequences of those linkers (see for instance Powell et al PNAS 83: 6465, in 1986 and Mishina et al in Nature 307:604 in 1984).
- Professor Brammar also expressed some doubt about the identity of a particular fragment, but agreed this was "a very minor" matter. I agree, and do not consider that it begins to justify a challenge to the sufficiency of Example 7A.
- Example 7A identifies a small fragment comprising nucleotides from "2553" to "2770". Professor Brammar expressed the view that "2553" is a mistake, because it is neither on or near a convenient restriction enzyme target site, and he assumes that "2533" was intended. In cross-examination, he explained that he had "[drawn] attention to it simply because it is an issue and it is the kind of error that some beginner might fall into". He also described it as a "fairly minor matter". I would accept that the mere fact that a mistake is a misprint, even, at least conceivably self-evidently a misprint, does not automatically mean that it cannot amount to an insufficiency. However, where the mistake is self-evidently a misprint which only might mislead a beginner, that cannot amount to an insufficiency. As I have mentioned, the notionally skilled man is not the lowest common denominator in the field; even less is he merely a beginner. As I understood Professor Brammar, he thought that any problem thrown up by the misprint should not have taken the ordinarily skilled person more than 24 hours to sort out.
- The next criticism of Example 7A is the lack of information as to how two of the DNA fragments would be ligated. It is clear that this would be achieved through the medium of a suitable linker, and it seems to me that the failure to identify the residual sequence of the linker between the promoter and the inserted DNA is not relevant so far as the construction of a suitable vector is concerned. In this connection, Professor Brammar did "not know of a specific problem", and said he was merely identifying "a potential risk" which he accepted amounted to a "minor criticism".
- The final matter I should mention in relation to Example 7A is the "worry" expressed by Professor Brammar because of the transcription which apparently took place from the SV40 late promoter upstream. I am slightly puzzled by this point, in the sense that Professor Brammar's concern appears to have been based on the proposition that the skilled person might not follow the teaching of Example 7A, and that he might fear that this would have undesirable consequences. In the first place, it seems rather odd to found an insufficiency in a patent on an assumption that if the notional reader would not follow its teaching. Secondly, it seems a little difficult to found an insufficiency on a double uncertainty, namely what the skilled person might do, and what he would fear it might lead to. There is no evidence that, if the skilled person did not follow the teaching of Example 7A precisely, and instead did what Professor Brammar thought he "might" do, that would have resulted in any problem. Quite apart from this, if such a problem had arisen, Professor Brammar accepted that the skilled person could simply "redesign". At worst, and assuming everything in favour of a problem in this connection, Professor Brammar accepted that the construction of an alternative expression vector would involve only "a few weeks".
- I turn then to the additional criticism advanced by Professor Brammar of Example 7B. He raised a number of criticisms similar or identical to those I have already considered and rejected in relation to Example 7A, and I do not propose to repeat my reasons for rejecting essentially similar or identical points that I have already disposed of. However, it is right to mention that, even if all the criticisms were justified, Professor Brammar accepted that "a few weeks, several weeks, but not much longer" would be involved if one was "left to design [one's] own version of the expression construct using COS cells".
Example 7B: The mistake in Table VI
- The only separate criticism of Example 7B which I ought to consider arises from a small, but potentially significant, inaccuracy in the sequence information contained in Table VI of 605. At position -44(i.e. 44 bases upstream of the five prime end of the first exon) Table VI shows a G instead of a C. Professor Brammar explained this would lead to the design of a linker with a mismatch. Based on his own view, supported to some extent by a paper published in 1987 by Wiaderkiewicz and Ruiz-Carillo in Nucleic Acid Research 15(19): 7831, he suggested this mistake would lead to a "C/C mismatch ligation" which was inefficient, and quite likely to lead to a failure to ligate, so that the skilled man would have to solve the problem himself.
- While accepting that there was a problem, Dr Gait thought that the mismatch would be tolerated to the extent that there would be significant ligation. In this connection, it is to be noted that in his "summary of the mismatches tested" Wiaderkiewicz stated that, at least as a generality, there would be ligation in the case of a C/C mismatch. The evidence, essentially based on this paper, suggests to me that there would be ligation, but the existence of the C/C mismatch would mean that the amount of ligation would be reduced. I accept Dr Gait's evidence that one "would still have expected a reasonable number of clones", in the sense that there would have been a sufficient degree of cloning for the purpose of achieving the end and purpose of Example 7B. Although Professor Brammar, an unusually careful witness, took a slightly more cautious view than Dr Gait, I do not think that the overall effect of his evidence called into question the conclusion expressed by Dr Gait which I have quoted and explained above.
Expression in human cells
- Although Professor Proudfoot appeared to suggest that Roche might be contending that 605 did not extend to expression of EPO in human cells, I do not understand the point to be advanced by Mr Thorley. In my view, there is no reason why a human cell is not a "eukaryotic cell" within the meaning of Claim 1 or a "host cell" within the meaning of Claim 19. I think it would require a pretty clear indication to the contrary in the specification before such general expressions were held not to apply to human cells. There is nothing in the specification to indicate that human cells were intended to be excluded. On the contrary, on page 48, lines 33 to 35, it is stated that "expression systems involving vectors of homogenous origins applied to a variety of ... mammalian cells in culture" where "[c]onspicuously comprehended" "within the contemplation of the invention". Plainly, "mammalian cells" is a term which includes human cells.
- TKT contend, however, that the 605 patent is insufficient in this connection. In the course of his attractive submissions, Mr Kitchin points out that the draftsman of the patent identifies early on the desirability of obtaining EPO "in quantity" to enable the treatment of over 1.5m patients "in the United States alone" (page 5, lines 18 to 19) and that:
"the best prospects for... providing large quantities of [mammalian EPO] for potential diagnostic and clinical use involve successful application of recombinant procedures to effect large scale microbial synthesis of the compound" (page 6, lines 54 to 56).
- He draws attention to the fact that, consistent with this, the draftsman claims at page 9, lines 32 to 35, that:
"Vertebrate (e.g. COS-1 ... and CHO) cells provided by the present invention comprise the first cells ever available which can be propagated in vitro continuously and which upon growth in culture are capable of producing in the medium of their growth in excess of a 100U (preferably in excess of 500U...) of erythropoeitin per 106 cells in 48 hours as determined by [RIA]."
- In fact, however, once one looks at the teaching of the patent, the only way of getting EPO in greater quantities which is specifically taught, is by the use of DHFR- CHO cells. In particular, Mr Kitchin contends that there is no teaching as to how to achieve the result in human cells. Accordingly, although the essential point which TKT wish to establish is that the claims are insufficient in so far as they seek to extend to expression of EPO in human cells, their argument amounts to contending that, subject to any question of common general knowledge, any claim which extends beyond DHFR- CHO cells is invalid on the grounds of insufficiency. In this connection, there is obvious force in the contention that an amplified level of expression is of the essence of the patent, given that EPO was known to exist in nature, and had, indeed, been isolated, albeit in very small quantities, from urine.
- In his evidence, Professor Proudfoot pointed out that the only Experiment in the patent which produced EPO at the level promised on page 9 of the 605 patent was in amplified DHFR- CHO cells. In particular, Amgen failed in unamplified CHO cells, in COS cells and in human cells. He also stated that, although subsequent technology has enabled EPO to be produced in a DHFR+ cell, the patent provided no way of amplifying such a cell, and that the consequent requirement that the cell be DHFR- at the date of the patent, limited the technique described in the patent to CHO cells, because there were no other generally available DHFR- cells at the time (albeit that the position subsequently changed). In effect, therefore, he stated that, even if the 605 patent enabled the production of EPO in human cells, it did not enable it in amplified or enhanced quantities.
- So far as using the information in the 605 patent to obtain some expression of EPO in human cells is concerned, Professor Wall relied on the fact that, as at the relevant date, use of certain human cells for that type of technique was well known. Those cells were the so-called HeLa and 293 human cells. He therefore said that it would have been within the common general knowledge that the disclosure of the 605 patent would have enabled the expression of EPO in such cells. I agree. His view is supported by the fact that scientists employed by Amgen used the technology described in the 605 patent, including a specific plasmid therein specified (the so-called pUC8-HuE plasmid - see page 27) to achieve the expression of rEPO, apparently indistinguishable from uEPO (on SDS-PAGE with RIA testing) in 293 cells. He accepted that this work did not involve "high level expression", but suggested that procedures to achieve that would have been "routine and standard practice".
- However, I do not consider that, subject to one piece of evidence he produced, Professor Wall made that last statement good or called Professor Proudfoot's view as to the enablement of the expression of enhanced quantities of EPO into question. Thus, he agreed that the reader of the patent at the relevant date would understand that the identification of stable transformants and amplification was essential to achieving the higher levels of expression promised, that DHFR was the only contemporary amplification marker available, at least as far as he could recollect, and that CHO cells were the only known DHFR- cells at the time. He also agreed that Amgen's experiments with human cells, which were with one type of such cell, namely the so-called human 293 cell line, did not produce high level expression.
- The one piece of evidence which Professor Wall did produce, was a paper by Murray et al. published in 1983 in Molecular and Cellular Biology 3(1):32 ("Murray"), before the relevant date, which described the use of certain DHFR+ cells for amplification. However, I do not consider that that assists Amgen on the issue of enhanced or amplified expression in cells other than CHO cells. First, I do not consider that Murray was part of the common general knowledge. Professor Wall, who raised it, only produced it in his evidence in reply, and he had not come across it until it he had been instructed in connection with these proceedings. Further, Professor Wall accepted that the amplified genetic material reported in Murray would be unstable and could only be maintained by continued MTX treatment. Accordingly; in my view, even if the notional addressee of 605 would have thought of Murray, he would have dismissed it as unreliable.
- It therefore follows that, subject to reconsidering this aspect in light of the guidance given by the House of Lords in Biogen [1997] RPC 1, I consider that this ground of classic insufficiency is made out, so far as enhanced or amplified expression of EPO in human cells is concerned.
"Stringent conditions"
- Roche (with the support of TKT) contend that whether or not certain conditions were investigating whether hybridisation is occurring are "stringent" in Claim 1(b) is an uncertain matter upon which views could differ. The point is perhaps well illustrated by considering an addressee who is seeking to discover whether a certain polynucleotide sequence (not falling within integer (a) of Claim1) falls within Claim 1(b). He would seek to investigate whether the sequence hybridises in accordance with integer (b), and he may test it at, say, three levels of stringency, which I will call (in ascending order) levels 1, 2 and 3. At level 1, which he believes to be low stringency, it does hybridise; at level 3, which he believes to be high stringency, it does not hybridise; at level 2, which some might regard as high stringency but others might not, it does hybridise. In those circumstances, if level 2 is "stringent", then his sequence would appear to fall within Claim 1(b), but if it is not, it would not appear to do so. Roche contends that there would clearly be circumstances in which such an addressee could not by any means always be confident whether a particular hybridisation level was "stringent" and, therefore, that Claim 1, or in the alternative integer (b), is insufficient.
- It seems clear that it is and was common to identify in specific terms the nature, and hence the stringency, of the conditions of hybridisation. Thus, in Example 3B on page 14, lines 48-54 of the 605 patent, the essential features of the conditions to enable one to determine the degree of stringency of the hybridisation and washing conditions, namely the identity and concentration of the salt, and the temperature and the time for hybridisation, are specifically defined. Similarly specific identification of the hybridisation conditions are to be found in Example 4 (page 19, lines 9 to 14). Further, on page 10 (lines 10 to 11), where the draftsman is summarising the DNA sequences to which the patent extends, he identifies in paragraph (a) the sequences set out in Tables V and VI - reflecting Claim 1(a) - and then refers in paragraph (b) to:
"DNA sequences which hybridise (under hybridisation conditions such as illustrated herein or more stringent conditions)."
The passage immediately goes on in paragraph (c) to refer to DNA sequences which, in effect, fall within integer (c) of Claim 1.
- In these circumstances, the argument that the reference to "stringent conditions", without any guidance as to what that expression means, is insufficient has obvious force. However, I have come to the conclusion that the argument should be rejected. The Courts have demonstrated a degree of reluctance to hold a claim in a patent insufficient merely on the ground that it purports to define its extent by reference to a qualitative requirement, as opposed to a quantitative one. Thus, in British Thomson-Houston Co Ltd -v- Corona Lampworks Ltd 39 RPC 49, the House of Lords unanimously held a patent relating to an incandescent electric lamp valid despite an argument that the stipulation that the filament be "of large diameter or cross section" was insufficient. Lord Dunedin said this at 39 RPC 84, lines 4 to 16:
"The criticism is made, that in specifying the diameter of the filament, "large" is not defined. Now, when any epithet is used which has inherent in it the notion of comparison, and when no particular standard with which to compare is given, it is, I think, consequential that the standard which... is employed is the average. The average thickness of filaments in vacuum lamps is well known. I think the Patentees meant anything larger than that. ....A further definition of "large" does not seem to me to be necessary."
- In Cleveland Graphite Bronze Co -v- Glacier Metal Co Ltd 67 RPC 149, the House of Lords held that a claim referring to "a bearing liner of thin and flexible metal" was valid. Lord Normand, with whom all but one member of the House of Lords agreed, gave this analogy at 67 RPC 153, lines 27 to 31:
"A shoemaker who is ordered to make a pair of shoes in thin and flexible leather knows that the customer wants thin and flexible shoes. It is true... that shoes which are thin and flexible for a grown man may be thick and stiff for a young child. But the intelligent shoemaker would appreciate this and would choose his leather accordingly."
- Of course, the mere fact that the House of Lords has, on two separate occasions, held a claim in a patent to be sufficient, notwithstanding the fact that the claim has used an arguably vague adjectival test does not mean that in every case where such a test is contained in a claim the Court will reject an insufficiency argument. Ultimately, whether or not a claim is classically insufficient must be a question of fact. Nonetheless, it would be wrong not to consider the instant argument without bearing in mind the approach of these two well established and authoritative decisions.
- In the present case, it appears to me that, unlike in the two cases to which I have referred, there is some guidance as to the meaning of "stringent" in the patent itself. I have already referred to the passage on page 10 of the patent. It includes three categories of DNA which appear pretty clearly to "track" integers (a), (b) and (c) of Claim 1. That point is reinforced by the fact that the three categories of DNA sequences identified on page 10 are introduced by the words "Novel DNA sequences of the invention include all sequences used in securing expression [of EPO or EPO-like products] which are comprehended by...". In other words, the passage on page 10 has the same informal purpose as the formal function of Claim 1, "conditions such as illustrated herein or more stringent conditions" and it appears to me to tie in fairly clearly with integer (b) of Claim 1. The "hybridisation conditions... illustrated herein" are the specific conditions identified in Examples 3(b) and 4. The "more stringent conditions" are the sort of stringent conditions which integer (b) of Claim 1 has in mind, and therefore guidance is given within the patent itself to the reader.
- As I see it, a provision such as integer (b) of Claim 1 has to be read in the context of the patent as a whole, and, while it may not be possible to allow the rest of the patent in effect to contradict or even qualify the plain words of a claim, it seems to me that it must be permissible to invoke a relevant part of the specification as an aid to the construction of an otherwise ambiguous or unclear aspect of a claim. It is particularly powerful in a case such as the present, where, if the passage on page 10 does not assist, then the relevant part of the claim is said to be insufficient. In such a case, as I have already mentioned, the court should, to my mind at any rate, plainly lean in favour of a construction which "saves" the document or a particular provision therein. In my view, that is how a reasonable addressee would understand integer (b) of Claim 1. He would also see that Claim 1 included the additional requirement of being EPO-producing (or producing protein with EPO-like characteristics).
- It is right to mention that Professor Wall dealt with this issue in his first report and in cross examination. He referred to the Maniatis manual, which he explained would be within the common general knowledge, as is accepted by Roche and TKT. In Chapter 7, entitled "Synthesis and Cloning of cDNA", the manual, under the heading of "Hybridization of Southern Filters" at pages 387 to 388 specifies hybridisation and washing conditions and states that "if the homology between the probe and the DNA... is inexact, the washing should be carried out under less stringent conditions". Professor Wall said that the concept of stringent condition "would... not be something that would be a great mystery to a person of skill in the art". I accept that evidence, although I also accept the argument that there would still be conditions which some might regard as "stringent" and others would not (or, to put it another way, which the reader would be uncertain whether they were "stringent" or not). However, I think these conditions would be a comparatively small band on the spectrum of stringency.
- In order to dispel the suggestion that there was any such possibility of uncertainty, Amgen rely on the further evidence of Professor Wall, to effect that there would be unique optimum conditions for each probe, and that the strength of hybridisation between two pieces of DNA would depend upon the length, area and degree of complementarity as well as upon the GC, as opposed to AT, content. I was unimpressed by this suggestion which amounts to saying that, because a person skilled in this field would normally optimise the conditions for any given system, he would know if he had achieved "stringent" conditions by looking at the result. The point, as I understood it, was that one would have achieved "stringent" conditions if the probe hybridised and was not washed off. However, I agree with Mr Thorley, who suggests that this simply defines "stringent" by reference to the conditions in which the probe hybridises, and, as the whole point of the test in Claim 1(b) is to see if there is hybridisation under stringent conditions, the exercise proposed by Professor Wall is self-fulfilling or circular.
- I am thus of the view that integer (b) of Claim 1 does not suffer from insufficiency. Even with the specific guidance on the specification and from items with common general knowledge, such as the Maniatis manual, I accept that there could be hybridisation conditions which might or might not be "stringent" within the meaning of Claim 1(b). It is nonetheless the case, in light of the evidence I heard and the contents of the specification and common general knowledge that the great majority of hybridisation conditions will either be of such a stringency that no reasonable reader of the 605 patent would be in any doubt that they were not "stringent", or will be such that the reader could not reasonably doubt but that they were "stringent". In those circumstances, it does not appear to me that Claim 1(b) could properly be said to be insufficient.
- If a particular piece of DNA hybridised in conditions which were on any reasonable view "stringent", then it would fall within Claim 1(b); if it only hybridised in conditions which on any reasonable view were not "stringent", then it would plainly not infringe. The only problem which could arise would be where a sequence hybridised in conditions which were plainly not stringent, did not hybridise in conditions which were plainly stringent, but did hybridise in conditions which were in the middle and uncertain area of "stringency", namely which some might reasonably regard as stringent, while others might not. It would only be in such a case that Claim 1(b) would be unclear, and, in such a case, the issue of infringement would clearly have to be resolved in favour of the alleged infringer and against the patentee. I believe that view is consistent with common-sense. I also believe that it is consistent with, indeed a fortiori, the reasoning of the Court of Appeal in Scanvaegt International A/S -v- Pelcombe Ltd [1998] FSR 786 where at 797 Aldous LJ said that "Ambiguity is no longer a ground of invalidity". He then went on to say:
"[L]ack of clarity is no longer a matter that it can result in a patent being invalid, it cannot result in the patentee being unable to establish infringement. If you cannot define the invention claimed, you cannot conclude that it is being used."
- I think that my view is probably also consistent with the observation of the Court of Appeal in General Tire at [1972] RPC 511 about "puzzles set at the edge of the claim". Indeed it also ties in with what I have referred to as "fuzzy edges" which are encountered in science and technology. Such puzzles or fuzzy edges are inevitable or inherent on occasions. It would be unrealistic and disproportionate if they invalidated a claim in a patent, even in circumstances where they could be avoided. It would be inconsistent with fair protection to the patentee; it would not be called for by certainty for third parties. If the alleged infringement fell within the fuzzy edges then certainty for third parties would require a conclusion of non-infringement; again, fair protection for the patentee would not suggest otherwise.
- The conclusion that, even if it is incapable of being tied into the reference to stringent conditions on page 10 of the patent, the reference to stringent conditions does not invalidate Claim 1 in terms of classic insufficiency, is consistent with the approach of the Board. In T301/87 Alpha-Interferon/Biogen OJ EPO 1990, 335, the Board said this:
"Such macromolecular precursors may in appropriate cases be defined as a class by the characteristics of the end products they relate to and by some structural characteristics such as similarity based on capability of hybridisation with available structures, without necessarily creating uncertainty" (paragraph 4.5).
- As the Board went out to point out in paragraph 4.7:
"Unless claims with such functional connotations are allowable no worthwhile protection is provided against a third party which faithfully repeats the process of the patent and obtains new but equally useful variants of the invention."
- In The European Patent Office's Case Law on the Patentability of Biotechnology Inventions Dr Jaenichen, after discussing this case at pages 81 to 83, suggests that a claim in the following form would be allowable:
"A DNA sequence hybridising to the DNA sequence of Claim X and encoding polypeptide having the biological activity Z."
However, it is right to add that, on the same page, the author goes on to suggest that "in order to be on the safe side" it would be sensible to add something along these lines;
""Hybridisation" refers to conventional hybridisation conditions, preferably to hybridisation conditions under which Tm value is between Tm... to Tm... Most preferably, the term "hybridisation" refers to stringent hybridisation conditions."
- This latter passage indicates how a patentee can get himself greater protection on the insufficiency issue. However, even bearing that in mind, it appears to me to cut both ways. Although Roche can fairly point out that it does require identification of the hybridisation conditions, it is nonetheless the case that the term "stringent hybridisation" is included in this latter passage without giving specific guidance.
- Of course, as my conclusion as to the effectiveness of product-by-process claims where the process is novel in this very case emphasises, I am not bound by decisions, or even the approach, of the Board. Nonetheless, as has been said in a number of cases, the Board's decisions at least deserve respect. Furthermore, I note that it is not unusual to make claims to DNA in biotechnology patents by reference to those sequences which hybridise with the specific sequences claimed (or their complements) "under stringent conditions", or some other expression. Indeed, I was referred to a number of United States patents granted to GI substantially in such terms. I also note that Claim 11 in Chiron which included the expression "selectively hybridisable" was not rejected for lack of clarity or insufficiency, but because of the lack of any limitation to sequences having the appropriate biological activity.
- It is also right in this connection to refer to my decision in Novo Nordisk A/S -v- DSM NV (unreported 21st December 2000), which was said at least to cast doubt on the view that the reference to "stringent conditions" was sufficient. That case does not cause me to change my view. The essential problem for the patentee in Novo Nordisk was that the relevant passage in the claim referred to "DNA sequences hybridising at low stringency conditions" and to washing conditions which were of lower stringency than the hybridisation conditions. The patentee contended, in effect, that this was not to be construed literally, and that the addressee would increase the stringency of the conditions as a matter of routine. The effect of my decision on that issue was that the absence of any teaching to that effect, and the inevitable uncertainties and delays which it involved, resulted in the claim concerned being insufficient. It is also right to say that a number of points were made on behalf of the patentee in the present case in relation to this issue which were not made on behalf of the patentee in Novo Nordisk (and I do not mean to imply any criticism of the way in which the case was presented in Novo Nordisk: the patentee would probably not have been assisted by any of the additional points made on behalf of Amgen in relation to this issue in the present case).
- Accordingly, I reject the contention that Claim 1(b) is insufficient on this ground. This conclusion applies equally, and for the same reasons, to the incorporated reference to stringent conditions in Claim 1(c).
J. CLASSIC INSUFFICIENCY: CLAIMS 19 AND 20
Introductory
- Roche and TKT contend that that aspect of Claim 19 which stipulates that, in order to be within the Claim, recombinant EPO ("rEPO") has to have "higher molecular weight by SDS-PAGE from erythropoeitin isolated from urinary sources" ("uEPO") suffers from insufficiency. When considering the construction of Claim 19, I reached the conclusions that:
1. The closing words of Claim 19 require that there was a detectable overall difference between the performance of the two types of EPO on SDS-PAGE;
2. The uEPO need only be isolated to the extent necessary to carry out the SDS-PAGE exercise: in particular, it need not be completely pure;
3. The uEPO need not be from a pooled source;
4. The uEPO could be isolated in accordance with any technique referred to in the patent (or within the common general knowledge at the relevant time) with or without any workshop modification.
- The same argument is mounted against Claim 20 which incorporated Claim 19. Additionally, Claim 20 is said to be insufficient in relation to its reference to differential glycosylation.
Claim 19: Amgen's contemporary experiments
- Not surprisingly, scientists at Amgen had carried out a number of experiments testing the performance of a number of different batches of rEPO against a number of different batches of uEPO, on SDS-PAGE. The individual who carried out the bulk of this work during the 1984 and 1985 was Dr Joan Egrie, and I saw extracts from her contemporary entries in laboratory notebooks, which included results in the form of photocopies and/or originals of the SDS-PAGE experiments, including blotting and RIAs. This work was the subject of three publications in the name of Egrie and others (Experimental approaches for the study of Hemoglobin Switching p339, Tenth Annual Frederic Stohlman Symposium, and Immunobiol. 172:313) and two other published papers, one published in 1986 in the name of Browne and others (Cold Spring Harbor Symposia on Quatantitative Biology Vol. LI, page 693) and the other in the name of Eschbach and others in 1987 in New England Journal of Medicine 316(2):73.
- Amgen contends that one can draw adverse inferences from the fact that TKT and Roche rely on these experiments, including what was said about them in Dr Egrie's notebooks and in the published papers, rather than carrying out the same or similar experiments themselves for the purpose of these proceedings. I reject that contention. As has often been observed, the carrying out of experiments for the purpose of legal proceedings can be expensive and time consuming. If one party is content to rely upon experiments carried out by or on behalf of another party (whether or not for the purpose of the proceedings) then it seems to me that the court should, if anything, deprecate the idea that he should incur further cost and time in repeating, or seeking to repeat, those experiments. Relying on an experiment carried out on behalf of another party has a benefit and a risk. The benefit is that, while one will normally not be able to rely upon the experiment as an admission in technical legal terms, the fact that it was carried out by the other party may well make it more difficult for the other party to impugn it. The risk of relying upon the other party's experiments is that the other party may know of special facts, or may be able to call one of the people involved in carrying out the experiment to cast doubt, or worse, upon it. From the point of view of the court, a potential advantage of an experiment carried out by or on behalf of one party, but not in the context of the proceedings, is that the obvious ground for caution when considering the nature or the alleged result of the experiment (namely the desire to succeed in the proceedings) does not exist. However, there may well be other factors which could influence the perspective of the person carrying out and analysing experiments not in the context of proceedings.
- So far as the result of Amgen's experiments in 1984 and 1985 are concerned, I consider that, at the risk of appearing arrogant, the results were tolerably clear. I express myself in this slightly diffident way, because a significant proportion of the expert evidence, and, albeit to a lesser degree, of the argument, was given over to this topic.
- Amgen's work involved three types of urinary EPO and two types of recombinant EPO. The three types of uEPO were:
1. "Goldwasser uEPO", which was uEPO isolated from pooled urinary sources by Dr Goldwasser in accordance with the teaching of Miyake;
2. "Lot 82 uEPO" which was uEPO isolated substantially in accordance with the teaching of Miyake, but in respect of which there was a single source (i.e. the urine all came from one patient);
3. Alpha Therapeutics uEPO, which was uEPO from urinary sources, by a method of isolation which was not specified.
The two types of recombinant EPO used in Amgen's experiments were expressed substantially in accordance with the teaching of 605, in COS and CHO cells respectively (i.e. "COS rEPO" and "CHO rEPO").
- Dr Egrie ran a number of comparative SDS-PAGE experiments. She ran various urinary EPOs against each other, most notably Lot 82 uEPO against Goldwasser uEPO on a number of occasions, and also Alpha Therapeutic uEPO against Goldwasser uEPO. She also ran CHO rEPO against Lot 82 uEPO on at least three occasions and COS rEPO against Goldwasser uEPO on at least three occasions.
- The results of the Amgen experiments, at least as interpreted by Dr Egrie at the time, were:
1. Lot 82 uEPO had a higher apparent molecular weight than Goldwasser uEPO;
2. COS rEPO had the same apparent molecular weight as Goldwasser uEPO;
3. CHO rEPO had the same apparent molecular weight as Lot 82 uEPO;
4. Alpha Therapeutics uEPO had the same apparent molecular weight as Lot 82 uEPO.
- In my judgment, Dr Egrie's views, as I have just summarised them, were correct. Professor Cummings (who did not agree with Dr Egrie) and Professor Matsudaira (who did) both agreed that Dr Egrie was an experienced and careful scientist, and her contemporary conclusions are expressed in confident terms in her notebooks. These views were repeated in the articles to which I have referred, and those articles were not written by Dr Egrie alone, but included more senior scientists working for Amgen. Most of the articles were in prestigious journals or books, and they were all peer-reviewed. Each of the papers contained an unambiguous statement to the effect that one or other of the two types of rEPO migrated on SDS-PAGE effectively identically to one of the types of uEPO. In addition, in its own submissions to the FDA, Amgen stated that "the r-HuEPO migrates identically to the pure urinary EPO".
- In the end, of course, the decision as to whether Dr Egrie's conclusions were correct, must depend upon the evidence as a whole. In that connection, I have seen the results of the SDS-PAGE experiments, which are sometimes photographs of the results after blotting (normally with antibodies), sometimes the rather faded originals after blotting, and sometimes the autoradiograms or photocopies of them. Further, there has been the expert evidence as to how to interpret these results. Professor Cummings made a number of criticisms of the conclusions reached by Dr Egrie. I am bound to say that he appeared to regard this to be that of trying to think up as many conceivable uncertainties or problems with Dr Egrie's experiments and conclusions, so as to enable Amgen to contend that, viewed as a whole, her experiments could not be relied on to draw any of the conclusions which she identified. As an exercise in advocacy, it was moderately impressive, particularly bearing in mind the clear contemporary conclusions reached by Dr Egrie and, apparently, so many other experienced scientists employed by Amgen. However, as an example of the proper approach of an expert witness, it left much to be desired. It would be inappropriate for me to deal with every aspect of his evidence on this aspect, and I hope that a couple of examples suffice.
- First, Professor Cummings suggested that Dr Egrie's conclusion that Lot 82 uEPO and Alpha Therapeutics uEPO showed a higher apparent molecular weight than Goldwasser uEPO was based upon possible differences in the degree of iodination between the two samples (the iodination being the attachment of radioactive iodine to the protein so that it "shows up" on an autoradiogram). While that, of course, could be a possible explanation in theory for the difference in performance, it is clear from Dr Egrie's own notes that she had compared the X-ray film of the iodinated uEPO with un-iodinated uEPO by superimposing the autoradiogram over the Western blot. She then specifically recorded that "this comparison... definitively shows the difference in migration [between Goldwasser EPO and Lot 82 EPO]". Indeed, Professor Cummings effectively accepted that Dr Egrie had carried out "a sensible control", but said, without giving any reason (as Mr Kitchin points out), that "she may not have done it correctly". The notion that differential iodination could have been the cause of the difference in the performance of the two types of uEPO is further called into question by the fact that the two types demonstrated the same apparent molecular weight on SDS-PAGE after they had been deglycosylated: iodination involves attachment of iodine to the protein backbone, and not to the glycans. Accordingly that test confirms Dr Egrie's conclusions and is inconsistent with Professor Cummings's criticism.
- Furthermore, Professor Cummings was prepared to doubt the results of certain SDS-PAGE experiments on at least one ground which appear to apply equally to an SDS-PAGE experiment upon which he relied. He explained that a main reason for not regarding one of Dr Egrie's experiments as reliable was that two lanes which she compared were not close to each other on the gel: there were several lanes between them. However, when dealing with the one published paper which arguably assisted Amgen's case on this issue (Kung & Goldwasser in Protein Struct. Func. Gen 28:94, 1997), where the two lanes on SDS-PAGE which were to be compared were more than seven lanes apart, he said that the comparison "clearly showed" a difference in the apparent molecular weight of the substances in the two lanes.
- I should mention two other factors which Professor Cummings suggested might render Dr Egrie's conclusions wrong. First, the apparent overloading of some of the lanes in some of the experiments, and secondly the duration for which the SDS-PAGE experiments may have been run. The second point is not of significance, as Professor Cummings himself accepted. In relation to one or two of the lanes in one or two of the experiments carried out by Dr Egrie, the difference in loading, or, to put it more accurately, the apparent overloading, was something which, to my mind, could fairly be identified as rendering the performance of the material in the particular lane unreliable. However, I agree with Mr Kitchin, that one only has to look at the overwhelming majority of the results of the experiments to see that there is nothing in this point, in that the one or two occasions where it does apply do not in any way seriously be said to call Dr Egrie's conclusions into question.
Claim 19: Dr Strickland's and TKT's experiments
- Apart from the experiments of Dr Egrie, I should mention other experiments of relevance to this issue. During 1995, and in connection with the current proceedings, Dr Thomas Strickland of Amgen carried out the process of purifying EPO from urine in accordance with the teaching of Miyake. In this connection, it should be explained that the isolation of urinary EPO in accordance with the teaching of Miyake involved seven stages. As is usual when carrying out any process of purification or isolation, each stage should involve a mixture with an increasing proportion of the substance one is seeking to isolate, in this case EPO. At one stage in the Miyake procedure, namely DEAE-agarose ion-exchange chromatography, three fractions are produced, known as the 5 millimolar (5mM), the 17mM and the 30mM fractions. (Under the procedure, calcium chloride solutions of increasing concentration, measured by millimoles, is used to wash protein off the agarose column). Dr Strickland found that 85% of the protein was in the 30mM fraction, which was 2.6 times more pure or concentrated than the material loaded onto the column, and the 17mM fraction only contained 9.9% of the protein and was in fact only a quarter as concentrated as the material loaded on to the column.
- Nonetheless, Dr Strickland proceeded to seek to isolate the uEPO by the subsequent steps in the Miyake procedure from each of those two fractions. He then ran the 30mM fraction on SDS-PAGE against the GA-EPO made by TKT, which, according to Amgen infringes 605. In accordance with his conclusion, it seems to me that the urinary EPO (the "30mM uEPO") migrated a little faster than the GA-EPO. Accordingly, at least as against the 30mM uEPO, GA-EPO had a somewhat higher apparent molecular weight, and this was also found to be the case by TKT in its experiments.
- However, the uEPO obtained from the 17mM fraction (the "17mM uEPO") performed rather differently as against GA-EPO, according to TKT's SDS-PAGE experiments, as analysed by Professor Matsudaira, with whose view I agree. It seems to me that those experiments show TKT's GA-EPO migrating to the same point on SDS-PAGE as the 17mM uEPO, thereby suggesting that they enjoyed the same apparent molecular weight.
- Two criticisms are made of TKT's SDS-PAGE experiments so far as comparison with the 17mM uEPO are concerned. First, it is suggested by Amgen that the results are not to be relied on because, in order to identify the point to which the EPO had travelled, TKT used a polyclonal antibody, rather than a monoclonal antibody. I accept that it would be better normally to use a monoclonal antibody, because it is virtually certain to be specific to a particular protein: accordingly, one can be pretty confident that the band which appears as a result of the antibody fixing on to a protein is the specific protein one is looking for. Polyclonal antibodies are liable to be less specific, and therefore there can be more uncertainty as to whether they have in fact fixed on to the protein for which one is looking.
- However, in relation to TKT's experiments, I am unimpressed with this criticism. The trade literature relating to the polyclonal antibody used in TKT's experiments demonstrate that it is relatively specific, and Professor Matsudaira, a reliable witness, said that he often used polyclonal antibodies if he thought they were reliable. Further, if the apparent result of TKT's experiments was wrong because of the use of a polyclonal antibody, it could only arise from the most extraordinary coincidence. The experiments showed a single band, of normal width and character for a glycoprotein, indeed for EPO itself. Thus, if there was any other protein on to which the antibody fixed, it would have had to have had more or less exactly the same apparent molecular weight as EPO. Quite apart from this, even then, Amgen's indictment of the experiment would only be justified if one could conclude that there was no EPO present, merely another glycoprotein with more or less the same apparent molecular weight, which reacted to this polyclonal antibody, which, at least judging by the literature, was pretty specific. Thus the criticism of TKT's experiments requires such a combination of unlikely coincidences that I conclude that it can be confidently rejected.
- The second criticism of TKT's experiments relating to the 17mM uEPO is, in my judgment, more formidable. At first sight, at any rate, it would be almost perverse to take the 17mM uEPO, as opposed to 30mM uEPO. The great majority of the uEPO was in the 30mM fraction, and, indeed, the concentration of EPO in the 17mM fraction was actually lower after the fractionation than before. On the face of it, the natural course, if one had to choose between the fractions, would be obvious: one would go for the 30mM fraction. Sir John Walker expressed that view in robust terms.
- However, I do not think the point is quite that simple. The teaching of Miyake in his paper (op. cit.) involves taking the 17mM fraction, and not the 30mM fraction. Not surprisingly, that is because, as his paper reports, Miyake obtained a much greater concentration of EPO in the 17mM fraction than in the 30mM fraction (unlike Dr Strickland). I have accepted Amgen's construction of the closing words of Claim 19, in the sense that it refers to isolation in accordance with methods referred to in the patent itself (or within the common general knowledge). It seems to be tacit common ground that the most satisfactory or common of these methods is that reported by Miyake. There is thus obvious force in the contention that a person in Dr Strickland's position seeking to follow the teaching of Miyake in order to apply the test laid down at the end of Claim 19, would find himself in a state of some uncertainty: should he follow the 17mM fraction, in accordance with the literal teaching of Miyake, or should he follow the 30mM fraction, in accordance with what might be said to be the spirit of Miyake?
- Although alternative extremes were adopted by witnesses, it seems to me that the correct approach was that of Professor Matsudaira, whose evidence was to the following effect. He saw no reason why the skilled addressee should have to choose between the two fractions: given the dilemma, he would follow both. However, he accepted that, if one had to choose between the two fractions, one would "follow the protein", to use Mr Watson's expression, and therefore one would go with the 30mM fraction. In my judgment, a careful person carrying out the work, as Dr Strickland did, would have thought the safest course was to continue the Miyake process in relation to both fractions. I draw support for that conclusion from the fact that this is what Dr Strickland of Amgen actually did. It is suggested on behalf of Amgen that he merely did this because he was carrying out his work in the context of the present litigation, and, in particular, on the instruction of the lawyers acting for Amgen. There was no reliable evidence which even starts to support this contention, in my view. Indeed, there is contemporary documentary evidence to suggest the contrary. There is no record of any relevant discussions with any lawyer (whether in this country or the United States); no lawyer was called to give evidence on the point; Dr Strickland could not recall any relevant discussions with lawyers. Following both fractions seems to have resulted from discussions he had with Dr Egrie. The two doctors were relatively disinterested and relatively skilled scientists (albeit working for Amgen) and not lawyers. I believe my conclusion receives a little support from the fact that urinary EPO is so very difficult to obtain and therefore so scarce. That would be an additional practical reason for not discarding the 17mM fraction.
Claim 19: more recent publications
- There was further evidence about the performance of urinary EPO according to papers which were published after the relevant date, in particular, Imai et al. in Eur. J. Biochem. 194:457 (1990) ("Imai") and Inoue et al. in Biol. Pharm. Bull. 17(2):180 (1994) ("Inoue"). Inoue reported that if uEPO is purified from urine without the use of phenol (which was a treatment included in the process taught by Miyake) then it migrated more slowly on SDS-PAGE than if phenol was used. Inoue assessed the difference as showing a difference in apparent molecular weight in the region of 2kDa. Inoue also demonstrated that the in vivo performance of the uEPO isolated without the use of phenol was somewhat greater than that of the uEPO where phenol was used. A similar result was vouchsafed by Imai, but he also reported that uEPO prepared without the use of phenol treatment had the same mobility on SDS-PAGE as CHO rEPO.
- Amgen contend that this evidence is not reliable, because it merely amounts to two reports of experiments, which have not been repeated for the purpose of these proceedings. I accept that it can be dangerous to rely upon reports in scientific journals, even when they were peer-reviewed and when they include photographs of the relevant experiments (as both Imai and Inoue do). However, particularly in light of the time and cost involved in carrying out experiments, I am of the view that the court should be prepared to take into account the results of experiments such as those reported in the two papers I have mentioned. Plainly, care must obviously be taken in light of the fact that the person seeking to challenge the experiments cannot cross examine the person who carried them out, cannot have inspected when they were carried out, and cannot reasonably be expected to repeat them (given that the person relying on them has not chosen to do so). I see no reason to doubt the results reported by Imai and Inoue, particularly as no witness seriously called into question the reliability of the results so far as they could be judged from the contents of the papers.
- I have already referred to the paper by Kung and Goldwasser published in 1997. This paper involved a comparison between CHO rEPO and uEPO prepared in accordance with the teaching of Miyake. The paper concluded that the recombinant EPO was more easily iodinated than the urinary EPO, which in turn suggested that there was a difference in conformation (i.e. three dimensional shape) and/or glycosylation between the two types of EPO, on the basis that the amino acid which is iodinated was perhaps more difficult for iodination access in the urinary, than in the recombinant, EPO. I should interpose to say that this does not call into question my conclusion that there was no difference in the iodination of the different EPOs carried out by Dr Egrie. First, she was comparing two types of urinary EPO; secondly, she competently carried out an experiment which established to her contemporary satisfaction, and indeed to my satisfaction (as well as to that of at least one of the expert witnesses) that any difference in the apparent molecular weight between the types of urinary EPO she was experimenting with was not attributable to differential iodination. Indeed, I was referred to a paper by Tsuda et al. published in 1988 in Biochemistry 27(15) 5646, which supported the conclusion that uEPOs from different pooled sources, isolated by the same procedures, can have different apparent molecular weights.
- The experiments carried out by Dr Egrie establish the basic facts as I have summarised them, by reference to Goldwasser uEPO and Lot 82 uEPO, one of which was prepared in accordance with Miyake's teaching, and the other (subject to the urine having been obtained from the single source) also used Miyake's method subject to a minor workshop modification. The work carried out by Dr Strickland and TKT in connection with these proceedings complicates matters further, in that it indicates, at least to my mind, that following the teaching of Miyake can result in urinary EPO being taken from two different fractions at one of the stages of the purification process, and the two lots of uEPO having different apparent molecular weights, and therefore producing inconsistent results as against a sample of recombinant EPO with regard to the test in the closing words of Claim 19. The position is further complicated by the reports of Imai and Inoue: both of them showed that a minor workshop modification to the Miyake process results in the uEPO thereby produced having a significantly higher apparent molecular weight by SDS-PAGE than uEPO isolated strictly in accordance with the teaching of Miyake, the modification being the non-use of phenol at one of the seven stages taught by Miyake. Although I heard evidence and arguments as to the reason and effect of using phenol, I do not consider that it is necessary for me to go into that aspect: Miyake used phenol, and because it was suspected that phenol might have an effect on the ultimate isolated product, Imai and Inoue dispensed with it. As its use was unnecessary, it seems to me, standing at the relevant date, it would have been an obvious modification in the eyes of the skilled addressee to omit the phenol.
- There is no reason to doubt the conclusions of Imai and Inoue. Those conclusions are particularly significant because the uEPO produced without the use of phenol has a reported apparent molecular weight 2kDa greater than the uEPO isolated strictly in accordance with the teaching of Miyake, and that sort of difference is similar to, or greater than, any difference in apparent molecular weight recorded between rEPOs and uEPOs in the experiments reported in published papers, and conducted by Dr Egrie and by Dr Strickland. In particular, as I have mentioned, Kung and Goldwasser, upon whom Amgen rely, suggest that CHO rEPO has an apparent molecular weight 1.2kDa greater than uEPO purified in accordance with the teaching of Miyake. I accept that it can be dangerous and unsafe to compare the results of two independent experiments in this way, particularly when they are only contained in reported papers (albeit in respected peer-reviewed journals). However, it would seem from these two papers that, while CHO rEPO has a higher apparent molecular weight than uEPO prepared strictly in accordance with the teaching of Miyake, it has approximately the same, and if anything a lower, apparent molecular weight than uEPO isolated in accordance with the teaching of Miyake subject to a small, almost trivial, modification.
Claim 19: conclusions
- In my judgment, the aggregate effect of all the experiments and reports to which I have referred, involving CHO rEPO and COS rEPO, and various batches of uEPO prepared strictly or substantially (i.e. with minor workshop modifications) in accordance with the teaching of Miyake establishes the following. First, some rEPOs have a higher apparent molecular weight by SDS-PAGE than some uEPOs; secondly, some rEPOs have the same apparent molecular weight as some uEPOs; thirdly, no rEPOs have a lower apparent molecular weight than any uEPOs. However, it is clear from the paper of Wojchowski (op. cit.). to which I have not so far referred in the current connection, who expressed recombinant EPO in insect cells, that recombinant EPO of a type specifically referred to in the patent, and which otherwise would fall within Claim 19, can have a lower apparent molecular weight than most, indeed probably all, urinary EPOs. His experiments suggested that recombinant EPO expressed in a fruit fly cell had an apparent molecular weight of some 23 KDa.
- In light of these conclusions, it appears to me that Claim 19 is incapable of being infringed. According to the teaching of the 605 patent (in Example 10 on page 31), and indeed the evidence I have heard, any difference in apparent molecular weight is small. It appears to me that the variations in apparent molecular weight between different batches of urinary EPO, coupled with the fact that it is clear that many recombinant EPOs do not satisfy the test, would put the skilled addressee seeking to discover whether his product was within Claim 19, and seeking to discover this in a reasonable way, in an unsatisfactory, indeed, an impossible, position.
- In my judgment, it would be unreasonable to expect the skilled addressee seeking to discover whether his product is within Claim 19, to carry out the exercise required of him by the closing words of that Claim by running his product against more than one sample of urinary EPO. Indeed, I do not think anything else is contemplated by the Claim. The closing words of Claim 19 suggest that the draftsman believed that urinary EPOs have effectively the same apparent molecular weight, irrespective of the method by which they were isolated. Quite apart from this, it would be very onerous to expect the reader of 605 to get more than one sample of isolated urinary EPO, in light of the difficulty of obtaining samples of that product, a problem which would have been clear from the patent and is clear from the evidence in this case.
- Suppose, however, the skilled addressee could be expected to run his product on SDS-PAGE against three samples of uEPO from different sources. If his product displays a lower apparent molecular weight than, or similar apparent molecular weight to, one or more of the samples of urinary EPO, then his product would not be within Claim 19, given my conclusion that Claim 19 requires the product to have a higher molecular weight than all urinary EPOs isolated in accordance with any teaching referred to in the patent or within common general knowledge. If, however, his product has a higher apparent molecular weight than all three samples of uEPO, that would not be the end of the matter. Given the different performance of different uEPOs, depending on their source and the precise manner of isolation (and possibly other factors), and bearing in mind the very small difference in apparent molecular weight which would have been seen by the skilled addressee, he could not be confident that his product was within the Claim. He may well, for instance, not have run his product against urinary EPO isolated in the same way as Imai and Inoue isolated their product. In my judgment, it would be a long way away from giving "a reasonable degree of certainty for third parties" if the closing part of Claim 19 did not result in that claim being incapable of infringement.
- It does not seem to me that this conclusion is inconsistent with the conclusion I would have reached in relation to the apparently similar problem thrown up in relation to the requirement of "stringent conditions" in Claim 1(b) of the 605 patent. In order to see whether there is hybridisation in "stringent" conditions, the skilled man would carry out a routine series of experiments, involving increased stringency, and if hybridisation only occurs in conditions which some might reasonably regard as not being "stringent", then his product is not within the claim: the uncertainty thrown up by the word "stringent" in that event is resolved in favour of the alleged infringer, so that if the relevant conditions might be considered by some, but not others, to be "stringent", there is no infringement. I do not think that Amgen can say that the same result obtains in relation to the uncertainty thrown up by the closing words of Claim 19. The position might appear the same, in that the fact that one type of urinary EPO may have the same apparent molecular weight as (or, indeed, a higher molecular weight than) the reader's recombinant EPO would, as I see it, mean that there was no infringement, and it would only be if the alleged infringer's recombinant EPO had a higher apparent molecular weight than urinary EPO isolated from any source in accordance with one of the methods described in the patent that he would infringe. However, unlike the requirement of "stringent conditions" in Claim 1(b), which requires a set of routine experiments which could, indeed arguably would in any event, be carried out by the skilled man, the requirement of the closing words of Claim 19 would involve him potentially carrying out SDS-PAGE comparisons between his recombinant EPO and an indeterminate number of different urinary EPOs, all of which would be pretty difficult to obtain. The result of the comparisons would either be that he did not infringe or that he could not be sure that he infringed.
- The more difficult question, to my mind, is whether my conclusion as to the effect of the closing words of Claim 19 not merely renders that Claim incapable of infringement, but renders the Claim invalid on grounds of insufficiency. I revert to the words of Section 72(1)(c) of the 1977 Act: can it be said that the closing words of Claim 19 result in "the invention [not being disclosed] clearly enough and completely enough for it to be performed by a person skilled in the art"? As I have mentioned when considering Claim 1(b), it has been held by the Court of Appeal that lack of clarity or ambiguity do not render a claim insufficient, albeit that they may render it incapable of infringement either in whole or to an extent. Thus if a claim was ambiguous, in the sense that it had two meanings, it would be incapable of infringement if an allegedly infringing product could not fall within both meanings. On the other hand, it might nonetheless be capable of infringement if an infringing product fell within both meanings.
- I have come to the conclusion that the problem raised by the closing words of Claim 19, does not merely render the Claim incapable of infringement; it renders the Claim invalid on grounds of insufficiency. It is true that, the Claim is conceptually clear: it is a claim to all recombinant polypeptides which satisfy the other requirements of the Claim, and which have a higher apparent molecular weight than EPO isolated from any urinary source in accordance with a method referred to in the patent or within the common general knowledge. However, in so far as the Claim is or embodies the invention for the purposes of Section 72(1) of the 1977 Act, it appears to me inescapable that the disclosure is not at all "clear", in so far as it is "complete", within the meaning of the section.
- The clarity of the disclosure required by Section 72(1)(c) is not merely conceptual clarity. If it were, then, as I say , Amgen would have no problem with insufficiency on Claim 19. The clarity and completeness have to be such as to enable the skilled person to "perform" the invention. The word "perform", in this context, carries the implication of work which is non-inventive, routine, not prolonged and not involving research. Even if the skilled addressee finds his product appears to satisfy Claim 19 when run against one sample of uEPO on SDS-PAGE, he might still have to check it against any number of other EPOs. There is no teaching in the patent to suggest that he would have to do this. Further, that he might have to do it was not within common general knowledge . To obtain a number of uEPOs from different sources (whether to find out which has the highest apparent molecular weight or whether to run each of them against his product) would involve the addressee in prolonged work; it would not be routine. Further, it is an exercise which he could never be sure he had completed, unless and until he found that he was not within the Claim (because he identified urinary EPO which had the same or a higher apparent molecular weight than his product). His investigation could be never ending, unless it established non-infringement. His research, even if it ended with a conclusive result, could be time consuming and difficult. His research would not be routine, let alone taught, by the patent.
Claim 20
- There is no specific teaching in the patent to the effect that some or all recombinant EPOs differ in their glycosylation characteristics from some or all EPO isolated from urinary sources. The nearest one gets to any such teaching is in the last two paragraphs of Example 10 (page 31, lines 10 to 22) which describes a "preliminary attempt... to characterise recombinant glycoprotein products from... COS... and CHO cell expressions... in comparison to human urinary isolates". The disclosure that CHO rEPO had a "somewhat higher molecular weight" than COS rEPO, which in turn had a "slightly larger" molecular weight than pooled urinary EPO would be understood to suggest that the aggregate apparent molecular weight of the glycans on the two types of recombinant EPO was greater (albeit not by much) than on urinary EPO. Beyond that, however, the patent gives no guidance as to the nature of the difference in the "average carbohydrate composition" between the "glycoprotein polypeptide" and the "human erythropoeitin isolated from urinary sources" referred to in Claim 20. Furthermore, as I shall discuss in a little more detail when considering the issue of novelty, there is, despite Professor Cummings's suggestions to the contrary in his written report, no satisfactory evidence, let alone specific teaching, even today, as to what differences exist in the glycosylation of some or all rEPOs and some or all uEPOs. Accordingly, I think Claim 20 is insufficient on that ground. In any event, it is insufficient for the reasons given in relation to Claim 19, given that it incorporates that Claim by reference.
K. BIOGEN INSUFFICIENCY & BREADTH OF CLAIM AND CONCLUSIONS
Introduction
- The two main insufficiency arguments raised by TKT and Roche which I shall consider in this part of the judgment are as follows. First, Claim 1 (as well as many of the other Claims) extends to amplified or enhanced expression of EPO in eukaryotic cells generally, whereas the only teaching of the patent is in relation to DHFR- CHO cells, a point I have already considered when discussing classic insufficiency. Secondly, Claim 1 (again like many of the other Claims) extends not merely to EPO but to analogues of EPO (that is any polypeptide with possibly significant variations in the amino acid residues of EPO, which retains EPO-like characteristics) with virtually no teaching (other than one or two specific instances) as to what those analogues may be.
- In terms of what I might call classic insufficiency, there is an obvious case for saying that, in each respect, the Claims, in so far as they extend to either or both features, are insufficient. So far as the first aspect is concerned, the claim of the patent was not to have obtained pure EPO for the first time: this has already been done, albeit in very small quantities indeed, for instance by Miyake (op. cit.). What the patent purported to enable was a method of obtaining EPO in greater quantities, in particular in quantities sufficient for commercial and therapeutic use. The only specific way of doing this taught by the patent was in amplified DHFR- CHO cells. Amgen did not claim to have succeeded, and indeed had not succeeded, in obtaining amplified expression of EPO in human cells or COS cells. Unlike, for instance, TKT's technology (of which more below) the patent does not provide a way of amplifying a DHFR+ cell. Further, the only known DHFR- cells, at least at the relevant time, were CHO, as was accepted by Professor Wall in cross examination. Although Amgen's experiments with a certain type of human cell, the so-called human 293 cell line, produced EPO, it was not at a high level of expression. Accordingly, as I have already determined, the patent can be said to be classically insufficient so far as amplified or enhanced expression in human cells is concerned.
- As to analogues, Professor Proudfoot's unchallenged evidence was that the specific structures and activities of the proteins coded for are not predictable. EPO has 165 amino acids, and a change of one amino acid to another specific amino acid could be deleterious, beneficial, or make no difference; the number of different permutations involved in changing any two of the 165 amino acids runs into millions. Once one contemplates the possibility of changing, say, up to ten of the amino acids (and the evidence I heard suggested, albeit not specifically, that over ten changes in a protein with 165 residues may well not affect functionality, particularly if the changes were to residues not in the active sites of the protein), the permutations are, almost literally, approaching the infinite. Over and above this, the patent purports to cover deletions and additions. It seems to me that investigation as to which analogues (and therefore which encoding sequences) fall within Claim 1 would involve work of a routine nature, but it could not possibly be said that it would take a reasonable time. This is nicely encapsulated in an observation of Professor Wall:
"I know scientists who have spent a career taking a protein and changing every amino acid and seeing what its biological function is. They have long boring careers in my opinion."
The last adjective in that passage indicates the routine nature of the work, and the penultimate adjective shows that the time involved would not be reasonable.
- The 605 patent gives very little, if any, significant help on this aspect. There is nothing to indicate which amino acids might be changed and/or which amino acids could probably not be changed in the EPO sequence contained in Table VI. If the patent had revealed the three dimensional structure of EPO, and, perhaps even more, if it had revealed which of the internal sequences constituted the active sites, that would have given some assistance, possibly substantial assistance, to the reader. As I have mentioned, amino acid residues in the active sites of a protein are generally more likely to be essential to the functioning of the protein than residues outside the active sites.
- The evidence of Dr Browne of Amgen, indicated that Amgen had started an analogue programme, but it was not given high priority, compared, for instance, with expanding the effort on the development of mammalian cell lines producing higher levels of EPO. He also said that Amgen knew very little about the three dimensional structure of EPO or its active sites. He accepted that the investigations necessary to give any real guidance as to which amino acid residues in EPO could be varied without the resultant polypeptide losing its EPO-like characteristics would have involved "a research programme", the very thing which has been said to give rise to classic insufficiency.
- Amgen's answer to this attack is that the disclosure of the 605 patent is such that, even though the Claims extend to human and other cells, and are not limited to DHFR- CHO cells, and to all analogues of EPO, and are not limited to the natural sequence of EPO, does not amount to insufficiency.
Amgen's argument based on Biogen [1997] RPC1
- Adopting the language of Lord Hoffmann in Biogen [1997] RPC 1 at 48, when explaining the Board's decision in Genentech I/Polypeptide Expression Amgen's argument effectively amounts to contending that its "invention discloses a principle of general application" rather than "a number of discreet methods or products". As Lord Hoffmann also explained at [1997] RPC 48, if the invention is of the former quality "the claims may be in correspondingly general terms", whereas if of the latter quality, "the patentee must enable the invention to be performed in respect of each [discreet product or process]".
- It appears to me that the first step in considering this argument is to identify the technical contribution to the art made by the disclosure of the patent in suit. In this connection, I start with the fact that it is common ground that the disclosure of the patent was not obvious, and with the assumption that the arguments that the disclosure was in some way anticipated have been rejected. I can therefore proceed on the basis that the disclosure of the patent has made a contribution, and it is therefore a question of identifying the nature and extent of that contribution. All parties eschewed the notion that one should take into account the degree of inventiveness, a view with which I would agree. Obviously, in many cases the degree of inventiveness and the contribution to the art of a particular invention have a degree of equivalence, but in some cases they may not. It cannot be the function of the court to decide how clever or ingenious a particular invention was. First, it is obviously a particularly difficult and, to a significant extent, a subjective, exercise. Secondly, it would lead to a substantial degree of unpredictability in the law, which is always undesirable, and, I would have thought, particularly so in the field of patents. Thirdly, it would appear to me to be contrary to principle for the Court to consider the degree of inventiveness of a particular contribution to the art: once it has decided that there is an inventive contribution, then one goes on to look at the nature of the contribution, not the nature of the inventiveness.
- So far as the nature of the contribution is concerned, Amgen's case, in a nutshell, is that the contribution of the 605 patent is encapsulated in the disclosure contained in Table VI of the 605 patent. This Table identifies the precise amino acid sequence of EPO, and the great majority of the EPO gene (including much of the upstream sequence including the two start sites, the whole of the encoding exons, the whole of the intervening introns, the splice donor sites and the whole of the downstream sequence) coupled with enabling teaching as to how enough of that DNA sequence could be isolated and used in a cell to express EPO in far greater quantities than would be achieved naturally. The disclosure enabled something which could not have been achieved before, something which was plainly desirable and beneficial in commercial and humanitarian terms, and something for which eminent groups of scientists had been searching without success, despite substantial financial backing, over the previous five years or so.
- So far as the last point is concerned, there was no dispute that Dr Lin and Amgen had won a race in which they had a number of formidable competitors. Biogen had a US$6m project between 1980 and 1984, employing some of the most able molecular biologists of the time, who were seeking throughout that period to do just what Dr Lin did. For three years, a group of researchers, including Professor Stuart Orkin of the Harvard Medical School were engaged on the same project, and, according to Professor Orkin, it was one of the few projects on which he was not successful. He said that he regarded Dr Lin's work as "an outstanding achievement". It is also clear from the evidence that GI (one of the Roche parties) was carrying out similar work between October 1982 and March 1984. Further, from the evidence given by Dr Alex Ullrich in the United States proceedings, it is apparent that other institutions, including Colombia University and Genentech were seeking to clone the EPO gene in the early 1980s. Indeed, he said that to his knowledge "there were at least five or six good molecular biology laboratories attempting to clone this gene".
- It seems clear from the evidence that these eminent researchers failed, or at least had not succeeded by the relevant date, because of the very limited, indeed in some respects erroneous, information regarding EPO. The only significant available information was that published by Sue and another (op. cit.) which purported to record the first 27 amino acid residues, but two of them were not identified, two of them turned out to be wrong, and the sequence was degenerate; ironically, the least degenerate area included one of the errors. As Professor Brammar said, it was "astonishing how many people were misled by the available erroneous information.". The contrast of this position with that following the disclosure of Table VI, with all the coding sequences, all the intervening introns, the whole of the 3'prime non-coding region, and much of the 5'prime non-coding region (including the two endogenous transcription start sites) speaks for itself.
- It is clear that the information as to the sequence of the EPO gene (and in particular of the encoding regions, and the location of the splicing regions) as a result of the publication of 605, facilitated (to use a neutral word) something which had previously eluded a number of dedicated experienced and well-financed groups of researchers. Professor Brammar accepted that, once the EPO gene had been cloned and its sequence made available, "it was straightforward for someone to clone and express this gene". He accepted that, once GI and Roche had the information in Table VI of the 605 patent, "they would have been able to clone EPO sequences". He indicated it would have taken a matter of "a few weeks". Professor Proudfoot expressed himself similarly, when he said that "once the EPO gene had been cloned, it would have been a matter of course for the skilled worker to apply each of the standard techniques described to that cloned gene". Given the ability to make what he called "unique sequence probes" as a result of the disclosure of 605, Professor Proudfoot also said that "the rest of the exercise... is relatively straightforward" and it would "not surprise" him if it took only "a few months".
- Professor Wall, for Amgen, gave evidence to much the same effect. He said this:
"The disclosure of the 605 patent enabled skilled workers for the first time to beneficially undertake recombinant techniques to target and manipulate the EPO gene and cells containing the EPO gene to produce abundant quantities of EPO in a variety of ways... The patent represented a genuine breakthrough and its contribution to patient treatment was immense.... On the basis of the information in the patent it became possible to clone EPO gDNA and EPO cDNA and to express the EPO protein. This would have been possible based on the sequence set out in Table VI, whether the particular examples given in the patent had been followed or not."
He also said that "with the gene [in Table VI] in hand, using very standard methods... one could obtain expression" and that any errors or omissions in the specification of 605 "would [not] keep the skilled person from reproducing the invention". Although Professor Wall struck me as a mildly partisan witness, I do not consider that he was exaggerating when he made those observations. Dr Brenner, in a characteristic turn of phrase, said "once you have got the 605 patent disclosure, you have the blueprint for the production of EPO".
- The characterisation of Dr Lin's achievement abounds with metaphors. In T223/92 Genentech/HIF-gamma the Board said this (albeit in connection with a different patent):
"By identifying the DNA-sequence the respondents so to speak provided a guide rope to the peak which enabled others to be certain of getting to the same peak with much less trouble."
In relation to the Australian equivalent of the 605 patent, Counsel, in the Australian Federal Court in Genetics Institute Inc. -v- Kirin-Amgen Inc., 25th June 1996, likened Dr Lin's disclosure to a map giving directions to buried treasure. Heerey J preferred to invoke the analogy of a treasure in a castle which had many gates, each with a combination lock, where the essential or critical knowledge was the combination of the lock: without that knowledge, the castle could not be entered, but once one had that knowledge, one can obtain entry through any gate and with no more than a routine amount of time and effort, the treasure will be discovered.
- Heerey J took the view that cDNA or genomic DNA represents the gate and the combination sequence of the lock was to be found in Table VI. Indeed, at page 23, Heerey J referred to Biogen [1997] RPC 1 at page 23, and said this on the following page
"The fundamental difference which distinguishes the present case from Biogen is that in the Amgen patent the coding sequence is defined. The patent thus discloses a "principle capable of general application" and discloses a beneficial property which is common to the class. It cannot be said of it that it discloses no principle which would enable other products [of the class] to be made."
- The Netherlands Court of Appeal took the same view, also relying on Biogen [1997] RPC 1. At paragraph 13.5 of its judgment, the Court said this:
"In Table VI of the patent the inventor has disclosed for the first time the complete error-free sequence of the coding regions (exons) of genomic DNA... By demonstrating the exons the inventor therefore provided the essential genetic information for obtaining the object aimed at: the production of EPO by recombinant means. It is therefore reasonable that the inventor is allowed a process claim in general terms for applying any "means" in which use is made of (at least a part of) this genetic information. That in finding those alternative "means" inventions with independent merit may be involved, does not detract from the above"
- In my judgment, in agreement with the Australian and the Netherlands Courts, Amgen's contention is well-founded. Dr Lin delivered the goods, in the sense of providing all the necessary teaching which thereafter enabled biotechnologists to express EPO in cells using exogenous EPO-encoding DNA in accordance with routine methods, as they existed at the relevant date and developed from time to time over the life of the 605 patent. To that extent, he is entitled to commensurate "fair protection" under the Protocol. Suppose a third party invented and patented a new method of transfecting a human cell with exogenous encoding DNA, for example. It would be unfair on Dr Lin if the public could use the new technique to express EPO with the licence of the third party alone - just as it would be unfair for the third party if that could be done with the licence of the 605 patent alone. It would be fair on the third party and Dr Lin, if the public could not use the third party's technique to express EPO without infringing the third party's patent (in relation to use of the particular technique) and the 605 patent (in relation to the particular product). In such a case, although the new technique would "owe nothing to the teaching of the [605] patent or any principle it disclosed", the ability to express EPO by the new technique would do so. Identifying the extent of the monopoly to the expression of EPO, but not to any particular technique, appears to me to accord to Dr Lin a monopoly which "correspond[s] to [his] technical contribution to the art" as embodied in the 605 patent.
- Roche rely on a passage in the speech of Lord Hoffmann in Biogen at [1997] RPC 51 when he said this:
"[T]here is more than one way in which the breadth of a claim may exceed the technical contribution to the art embodied in the invention. The patent may claim results which it does not enable, such as making a wide class of products when it enables only one of those products and discloses no principle which would enable others to be made. Or it may claim every way of achieving a result when it enables only one way and it is possible to envisage other ways of achieving that result which make no use of the invention."
This passage must, I think, be read as a reference to a claim which is to one or more of "a number of discrete methods or products" as opposed to a claim which is to "a principle capable of general application" - [1997] RPC 48.
- In a sense, it might be said that neither of the two alternatives embodied in that observation apply precisely here. So far as the third sentence of the passage, upon which Roche rely, is concerned, all I need to say is that, even now, more than 16 years after the relevant date, nobody has suggested a way of expressing EPO in commercially viable quantities in cells "which makes no use of the invention". In particular, the technology used by Roche, and the much more modern and sophisticated technology used by TKT, both depend on information first disclosed in the 605 patent, information which the various groups of well supported and eminent scientists were looking for without success before Dr Lin. One can see how it could be said that there is "no principle", as contemplated in the second sentence, in the sense that there is no guidance as to how to express EPO in amplified quantities other than in DHFR- CHO cells, and "no principle" as to how to identify the analogues. However, the word "principle" should not be interpreted too strictly. I think that Mr Waugh is correct to describe the disclosure of the patent as being how to make recombinant EPO using its gene sequence. That is the contribution to the art.
- It is true that, in the present case, people knew of the existence of EPO in nature, knew that it had been isolated in small quantities, knew that the EPO gene existed in the genome, and knew that a cDNA sequence could be made by reverse transcription of EPO mRNA, if it could be found. They also knew that DNA sequences could be isolated by screening libraries with probes, and that once DNA sequences were isolated they could be expressed to produce the encoded protein. Accordingly, as Mr Thorley puts it in his elegant argument:
"[T]he goal was known, the potential sources of DNA sequence were known, the techniques were available. The problem was finding a way to apply the techniques to achieve the goal."
- In so far as that is a helpful analysis, it seems to me to assist Roche more on an argument based on obviousness, than on insufficiency. However, as Mr Thorley's argument continues, Amgen's only "inventive contribution was to decide to do so and succeed. This was the only novel aspect of the patent". As is clear from those words, it is accepted that the patent is inventive and novel, and that means that I am not concerned with the presence or absence of inventiveness; what I must do is to assess "the technical contribution" for the purpose of the present analysis. In connection with the concession that the disclosure of the patent was novel, it is fair to say that this view is supported by Professor Brammar who described Dr Lin's screening of a genomic library with mixed oligonucleotide probes as "an unusual thing to do" viewed as at 1983, the time he did it.
- However, I do not agree with Mr Thorley, that this admittedly inventive notion is to be equated, for the purpose of assessing its contribution, with the idea Professor Murray had in Biogen [1997] RPC 1. His inventive idea was how to achieve something which people have been trying to achieve, but, once he achieved it, others could do the same thing in different ways which owed nothing to him: he revealed nothing of general application. Dr Lin's idea can be said to have a similar quality, in the sense of being an inventive way of achieving something which everyone was looking to do. However, what he contributed to the art was information, embodied in Table VI, which enabled other people to design methods of producing EPO through expression in cells, which they would not otherwise have been able to do. If all that Dr Lin had done was to achieve the expression of EPO in transfected DHFR- CHO cells, then there would have been a strong similarity between his disclosure and that of Professor Murray. However, albeit as part of the exercise enabling him to achieve the expression of EPO, Dr Lin's disclosure extended to a contribution of general application.
- As Lord Hoffmann emphasised in Biogen, biotechnology is a field which is rapidly expanding and developing. As a result, the law has to adapt, not in the sense of changing the applicable principles, but more in the sense of shaping, or even modifying, those principles to apply in a coherent and sensible way to the evolving technology in the field. Inherent in the concept of "fair protection for the patentee", which, together with "certainty for third parties", can be taken as one of the guiding lights in the field of patents, is a degree of value judgment. Although, there is inevitably an element of value judgment, in the context of Biogen insufficiency, the principle is better enshrined in the words of the Board as approved by Lord Hoffmann in Biogen at [1997] RPC 49 line 20, namely "the extent of the patent monopoly... should correspond to the technical contribution to the art in order for it to be supported".
Discussion
- The conclusion that the invention embodied in 605 "discloses a principle of general application" is not necessarily the end of the matter so far as the alleged insufficiencies with regard to types of cell and analogues are concerned. I turn first to the use of transfected cells (other than DHFR- CHO cells) with exogenous EPO-encoding DNA for the expression or amplified expression of EPO. It appears to me that the view I have formed must lead to the conclusion that the patent does extend to such other cells, albeit that achieving such expression might involve a fresh inventive step, which could lead to a new patent in its own right, or else an improvement or nesting patent. Whatever fresh inventive step was involved in achieving expression of EPO in the new type of cell, it would remain the case that it could not have been achieved without the disclosure given in the 605 patent, and in particular in Table VI thereof.
- In particular, this is relevant to expression in human cells. Techniques developed since the relevant date enable the disclosure of the 605 patent to be invoked to obtain amplified or enhanced expression of EPO in human cells. In light of my conclusion in the previous section of this Part of the judgment, the claims of 605 can properly extend to such processes as their product (without prejudice to the issue of whether the new process may itself also be patentable).
- I have more difficulty with the question of analogues, essentially for two reasons. First, more could have been done by Amgen to give guidance to the reader on this issue: in particular, as I have mentioned, the three-dimensional conformation, and possibly the glycosylation sites, and the active regions could all have been identified. Secondly, there is the reasoning and conclusion of the Court of Appeal in American Home Products [2000] IP&T 1308.
- Amgen rely on the comparatively worthless nature of a patent relating to a protein and its encoding gene if it could not extend to analogues of the protein. A few amino acids or even just one amino acid could be substituted with no significant change (or possibly even an improvement) to its activity; concomitant alterations could be made to the codons in the gene. As Mr Waugh says, it would be only too easy for someone to benefit from the teaching of the patent by proceeding, on a trial and error basis, to substitute one amino acid residue for another somewhere along the polypeptide chain of EPO, and then to construct a strand of DNA with an appropriately adjusted sequence of codons. If, as seems virtually certain, it would be possible to change, say, ten of the 168 amino acid residues in human EPO for any one of say ten other amino acids, without the EPO losing its biological properties, that would mean that there were over ten billion different variants. As indicated above, I suspect that that is a substantial under estimate, because it appears likely that more than ten amino residues could be changed (provided that they were not within the active sites) without the resultant polypeptide losing its biological properties, and each such one amino acid could well be substituted by most or even all of the other nineteen amino acids. It also appears that some of the amino acid residues can be lost without the resultant protein losing is biological properties. Mr Waugh, in the course of his painstaking submissions, argues with justification that it would be impossible in these circumstances, however great the resources available to a patentee, to test for each of these variants, and then to ensure that there is sufficient teaching in respect of them.
- I accept that Amgen could have carried out, and could have disclosed the results of, further work which would have given more information and guidance about the possible analogues. However, the result of such information would not have been exhaustive, and it would certainly still have been the case that the reader would have been in a state of uncertainty as to whether any particular change would have resulted in a loss or reduction in the EPO-effectiveness of the resultant polypeptide. He would know that there was a better chance of a particular change having no effect or having an effect, but that is all. Further, at least on the facts of the present case, it is unrealistic not to have regard to the fact that the pressure on Amgen, and indeed on the other researchers in this field, was to obtain the sequences of EPO and the EPO gene, and to use this to effect expression of EPO in a transfected host cell. It would have put the first person to win this particular race in an unreasonably hard position if he had to elect between revealing his disclosure but not being able to claim analogues, which would probably render his patent almost worthless, or to hold off applying for his patent while he investigated analogues, in which case he may lose out to a later competitor who applied for a patent earlier. At least on the present facts, I do not think "fair protection to the patentee" can fairly require that unpalatable choice of the inventor. Certainty or even fairness to third parties or the public does not appear to me to point the other way.
- I turn then to American Home Products [2000] IP&T 1308. There is no doubt that many of the observations I have quoted from the judgment of Aldous LJ in that case give considerable apparent support to the argument of Roche and TKT that, whatever my view would have been in the absence of that decision, 605's claims to analogues must fail on grounds of insufficiency in the present case. In my judgment, however, that is not correct, particularly when one compares the approach of the Court of Appeal in American Home Products to that of the Court of Appeal in Chiron.
- First of all, the scope of the contribution of Professor Calne in American Home Products was relatively limited: he simply discovered a new use for a product which was not merely well known, but which was available already in commercially viable and therapeutically valuable quantities, and its chemical formula was well known. That is to be contrasted with the present case, where Dr Lin's discovery and invention enabled EPO to be produced in commercially and therapeutically worthwhile quantities for the first time, and where he also identified for the first time the chemical structure of EPO and (with the exception of a part of the upstream sequences) the EPO gene.
- Secondly, the exercise involved in identifying which variations, in the form of additions, substitutions or removals, of the rapamycin molecule would have the same immunosuppressant activity as rapamycin was quite unpredictable and conceptually boundless. As Aldous LJ put it at [2000] IP&T 1325, in so far as molecules other than rapamycin were claimed, "such a claim does not reflect a class with a unifying characteristic" and at 1326, "it would take prolonged tests to find out whether [a variant] had the appropriate qualities". On the other hand, in the present case, although I accept that Claim 1, and indeed Claim 19, could extend, at least in theory, to an almost infinite number of variants, the class of substitutions for each amino acid residue in naturally occurring EPO would be limited, and the experiments involved in relation to any particular analogue would be relatively short and routine. Thus, while the number of permutations arising from the possibility of substituting each amino acid residue with one of the other 19 amino acid residues would lead to a virtually infinite number of different analogues to be tested, the reader would have appreciated exactly what the possible substitutions could be, and that the test that he would have to carry out in order to effect the substitution and in order to see whether the substitution produced a polypeptide which had EPO-like activity would be a routine exercise. It would simply involve removals or substitutions of amino acid residues. It appears to me that that is very different from the exercise required of the reader if the patent in American Home Products [2000] IP&T 1308 extended to variants of rapamycin, which was not a protein, and in respect of which there was no general knowledge and experience as to the types of variants which might be effective to the rapamycin molecule without it losing the characteristics essential for the disclosure of the patent in that case.
- It appears to me that this view is support by the reasoning, indeed the decision, of the Court of Appeal in Chiron [1996] FSR 153. I have already quoted extensively from the judgment of Morritt LJ in that case. The Court of Appeal held valid a claim which, as Morritt LJ put it, did not "differentiate between one polypeptide falling within the claim and any of the millions of others". However, the claim was held nonetheless to be valid essentially on the basis that any analogue falling within the claim "can be defined in the sequence by exposure to an antibody and can be made by routine methods in molecular biology". In other words, approaching the question of sufficiency on the classic basis, an analogue of EPO will only be within the claim if the analogue has the biological properties of EPO and its gene hybridises under stringent conditions to the coding regions of Table VI of the 605 patent. The evidence establishes that it only takes a few days to make an analogue and to test its biological activity. Similarly, there is no suggestion that any difficulty in terms of time or effort, would be encountered in seeing whether a gene sequence which encoded that analogue satisfies the hybridisation test. Accordingly, as in Chiron [1996] FSR 153, the reader of the patent would have no difficulty in finding out whether a particular analogue was, or was not, within the claims. In other words, as Professor Wall said in evidence, "making and testing such fragments and derivatives can be readily undertaken starting with the sequence information in the patent and using the routine tests and assays described in the patent". He also said that to make an analogue would take a few days, and certainly would not take more than a few weeks. There was nothing to suggest that looking for in vivo EPO-like activity would be difficult or time consuming. The position in this connection appears to be comparable to that in Chiron where (to quote from [1996] FSR 187) "other strains may be isolated and sequenced [by] processes... well known to the man skilled in the art and of a routine, if time consuming, nature". It is true that Dr Browne said that finding an analogue which had advantages over EPO would be a research project, but I do not think that is quite to the point.
- Dr Browne also explained that Amgen had found an analogue of EPO which they were considering marketing under the name ARANESP, which had the advantage over naturally occurring EPO of being metabolised from the blood less slowly (so that it needs to be injected into patients less often) and that this, self evidently, had taken some time to develop. The precise amounts of time and effort involved in this development was unclear. Further, it appears that the reason why the development of ARANESP took the time that it did is also unclear: the evidence suggests that it could well have been not very high on Amgen's list of priorities, which of itself could well have been due to the success they were encountering in relation to EPO with its naturally occurring sequence. However, even on the limited amount of evidence available, it is fair to say that the development of ARANESP involved, in terms of time and effort more than enough work to take it outside the description of routine, although it is fair to say that there is nothing to suggest the actual nature of the work was any different from Professor Wall's slightly contemptuous description.
- However, essentially for the reasons given by Morritt LJ in Chiron [1996] FSR 153, I do not think that this presents a problem for Amgen. As Professor Wall said, after referring to the passages in the 605 patent at page 47 where certain specific analogues and variants are referred to:
"I believe that those skilled in the art in 1983 could have designed a variety of such derivatives and easily have tested them for activity using standard assays available at the time."
He went on to say:
"[I]t was entirely predictable that there would be useful analogues and variants of the EPO sequence shown in Table VI, although it would be necessary to make and test such analogues and variants to assess the properties of these new compounds."
- It is true that, unlike in Claim 1 of 605, the claims in Chiron [1996] FSR 153, did not involve the variants (or epitopes as they were in that case) having to be tested in vivo. It is also true that there was a specific routine method for testing for epitopes as well as evidence that this testing had been carried out successfully. I do not regard those points as valid grounds of distinction. The added ingredient of in vivo testing does not, on the evidence, render the overall testing of analogues non-routine or lengthy. The fact that there is no evidence of any efficacious analogue of EPO is not in point: testing a particular analogue is not a problem.
- The disclosure of the 605 patent provided the guide rope to the peak, the route map to the buried treasure, the keys to the castle. Following the disclosure, it would have required routine work, not of a lengthy nature, to contemplate an analogue to obtain it, and to test for its efficacy.
- Professor Wall's view, in summary, was that "what is important... is that the [605] patent has published the starting point from which such work can proceed". I agree with that assessment. As a matter of law, applying the reasoning in Biogen [1997] RPC 1 and in Chiron [1996] FSR 153, it seems to me that extension of Claim 1 to biologically effective analogues of EPO is permissible. I do not believe that there is anything in American Home Products [2000] IP&T 1308 to call it into question.
Insufficiency: conclusions
- As I have already mentioned, six categories of insufficiency have been alleged against the 605 patent. My conclusions on those are as follows:
1. A source of human cDNA was available, in the form of transfected cells;
2. the teaching of the 605 patent is classically insufficient so far as human cells are concerned, but this does not render the patent invalid, on the basis of the reasoning in Biogen [1997] RPC 1;
3. the reference to "stringent conditions" in Claim 1(b) is not classically insufficient;
4. the claim to analogues of EPO does not render the patent insufficient in light of the reasoning in Biogen;
5. the comparison of recombinant and urinary EPOs at the end of Claim 19 is insufficient;
6. the carbohydrate content comparison in Claim 20 is also insufficient.
- The consequence of these conclusions is that the insufficiency arguments fail, save in relation to Claim 19 and Claim 20. However, because they are contingent on Claim 19 and Claim 20, Claims 21, 22 and 23 also fail for insufficiency. At least in so far as they rely on any of Claims 19 to 23 inclusive, Claims 29, 30, and 31 are also insufficient.
- The final issue I have to consider is whether the insufficiency of the claims to which I have just referred renders any of the other claims, or indeed the whole of the 605 patent, invalid on grounds of insufficiency. Where the court concludes that a particular claim, upon which some but not all other claims depend, is insufficient, it does not appear to me, at least as a matter of principle, that one can lay down any absolute rule as to whether the patent as a whole, and in particular, the claims which are not dependent on the insufficient claims, can survive. Again approaching the matter as one of principle, I would have thought that, if stripping out the insufficient claims (and any teaching directly related to those claims) leaves, both conceptually and in terms of the understanding of the notional reader, a valid and understandable patent, then there is no reason in principle or theory why the remainder of the patent should not survive.
- However, the point has to be determined by reference to the legislation. Under Section 72(1) of the 1977 Act, the court "may... revoke a patent for an invention" on various grounds including, in paragraph (c), "the specification of the patent does not disclose the invention clearly enough and completely enough...". Stopping there, the position would appear to be "all or nothing". However, it is necessary to go on to Section 72(4), which provides:
"An order under this section may be an order for the unconditional revocation of the patent or, where the court... determines that one of the grounds mentioned in sub-section (1) above has been established, but only so as to invalidate the patent to a limited extent, an order that the patent should be revoked unless within a specified time the specification is amended under section 75 below to the satisfaction of the court..."
- This would appear to suggest that, in a case such as this, provided I am satisfied that it is possible to strip out Claims 19 and 20, and all other Claims which are contingent thereon (or if some are only contingent in part, to the extent that they are so contingent) then I can do so, at least in accordance with the procedure set out in Section 72(4) of the 1977 Act. Such a conclusion appears to me to be consistent with Section 125(1) of the 1977 Act which requires the "invention for a patent for which an application has been granted" to be taken as being that "specified in a claim of the specification of the... patent". In other words, if one can establish that there are a number of independent, albeit possibly conceptually closely related, claims, the fact that one claim is insufficient does not thereby automatically invalidate another.
- In the instant case, it seems to me that it is relevant to bear in mind not merely that the claims other than Claims 19 and 20 and the Claims dependent thereon are conceptually and linguistically independent of Claims 19 and 20, but that the only reason that Claims 19 and 20 are invalid is due to a feature which is not to be found in any of the other Claims (other than those dependent on Claims 19 and 20), namely the SDS-PAGE comparison with urinary EPO and the comparison of the carbohydrate content. In my judgment, at least so far as the Claims are concerned, the Claims other than Claims 19 and 20 (and the Claims dependent thereon) should, if possible, be permitted to stand. The only part of the specification which may be, as it were, tainted is the two paragraphs in Example 10 which I have quoted. I am not convinced that the 605 patent need be amended by the removal of those two paragraphs. The first of the paragraphs seems plainly innocuous: it states at the very beginning that it merely reports the result of a "preliminary" experiment. Although that is not expressed in the second of the two paragraphs, I think that the notional reader would appreciate that what is stated there is based on the "preliminary" experiment described in the preceding paragraph. Indeed, it is hard to see how he could have any other view in the absence of any other experiments or figures being reported in the patent itself.
- In these circumstances, at least subject to further argument, it appears to me that the appropriate course in this case is to make an order in accordance with the second part of Section 72(4) of the 1977 Act, whereby the patent can be amended by the removal of Claims 19 and 20, and any other Claims in so far as dependant thereon or referential thereto. I should make two further points. First, at least in relation to Claim 19, it would seem to me very unsatisfactory if the closing passage relating to urinary EPO resulted in the patent as a whole being revoked. Neither 605A (the application for the patent) nor 605B (the patent as originally granted) contained any reference to urinary EPO. It was only because of the Board's view (which in my judgment was an erroneous view) that Claim 19, without the closing words, was an impermissible product-by-process claim, that the comparison with urinary EPO was added. Secondly, my conclusion on the insufficiency issue, namely that the patent is partly, indeed largely, valid, but that, in respect of a few claims it is invalid, must, of course, be subject to my conclusion on the other grounds of alleged invalidity, to which I now turn.
L. MERE DISCOVERY
- Section 1(2) of the 1977 Act provides:
"The following... are not inventions for the purposes of this Act, that is to say, anything which consists of -
(a) a discovery...
but the foregoing provisions shall prevent anything from being treated as an invention for the purposes of this Act only to the extent that a patent or application for a patent relates to that thing as such."
- Roche contends that, in so far as Claim 1 of 605 claims genomic DNA (and in particular human genomic DNA), it is not an invention: it is simply a discovery of what existed in nature, and therefore cannot properly be claimed. While not advancing any separate argument on this issue, TKT supports Roche's position on this issue.
- Clearly, it can fairly be said that it was the "discovery" of the gene sequence for EPO which effectively provides the basis for the whole 605 patent. Indeed, it could scarcely be otherwise given the way in which Amgen put their case on breadth of claim and Biogen insufficiency. However, that is not the end of the matter. In Gale's Patent Application [1991] RPC 305, Nicholls LJ cited at 324, lines 21 to 25 with approval an observation of Whitford J in Genentech Inc's Patents [1987] RPC 553 at 566, to the following effect:
"It is trite law that you cannot patent a discovery, but if on the basis of that discovery you can tell people how it can be usefully employed, then a patentable invention may result. This in my view would be the case, even though once you have made the discovery, the way in which it can be usefully employed is obvious enough."
- Nicholls LJ went on to say at [1991] RPC 324, lines 31 to 36:
"Thus, a discovery as such is not patentable as an invention under the Act. But when applied to a product or process which, in the language of the 1977 Act, is capable of industrial application, the matter stands differently. This was so held in Genentech Inc's Patent [1989] RPC 147. There, this court by a majority decision held that Section 1(2) did not depart from the established principle mentioned above."
- The law was also considered by the Board in T208/84 Vicom/Computer Related Invention [1987] 2 EPOR 74 and by the Court of Appeal in Fujitsu Limited's Application [1997] RPC 608. In the latter case, Aldous LJ said this at [1997] RPC 614, lines 40 to 46:
"[I]t is... a principle of patent law that mere discoveries or ideas are not patentable, but those discoveries and ideas which have technical aspect or make a technical contribution are. Thus the concept that what is needed to make an excluded thing patentable is a technical contribution is not surprising. That was the basis for the decision of the Board in Vicom. It has been accepted by this court and by the EPO and has been applied since 1987. It is a concept at the heart of patent law."
- It is interesting to note that Nicholls LJ in Gale [1991] RPC 305 and Aldous LJ in Fujitsu had "difficulty in identifying clearly the boundary line between what is and what is not a technical contribution" - see per Aldous LJ at [1997] RPC 616, lines 41 to 42.
- In the present case, in support of the contention that Claim 1 is invalid, at least in so far as it extends to genomic DNA, because it constitutes a mere discovery, Roche place considerable weight on the reasoning of the Court of Appeal in Genentech Inc's Patents [1989] RPC 147, and in particular the statement of Mustill LJ at 269 line 2 that a particular DNA sequence was an "existing fact of nature, newly discovered".
- Clearly, in so far as Genentech, a decision of the Court of Appeal, contains any reasoning or conclusion which can fairly be said to apply to the facts of this case, it is binding on me. However, I am of the view that it does not. First, reading the judgments of Purchas and Dillon LJJ at 208ff and 239 to 240 respectively, it seems to me that they took a different view on the point (which appears to be accurately reflected in the headnote at [1989] RPC 152, lines 18 to 21). Secondly, while it is fair to say that Genentech was concerned with a patent which claimed a DNA sequence, each case must to a significant extent turn on its particular facts. If the principles are tolerably clear, and there is no question of those principles having been altered as a result of Genentech (and that is not suggested by Mr Thorley), then my duty is primarily to apply the principles to the facts of the present case, rather than delving into detailed judgments in other cases merely to see how the principles were applied to the facts of those other cases.
- In my judgment, applying the well established principles summarised in the passages from earlier cases I have quoted above, I conclude that, while it is obviously the case that the essential feature of 605, and in particular of Claim 1, is a "discovery", namely that of the DNA sequence of the EPO gene, or at least a substantial part of that gene (including, crucially, the encoding regions, the introns and the splice sites), it was a discovery which clearly made a technical contribution. I have considered that contribution during the course of this judgment, perhaps most notably in connection with Biogen insufficiency and breadth of claim, and I do not propose to repeat myself in what is already a regrettably lengthy judgment. Claim 1 is to a DNA sequence which is "suitable for" the claimed purposes, and I accept Mr Waugh's submission that it is "plainly the application of the discovery which is capable of industrial application (whatever the origin of the DNA sequence)".
- I take comfort from the fact that, according to the submissions on behalf of Amgen (and not, I think, challenged by Roche), over the past 20 years or so, it has been the regular practice of the European Patent Office (and, I think, of the US Patent Office) to grant claims substantially in the form of Claim 1 of 605. Further, the sequences claimed in 605 has not been impugned in any other European country, including Germany and the Netherlands, where, as I have mentioned, 605 has been attacked on various other grounds.
M. LACK OF NOVELTY
General
- Roche and TKT each raise a number of arguments as to why Claim 19 of 605 is invalid for want of novelty, and TKT alleges that Claim 1 and Claim 26 also suffer from want of novelty. I do not think that I need deal with TKT's contention that Claim 1 suffers from want of novelty in light of my conclusion that, on its true construction, the DNA sequences claimed by Claim 1 must have been isolated from the cells in which they were found.
- Further, I have already dealt with one of TKT's arguments as to why Claim 26 is invalid for want of novelty, namely that it is a product-by-process claim, which should be interpreted, in accordance with the Board's leaning, as a claim to the product as such, and not merely to the product provided that it is made by the process identified in the Claim. For the reasons already given, I reject this argument of TKT.
- If I am correct in my view relating to product-by-process claims, then TKT's case as to why Claim 19 is invalid on grounds of lack of novelty fails. TKT contend that, if the Board's approach to product-by-process claims is correct, Claim 19 will be invalid on grounds of lack of novelty if the recombinant EPO it claims does not have a higher apparent molecular weight on SDS-PAGE than urinary EPO, as the higher apparent molecular weight is the only specific distinction which rEPO is said to have over uEPO. It is right for me to deal with that contention on the basis that I am wrong in not adopting the Board's view of product-by-process claims, in part because the point has been fully argued and in part because this case may well go further. I shall then deal with an argument raised by Amgen if TKT's point is otherwise correct, namely that the recombinant EPO claimed by Claim 19 has other characteristics which would distinguish it from urinary EPO. I shall then turn to TKT's contention that Claims 1, 19 and 26 are anticipated by the patent of Sugimoto referred to at page 6 of the 605 patent ("Sugimoto"), and finally turn to the effect of some publications. I should add that TKT's allegation based on Sugimoto and the published papers are only maintained if TKT's product, GA-EPO, infringes those Claims.
Higher molecular weight by SDS-PAGE
- This section proceeds on the assumption that I am wrong and that the Board is right in its view of product-by-process claims, such as Claim 19.
- I have already dealt in some detail with various experiments carried out for the purpose of these proceedings, and various papers, showing the performance on SDS-PAGE of recombinant EPO from various sources and urinary EPO from various sources, as well as discussing the evidence relating thereto. As I have mentioned, certain conclusions can be drawn from the evidence. First, it does appear that recombinant and urinary EPOs both produce relatively wide and fuzzy bands, as one would expect of a glycoprotein. Secondly, urinary EPOs can vary depending on their source, and even urinary EPOs from the same source, albeit from different fractions during the purification process, can vary from each other. Thirdly, recombinant EPOs can vary from each other, and indeed this is what one would expect bearing in mind that recombinant EPO can be made, as 605 teaches, in different types of cell, which effect different types of glycosylation.
- In relation to its case on lack of novelty, TKT rely primarily on the evidence which, at least to my mind, establishes that:
1. Recombinant EPO made according to the teaching of 605 in CHO cells ("CHO rEPO") has a similar apparent molecular weight to the so-called Lot 82 urinary EPO ("Lot 82 uEPO"); and
2. Recombinant EPO expressed in COS cells according to the teaching of 605 ("COS rEPO") has a similar apparent molecular weight as urinary EPO obtained from Dr Goldwasser, which had apparently been obtained according to the teaching of Miyake ("Goldwasser uEPO").
- TKT also rely upon other evidence to which I have referred when dealing with the insufficiency arguments on Claim 19, namely the experiments of Dr Strickland and of TKT and the papers written by Inoue and by Imai (opa. cita.). It is not necessary to consider this evidence, nor indeed the arguments relating to this evidence, in any detail so far as TKT's case on novelty is concerned, in light of the discussion on the evidence in relation to insufficiency.
- The position appears to me to be this. The closing words of Claim 19 set a standard which, on its face, identifies the recombinant EPOs claimed by reference to a factor which distinguishes them from naturally occurring EPOs. However, if the evidence in this case merely demonstrates that many recombinant EPOs do not fall within the Claim (either because they have a lower apparent molecular weight than, or the same apparent molecular weight as, urinary EPO), that would not mean that the limitation of the claim to recombinant EPO with higher molecular weight is invalid on grounds of lack of novelty. Far from it: the claim would only extend to recombinant EPOs which were different from naturally occurring EPO - the very requirement which would, if sufficient, bestow novelty.
- The evidence establishes that there may well be some difference between one EPO and another. However, not only can urinary EPOs vary from each other, but it is not sensibly possible to predict as at the relevant date (or even today) what the differences will be. On the limited evidence I have, the apparent molecular weight of CHO and COS rEPO would appear to be the same or higher than uEPO prepared in accordance with the teaching of Miyake (op. cit.) (with or without minor modifications). However, even with that difficulty (fatal to Claim 19 on insufficiency grounds) it does not seem to me to alter the force of Amgen's argument on the issue of novelty. If it is otherwise valid, Claim 19 only extends to EPOs which do differ from all natural EPOs in their apparent molecular weight whether one takes the average or the spread of different weights. I believe my reasoning on this aspect is consistent with the approach of the Board in T130/90 - see at [1996] EPOR 55, paragraphs 4.10 and 4.11.
Other alleged differences between rEPO and uEPO
- In his report, Professor Cummings stated that there were numerous differences between urinary EPO and recombinant EPO, quite apart from the fact that the former had an apparently higher molecular weight on SDS-PAGE than the latter. The differences he identified may be expressed in summary terms as being variations in performance in isoelectricfocusing and electrophoresis, differences in oligosaccharide and/or sialic acid composition, different N-glycan distribution, different proportions of branched polysaccharides, different sialic acid and/or glycan linkages, and different biological activity.
- In my judgment, neither as a matter of law nor as a matter of fact do any of these contentions assist Amgen, and I am bound to say that Professor Cummings did his credibility no favours by suggesting otherwise. As a matter of principle, it does not seem to me that it is open to Amgen to rely upon differences between recombinant EPO and urinary EPO which are not identified in Claim 19, or in the specification. First, the patent would be insufficient if it relied on a test which it did not describe, and the differences alleged by Professor Cummings were not known to exist (assuming that they do exist) at the relevant time. Secondly, Amgen led no evidence on these differences, other than references to papers. Thirdly, and perhaps most importantly, the invention of a patent is defined by its claims, as is emphasised by Section 125(1) of the 1977 Act: it is not defined by later evidence of actual or potential differences between the disclosure of the specification and the prior art. That, I believe, is a fundamental feature of patent law.
- Quite apart from this, I am satisfied as a result of hearing Professor Cummings cross examined, that (with one possible exception) none of the alleged differences are made out. The various papers cited by Professor Cummings in his report which were said to support the various alleged differences he was putting forward did not appear to assist the conclusion when examined more closely. Indeed, Professor Cummings was forced to accept this in the case of virtually every difference he alleged. The one exception was the paper by Kung and Goldwasser (op. cit.) which suggested that a sample of CHO rEPO and uEPO isolated in accordance with Miyake's teaching might have different conformations and/or glycosylation. The basis for this view was the differential rates of iodination of the two types of EPO. However, this evidence was recent, indirect, imprecise and speculative.
- I thought Professor Robbins's evidence on this issue was far more convincing, and indeed it was not really challenged in cross examination. He said "One can never generalise to all urinary and recombinant from any of the papers". I agree with that. In my judgment, none of the papers cited in support of Professor Cummings's alleged differences between recombinant EPO and urinary EPO bore out his case.
- There is one aspect of the alleged differences between recombinant EPO and urinary EPO with which I should deal in perhaps a little more detail. It is Amgen's contention that recombinant and urinary EPO will inevitably differ in their oligosaccharide content. It was suggested that the effect of the evidence of Professor Clausen and Professor Robbins was that there would inevitably be a difference in the structure of the glycans of recombinant EPO and of urinary EPO. Thus, Professor Clausen said that, as a result of increasingly sensitive and sophisticated techniques since the mid-1980s "it may be... now possible to show some differences between a sample of rEPO by these techniques". In the first place, that does not seem to me to support the contention which Amgen are advancing: it is directed to potential differential glycosylation of recombinant EPO. Secondly, it is a speculation, albeit informed speculation by an expert.
- So far as Professor Robbins is concerned, he said that his "general feeling" was that "the glycosylation pattern of erythropoeitin will vary from cell line to cell line depending on the glycosylation machinery available in the cell for glycosylation". He also said "one always does find small differences based on the source, cell line, and so forth". It seemed to me that he came nowhere near saying that, for instance, certain types of glycosylation would only occur in recombinant EPO or urinary EPO. On the contrary: he said that he would be "surprised to ever find that any [oligosaccharide] structure one finds in one preparation of erythropoeitin would be totally absent from another. It will be a matter of proportion probably". In other words, as I understand it, Professor Robbins's view was that each type of EPO will contain many of the same glycans, and the normal explanation for any differential in the apparent molecular weight is due to a difference in the relative proportion of different types and degrees of glycosylation. Quite apart from that, even more than 15 years after the relevant date, the experts were unable to do more than speculate on this issue.
- In what I can only characterise as something of a last-ditch argument on this issue, Amgen placed reliance on a test apparently used at the Sydney Olympics, because some athletes would be tempted to take recombinant EPO in order to increase their oxygen production. The limited information provided on this test was scant and imprecise, and I found it of no assistance. In any event, as Mr Kitchin contends, "technology invented 16 years after the patent can hardly be used to construe it or render it sufficient", a submission which I quote partly because it tends to support my view that, when construing the reference to EPO isolated from urine in Claim 19, one should do so by reference to methods which existed as at the priority date.
Sugimoto
- Sugimoto purports to teach the preparation of fused cells which produce EPO. The fusion is of a cell from a particular type of tumour and a lymphoblastoid cell. The former cell is from a tumour which produces EPO, and the latter is a type of cell which expresses protein at a high level and will grow continuously. In a nutshell, TKT contends that the effect of what was taught by Sugimoto is that the cells, including their nuclei, will fuse, that this will result in a rearrangement of the DNA in their chromosomes, and that this involves the expression of "recombinant" EPO from "exogenous" DNA in "host cells", at least if those expressions as used in Claims 1 and 19 of 605 have the wider meaning for which Amgen contend. I have held that, as a matter of construction, they have the narrower meaning. Accordingly, this point is, on my view of construction, academic. However, I assume that Claims 1 and 19 have the wider meaning for which Amgen argue. On the basis of that assumption, I turn to consider whether there is anything in TKT's contention of lack of novelty based on Sugimoto.
- I do not consider that any claim of 605 can be fairly said to have been anticipated by Sugimoto, even on this basis. First, as a matter of fact, while it is true that Sugimoto makes certain claims as to what has been, or could be, achieved by its teaching, I consider, on the balance of probabilities, that following the teaching of Sugimoto would not result in the expression of EPO as a result of the fusion of the two cells. It appears that, in connection with the United States proceedings, TKT tried to replicate the results of the teaching of Sugimoto, and it appears that they failed, save to the extent of fusing a lymphoblastoid cell line with TKT's HT1080 cells. However, that takes matters little further in this connection, because these HT1080 cells are the ones used by TKT for the production of its GA-EPO, in which the EPO gene had already been "switched on" artificially.
- Furthermore, given the pretty intensive search which was going on in a number of laboratories around the world for an artificial way of producing EPO in relatively substantial quantities, it is hard to understand why the teaching of Sugimoto was not eagerly taken up and implemented, if its claims were justified. As I have mentioned, it was filed in the United States in August 1981 (and was filed in Japan almost exactly a year earlier) and it appears to have been published in the United States in March 1983. Despite that fact, there is no evidence of anyone trying to follow, let alone succeeding in following, its teaching. The EPO production claimed in Example 6 of Sugimoto is of the same order as Amgen's commercial strain, which makes it more remarkable that there appears to have been no report of EPO ever having been produced or marketed according to the teaching of Sugimoto, given that there was such a strong commercial incentive to produce EPO in large quantities.
- Additionally, there is no evidence that any EPO-like activity measured as a result of the apparent expression in the fused cell described in Sugimoto resulted from the fusion of the cells, or indeed, that any EPO or EPO-like material was produced in vitro. According to Sugimoto, following suspension, the cells had to be sonicated to release the protein which either was EPO or had EPO-like activity. On the basis of the description in Sugimoto, and the evidence and the arguments I have heard, it is at least possible, and it may well be probable, that the protein recovered from the suspension had been made while the tumours were growing in the animals. As Professor Wall stated when he was being cross examined, in order to see whether the cells which Sugimoto claimed to have fused were actually fused, and whether such fusion resulted in the expression of EPO, would require some sort of marker to identify the product of any such fusion. I accept his evidence that there were simply no data to show whether there had been a successful fusion, and whether the reported expression of EPO was coming from the fusion in any event.
- Further, even assuming that Sugimoto achieved the results it claimed, its teaching was not enabling in any event: in other words, I consider that it would have been classically insufficient. In this connection, I have already referred to the fact that it appears that, in connection with the United States proceedings, TKT tried to replicate the teaching and results of Sugimoto, and, indeed, tried to obtain the tumour cells therein described, but were unable to do so. In that connection, there is not only the evidence of Dr Heartlein of TKT, but also the evidence of Amgen. On 4th December 1995, they asked one of the alleged inventors of Sugimoto for "a sample of the erythropoeitin producing cells referred to in this patent as well as a sample of any erythropoeitin produced by such a cell line" and were met with the answer that "we are not in a position to provide such a cell line to any party having nothing to do with our research activities regarding erythropoeitin". While this correspondence obviously does not show that no such cell line or EPO exist, it does provide some support for Amgen's contention that Sugimoto is in any event insufficient, particular when one bears in mind TKT's inability to reproduce the teaching or results of Sugimoto.
- When considering whether any of the claims in 605 can properly be treated as anticipated by Sugimoto, it is, I think, helpful to identify the contribution to the art made by Sugimoto (assuming, in favour of TKT and Roche, that Sugimoto can be taken at face value) when compared with 605's contribution. It seems to me that the contribution of Sugimoto (if any) is pretty similar to that of Professor Murray as described by Lord Hoffmann in Biogen [1997] RPC 1: Sugimoto found a way of making erythropoeitin, and, on this hypothesis, if one followed Sugimoto, one would do just that. However, like Professor Murray, but in contradistinction to Dr Lin the inventor of 605, he did not characterise and sequence the EPO gene. Dr Lin's reported sequencing could then be used for the production of EPO in cells, whether by transfecting cells thought to be suitable in 1984 with an exogenous EPO gene, whether carrying out the same exercise in cells later discovered to be suitable, or whether using newer technology (such as using exogenous promoters to "switch on" endogenous EPO gene in a human cell). All this is what, at any rate on my view, can be said to have been effectively enabled by the disclosure of 605. Accordingly, I do not consider that Sugimoto could fairly be said to anticipate Claim 1 of 605, given that it claims a DNA sequence "suitable for use in securing expression in a ... host cell".
- Quite apart from this, there are difficulties for TKT in light of the drafting of the Claims of 605. So far as Claim 1 is concerned, while it is fair to say that I heard little evidence or argument on the point, I incline to the view that, assuming a protein with EPO-like characteristics was produced by Sugimoto, it would not have been the "product of the expression in a eukaryotic host cell of a DNA sequence", irrespective of the meaning of that expression one adopts. As I have indicated, on any view it seems to me that, to be a "host cell" the cell in question must have DNA with which it has been transfected. While the fused cell in Sugimoto can be said to be artificial in itself, it seems to me that the DNA included in it, whether originally from the lymphoblastoid cell or the tumour cell, would not be foreign or "exogenous" to the new fused cell, which hence would not be a "host" to the DNA which was natural to either cell. Otherwise one would reach the rather odd conclusion that all the DNA in the fused cell was "exogenous", and that it had no endogenous or natural DNA whatever. I appreciate that the fusion was effected by exogenous DNA, namely a Sendai virus, but it was not alleged that this rendered the fused cell a "host cell", and in any event there is no evidence that this virus played any part in the production of EPO once the fused cell came into existence, and therefore, in any event, it cannot be said, in my view, that the cell was, as it were, relevantly a host cell. As to Claim 19, there is no evidence that, assuming Sugimoto's claims are correct, it results in EPO which has a higher apparent molecular weight by SDS-PAGE than urinary EPO.
Cell line prior art
- Six articles in journals and a patent are also relied on by TKT as providing sources of EPO before the priority date claimed by 605. The majority were reported in 1983 or 1984. They are Sherwood et al in Clinical Research 31,323A, Ascensao, reported in Blood 62,1132, Katsuoka et al reported in Gann 74,534, Hagiwara et al, Blood 63(4), 828, Saito et al Exp Hematol 11(4) 228. There is also a report in 1979, Toyama et al Blood 554(1), 245, and a Japanese patent in the same year, JP-A 54-55790, Tajima being the first of its five recorded inventors and applicants. I believe that these can be dealt with relatively shortly.
- The possibility of reproducing the experiments reported in these papers appears to depend on the availability of the cell lines referred to, and there was no evidence that such cells were available. I accept that there was a practice or policy, at least in principle, that, when papers were published, the cells referred to in them should be available. However, as Professor Wall pointed out in cross examination, "there are a number of interesting ways to get around" that general rule, and there was "no way of enforcing that policy". Indeed, in relation to the work of Dr Sherwood (who produced the first of the papers, actually in abstract, to which I have referred) it appears that GI did not obtain a copy of the cell line, and merely reached an agreement that Dr Sherwood would "submit her cell line to independent testing". I turn to consider briefly the various papers to which I have referred.
- Sherwood is, as I have mentioned, an abstract, describing human kidney tumour cells derived from tumours passaged in nude mice, and which were reported as capable of producing EPO. Pursuant to the correspondence to which I have just briefly referred, GI scientists eventually obtained Dr Sherwood's cells, and made cDNA libraries therefrom, but were unable to detect any EPO-encoding mRNA. Further, despite three attempts, it appears that those carrying out the research at GI were unable to get any reliable figures in an EPO assay.
- GI's experience in seeking to follow the teaching of Sherwood, even after obtaining a sample of her cell line, indicates how unsafe it is to rely upon papers in this field, which have not been subject to independent verification by experiment, of which direct (or some other satisfactory form of) evidence is available at trial. At first sight, it might be said that it is unsafe to draw this conclusion in relation to other papers simply from what happened in relation to Sherwood, not least because Sherwood is an abstract. However, in this connection, it is important to bear in mind that Dr Sherwood and her colleague subsequently published their work in a peer-reviewed article in the prestigious Proceedings of the National Academy of Sciences - PNAS 83:165 (1986).
- Ascensao describes a human testicular germ cell line which apparently produced EPO-like activity, but this only occurred after the cultures had been starved for some ten days. Katsuoka reported production of small amounts of EPO in fairly similar circumstances to those described by Sherwood, as did Hagiwara. Saito is a short abstract reporting two experiments, one of which again is similar to Sherwood, and the other described the injection of mRNA derived from tumour cells into frog oocyte cells, which produced something with EPO-like activity. Toyama is another example of work similar to that of Sherwood (albeit earlier) and Tajima also describes a similar activity. I do not consider that, without further ado, any of these reports can be fairly relied on as producing what they allege, particularly in light of GI's experience with Dr Sherwood's cell line, and, indeed, the fact that no one appears to have taken any of this work further, despite the clear desirability of obtaining EPO in appropriate quantities.
- Quite apart from this, it appears to me that, even if any of these papers could be relied on by TKT in principle as being novelty-destroying, they would run into the difficulty in relation to Claims 1, 19 and 26 as would Sugimoto.
N. ADDED MATTER
- By virtue of Section 72(1)(d) of the 1977 Act, if the matter disclosed in the specification of a patent extends beyond that disclosed in the application for the patent as filed, then the patent is invalid. The principles applicable to the determination of whether a patent is invalid on the grounds of added matter are not in dispute. The court must decide whether, when viewed through the eyes of the relevantly skilled addressee, any subject matter relevant to the invention has been added to the disclosure afforded by the application. Although an allegation of added matter, as its name suggests, normally involves the addition of fresh material of some sort, it can apply to deletion. The threefold exercise laid down in Bonzel -v- Intervention Limited (No. 3) [1991] RPC 553 at 574, helpfully explains the approach the court should adopt. In that case, Aldous J said this:
"The task of the Court is threefold:
(1) To ascertain through the eyes of the skilled addressee what is disclosed, both explicitly and implicitly in the application.
(2) To do the same in respect of the patent as granted.
(3) To compare the two disclosures and decide whether any subject matter relevant to the invention has been added whether by deletion or addition. The comparison is strict in the sense that subject matter will be added unless such matter is clearly and unambiguously disclosed in the application either explicitly or implicitly."
- In the present case, the contention that the 605 patent is invalid because of added matter is advanced by Roche (with TKT's support) on the basis of what I have called the deleted matter, namely the paragraph included in Example 10 in 605A and 605B, but deleted as a result of argument and evidence before the Board, and consequently not in 605. The argument is based on the contention that the effect of the inclusion of the deleted matter in Example 10 affected the construction of Claim 19 so as to cut down the scope of protection afforded to the patentee by that Claim, and, accordingly, the subsequent deletion of the deleted matter has resulted in Claim 19 (and indeed the claims depending on Claim 19) being broader in scope in 605, the patent, than in 605A, the application for the patent and 605B, the patent as originally granted. Accordingly, references in this part of the judgment to 605A also extend to 605B .
- As I have explained, when construing Claim 19, it appears to me that, particularly when one reads the Claim through the eyes of the appropriately skilled man, the requirement, that the rEPO has "higher molecular weight by SDS-PAGE" than uEPO, is satisfied if, on SDS-PAGE, the uEPO band, viewed as a whole, runs ahead of the rEPO band, even if there is a substantial degree of overlap between the two bands. Mr Thorley on behalf of Roche (with the support of Mr Kitchin on behalf of TKT) contends that, with the benefit of the deleted matter in Example 10 in 605A, the skilled addressee would have read the patent as indicating a much more substantial difference between the respective apparent molecular weights by SDS-PAGE of uEPO and rEPO, so that, in particular, he would expect there to be no overlap between the bands. In other words, with the benefit of the deleted matter in Example 10, he would expect the uEPO band to be running so much faster than the rEPO band that there would be no overlap between the trailing edge of the uEPO band and the leading edge of the rEPO band.
- This argument is entirely based on the proposition that the skilled man, reading 605A, and in particular the deleted matter in Example 10, would appreciate that the high hexose ratio reported for rEPO, when compared with that reported for uEPO (namely 15.09 against 1.73), meant that there was a substantial difference in the actual, and therefore the apparent, respective molecular weights of rEPO and uEPO. This would have been on the basis that they would both have the same amino acid sequence (and therefore the same "bare" protein molecular weight) but the rEPO would be, to put it simply, more heavily glycosylated, and hence would have a much higher molecular weight.
- In my judgment, the argument that 605 is invalid on the grounds of added matter should be rejected. I consider that the appropriately skilled team (and in particular the post doctoral biochemist with experience of glycoproteins) would have appreciated that the analysis reported in the deleted matter was inaccurate. The suggestion of a hexose ratio of 15.9 (using the N-acetylglucosamine - GlcNAc - as the base 1 level) for rEPO would have been far too high to be believable. In this connection, the evidence of the expert witnesses was virtually unanimous. In the US proceedings, Dr Fritsch described "the carbohydrate composition" described in the deleted matter as "plainly inaccurate". In his evidence before me, Dr Robbins said that one "could really give... no weight at all" to what was reported in the deleted matter. He also confirmed what he said in his deposition in the US proceedings, namely that "any person of ordinary skill in the art of glycobiology in 1984 would understand [the information in the deleted matter] to be grossly inaccurate". Professor Cummings stated that he thought that he "would have probably found it difficult to convince anybody in the field, who [had] analysed glycoproteins, that [he] had found one [sc. a glycoprotein] that had that ratio [sc. namely hexose 15.09] even if I had repeated the analyses". Professor Clausen said that he would have been highly suspicious of the 15.09 figure, and I infer from his evidence that he would not have given the contents of the deleted matter any significant weight. Accordingly, at least if the notional addressee is to be equated with any or all of these witnesses, he would not have believed this figure, and would therefore have effectively discounted the results contained in the deleted matter.
- However, Mr Thorley contends that Dr Fritsch, Professor Robbins, Professor Cummings and Professor Clausen were substantially more skilled, and indeed more specialised in the field of glycosylation of proteins, than any member of the notional team of ordinarily skilled people to whom 605 would have been addressed. He suggests that the notionally skilled team of addressees would not have included someone sufficiently well informed about glycoproteins in 1984 to take this view. He draws support from what was said by Professor Clausen and by Professor Cummings, and also by the reaction of those working for Amgen who commissioned and reported on the work which resulted in the information contained in the deleted matter. Professor Clausen said that he would have had problems with the deleted matter because, as at 1984, he had had two years of training in the world's best carbohydrate laboratory, but he would not expect someone with experience in cell biology, as opposed to an expert in carbohydrate biochemistry, to spot the difference. Professor Cummings accepted that expression of proteins in mammalian cells was in its infancy in 1984 and the kind of glycans that one might see on proteins was unpredictable and had not been the subject of substantial publication. It also is apparent from Amgen's disclosure that none of the people working for Amgen on this project, including Dr Yu, who was accepted by Professor Cummings as a "reasonably respectable carbohydrate chemist", and who performed the analysis which resulted in the information in the deleted matter, raised any question about the 15.09 hexose figure.
- These points all have force, but in the end I am not persuaded by them. Resolving an argument as to whether or not the notional skilled team includes someone with a particular expertise is difficult, not least because most of the arguments appear to me to involve a degree of circularity. In my judgment, one of the members of the notional team to whom 605 was addressed would have had sufficient knowledge of glycoproteins to have serious concerns about the accuracy of the contents of the deleted matter, and that concern would have been sufficiently great to discount its effect. There is no reason to think that Dr Yu, who carried out, and reported to Amgen on, the experiments which resulted in the information contained in the deleted matter, was asked to advise on or consider, the effect of the evidence that he apparently obtained.
- In any event, it appears to me that Roche's argument on added matter faces a number of difficulties. First, subject to one possible point, such a high figure as 15.09 for hexose ratio in rEPO (compared with the unexceptionable level of 1.73 in the case of uEPO) would have resulted in the rEPO having a much larger molecular weight than the uEPO. Professor Cummings estimated that such a high hexose level would roughly double what would otherwise be the molecular weight of the EPO: he said it would increase it by over 30kDa; this can be compared to its actual molecular weight of around 36 kDa. Yet at the end of the immediately preceding paragraph of 605A the rEPO is described as having a "slightly larger" molecular weight than the uEPO. Mr Thorley's argument involves assuming that, while the addressee of 605A will not be sufficiently skilled in glycosylation of proteins to appreciate that the deleted matter is inaccurate, he will be sufficiently skilled in glycosylation of proteins to appreciate that the reported hexose ratio of 15.09 means that rEPO has a substantially greater molecular weight than uEPO, as otherwise he will merely note what is in the preceding passage, and conclude that the difference in apparent molecular weight is not very great. In my view, Mr Thorley's argument therefore faces a squeeze. If the addressee is sufficiently well informed to appreciate that the figure of 15.09 in the deleted matter results in rEPO having a very much higher molecular weight, then he will also have sufficient knowledge to appreciate from his general knowledge and from the previous paragraph that the figure is unreliable. On the other hand, if he does not appreciate that the figure is unreliable, he will not be sufficiently skilled to appreciate that the 15.09 means that rEPO has a much higher molecular weight than uEPO, and will therefore rely on the reference to the difference being "slight". I do not think that Mr Thorley's argument is sustainable in light of this squeeze. In any event, as I have indicated, it appears to me that this notional addressee will appreciate the inaccuracy of the contents of the deleted matter.
- Quite apart from this, even if Mr Thorley's argument can avoid this "squeeze", and the addressee thinks that the deleted matter results in rEPO having a substantially higher molecular weight than uEPO, he will appreciate that there is a clear inconsistency between the two successive paragraphs of Example 10 in 605A. The first expressly states that there is a "slight" difference between the two molecular weights, whereas the second merely implies that there is a large difference between the two molecular weights. Faced with an inconsistency between express and an implied teaching, I would have thought that the reader would either assume that the express teaching is correct, or would regard both statements as being unreliable for obvious reasons. Whichever view he took, the deleted matter could not affect his interpretation of Claim 19.
- That is not the end of the difficulties faced by Roche in its added matter argument. First, it appears to me that the whole basis of the argument falls foul of the fourth principle identified by Staughton LJ in Glaverbel. The added matter argument involves invoking the deleted matter, which is contained in a paragraph in an example in the specification, to cut down Claim 19, which, in this connection, I regard as "expressed in clear language", at least once one understands the heterogeneity of the glycans in glycoproteins, and the consequence on SDS-PAGE.
- Quite apart from this, the whole of Mr Thorley's argument in relation to added matter (and indeed the discussion in this judgment so far) proceeds on the assumption that the hexose ratio of 15.09 reported in the deleted matter in relation to rEPO would inevitably result in the rEPO having a much higher molecular weight. Although that was the primary evidence of Professor Cummings, he said at one point in his evidence that there was one possible (albeit, as I understood his evidence, unlikely) way in which a skilled person who read and understood the implications of the deleted matter could nonetheless conclude that the reported hexose ratio of 15.09 was accurate. This would be on the assumption that there was only one N-glycan in rEPO. As I say, I do not think that is what a reader of 605A who understood the implications of the deleted matter would have concluded, not merely because it did not convince Professor Cummings, but also because it was scarcely consistent with the evidence of the other experts as to the view they would have taken on the deleted matter in 1984. However, if, contrary to my view, that is the conclusion which would have been reached by the addressee, then the consequence would have been, as Professor Cummings explained, that the molecular weight of rEPO, with only one N-glycosylation site, would not have been perceived as likely to be very much greater than that of uEPO with a hexose ratio of 1.7.
- Accordingly, my primary conclusion is that the deleted matter would have been discounted by the notional addressee of 605A, because it was simply not believable and was inconsistent with other passages in the patent, so it would have been given no effective weight. If it would have had no effective weight in the mind of the reader of the patent in question, then its removal cannot by definition, have any effect on the construction and teaching of the patent.
- Even if I am wrong in the view that the deleted matter would simply be rejected by the notional addressee, when reading 605A, I would still reach the same conclusion on the argument that 605 is invalid on the ground of added matter. Claim 19, whether in 605 or in 605A, seeks to distinguish the performance of uEPO from that of rEPO by reference to their apparent molecular weights on SDS-PAGE, not by reference to any difference in the nature or extent of their respective glycosylations. Accordingly, construing that aspect of Claim 19 by reference to the rest of the patent, it appears to me that the essential aspect of Example 10 of 605A is to be found in its reference to rEPO having a "somewhat higher" apparent molecular weight than uEPO, and not in the passage relating to the apparent differences in glycosylation.
- In these circumstances, the exercise laid down in Bonzel appears to me to be tolerably easy, because, although, through the medium of the deleted paragraph, information is disclosed in the application which is not disclosed in the patent as granted, the effect of that information on the skilled addressee is such that the meaning of Claim 19 (and of Claim 20) in 605A is identical to the meaning of those Claims in 605, and therefore no real question of added matter arises.
O. DO ROCHE INFRINGE 605?
- Amgen contends that Roche infringes by disposing a product which had been obtained by means of the processes of Claims 27 to 29, and/or by disposing a product within Claims 19 to 21, 23, 26, 30 and 31, of 605.
- So far as Amgen's case based on Claim 26 is concerned, the issue between the parties really relates to Claim 1 and Claim 2, which are, of course, incorporated by reference into Claim 26. In light of the nature of the dispute between Roche and Amgen in this connection, it is unnecessary for me to describe in any detail the process carried out by Roche. Suffice it to say that the dispute between the parties arises from the HuEPO cDNA sequence used by Roche. The encoding regions of that sequence are identical to those shown in Tables VI of 605. The downstream 3' untranslated region is also identical; however there are changes in some of the untranslated upstream 5' region.
- Accordingly, Amgen's primary case is that Roche infringe because the DNA used in that process falls within integer (a) of Claim 1 as a matter of construction. I determine that issue against Amgen, essentially for the reasons given in this judgment for favouring the narrow, at first sight perhaps startlingly narrow, construction of that integer of Claim 1. I have concluded that integer (a) is limited to DNA sequences which are specifically identified in that integer, namely only those sequences set out in Tables V and VI of 605. Roche use a DNA sequence which, while it has complete identity with the corresponding protein encoding region and the downstream sequence shown in Table VI (with the exception of the misprint to which I have already referred), has differences in the sequences which are upstream of the first exon. Accordingly, the DNA sequence used by Roche does not fall within the strict words, as I have construed them, of integer (a) of Claim 1.
- If integer (b) and integer (c) were not included in Claim 1, so that it was, in effect, limited to integer (a) alone, and if Amgen had included the point in their opening, I would have thought it right to go on to consider whether the application of the three Improver or Protocol questions would nonetheless lead to the conclusion that Roche infringed. It is unnecessary (and arguably inappropriate) to embark upon that exercise, but it is right to say that there would have been an obviously attractive argument in favour of infringement in such circumstances. As Mr Waugh says, at least in the context of Roche's technology, the essential ingredient for the purpose of expression, whether one is using cDNA or genomic DNA, are the encoding regions, and Roche uses precisely the encoding sequence disclosed in Table VI, nucleotide for nucleotide.
- However, consideration of the three Protocol questions is, in my judgment, unnecessary in this context. It seems to me quite clear that, in conditions which anyone skilled in the art would regard, and would have regarded, as stringent, the DNA used by Roche would hybridise in accordance with the requirements of integer (b). As I have said, the encoding regions identified in Table VI and the encoding regions of the DNA used by Roche are identical, and extend to around 500 nucleotides. The idea that one of these DNA strands would not hybridise in stringent conditions with the complement of the other seems to me fanciful. Accordingly, even if (as I believe is the case) there is room for argument as to whether some conditions are "stringent", it appears to me that integer (b) of Claim 1 is infringed by Roche, as a result of which Roche's defence to the contention that it infringes Claim 26 must fail, even if I am wrong on integer (a).
- In these circumstances, it is unnecessary for me to consider the same argument in relation to Claim 2. However, it is right to add that, in my judgment, Amgen would also succeed on Claim 26 in so far as it incorporates Claim 2: the DNA used by Roche seems clearly be a DNA "encoding for human erythropoeitin". Mr Thorley suggests that that expression is unclear, but I cannot see why that should be. I have discussed the meaning of Claim 2 when dealing with construction. It appears to me to contain a perfectly satisfactory description of any DNA sequence, such as that used by Roche, which, if inserted into an appropriate cell in appropriate conditions, will cause the expression of the protein, human erythropoeitin.
- In light of this conclusion, I believe that it follows that Roche's process also infringes Claim 27, in so far as it is based on Claim 1 or Claim 2. Similarly, I believe that Roche infringes Claim 28, in so far as it based on Claims 12, 13, 14 and 16, in so far as those Claims in turn are based on Claims 1 and 2.
- In so far as Amgen
• contends that Roche infringes Claim 29, the allegation is only to the extent that Claim 29 is based on Claim 19, 20, 21, 23 and 26. So far as the first of those four Claims are concerned, it appears to me that they are invalid on grounds of insufficiency, and, even if that is wrong, they are incapable of infringement. However, in light of my reasoning, I believe Roche infringes Claim 29 but only in so far as it is based on Claim 26. Roche also infringes Claims 30 and 31, but only in so far as they are based on Claim 26.
P. DO TKT INFRINGE 605?
Introductory
- Amgen contend that TKT infringe Claims 19, 20 and 22. They also contend that TKT infringe Claim 26, in so far as it is based on Claim 1, 2, 5, 6 and 7. The claim against TKT based on Claims 19, 20 and 22, it must fail. Even if I am wrong in my view that Claim 19 is invalid, it appears that the findings of fact I have made when considering the insufficiency of Claim 19 would render it incapable of infringement. Given that Claims 20 and 22 are each based on Claim 19, it must follow that the most important claim for infringement against TKT is under Claim 26.
The issues raised by the allegation of infringement against TKT involve a closer understanding of the technology used by TKT, and, in light of my conclusions as to the proper construction of Claim 1 (being the principally relevant claim upon which Claim 26 is based), consideration also has to be given to the Improver or Protocol questions. Further, although I have gone over much of the ground when considering the permissible breadth of the claims of the patent, I believe that it is also appropriate to consider the guidance given by the House of Lords in Biogen [1997] RPC 1 in this connection. I propose, therefore, first to explain what is involved in TKT's technology, being the technology whose product is said by Amgen to infringe Claim 26 of 605. I will then turn to the application of the Protocol questions to that technology and its product. I will then discuss the consequences, including consideration of the impact of the reasoning in Biogen [1997] RPC 1.
TKT's technology
- In summary terms, TKT's activity involves gene activation technology, which I have briefly explained when considering common general knowledge, and in particular homologous recombination. In most human cells, the EPO gene (like almost any other gene) is "switched off" by a negative regulatory element ("NRE"). TKT introduce into the genome of a human cell, upstream of the EPO gene disclosed in Table VI of 605, a nucleotide sequence which effectively overrides the NRE, and "switches on" the EPO gene. This nucleotide sequence is a DNA "targeting construct" which does not include the encoding region. This construct contains a powerful viral promoter, called the CMV promoter, which is more effective than the SV40 promoter referred to in the 605 patent, and in particular in Examples 7A and 7B. Thus, after this treatment, TKT's cell line contains endogenous EPO DNA so far as the coding regions are concerned, but it also contains an exogenous promoter construct inserted upstream of that endogenous DNA.
- What I have just described is a somewhat over-simplified version of TKT's technology whereby they produce their product, known as GA-EPO. It involves the process known as homologous recombination, which I have already briefly mentioned. As explained by Dr Brenner, it is "the process whereby two DNA sequences which have a region of identity can exchange material at that region." As he explained, it is a natural process used by all living systems to generate new combinations of genetic material. Targeting of mammalian genes using homologous recombination, i.e. adapting this natural process to enable genes to express proteins artificially, was first disclosed in 1985 by Smithies et al in Nature 317:230. Before that time, and therefore before the relevant date for the purpose of the 605 patent, (as I have already indicated) it was a procedure which had not been used.
- It is necessary to describe TKT's process in a little more detail. The targeting construct is a piece of DNA which is constructed outside the cell in which the EPO is to be produced. The targeting construct can conveniently be divided up into seven sequences. First, a so-called targeting sequence of some 3,000 nucleotides upstream of the EPO gene (-5787 to -2482). Secondly, a DHFR gene, whose function is the same as in Amgen's technology. Thirdly, a CMV gene promoter for activation of the endogenous coding regions. Fourthly, a sequence encoding the first three and a third codons of the human growth hormone ("hGH") peptide. Fifthly, a selectable marker gene, neomycin phosphotransferase ("neo") to allow selection of stably transfected mammalian cells. Sixthly, an EPO splice donor site, to enable correct processing of the primary transcript following transcription. Seventhly, another targeting sequence of some 1800 nucleotides upstream of the EPO gene (-2481 to -675).
- This targeting construct consisting of these seven components is known as pREPO 22. TKT's technology involves introducing it into a tumour cell line known as HT-1080. As the targeting construct contains sequences at each end of more than 1500 nucleotides in length which are precisely homologous to sequences upstream of the natural EPO gene, the genome will recombine with these sequences by homologous recombination. Ideally, therefore, the targeting construct is effectively inserted into the genome a few hundreds of bases upstream of the EPO encoding sequences in chromosome 7. In other words, the targeting construct is inserted just upstream of the beginning of the sequence disclosed in Table VI of the 605 patent. The insertion is effected by the DNA in the first and seventh components being wholly homologous with these endogenous upstream sequences and therefore combining with them. The "switching on" of the endogenous DNA encoding sequences is achieved by the CMV promoter. The function of the other components of the construct has already been briefly described.
- Although the targeting construct is intended to be inserted into chromosome 7 (being the chromosome in which the EPO gene is located), homologous recombination in mammalian cells is not efficient or reliable, and there will be insertions at sites in other chromosomes. The neo can be used to separate out those cells which have taken up the targeting construct, and TKT then screens those transfected cells for EPO production, because it will only be those cells where the construct has been inserted upstream of the EPO coding sequences where this will occur. Thereafter, TKT treat the cells with MTX, a standard procedure which I have already described and, indeed, is part of the 605 patent's teaching, given that it was a standard technique in 1983.
- As I have already explained, the cell will amplify the inserted DHFR gene, and, because of its proximity thereto, the EPO gene will similarly multiply (again, this is described in the 605 patent, in Example 10). The normal type of cell used in this procedure was the DuX-B11 CHO cells, also referred to in Example 10. Only those cells which contain sufficient copies of the DHFR gene, and, therefore, also multiple copies of the EPO gene, will survive the MTX treatment. In TKT's technology, those cells were designated as R223 cells.
- After homologous recombination of the exogenous pREPO 22 construct and the endogenous DNA, the CMV promoter and hGH fragment are integrated into the genomic DNA upstream of the natural promoter of the EPO gene. Subsequent thereto, the following steps occur in the R223 cell line:
1. Transcription of the EPO gene is initiated by the CMV promoter and proceeds along to the hGH and EPO splice donor site sequences upstream of the natural promoter, and along the five exons of the endogenous EPO gene, thereby producing the primary transcript;
2. The primary transcript is then spliced resulting in a variant of natural EPO mRNA. The splicing involves the exogenous hGH sequence being spliced to the next available downstream splice acceptor site. This is so designed as to be the sequence flanking the second exon of the endogenous EPO gene. Accordingly, the first encoding exon is not present in TKT's EPO mRNA: it is spliced out. It is not required, because it is part of the leader peptide (the rest of the leader peptide being the upstream part of the second encoding exon) which, as I have explained, is cleaved at an early stage, resulting in relatively "mature" EPO;
3. Translation of the EPO mRNA produces immature EPO whose leader peptide sequence is a hybrid between the first three and a third codons translated from the hGH fragment and the first twenty two and two thirds codons translated from the second encoding exon of the endogenous gene;
4. Finally, the leader peptide is removed, and mature EPO is secreted from the cell.
Differences in TKT's technology
- Mr Kitchin says that there are a number of differences between TKT's technique, and the teaching of the 605 patent, so far as the production of EPO is concerned. First, as I have already mentioned, Amgen's process involves the isolation and cloning of an exogenous DNA sequence encoding EPO, which is then used to transform or transfect a host cell. On the other hand, TKT's technique does not involve the isolation or cloning of any EPO-encoding sequence: they use the endogenous encoding sequence in human cells. Dr Brenner described this as "the fundamental difference" between the two techniques, and Professor Proudfoot considered that it was "too big a difference to just dismiss as being an improvement". Whether or not Professor Proudfoot is right, this is clearly a significant and marked difference.
- Secondly, TKT's technique is of general application, in the sense that it can be applied to any endogenous gene. On the other hand, the traditional cloning approach of Amgen requires transformation of a cell with a gene of interest, and this can only be done with genes which could be inserted into cloning vectors, which is a limiting factor, in that the transformation of a cell with an exogenous gene of more than 10,000 bases is difficult, and even more difficult with genes of more than 20,000 bases. I do not consider that this is a particularly telling point, in the sense that it does not seem to me to take the argument much further than the first point takes it. I suppose it can be said to emphasise the existence and extent of the difference between the two techniques.
- Thirdly, TKT's technique enables the promoter to be placed at a range of positions upstream of the endogenous gene, so that, as the evidence suggested, its position can be varied with a view to finding the optimum location for ultimate maximum expression. The technique described in the patent requires the promoter to be relatively close to the coding sequence, to avoid the possibility of a false translation initiation or an intermediate stop codon. The evidence as to the real benefit of being able to change the position of the promoter using TKT's technology appeared to me to be a bit dubious, in practice, but it is a feature of difference. However, it seems to me that this third factor is of little weight. Like the second factor, it can be said to underline the difference between the two techniques, but once one understands the two techniques, it does not seem to me that this is a factor which takes matters much further: it is a potential advantage arising from the nature of one of the techniques. Mr Kitchin accepts, quite rightly in my view, on behalf of TKT that if, for instance, TKT was to use the technology described in the 605 patent, but using the more powerful CMV promoter instead of the SV40 promoter described in the Examples of 605, that would involve infringing the patent. This underlines the point that the mere fact that an improvement, or possible improvement to the process disclosed by the patent, such as the ability to relocate the promoter, does not of itself render a new process non-infringing.
- I think this point also casts doubt on the fourth factor relied on by Mr Kitchin. Amgen's teaching involves the removal of the natural EPO promoter regulatory sequences from the exogenous gene and coupling that gene to an exogenous promoter, which is then integrated randomly into the chromosomes of transformed cells. As against this, TKT activate the EPO gene in its natural environment by targeting an exogenous promoter to a specific location upstream of the endogenous encoding region. In any event, it appears to me that this is not really a new difference, but is a combination of the first and third differences.
- The fifth difference identified by Mr Kitchin is that Amgen's technology, at least as taught in the 605 patent, could only be successfully undertaken in DHFR- CHO cells, whereas TKT's technology appears to allow one to select from a virtually limitless number of cell lines, provided they contain the endogenous EPO gene. This appears to me to be a difference which, at least in the main, is really inherent in the nature of the two techniques. Quite apart from this, as I have already mentioned, I believe that the 605 patent is entitled to extend to all eukaryotic cells.
- Sixthly, it is argued on behalf of TKT that there is a difference in the EPO produced by the two techniques and, indeed, that the DNA which indirectly expresses the EPO is also somewhat different under the two techniques. The evidence on this point went little further than speculation. In the United States proceedings, the TKT witnesses stated that they were not aware of any advantages of TKT's product over Amgen's product, and in these proceedings Professor Proudfoot could not point to any advantage. According to their own internal documentation, TKT regarded the product as "me too" and they described there being "nothing new here".
- I accept that it is possible that the nature and degree of glycosylation between the two types of EPO could vary. I have already referred to experimental evidence, papers in journals, and opinions of witnesses given in this case, which relate to that issue. Thus, there appears to be a small difference in apparent molecular weight between CHO rEPO and COS rEPO, which suggests differences in the aggregate molecular weights of the glycans attached in the two types of cells. There is plainly far less glycosylation of recombinant EPO produced in at least insect cells (according to the 1987 paper of Wojchowski (op.cit.)). However, it is fair to say that the SDS-PAGE experiments carried out on TKT's GA-EPO against urinary EPO suggests that TKT's product differs very little, if at all, in apparent molecular weight from COS or CHO rEPO produced in accordance with the teaching of 605.
- On the basis of the evidence I have seen, it appears to me that any difference in glycosylation between GA-EPO, and COS or CHO rEPO produced in accordance with the teaching of the 605 patent, is non-existent in biochemical and therapeutic terms, and any such difference is speculative and, if it does exist, of no apparent significance in chemical terms. So far as therapy is concerned, GA-EPO has the same effect as EPO produced in accordance with the teaching of 605. On the evidence I have heard, there is nothing to suggest that the difference between GA-EPO and COS rEPO or CHO rEPO is any more significant than the difference between rEPO produced from COS cells and rEPO produced from CHO cells, both in accordance with the teaching of 605. Chemically speaking, in light of the evidence of Professor Robbins, to which I have referred when discussing the insufficiency of Claim 19 of 605, it seems to me quite possible that the heterogeneous nature of their glycosylation of GA-EPO is no different, or at any rate little different, from that of recombinant EPO produced in accordance with the teaching of 605.
- So far as the difference in the EPO gene is concerned, Professor Proudfoot suggested that in TKT's R223 cells, the EPO gene would be methylated, whereas in the cells containing exogenous DNA expressing EPO in accordance with Amgen's teaching, the gene would not be methylated. As Mr Watson submits, this may be of interest to academic scientists, but there was no suggestion in the evidence that methylation has any relevant effect. Indeed, Dr Brenner stated that it would have no effect, and I do not consider that there are any grounds for doubting that.
- Seventhly, it is contended on behalf of TKT that its technique has miscellaneous advantages over that described by the 605 patent. Those advantages may be summarised as follows. First, there may be advantages in having a protein expressed from an endogenous gene in its local chromosomal environment: the evidence on this was vague, and not supported by any specific facts or features, even though the two techniques have been used for some time. Secondly, the level of expression per gene copy in TKT's R223 cells is significantly higher than that recorded in the patent. However, this does not result in any change in the nature of the final product, and there is little difference in the production of EPO per cell under the two systems. Thirdly, there is the advantage of being able to move the exogenous promoter in TKT's system; I have said, the evidence on this was not satisfactory or clear, but I accept that there is a potential benefit in this connection. Fourthly, TKT suggest that there was some advantage in their system in that they did not have to amplify to the extent described in Example 10 in the 605 patent. I think that is a matter of little significance; on the basis of Professor Proudfoot's evidence, it seems to me that, at best from the viewpoint of TKT, the difference between the two processes could be no more than a month if Amgen's process required an extra amplification step.
- Quite apart from the fact that these advantages do not seem to me to be of much significance in themselves, it is right to say that I am not convinced that advantages of this sort can make any difference to the question of whether the process by which GA-EPO is produced, or GA-EPO itself, infringes 605. It appears to me that, having considered the various points raised by Mr Kitchin, the important and essential difference, and I can well see how it could be called a fundamental difference, between the process used and taught by the 605 patent, and the process undertaken by TKT, is encapsulated in the first difference I have identified. That distinction, "the fundamental difference" as Dr Brenner put it, is between isolating and cloning an encoding exogenous sequence and transfecting a host cell with it, and inserting an exogenous promoter and facilitating sequences into a host cell to "switch on" the endogenous coding sequences.
Does TKT infringe Claim 26: the Protocol questions
- In light of my conclusions on the issues of construction relating to Claim 1, TKT do not infringe Claim 26, on a literal construction of the patent. Although it can be said that the targeting construct is a "DNA sequence for use in securing expression [of EPO] in a .... eukaryotic host cell" in the sense that the targeting construct is "a DNA sequence" and it is "used in securing expression" of EPO in what is to the targeting sequence "a eukaryotic host cell", I do not think that it falls within the composite expression contained in the first part of Claim 1 of the 605 patent. As explained when discussing questions of construction, any set of words has to be construed in its context. What the 605 patent is concerned with, and what it teaches in its specification, is the expression of EPO by means of exogenous EPO-encoding DNA, and, equally, a cell which is host to that exogenous encoding DNA. It is inappropriate to give the words "for use" the wider meaning for which Amgen contend, and without which TKT's technology would be outside the scope of Claim 1 as a matter of literal interpretation. The patent contains no teaching whatsoever so far as using exogenous non-encoding DNA to "switch on" endogenous EPO-encoding DNA. It makes no reference to such technology; indeed, this is scarcely surprising as, although it may have been hoped for at some time in the future, it was at that time simply not part of the state of the art. It would not have been in the mind of the skilled addressee, and he would not have read the claim as extending thereto.
- However, that does not automatically mean that TKT do not infringe. As is clear from the discussion in American Home Products [2000] IP&T 1308, it is then necessary, in accordance with the requirements of the Protocol, for the Court to pose itself the question characterised by Hoffmann J in Improver [1990] FSR 181.
- The three questions which have to be asked were set out by Hoffmann J in Improver at [1990] FSR 189. Basing himself on the reasoning of the House of Lords, in Catnic [1982] RPC 183, Hoffmann J said that the proper approach is as follows:
"If the issue was whether a feature embodied in an alleged infringement which fell outside the primary, literal or acontextual meaning of a descriptive word or phrase in the claim ("a variant") was nevertheless within its language as properly interpreted, the Court should ask itself the following three questions:
1. Does the variant have a material effect upon the way the invention works? If yes, the variant is outside the claim. If no -
2. Would this (i.e. that the variant had no material effect) have been obvious at the date of publication of the patent to a reader skilled in the art? If no, the variant is outside the claim. If yes, -
3. Would the reader skilled in the art nevertheless have understood from the language of the claim that the patentee intended that strict compliance with the primary meaning was an essential requirement of the invention. If yes, the variant is outside the claim."
- It may be that the way in which these questions were expressed would seem to be a little narrow to those who practice in jurisdictions where there is a sharp demarcation between construction and infringement (for instance in Germany, where those issues are tried separately, indeed by different Courts). In a sense, by tying oneself down to construction, it might be said that one was concentrating more on form than on substance. I suspect that, free of the binding constituents of Catnic [1982] RPC 183, the questions might not have been expressed in terms of construction. After all, the claim in Improver [1990] FSR 183 referred to a "helical spring": as a matter of language that could scarcely include a "rubber rod", as Hoffmann J recognised at [1990] FSR 187. Yet Catnic [1982] RPC 183 enabled him to conclude that the rubber rod could be within the claim. It may be that, shorn of authority, the question should be characterised as whether the rubber rod infringed the claim bearing in mind the technical contribution of the patent and not merely the words of the claim.
- However, this is perhaps playing with legal categorisation. There is no doubt that the three questions are useful, important and authoritative (having been approved and applied by the Court of Appeal on a number of occasions, most recently, perhaps, in American Home Products, as I have mentioned). However, it is worth remembering that they are based upon Lord Diplock's observation in Catnic [1982] RPC 183 at 242 that the language of a claim "should be given a "purposive" and not necessarily a literal construction" (per Hoffmann J in Improver at [1990] FSR 188 to 189). While, particularly at first instance, it would be brave or worse to depart from the three questions, I do not find it entirely easy to accept that it can be right that, where there is no infringement on a literal construction, the Court is in every case to be forced into the straightjacket of answering the three specific questions in order to decide issues of construction or even issues of infringement.
- Before answering the three questions, it is, of course, necessary to identify what Hoffmann J called "the variant". This, involves, I think, a three part exercise, namely analysing what the patent in suit claims, identifying the alleged infringement, and then distilling the essential variation or variations between the two techniques or products. Earlier in this judgment, and in the preceding section of this part, I have tried to undertake the first two parts of the exercise. I shall concentrate on Claim 26, because, unlike the other Claims which TKT are said to infringe, it is, at least in my view, valid and capable of being infringed. Claim 26 really takes one back to Claim 1.
- The nature of the variant on Claim 1 in TKT's technology is that the EPO-encoding DNA has not been isolated and the encoding DNA is not in a "host cell" (these two features being, at least to an extent, different ways of putting the same point). The "fundamental difference" between the two techniques has been identified by Dr Brenner, as I have mentioned. Amgen's teaching involves transfecting a cell with exogenous encoding DNA (together with its natural exogenous promoter or an artificial promoter) and this exogenous DNA is then incorporated into the endogenous genome, where it serves to express EPO. TKT's technology involves constructing a DNA sequence with various components, including an artificial promoter, but not including any EPO-encoding sequence, and inserting the construct into a human cell with a view to the construct being incorporated into the genome upstream of the EPO-encoding sequence, whereupon that sequence is "switched on" and results in the expression of the EPO. It is accepted by TKT, as I have mentioned, that the identity or location of the exogenous promoter is not a relevant variant.
- I turn then to the first question. As Hoffmann J pointed out in Improver at [1990] FSR 191, the answer to the first question "depends upon the level of generality at which one describes the way the invention works." He went on to say, on the following page, that "the right approach is to describe the working of the invention at the level of generality with which it is described in the claim of the patent." In this context, it appears to me that Claim 1, consistently with the whole thrust of the specification, and, indeed with commercial common sense, indicates that the patentee is getting at the production of EPO. The obtaining of the amino acid sequence of EPO and of the sequence of nucleotides in the EPO gene is, in the context of the patent, a means to an end, even if, to an academic scientist, it could be said to be an end in itself. Claim 1 does not merely identify the DNA sequences claimed, but it identifies those which are suitable for a certain purpose, namely the production of EPO. That is even clearer when one goes to Claim 26, which is concerned to claim the EPO produced in this way.
- In my judgment, the variant involved in TKT's technology does not have a "material effect on the way the invention works". In each case, one has what was referred to in the evidence as an "identical string of DNA", namely the encoding regions of the EPO gene, expressing EPO in the conventional and natural (albeit artificially massaged) way.
- Mr Kitchin in his effective submissions, contends that the variant does have a material effect on the way the invention works and relies on the contention that "an important part of [Amgen's] invention is the disclosure of a route to the expression of EPO" and that that route "is fundamentally different to that used by TKT". That point obviously has attractions, but I do not accept it. First, it seems to me to concentrate more on the teaching of the patent in the specification, and not on what the relevant claims assert. In this connection, it is true, of course, that, as a matter of literal construction, the claims do not extend to TKT's technique, but that cannot be the determinative factor when considering the first question. If it were, then the answer to the first question would always be in the negative, and the whole exercise of going through the three questions would be pointless.
- My broader reason for rejecting Mr Kitchin's contention in this connection is, at least to an extent, substantially along the same lines as my reasoning for favouring Amgen's arguments on the issues of Biogen insufficiency and breadth of claim. The essence of the invention embodied in 605, its contribution to the art or what one might call its inventive concept, is the disclosure encapsulated in Table VI, which contains the whole of the encoding sequences, the whole of the intervening introns, and a large proportion of the upstream and downstream sequences. It enabled that to be done which was previously impossible, namely the production of EPO in accordance with biotechnological methods, as they existed at the relevant date as they would have been expected to develop and improve over the ensuing years. It seems to me that, in a fast developing technology such as that involved here, it would have been inconceivable to the notional reader of the 605 patent at the relevant date that there would not be significant developments and changes in the technology of genetic engineering over the life of the 605 patent. What TKT have done is to use a new technique, homologous recombination, to achieve EPO expression by the natural EPO-encoding sequences.
- Indeed, this is consistent with the evidence of Professor Proudfoot who said there was nothing new about "an SV40 promoter, the gene of interest, DHFR for amplification placed in a CHO cell, that was all standard in 1984". He accepted that "the basic invention was the isolation and sequencing of... the EPO gene followed by the disclosure of a route to its expression." He said that TKT "certainly uses newer technology to achieve that result". While he immediately went on to say that he did not think that was the only difference, it appeared to me that the other differences he identified were either inherent in TKT's technology (e.g. the ability to target the promoter) or were features admitted on behalf of TKT not to assist on the issue of infringement (e.g. the stronger promoter). In other words, the only significant variant is the use of newer technology, which did not exist in 1984
- The teaching of the 605 patent and TKT's technology involve many of the same essential features. They employ the same EPO encoding sequences; they involve expressing the same EPO artificially; the do so in a eukaryotic cell; they employ an exogenous promoter; the biochemical/chemical way in which they express the EPO is substantially the same. Neither could be achieved without the essential disclosure - the contribution to the art - of the 605 patent itself (as I shall explain in a little more detail below).
- I turn to the second Protocol question. In Improver, at [1990] FSR 192, Hoffmann J emphasised that this question did not involve limiting possible infringement to a variant "which would have suggested itself to the skilled man as an obvious alternative to the thing denoted by the literal meaning." He went on to say:
"In my view the question supposes that the skilled man is told of both the invention and the variant and asked whether the variant would obviously work in the same way. An affirmative answer would not be inconsistent with the variant being an inventive step."
Accordingly, the fact that TKT's technology would not have been known to the notional addressee of the 605 patent, and, indeed, the fact that the technology involved in TKT's technique was inventive as at the relevant date (as I strongly suspect it to be) does not go to the determination of the second question.
- TKT contend that the answer to the second question is no, because the reader of the 605, skilled in the art as at the relevant date, would not know that the technique employed by TKT would work, and, even if one is to assume that he believed that it could work, he would not know that it would have no material effect upon the way in which the invention claimed in 605 would work. Thus, Mr Kitchin points out that the evidence supports the view that a person skilled in the art as at the relevant date would not know that splicing would work. He also says that there is nothing to suggest that he would appreciate that the introduction of a targeting construct such as pREPO 22 would have no material effect upon the level of or manner of expression of the endogenous encoding EPO gene in a human cell, when compared with the level and manner of expression of an exogenous encoding gene and associated promoter introduced in to a CHO cell.
- Mr Kitchin's first point raises an issue of principle in relation to the second Protocol question. It is whether one should assume that the reader of the patent is not merely told of the variant, but, if it is necessary, he is also to be told that it works. In my view, it should be assumed that the notional reader is so informed. The point did not arise in Improver, and it was not suggested that any other authority in which the questions were considered (including American Home Products at [2000] IP&T 1320 to 1322) dealt with this particular issue.
- In my judgment, it is to be assumed that the reader of the patent is told that the variant works, at least in a case such as this, if he would not in fact have known. The point is not an entirely easy one, not least because, if my conclusion is right, it could be a matter of argument as to quite how much the notional reader is told. It can be said with some force that the more he is told, the more likely the second question is to be answered in favour of the patentee, on the basis that, in the ultimate analysis, he will be told everything including the answer to the first of the three questions. However, I do not regard that as a particular difficulty, because it seems to me that all that the notional reader need know is the nature of the variant and the fact that the variant works, i.e., in the present case that it results in the expression of EPO in the human cell by virtue of the endogenous EPO-encoding DNA being "switched on".
- I believe that the opposite view would be inconsistent with the underlying intention of the Protocol, particularly in the context of fast moving technology, such as that involved in the present case. If I am right in the present case in holding that 605 can validly extend to analogues of EPO and variants of the natural EPO gene, it would seem somewhat inconsistent if the Claims were incapable of extending to subsequent inventive techniques, or even to inventive new versions of existing techniques, subsequent to the date of the patent, simply because the notional reader skilled in the art at the relevant date would not have known whether they would have worked or not. That seems inconsistent with "fair protection for the patentee". Obviously, there must also be "fair protection" for any subsequent inventor. However, if his invention involves a new way of doing that which could not be done without the disclosure of the patent, it is not apparent to me that the new inventor should be able to "scoop the pool", thereby obtaining the benefit of the old invention, which properly belongs to the original patentee, as well as obtaining the benefit of the new invention, which does properly belong to him, and which can be protected by way of a new patent.
- The second point raised in relation to the second question is whether, in this case, the reader skilled (albeit non-inventive) in the art at the relevant date, on being told of the nature and feasibility of TKT's technology, would have thought it obvious that it would not affect the way in with the invention disclosed by 605 works.
- While I accept that Mr Kitchin is right in his argument that the reader would have expected, or at least suspected, that there might well be differences in details, such as rate of expression, and nature and extent of glycosylation of the EPO expressed, I do not consider that he would have had doubts about the existence of any variation in what I might call the essential features of the invention embodied in the 605 patent (and Claim 1 in particular), or the essential features of the way in which that invention works. As with the first question, it appears to me that, in order to identify "the invention" in the context of the second question, one must go to the claim or claims or the patent in suit. In the present case, Claim 1 does not condescend to the sort of detail which Mr Kitchin raises as a ground for answering the second question in the negative. Even if that is too narrow a basis for considering the second question, I would still answer it in the affirmative. The way in which the invention works in this case (whether one looks at the patent in suit or at TKT's technology) is the use of the natural EPO-encoding DNA sequence resulting in the expression of EPO in a cell where the genome has been artificially manipulated. It is the nature of the artificial manipulation which is the difference between the two systems: in the teaching of the patent, the traditional method of inserting the encoding sequence (possibly plus introns and an artificial promoter) is used, whereas TKT's newer technology involves switching on the endogenous encoding sequence by means of an inserted artificial exogenous sequence including an artificial promoter. If this latter technology were explained to the notional reader at the relevant date, and he was also told that it worked, I consider that he would have concluded that it was obvious that it worked in the same way as the more traditional technique described in the patent. The essential point, as I see it, is that EPO is expressed in a eukaryotic cell through the medium of a DNA sequence which is in each case identical, being the naturally occurring EPO-encoding DNA sequence.
- While the reader may well certainly have had doubts as to whether the glycosylation of the EPO produced by the two techniques was the same, that does not seem to me to be relevant. While the patent in suit only teaches a technique in relation to two specific cell lines, namely the COS cell and the CHO cell, only the latter of which can be described as wholly successful, it is clear that the patent envisages (and the notional reader would have expected) that, in accordance with advances in biotechnology, it would be possible to use its disclosure to express EPO in other types of cell. If, after the relevant date, a naturally occurring cell line, not previously known to be suitable for the sort of technology described in the patent, had been discovered as being suitable for that purpose, it seems to me that a person following the teaching of the patent, in all respects, save that he uses the new cell rather than the CHO cell, would infringe the patent. I find the argument that he would not do so because the glycosylation of the EPO expressed in the new cell was different from that of the EPO expressed in the CHO cell or COS cell unconvincing. First, apart from the rather vague guidance in the two paragraphs in Example 10 (which amount to no more than the result of a preliminary experiment) coupled with the closing words of Claim 19, there is nothing in the patent to suggest that the nature of the glycosylation of the EPO is part of the invention, or that it is a necessary feature of the EPO produced in accordance with its teaching. Secondly, it is clear that the nature and extent of the glycosylation varies from one type of cell to another, and, indeed, can vary within cell lines. Thirdly, as the passage in Example 10 shows, there is a variation in the glycosylation of the EPO produced in the two types of cell specifically referred to in the patent, namely the COS cell and the CHO cell. Fourthly, apart from the reference to the results of preliminary experiments in Example 10 and the closing words of Claim 19, there is nothing, particularly in Claim 26, to suggest that the glycosylation of the resultant EPO is significant.
- I turn, then, to the third question, namely whether "strict compliance with the primary meaning [of Claims 1 and 26] was an essential requirement of the invention". In this connection, as Professor Proudfoot accepted, "there is nowhere [in the 605 patent] where ["the use of a system such as TKT use" namely "homologous recombination"] is excluded specifically". Although that is a point relied on by Mr Watson, it seems to me to be pretty weak. The essence of Amgen's case, as I see it, is that homologous recombination was not referred to because, at the relevant date, nobody had achieved it, and even an unusually skilled and inventive person would merely have hoped, indeed, probably, expected, that it would be achievable in the future. It would therefore be surprising if it was specifically excluded, almost as much as it would have been surprising if it had been specifically included. However, it is fair to say that, given that the third question involves considering the language of the patent, and particularly of the claims, the point is not devoid of force, albeit that I think that it amounts to a shield, rather than a sword, so far as Amgen is concerned.
- It seems to me that, unlike in Improver [1990] FSR 181 (where the concept of a rubber rod could be said to have existed), and unlike in Catnic [1982] RPC 183 (where the possibility of an angle other than 90o could be said to have existed), it cannot be said that the variant existed at the date the patent was applied for, or even when it was published. In the previous case the variant existed but had not been thought of by the patentee. In the present case, homologous recombination did not exist at the relevant date. Roche and TKT cannot, therefore, argue, as could have been argued (and presumably was argued) in the earlier two cases, that there was an apparent and presumed intention to exclude the variant on the part of the patentee, because the basis of the variant (namely the use of a different, albeit well known, material in Improver, and the slight change of angle in Catnic) could have been considered by the patentee, and therefore must have been intentionally excluded. In the present case, homologous recombination in eukaryotic cells, the technology employed by TKT, had not been achieved by the relevant date.
- As in Catnic, to quote Lord Diplock, "no plausible reasons have been advanced why a rational patentee should want to place so narrow a limitation on his invention". It is fair to say that this point may be tied up with my conclusion on breadth of claim and Biogen insufficiency. If I am right on that issue, then there is indeed no reason why the patentee in the present case should have wished to limit the extent of his claims so that it would not extend to variations and improvements in genetic engineering over the life time of the patent. If I am wrong in that view, then TKT does have the answer to the point: the patentee would have been concerned to ensure that his claims were not cast so widely as to include techniques in respect of which the patent contains no teaching, and therefore as a result of which the patent, or at least some of its claims, might be determined to be insufficient.
- Further, various passages in the 605 patent (e.g. at page 48) indicate that the patentee appreciated that the disclosure he had given would in due course result in people being able to express EPO in cells other than those enabled by the specific teaching of the patent. It is to be noted in those passages the patentee not merely showed that he appreciated that, with inevitable advances in the biotechnology field, expression of EPO in other cells, and presumably with different techniques than those identified in the patent, would be achieved, but that he envisaged that the disclosure he has given would enable, or at least would be a necessary contribution to, the expression of EPO in other cells.
- In Palmaz's Patents [1999] RPC 47, Pumfrey J rejected the patentee's contention that the third question in that case should be answered in the negative, and said this:
"This is not just a departure from some descriptive word or phrase or a matter of degree: it is the omission of whole features of the claim."
- In the present case, TKT say that the use of endogenous encoding DNA, albeit "switched on" and promoted by exogenous DNA involves a departure from a feature of Claim 1, namely the use of exogenous encoding DNA. However, once one has identified the essential inventive feature of the Claim, which, as I have mentioned when considering Biogen insufficiency and breadth of claim, is the disclosure embodied in Table VI of the 605 patent (coupled with the fact that it can indeed be used for expression of DNA in a eukaryotic cell) it appears to me to follow that the departure involved in TKT's technology is "a matter of degree". I do not think one could go so far as to say that there were "trivial difference", to quote Mr Watson's description, between TKT's technology and that described in the patent. However, it does seem to me that, when one looks at the difference in the technology, in the context of the nature of the disclosure and claims of the patent, the difference is indeed one of degree. The essential nature of the disclosure of 605 is that it enables the expression of EPO in a eukaryotic cell to be artificially achieved by disclosure of the EPO gene sequence.
- There is a further point with which I should deal. TKT argue that the DNA sequences used for expression is not within any of the three integers of Claim 1. I do not agree. As the targeting sequence is upstream of the DNA sequence shown in Table VI of 605, it seems to me that the natural EPO gene - to the extent shown in Table VI - is used for expression of EPO - i.e. integer (a) applies. Even if it does not do so, I consider that the three Protocol questions are satisfied. Quite apart from this, I believe TKT's technology to be within integer (b). Professor Wall said that TKT's EPO gene would hybridise with the DNA sequence in Table VI. In any event, the DNA sequence used by TKT falls within Claim 2 of 605.
The teaching of the 605 patent and TKT's technology
- It is clear that TKT's technology involved drawing on the disclosure of the 605 patent to a significant extent. Indeed, were it not for the disclosure, the contribution to the art, of the 605 patent, TKT could not have achieved the expression of EPO, whether in mammalian cells or otherwise. Before turning to the specific details, it is right to deal with Mr Kitchin's point that, in so far as TKT did obtain relevant information, it was from a source other than the 605 patent. It seems to me that that is not quite the point: 605 disclosed for the first time the amino acid sequence of EPO and the nucleotide sequence of the EPO gene (save that the uppermost part of the upstream sequence was not included) and the use of this to obtain artificial and amplified expression of EPO in a eukaryotic cell. This enabled that which the patent described as being highly desirable, namely production of EPO on a commercially useful and therapeutically worthwhile scale. The facts that others (including Jacobs and his co-workers at GI) published the same or similar information, arrived at by a slightly different route, thereafter, and the fact that TKT derived its information from another such source, do not alter the relevance of the fact that it was the patent in suit in which the information was first revealed. As Mr Waugh points out, this is not a copyright case, where the copyright owner must establish that it is his work which has been copied (directly or indirectly); it is a patent infringement case, and the question therefore is whether matter disclosed in the patent (whether or not derived from the patent) has been drawn on or used by the alleged infringer. Even if the infringer can show that he arrived at the same result by the same route quite independently, or even if he can show that he arrived at a better result by a cleverer route, that, of itself, is no defence to an action for patent infringement.
- I turn, then, to what was first disclosed in the 605 patent, without which TKT could not have developed the technology it used for the production of EPO. First, and perhaps most importantly, to achieve the homologous recombination which is the fundamental feature of TKT's technology, it was necessary to have the two targeting sequences, running to a total of some 5,000 nucleotides. In order to insert the CMV promoter and the DHFR gene (to effect promotion and amplification respectively) upstream from the endogenous EPO-encoding region in the HT1080 cells, with a view to "switching on" and amplifying that gene, it was necessary to know the sequences of approximately 5,000 nucleotides of the EPO gene upstream of the coding region. It is true that the downstream segment of TKT's targeting construct ends 675 nucleotides upstream of the EPO translation initiation codon, and that Table VI does not disclose this sequence, since it begins at 620 nucleotides upstream of that codon. However, TKT's US Patent (No. 5,641,670) indicates that TKT created probes having part of the sequence set out in Table VI in order to clone the human EPO gene from a genomic DNA library. This enabled TKT then to carry out further sequencing work upstream of the -620 nucleotide in Table VI, and that could not have been achieved without the disclosure of 605.
- The evidence of Professor Proudfoot confirms this conclusion. He accepted that one "cannot think about using the TKT approach to express EPO... without knowing the exon arrangement", and that, in order to implement TKT's technology, "one needed to have information about the EPO gene and its sequence". Further, although it so happens that TKT's targeting construct ends at -675, there is no reason why it could not have included the additional downstream sequences (albeit upstream of the start codon) set out in Table VI of the 605 patent. Indeed, TKT's US Patent suggests that they started their project using targeting sequences which included sequences set out in Table VI.
- The second feature of the disclosure of the 605 patent relied on by TKT in its technology was the nucleotide sequence for the first three and a third amino acids from hGH. The final one-third part of the sequence is a G which links up with the first two nucleotides of the second exon in Table VI to give the codon GAA (which happens to code the amino acid glutamic acid). For the TKT technology to work, the first encoding exon (which is the exogenous hGH sequence) must be spliced to the second encoding exon of the EPO gene in the correct phase: if it was out of phase, then the first two nucleotides in the second exon would be translated incorrectly, and this error in phasing would perpetuate, as a result of which virtually every single encoding codon would be out of phase and would therefore encode for the wrong amino acid. Accordingly, knowledge of the phasing of the second exon, which was provided by Table VI of 605, was essential to TKT.
- This second feature is considerably less important than the first feature, because, if everything else were known to TKT, it would only have taken relatively little time and work of a fairly routine nature, essentially on a trial and error basis, to arrive at the correct phasing. However, the assistance on this aspect given by the disclosure of 605 did not merely extend to phasing: it also enabled TKT to ensure that the one third codon of their new first exon matched the two thirds codon on the second exon to give the correct amino acid.
- The third feature disclosed by the 605 patent and used in TKT's technology arose from the disclosure of the splice donor site sequence immediately following exon 1 in Table VI. The importance of this to TKT was, as in relation to the second feature, to enable the final third codon of hGH being spliced into the two thirds codon of the second exon.
- There was a dispute between Dr Brenner and Professor Proudfoot as to whether the splice donor site consisted of ten bases or six bases. They are both highly distinguished scientists in this field, and I do not find it easy to choose between them. However, in TKT's US Patent, to which I have referred, it is clear that TKT themselves expressed the view that it consisted of ten bases, and that TKT used the ten base pairs from the first intron. Professor Proudfoot accepted this, while it did not cause him to change his view as to what constituted the splice donor site. I also note that TKT clearly indicated to the FDA that the splice donor site was the ten bases derived from Table VI of the 605 patent.
- In these circumstances, while not doubting the genuineness (or, indeed, the reasonableness) of Professor Proudfoot's view, I have come to the conclusion that, in so far as it is necessary for me to conclude as a matter of general principle what the splice donor site amounts to in the present case, it consisted of ten base pairs, as Dr Brenner said, rather than six base pairs, as Professor Proudfoot indicated. More centrally, it is clear that, whatever the splice donor site may be, TKT proceeded to develop their successful technology by reference to a splice donor site consisting of ten base pairs, and this was first revealed by Table VI of 605.
- It is right to add that, whatever the correct view as to the extent of the splice donor site there were publications at the relevant date which gave pretty good guidance as to the likely splice donor site sequences in genes generally - e.g. Lodish et al. Molecular Cell Biology (although I was only referred to the 4th, 1999, edition, at page 416). However, the disclosure of the 605 patent gave certainty where it did not exist previously, on this issue as on any other issue which depended on the sequence of the EPO gene.
- I turn to the fourth feature. As Dr Brenner said, and Professor Proudfoot accepted, the disclosure in the 605 patent that the first encoding exon encoded entirely of amino acids which were in the leader peptide, and that the leader peptide extended into the first part of the second exon, "was certainly helpful to the TKT process". Without knowing that the whole of the first encoding exon encoded part of the leader peptide (and therefore not part of the mature EPO polypeptide) TKT was able to develop its technology, which involved effectively splicing out that first exon. If any part whatever of that first exon had encoded the mature polypeptide (i.e. not part of the leader peptide) then the technology could not have worked.
- As Mr Kitchin argues, the mere fact that TKT's technology relied on, indeed could not have been developed, without making use of information first vouchsafed in the 605 patent, does not of itself necessarily mean that TKT infringe. Authority indicates that the issue as to whether the process or product claims of a patent are infringed can only be answered by considering whether the infringing product or process is within the literal language of the claim or, if it is not, whether the patentee can satisfy the three Protocol questions in respect of the allegedly infringing product or process. However, although claim and infringement must not be confused, I do not think that, contrary to Mr Kitchin's submission, it is irrelevant to consider the existence and extent of the debt, if any, of the allegedly infringing product or process to the disclosure of the patent which it is said to infringe, at least in this case, particularly when one is considering an alleged infringement by reference to the Protocol questions. In my view, one should look at the contribution to the art of the patent, the way in which its claims are phrased, the extent to which the alleged infringer relies on the contribution and the nature of the alleged infringement when compared with the claims.
- Although there is no suggestion to that effect in Improver [1989] FSR 183, perhaps because it was not relevant in that case, it appears to me that, before I conclude that there is an infringement on the basis of the Protocol questions, I should enquire whether the alleged infringement actually draws on the disclosure of the patent. Where, as here, the contribution to the art of a particular patent is to a large extent a discovery, it might cause one to reconsider a conclusion that there had been an infringement, where the alleged infringer uses modern technology not contemplated by the patent and does not draw on the disclosure or contribution of the patent. After all, such a conclusion would involve the claim, in effect, extending to a technique for expressing EPO which "owed nothing to the teaching of the patent" or "which could... have been envisaged without the invention", even so far as the expression of EPO is concerned. On the other hand, for the same sort of reason, if the alleged infringement does rely on the patent's contribution, that might cause one to reconsider a conclusion that there was no infringement, albeit perhaps without quite so much concern.
- It does appear to me that, in some ways, my reasoning is encapsulated in a passage quoted in the speech of Lord Hoffmann in Biogen [1997] RPC 1, which I have already quoted. At [1997] RPC 48 lines 17 to 21 he cited with approval a passage from the Board's decision T292/85 [1989] OJ EPO 275 which includes this:
"Unless variants of components are also embraced in the claims, which are, now or later on an equally suitable to achieve the same effect in a manner which could not have been envisaged without the invention, the protection provided by the patent would be ineffectual."
(Lord Hoffmann specifically adopted part of this in a passage in his speech at [1997] RPC 53, lines 13 to 14, quoted above).
- As I have mentioned, the extent of the claim and the existence of infringement, while conceptually distinct, appear to me to overlap to some extent when one is having to deal with the Protocol questions: that is apparent from the way in which the questions were formulated, and the circumstances in which they were formulated in Improver [1990] FSR 189 itself. At least in the present case, it does appear to me that the issue of breadth of claim and Biogen insufficiency and the issue of whether TKT's process and product infringe cannot be regarded as independent. In my judgment, they both involve considering the nature and extent of the contribution to the art as disclosed by the patent. I believe this is supported by the way in which Lord Hoffmann dealt with the question of sufficiency in Biogen [1997] RPC 1. In this connection, I have in mind in particular the observations at [1997] RPC 50, where he referred to a claim not being permitted to extend to "ways which owe nothing to the teaching of the patent or any principle which it disclosed".
- Another way of looking at the question of whether TKT infringes is this. The technical contribution to the art of the 605 patent is the disclosure of the EPO gene (albeit that part of the furthest upstream sequence is not included). It is not suggested that the subsequent aspects of the teaching of the 605 patent are inventive. In particular, the way in which Dr Lin used the disclosure of the EPO gene and amino acid sequence to achieve artificial expression of EPO, namely by heterogeneous recombination, was relatively routine. It does not seem unreasonable if a new method of achieving artificial expression of EPO, which, as a method owed nothing to the teaching of the patent, but which depended on the disclosure of the patent to express EPO, infringed the patent albeit only to the extent that it involved being adopted specifically to express EPO.
Conclusion on TKT's infringement
- In these circumstances, I conclude that the case that TKT infringe the 605 patent is made out, but only in connection with Claim 26. If it were the case that TKT's technology was inventive, then it seems to me that it may very well be that TKT could claim a patent in relation to that technology. However, in so far as that technology is employed to express EPO I believe that it infringes the 605 patent. Fair protection for the patentee of 605, fair protection for the hypothetical inventor of TKT's technology, certainty for the public, and the desirability of not stifling research appear to me to be satisfactorily served by that conclusion. To conclude that the hypothetical inventor of TKT's technology would, for instance, be entitled to a monopoly of manufacturing EPO in accordance with that technology would appear to me to be less than fair to the 605 patentee and, if I can put it this way, more than fair to the hypothetical inventor of TKT's technology.
- Although I have held that Claim 19 is invalid on grounds of insufficiency, and, even if valid, it is incapable of being infringed, it is right to mention that I would have held that, at least as a matter of literal construction, TKT's product did not infringe Claim 19. I do not consider that its product is a "recombinant polypeptide" within the literal meaning of that expression in Claim 19; it does not seem to me that it is "the product of eukaryotic expression of a exogenous DNA sequence", essentially for the reasons I have summarised for rejecting the contention that TKT infringes Claim 1 on a literal construction, as amplified when dealing with the questions of construction in Claim 1 and Claim 19. However, these conclusions would probably be vitiated by applying the three Protocol questions to the issue.
- However, problems would still arise from Amgen's infringement argument in light of the closing part of Claim 19. Even if the last part of Claim 19, the urinary EPO comparator for apparent molecular weight purposes, is valid and capable of infringement, I do not consider that TKT's product falls within it. Although, as I have accepted, there is a real case for saying that Dr Strickland's 17mM Fraction might have been discarded, I consider that, bearing in mind the teaching of Miyake, the skilled addressee seeking to discover whether his product falls within Claim 19, and isolating a urinary EPO for that purpose in the same way as Dr Strickland, would have followed the 17mM Fraction as well as the 30mM Fraction. On that basis, TKT's product would not have infringed Claim 19, because it has the same apparent molecular weight on SDS-PAGE as the uEPO isolated from the 17mM Fraction, even though it has slightly higher apparent molecular weight than the uEPO isolated from the 30mM Fraction.
Q. THE 678 PATENT
The contents of the 678 Patent
- The first application made by GI (one of the Roche parties) in the United States for the grant of a patent relating to EPO was filed on 4th December 1984. It was followed by a second application filed on 3rd January 1985 and a third application filed on 22nd January 1985. The application for 678 itself was filed at the European Patent Office on 3rd December 1985.
- The issues relating to 678 are within a much more limited compass than those relating to 605. Accordingly, it is unnecessary to go into the contents of 678 to anything like the extent necessary for 605. 678 claims a priority date of 22nd January 1985, and purports to describe a "Method for the Production of Erythropoeitin". It appears that the applicant for 678, GI had succeeded in cloning the EPO gene around a year after Dr Lin of Amgen had done so. The "Field of the Invention" claimed by 678 is described on page 3 of the Patent as being:
"Directed to cloned genes for human erythropoietin that provide surprisingly high expression levels, to the expression of said genes and to the in vitro production of active human erythropoietin".
- Claims 8 and 9 of 678 are in these terms:
"8. Method for producing recombinant human erythropoietin (hEPO) by the steps of
(a) culturing, in a suitable medium, CHO cells which contain, operatively linked to an expression control sequence, a DNA sequence encoding hEPO, and
(b) recovering and separating the recombinant hEPO produced from the cells and the medium,
characterised in that CHO cells are used which have the capability of producing N- and O- linked glycosylation with incorporation of fucose and N-acetylgalactosamine and that recombinant hEPO with N- and O- linked glycosylation is recovered and separated from the cells and the medium.
9. Method according to claim 8, wherein the recombinant hEPO has a glycosylation pattern comprising relative molar levels of hexoses to N-acetyglucosamine (Nacglc) of 1.4:1 specifically galactose: [GlcNAc] = 0.9:1, and mannose: [GlcNAc] = 0.5:1."
- The Specification contained two paragraphs of relevance which were not included in the third application for 678. Paragraph 36 (page 12, lines 37ff) refers to purifying "recombinant EPO produced in CHO cells as in Example 11" and thereafter analysing "the relative amounts of sugar present". This analysis records the presence of GlcNAc, Gal, Man, NeuNAc and Fuc and a small amount of GalNAc. This last sugar is significant, as it confirms the presence of a degree of O-linked glycosylation (as mentioned above). That point is expressly made in paragraph 37 (lines 54ff) which records that O-linked glycosylation was confirmed by SDS-PAGE analysis. Paragraph 37 also draws attention to the presence of fucose.
- Example 10 of 678 is entitled "Expression of EPO in CHO cells - Method I". It confirms expression of EPO in DHFR- CHO cells which have been transfected with exogenous recombinant DNA ligated with a specified plasmid and an artificial promoter, with amplification. Example 10 states that "the cell line of choice for EPO production" has been deposited with the ATCC and the reference number is given. Example 11 is entitled "Expression of EPO in CHO cells - Method II". It describes a similar exercise to Example 10, but in DHFR - CHO cells and a different plasmid. Unlike Example 10, there was no deposit of the Example 11 cell line. However, it is stated that the plasmid described in Example 11 had been deposited with the ATCC.
- There is nothing of relevance in the filed application for 678 which differs from the contents of 678 for the purpose of these proceedings.
The issues
- There was no argument relating to the issue of whether Amgen infringes the 678 patent. As I understand it, Amgen conceded infringement if 678 is valid. The only live issue is whether 678 is valid or not. The validity of 678 is primarily challenged by Amgen on the basis of lack of novelty and/or obviousness. In this connection Amgen's case principally relies upon the disclosure of the application for the grant of 605, namely 605A. The issue is complicated by the fact that there is a dispute as to the priority date which can be claimed for 678.
- Only Claim 8 and Claim 9 of 678 are relied on by Roche as being valid and infringed, and Claim 9 is not intended to have validity independent of Claim 8. Accordingly, I shall limit myself to considering Claim 8. Claim 8 of 678 claims expressing recombinant human EPO in CHO cells which had been cultured, and which contained "operatively linked to an expression control sequence, a DNA sequence encoding" for human EPO. The centrally relevant feature of Claim 8 for present purposes is that the CHO cells concerned "have the capability of producing N- and O-linked glycosylation with incorporation of fucose..." and that the recombinant EPO has "N- and O-linked glycosylation" when "recovered and separated from the cells...". This is the essential ingredient for the purpose of Roche's case, because, given that Amgen had filed the application for 605, namely 605A, before the date on which Roche claim priority for 678, 22nd January 1985, Roche cannot rely on 678 to claim a monopoly in respect of anything disclosed in 605A, provided, of course, that the disclosure relied on by Amgen in 605 is enabling.
- Amgen contend that, even if Roche are right in contending that 678 is entitled to a priority date of 22nd January 1985, Claim 8 of 678, and in particular the essential ingredients of O-glycosylation and fucosylation, were anticipated by the disclosure of 605A, and in particular Example 10 thereof: in other words, Claim 8 is said to fail on the ground of want of novelty. If Amgen are correct as to the relevant priority date for 678, namely 3rd December 1985, then Amgen's reliance on 605A, and in particular Example 10, rests not merely on anticipation, but also on obviousness. If Amgen are right on the priority date issue, they also rely on the paper published by Jacobs et al in 1985 in Nature 313:806 ("Jacobs") as supporting the case on anticipation and obviousness in relation to Claim 8 of 678. Of course, Amgen's reliance on 605A, and Example 10 in particular, depends on its being properly enabling (which I have held it to be when considering the issue of classic sufficiency of 605).
- It is right to add that Amgen rely, albeit very much as "long stop" arguments, on the contentions that, in any event, Claim 8 of 678 is invalid on the ground that it is insufficient and on the ground that it constitutes a mere discovery. I do not propose to say anything further on those two issues, because Amgen accept that, if they succeed in fighting off Roche's attacks on 605 on similar grounds (as, in my view, they do) then they can scarcely hope to be successful in their attack on 678 in so far as that attack involves relying on similar grounds.
- I propose first to consider whether 678 is entitled to claim a priority date of 2nd January 1985, as opposed to 3rd December 1985. I shall then deal with the issue of whether, assuming that Roche are correct in saying that 678 is entitled to the earlier priority date, Claim 8 of 678 ("Claim 8") was nonetheless anticipated by Example 10 of 605A ("Example 10"). I shall then turn to the issue of whether, if 678 is only entitled to a priority dated of 3rd December 1985, Claim 8 is invalid on grounds of obviousness over Example 10, and/or over Jacobs.
- However, before doing this, I should perhaps explain briefly why 605A can be relied on by Amgen to support the contention that 678 is obvious or anticipated if 678's priority date is 3rd December 1985, but can only be relied on to support a case of anticipation, and not one of obviousness, if 678 can claim a priority date of 22nd January 1985. The priority date which can be claimed by a patent is governed by Section 5 of the 1977 Act. Prima facie, it is the date of filing - Section 5(1). However, by virtue of Section 5(2), if the invention claimed is "supported by matter disclosed in [an] earlier application" less than twelve months before the date of filing, the priority date may be effectively backdated to that disclosure.
- By 22nd January 1985, Amgen had filed its application for 605, i.e. it had filed 605A. However, 605A had not yet been published. Accordingly, at least as a matter of principle, it is open to Amgen to rely upon 605A as anticipating 678, because of Section 2(3) of the 1977 Act which provides:
"The state of the art in the case of an invention to which... a patent relates shall be taken also to comprise matter contained in an application for another patent which was published on or after the priority date of that invention, if the following conditions are satisfied, that is to say -
(a) the matter was contained in the application for that other patent both as filed and as published; and
(b) the priority date of that matter is earlier than that of the invention."
- However, if the correct priority date for 678 is indeed 22nd January 1985, then it would not be open to Amgen to rely on an application such as 605A to found a basis for alleging obviousness. This follows from Section 3 of the 1977 Act which provides in the case of obviousness:
"An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art having regard to any matter which forms part of the state of the art by virtue only of Section 2(2) above, and disregarding Section 2(3) above" (emphasis added).
- The rule that a person attacking a patent can rely upon a filed patent application to substantiate a case for anticipation, but not for the purpose of alleging obviousness, arises because, unlike the United States (where the principle is first to invent), the UK, and other signatories to the EPC, follow the first to file principle. Accordingly, a problem is thrown up by documents which, as at the date of an application for a patent, have been filed, but have not yet been published so as to make them available. This problem is dealt with by deeming such documents to have been published at the date of filing for the purpose of anticipation only, thereby substantially avoiding the problem of concurrent grants - in this connection, see the discussion in the House of Lords in Asahi Kasei [1991] RPC 485.
- However, by 3rd December 1985, 605A had not merely been filed: it had been published (namely on 17th July 1985). Accordingly, Section 2(3) of the 1977 Act is not needed by Amgen. The information in the filed patent application, 605A, had been in the public domain, in the sense that the application had been published for more than four months before the priority date which, on this hypothesis, 678 could claim. In these circumstances, there is no reason why 605A could not be relied on by Amgen to support a case of obviousness, as well as a case of anticipation. Equally, Amgen could rely upon any other publication prior to 3rd December 1985, such as Jacobs, which was published after 22nd January, but before 3rd December, 1985.
The priority date of 678
- The third application in the United States of 22nd January 1985 ("the third application"), which is the document upon which Roche's arguments as to priority date depends, describes the production of EPO from CHO cells specifically DuX-B11 cells. It discloses methods of transforming the genetic material in such CHO cells with vectors carrying the human EPO gene, and, as a result, effecting the expression of EPO. It is stated in the application that that cell line has been deposited with the ATCC. Roche contend that anyone could have obtained this cell line from the ATCC, and, having grown it in culture, expressed EPO which, on analysis, would have been found to be O-glycosylated (as well as N-glycosylated) and fucosylated. However, there is nothing in the third application which teaches or discloses that the EPO thereby produced would be O-glycosylated or fucosylated. The earliest document upon which Roche can rely in which O-glycosylation and fucosylation of recombinant EPO is mentioned, was in the application for the grant of 678, which, as I have said, was filed on 3rd December 1985.
- In these circumstances, Amgen's contention that a claim whose alleged novel feature is the presence of O-glycosylation and fucosylation can only claim a priority date by reference to a document in which this element is first disclosed, has the merit of appealing to common sense and logic. It would seem somewhat surprising if a process claim, which had a single feature which was said to render it novel or unobvious over a specific disclosure, could claim a priority date by reference to documents which, whatever else they revealed, did not make reference to, let alone claim or identify, the very feature upon which the alleged novelty or unobviousness depends. That proposition appears to me to be supported by Article 88(4) of the EPC, which is in these terms:
"If certain elements of the invention for which priority is claimed do not appear among the claims formulated in a previous application, priority may nonetheless be granted, provided that the documents of the previous application as a whole specifically disclose such elements" (emphasis added).
- I also draw support from Beloit [1995] RPC 705 at 732, where Jacob J said this:
"Matter disclosed in the priority document really means "information concerning the invention disclosed in that document". So the question is whether the claim is supported by the information disclosed in the priority document."
- Accordingly, at any rate at first sight, Roche's contention that the priority date of 678 is 22nd January 1985 must be wrong: in accordance with Article 88(4) and Jacobs J's observation I have quoted, the priority date would appear to be the date of the filing of the application of 678, when the O-glycosylation and fucosylation of the recombinant EPO was first referred to and explained.
- In answer to this, Roche rely, as I have mentioned, upon the deposited cell line referred to in the third application and in Example 10, and therefore in the filed application for 678. As a matter of principle, it appears to me that it could be open to Roche to rely upon what the cell line deposited with the ATCC, and referred to in the third application, would have revealed. In my view, the reference to "documents" which "specifically" make the relevant disclosure in Article 88(4) of the EPC must, as a matter of ordinary language and common sense, be capable of extending to things referred to in those documents, at least where there is an official and well established procedure for depositing those things, as there is in relation to micro organisms pursuant to the European Patent Office's Notice on Deposits -see OJ EPO 1986 8 p269.
- However, I am far from persuaded that this means that it is open to Roche in the present case to rely upon the fact that, if treated in accordance with the teaching in the third application, the cell line referred to therein as deposited would produce EPO which, if it was characterised or analysed, would turn out to be O-glycosylated and fucosylated. There was nothing whatever about this feature in the third application, no teaching to characterise it, let alone to look for its glycosylation.
- Mr Thorley relies upon the reasoning and decision of the House of Lords in Merrell Dow Pharmaceuticals Inc. -v- H N Norton & Co Ltd [1996] RPC 76, a case which essentially turned on the meaning of Sections 2(1) and (2) of the 1977 Act (which reflect Article 54 of the EPC). These sub-sections provide:
"2(1) An invention shall be taken to be new if it does not form part of the state of the art
(2) The state of the art in the case of an invention shall be taken to comprise all matter (whether a product, a process, information about either, or anything else) which has at any time before the priority date of that invention been made available to the public... by written or oral description, by use or in any other way."
- In Merrell Dow at [1996] RPC 76 at 89, Lord Hoffmann had to consider the argument:
"That for a substance to be known so as to be part of the state of the art within the meaning of Section 2, it must be known (or to be readily capable of being known) by its chemical composition. No other description will do" (lines 1 to 5)
- He went on to say this at [1996] RPC 89 line 13 to 90 at line 38:
"I think that on this point the Patents Act 1977 is perfectly clear. Section 2(2) does not purport to confine the state of art about products T12/81 [1979-1985] EPOR Vol. B 308, 312 to knowledge of their chemical composition. It is the invention which must be new and which must therefore not be part of the state of the art. It is therefore part of the state of the art if the information which has been disclosed enables the public to know the product under a description sufficient to work the invention.
For most of the purposes of a product claim, knowledge of its chemical composition will be necessary to enable the public to work the invention. It is something they will need to know in order to be able to make it. ...Other decisions... seem to me to make it clear that, at least for some purposes, products need not be known under their chemical description in order to be part of the state of the art. In BAYER/Diastereoisomers Decision Twelve/81 1979-85 EPOR Vol. B 308312... the Technical Board of Appeal [decided that] if the recipe which inevitably produces the substance is part of the state of the art so is the substance as made by that recipe. [In] CPC/Flavour Concentrates Decision T303/86 [1989] 2 EPOR 95 [the Board's decision] proceeded on the basis that for the purpose of being part of the state of the art, a process for making flavour concentrates was sufficiently described by a recipe for cooking food which did not expressly refer to the flavour concentrates but would inevitably have the effect of making them."
- I am here concerned with an argument revolving around Section 5 of the 1977 Act. Mr Watson says that I should not automatically apply the reasoning which related to a different section, namely Section 2, of the same Act, even though the issue involved under the two sections may be similar. I am not convinced that that is right. What I think is more important, as is clear from the passages I have selectively quoted from the speech of Lord Hoffmann, is that there is no absolute rule as to the extent, if any, to which the chemical composition of a particular substance has to be disclosed before it becomes part of the state of the art. Where an aspect of the chemical composition is effectively the claimed inventive concept, it appears to me illogical to conclude that that aspect does not have to be revealed in some way by the earlier application which is relied on by the patentee to justify an earlier priority date for his patent under Section 5(2) of the 1977 Act: I rely in particular on the word "disclosed" in Section 5(2). I do not read Lord Hoffmann's observations in Merrell Dow as calling that view into question.
- As I have said, it appears to me that it is open in principle for Roche to rely upon the deposited cell line, because it is referred to in the third application. However, I do not believe that that is of itself enough, without more, to enable Roche to claim that a specific feature of the product resulting from the processing of that cell line in accordance with the teaching of the third application can be relied on as giving novelty over a previous disclosure (or obviousness over the prior art) to the process described in the application. While one must be careful of relying on an intuitive or knee-jerk reaction (as graphically illustrated by what Lord Hoffmann had to say in Merrell Dow at [1996] RPC 83 to 87) it does appear to me that, in the absence of clear authority, or some other good reason to the contrary, it would be surprising if it were otherwise. Before the feature could be relied on as having been disclosed through the medium of a cell line referred to in the claimed priority document, it seems to me that it would be necessary for the feature to be clearly (albeit conceivably impliedly) identified or referred to in the document. It may also be permissible if the feature is so obvious from the cell line that the skilled man would appreciate it even though it was not referred to in the document, provided he is unequivocally directed to the cell line.
- Indeed, it appears to me that Mr Thorley's submissions on behalf of Roche go close to accepting a not very different formulation of this principle in any event. He accepts (rightly in my judgment) that, in light of the absence of any teaching as to O-glycosylation or fucosylation in the third application, upon which his argument essentially relies, it is an essential element of his case that the skilled man, following the teaching of the third application, would characterise what he had got. However, in my view, if it be the case that the skilled man would inevitably have sought to characterise what he obtained, in the sense of investigating its chemical composition, that would not on its own be enough to enable Roche to establish that the third application enables them to claim 22nd January 1985 as the priority date for 678. They would have to show that the purpose of the characterisation, or at least one of the specific purposes of the characterisation, was to investigate whether the EPO thereby produced was O-glycosylated and fucosylated. The fact that the addressee might happen to discover that it was O-glycosylated or fucosylated would not be enough, because that would be no part of the teaching of the priority document: he would have no reason to know whether this was a feature of the substance he had made, because of the precise conditions in which he made it, or whether it was an essential ingredient of the teaching. Not only does that appear to me to be right as a matter of general principle, it is of particular force in connection with the instant feature: as I have mentioned, it is clear that the nature and extent of the glycosylation is very variable and, particularly as at 1984, unpredictable, and, so far as O-glycosylation was concerned, relatively uncharted territory.
- In these circumstances, it is my view that the priority date which can be claimed for 678 is 3rd December 1985, because the third application filed on 22nd January 1985 did not contain any teaching whatever in relation to the O-glycosylation or the fucosylation of the recombinant EPO resulting from the process taught in that application, or even suggesting that the addressee should characterise the nature of the glycosylation of the EPO produced in accordance with the teaching. It is right to add that, for the reasons given in the next section of this part of the judgment, I do not consider that a person following the teaching of the third application would have sought to characterise the O-glycosylation or fucosylation of the EPO he obtained. I shall nonetheless first consider the question of whether Claim 8 of 678 was anticipated by Example 10 of 605A, on the assumption, favourable to Roche and which I do not believe to be correct, that the priority date which can be claimed by 678 is 22nd January 1985; thereafter I shall turn to consider the validity of Claim 8 of 678 if its priority date is, as I believe, 3rd December 1985.
Anticipation by 605A
- In order to establish that a claim in a patent has been anticipated by an earlier disclosure, it is necessary to show that the earlier disclosure unambiguously points to what is purportedly disclosed by the patent in suit. In the course of his judgment in General Tire & Rubber Co. -v- Firestone Tyre and Rubber Co. Ltd [1972] RPC lines 41ff, Sachs LJ said this:
"If the prior inventor's publication contains a clear description of, or clear instructions to do or make, something that would infringe the patentee's claim if carried out after the grant of the patentee's patent, the patentee's claim will have been shown to lack the necessary novelty, that is to say, it will have been anticipated. The prior inventor... and the patentee may have approached the same device from different starting points and may for this reason... have so described their devices that it cannot be immediately discerned from a reading of the language which they have respectively used that they have discovered in truth the same device; but if carrying out the directions contained in the prior inventor's publication will inevitably result in something being made or done which, if the patentee's patent were valid, would constitute an infringement of the patentee's claim, this circumstance demonstrates that the patentee's claim has in fact been anticipated."
- At [1972] RPC 486 lines 9-10 Sachs LJ said:
"To anticipate the patentee's claim the prior publication must contain clear and unmistakable directions to do what the patentee claims to have invented."
- These observations were, of course, made in relation to the Patents Act 1949, but it appears clear that they represent the current law. They have been cited with approval in a number of cases since the 1977 Act came into force, including, for instance, by Lord Jauncey of Tullichettle in Asahi [1991] RPC 485 at 544. Further, the approach of the Court as embodied in those observations is consistent with Article 54(1) of the EPC with which, indeed, Section 2 of the 1977 Act is expressly drafted to correspond - see Section 130(7) of the 1977 Act.
- Roche's case is that Example 10 did not anticipate Claim 8 because the Claim is novel to the extent that it discloses that the recombinant EPO produced in accordance with its teaching will always be O-glycosylated and fucosylated, whereas this was not the case in relation to the teaching of the Example, and that this teaching can, as it were, be related back to the third application and the deposited cell line referred to therein.
- A number of preliminary points are taken by Amgen. First, it is said that there is no clear evidence that the deposited cell line referred to in the third application was live, or that it would certainly produce EPO which was O-glycosylated or fucosylated. I do not think that argument is a good one. In the first place, although not worded perfectly, it seems to me that there is an express admission by Amgen in these proceedings to the effect that the deposited cell line would express EPO which was O-glycosylated and fucosylated. In any event, pursuant to Rule 10.1(ii) of the Budapest Treaty Regulations, the ATCC is obliged to test the viability of deposits at reasonable intervals, and it is to be noted that the ATCC apparently keep deposited cell lines for at least 30 years. The fact that Roche at one time formally proposed to carry out an experiment which they subsequently apparently abandoned does not, count against Roche, in this connection despite Amgen's case to the contrary. Roche may well have concluded, in light of the admission by Amgen, in light of the costs, in light of advice as to where the onus of proof lay, or because there might be some other, albeit wholly unrelated, problem for Roche if the result of the experiment was revealed, not to proceed with it. Quite apart from this, so far as the ability to O-glycosylate or fucosylate is concerned, it seems to me that, in light of the evidence discussed in the next section of this part of the judgment, it is not merely likely on the balance of probabilities, but overwhelmingly likely, that any EPO expressed by cells in the deposited cell line would have been O-glycosylated and fucosylated.
- Of greater weight, in my judgment, is Amgen's contention that glycosylation of EPO in the deposited cell line is not conclusive on this issue because there is no direction to the reader of the third application that he should use those cells. In his cross examination, Professor Clausen accepted that a person skilled in the art who was following the teaching of the third application might decide to draw on the deposited cell line, or he might not. It seems to me that, before I go into further points, that presents a real difficulty for Roche, because it is fundamental to Roche's case that the teaching of Claim 8, to the effect that the recombinant EPO will be O-glycosylated and fucosylated, was disclosed by the third application, and that argument in turn is dependant on the contention that the third application refers to the deposited cell line and that this cell line will produce EPO which is O-glycosylated and fucosylated. If the teaching of the third application may well involve the skilled addressee using another cell line, then, as I see it, the basis for Roche's argument falls away. Virtually every argument upon which Roche rely for contending that following Example 10 of 605 may produce EPO which is not O-glycosylated or fucosylated would apply equally to following the teaching of the third application, if one uses cells which are not part of the deposited cell line. This argument is particularly telling as Example 10 teaches the use of the same cells, namely DuX-B11 CHO cells.
- Amgen also rely on the fact that it is fundamental to Roche's case that the skilled man following the teaching of the third application would characterise the product and in particular whether it was O-glycosylated and whether it was fucosylated. Mr Thorley relies on passages in the evidence of Professor Clausen and Professor Cummings to support the contention that a person obtaining the deposited cell line would not only use it to produce EPO, but would also characterise the EPO, and in particular its polysaccharide content in order to see whether it was O-glycosylated and/or fucosylated. I am unpersuaded that the evidence of either witness assisted this contention. Professor Clausen's evidence merely amounted to saying that, once one had the EPO, one would purify it. He did not refer to characterisation or analysis. While he referred to characterisation, Professor Cummings only said that it would be "a routine matter". No witness suggested that a person following the teaching of the third application would be very likely, let alone bound, to characterise the EPO, let alone that the characterisation would go so far as determining whether there were any O-glycans and any fucose.
- In light of the conclusions in the preceding paragraphs, I believe it is strictly unnecessary for me to consider Claim 8 further: taking the most favourable priority date from Roche's point of view, and bearing in mind that Mr Thorley accepts that it is essential to Roche's case on anticipation that the skilled man would investigate the extent if any of O-glycosylation and fucosylation of the EPO produced in accordance with the teaching of the third application, Roche's case fails. However, as the points have been fully argued, and as the issues turn on evidence and as this case may well go further, it is right to deal Roche's arguments on anticipation, on the assumption that the cell line referred to in the third application can be relied on as teaching that the EPO thereby produced would be O-glycosylated and fucosylated (or that it does not matter whether that cell line can be relied on).
- On behalf of Roche, Mr Thorley takes three points as to why Example 10, does not anticipate Claim 8. First, he contends that Example 10 was not enabling in any event, due to what he says would have been the undue burden of obtaining materials with which to repeat the experiment. Secondly, he claims that, on the evidence, Amgen has not established that Example 10 anticipated Claim 8. Thirdly, he argues that, in light of the deleted matter, which was included in 605A, although it is not of course in 605, Example 10 did not anticipate Claim 8.
- I do not need to deal with the first of these arguments, because it is the same as Roche's argument, which I have already discussed and rejected when considering whether 605 was invalid on grounds of classic insufficiency. Mr Thorley's second and third arguments essentially turn on the same fundamental issue. Claim 8 teaches that recombinant EPO is partially O-glycosylated and contains fucose, and Mr Thorley's second and third arguments rely on the fact that EPO produced in accordance with Example 10 may not be O-glycosylated and/or fucosylated.
- Mr Thorley's second argument is that following the teaching of Example 10 will not inevitably produce O-glycosylated rEPO. On behalf of Roche, he contends that following the teaching of Example 10 may produce rEPO which is solely N-glycosylated. On this issue, I heard evidence of experiments, as well as expert evidence. As Mr Thorley rightly says, the effect of the observations of Sachs LJ in General Tire is that, in order to anticipate, the directions in Example 10 must "inevitably" lead to something falling within the scope of the claim being attacked, namely Claim 8. Accordingly, as I see it, if I am satisfied on the evidence that there is a real chance that following the teaching of Example 10 could produce rEPO which was not O-glycosylated, this second argument for rejecting Amgen's case on anticipation would succeed.
- On the evidence in the case, I am satisfied that carrying out the experiment set out in Example 10 would inevitably lead to the production of rEPO which is partly O-glycosylated. The possibility that following the teaching of Example 10 could lead to rEPO which was not O-glycosylated was floated by Professor Clausen. Neither he nor anybody else had conducted an experiment which produced that result, and he was only able to point to one specific example which might give rise to such a result, namely that of recent CHO cell lines which have been mutated to activate dormant genes. Indeed, in cross examination he accepted that one could not deprive a cell, in which a glycoprotein was normally O-glycosylated and N-glycosylated (as in the case of EPO in the CHO cell), of the ability to effect N-glycosylation without also depriving it of the ability to effect O-glycosylation. He also thought that a cell with an inability to produce GalNAc, but which could normally do so, would not be a cell which was likely to be able to grow.
- Professor Cummings, who gave evidence on behalf of Amgen, also referred to the single reported instance of a specific type of CHO cell losing its ability to O-glycosylate protein. This was a cell line which had been specifically mutated to lose the ability to make GalNAc and was one which had been selected for the loss of the ability to O-glycosylate proteins. However, as Professor Cummings pointed out (and Professor Clausen accepted), in addition to the fact that this cell line was intentionally mutated for this purpose, it was also a cell line which could not survive without the addition of a specific monosaccharide which, if added to the culture, would have reintroduced the ability of the cell to O-glycosylate. It appears to me fanciful to believe that the skilled man reading Example 10 would specifically disregard the teaching in the Example to use the DuX-B11 cell line, and would instead choose an unstable cell line which had been specially mutated. The possibility that a cell line might spontaneously lose its ability to O-glycosylate was raised by Professor Clausen, but neither he nor any other witness was able to identify an instance of this having occurred.
- Although examples can be found in the literature of things apparently going wrong in cells, which lead to some sort of functional inactivation, I am not persuaded that they take matters any further. The fact that there are reported experiments of attempts to express glycoproteins in cells which have been transfected and/or amplified, does not mean that such generalised evidence can be fairly invoked to suggest that there is a real, indeed any, prospect of a specific type of glycosylation of a specific protein in a specific cell not taking place, particularly in circumstances where the evidence suggests that it has done so in all examples which have been reported or have been the subject of experiments. Further, most of the evidence relied on by Roche in this connection was of recent cell lines which had been mutated to activate dormant genes, and it seems clear from the evidence of Professor Clausen and Professor Cummings that gene activation was not the only possible explanation of what was reported in those papers. In any event, the argument that things sometimes go wrong in expression systems in CHO cells would apply equally to the alleged disclosure of Claim 8 as it would to the disclosure in Example 10.
- Further, the evidence of Professor Clausen that a mutation might occur accidentally during transfection or amplification, a result of which the cell might lose its ability to O-glycosylate or fucosylate was merely speculation. However, it was speculation which Professor Cummings attempted to quantify, and in this connection his quantification (contained in his written evidence in chief in reply to Professor Clausen) was not challenged. Taking Professor Cummings's calculations at their most favourable to Roche, his opinion was that the chance of a relevant mutation in a cell, that is one which would deprive it of the ability to O-glycosylate, would be around one in a million; however, as Professor Clausen accepted, this is significantly too high a figure, because one has to adjust it to take into account the prospect of all the transformants being disabled: in other words, each additional clone has to retain the inability to O-glycosylate. Mr Watson described the prospect of that occurring as "vanishingly small" and a "minute mathematical probability". I do not think those descriptions are exaggerated, even in relation to a chance of one in a million, and I certainly do not consider that they enable Roche to say that following the teaching of Example 10 would not "inevitably" lead to EPO thereby produced being O-glycosylated and fucosylated. It is right to mention that it was suggested on behalf of Roche that gene amplification might result in a single clonal cell line consisting of cells none of which could O-glycosylate and/or fucosylate. However, there was no evidence to support this, and indeed this was not a matter within the expertise of Professor Clausen.
- Although there is obvious force, at least in terms of literal interpretation, in the point that the use of the word "inevitably" in the observations of Sachs LJ in General Tire indicates that even such a very low possibility should defeat a case based on anticipation, I do not consider such an argument to be correct. The law of patents is ultimately concerned with practicality, although, as parts of this judgment probably demonstrate, the theoretical and academic have an important, indeed, in many cases, an essential, part to play. In my judgment, if an experiment reliably produces a particular result more than 99% of the occasions on which it is conducted, then, at any rate for the purpose of deciding whether the disclosure of that experiment can be said to anticipate a subsequent patent, that experiment "inevitably" leads to the result in question. I draw some support for this conclusion from observations, albeit they were obiter, in Fomento -v- Mentmore [1956] RPC 87 at 101, Letraset Ltd -v- Rexel Ltd [1976] RPC 51 at 60-61 and Letraset Ltd -v- Dymo Ltd [1976] RPC 65 at 92. In Fomento it was accepted that if the required result obtained "save in exceptional circumstances (as one might say, 99 cases out of 100)" it would not deprive the earlier disclosure of its anticipating quality. In Letraset at [1976] RPC 61, Russell LJ suggested that if the experiment produced a particular result 33 out of 36 times it would be "open to question" whether that result was "inevitable", and that "some proportion of failure could no doubt be regarded as de minimis or due to failure to perform the test properly".
- I turn, then, to Mr Thorley's third argument, namely that, in light of the contents of the deleted matter is Example 10 in 605A, it would have been apparent that the EPO thereby produced would only have been N-glycosylated and accordingly Roche EPO, which was O-glycosylated, as well as N-glycosylated, was not thereby anticipated. The basis for this argument is that the deleted passage refers to an experiment which showed that there was no N-acetylgalactosamine (GalNAc) in "the recombinant product" derived in accordance with Experiment 10, and that this would have led the appropriately skilled reader of 605A to deduce that this product had no O-glycosylation. This is because, as I mentioned earlier, the presence of GalNAc is indicative of O-glycosylation, and its absence therefore demonstrates no O-glycosylation.
- Having reached my conclusion on Roche's second point, I do not consider that the inclusion of the deleted matter in 605A assists Roche on this third point. First, if, as a matter of fact, the appropriately skilled person, following the teaching of Example 10, would actually produce a product, namely rEPO, which is O-glycosylated in part, then, once the application embodied in 605A had been granted (including Example 10, whether with or without the deleted matter) a person following the "clear instructions" of the Example would have "do[ne] or ma[de] something that would infringe the patentee's claim" to quote from Sachs LJ in General Tire at [1972] RPC 485. To put the point another way, again to adopt the language of Sachs LJ at [1972] RPC 486, once one has rejected Roche's second argument on this issue, Example 10 "contain[ed] clear and unmistakable directions", and if the skilled man followed those directions he would have produced rEPO which was in part O-glycosylated. The fact that an experiment included in the Example might well have led the reader of the document to think that the product would not be O-glycosylated or fucosylated does not mean that it would not in fact be partially O-glycosylated or fucosylated.
- Secondly, as I have already said on the issue of added matter, I do not consider that the notionally skilled addressee of 605A would have believed the results reported in the deleted matter. It is true that the specific aspect of the deleted matter which would have been rejected by the notionally skilled addressee is the hexose ratio of the recombinant EPO, which does not directly relate to the finding of no GalNAc (and hence no O-glycosylation) and no fucose (and hence no fucosylation) in the recombinant EPO. Accordingly, as Professor Clausen said in his evidence, one can see an argument for saying that, while the addressee might reject the reading for hexose in the recombinant EPO, he would not reject the other readings, and in particular the absence of GalNAc and fucose in the recombinant EPO. Nonetheless, I accept Professor Cummings's evidence to the effect that he, and the notional addressee, would have been so concerned about the 15.09 figure in relation to the hexose content of the recombinant EPO in the deleted matter, that he would have effectively rejected the entire analysis. Indeed, that is the way that GI, one of the Roche parties, expressed its view in the US interference proceedings, when it described "the [reported] carbohydrate composition disclosure" as "plainly incorrect".
Obviousness and novelty if the priority date of 678 is 3rd December 1985
- If, as I believe to be the case, the priority date which can be claimed for 678 is 3rd December 1985, then the validity of 678 is even more difficult indeed to maintain against 605A. Given that 605A can be relied on by Amgen to support a case of obviousness, it can clearly be said that it was obvious as at 3rd December 1985, in light of the disclosure of Example 10, to produce EPO from CHO cells. If, as is quite clear from the evidence, a possible result of carrying out such production was that the EPO so produced was O-glycosylated and fucosylated, then Claim 8 must be invalid on grounds of obviousness.
- Once again, Roche seek to rely upon the deleted matter in Example 10, and, for the reasons I have already given when considering whether Example 10 can be relied on by Amgen to defeat Claim 8 on grounds of lack of novelty, I do not consider that the deleted matter assists Roche. Indeed, Roche's reliance on the deleted matter is even more difficult on the issue of obviousness than on the issue of anticipation. In addition to the arguments open to Amgen on anticipation, they can also rely on the fact that, if the deleted matter is to assist Roche on obviousness, then Roche have to establish three further points. Those three points are that the deleted matter would have caused the reader of the third application to try to make EPO which was not O-glycosylated, that this is something which the reader could have done, and that, in the light of what he would have done, Claim 8 does not lack inventive step. I agree with Mr Watson's contention that none of these three points can be made good. Not only is there no evidence to support the suggestion that a skilled addressee would seek to make EPO which was not O-glycosylated or fucosylated, but I think it most unlikely that he would have been interested in the precise make-up of the glycans in the EPO they had produced. Secondly, Professor Cummings could not conceive how anyone could screen for CHO cells which did not O-glycosylate (or, presumably, fucosylate) EPO, and in light of the evidence that nobody had ever found such CHO cells, I share his scepticism. Quite apart from this, if one goes along with Roche's case to the extent of assuming in their favour that an appropriately skilled person produces EPO in accordance with the teaching of Example 10 some of which is not O-glycosylated or fucosylated, he will have to analyse the EPO he has produced, and separate that which is O-glycosylated and/or fucosylated from that which is neither (if any of the latter exist, which I very much doubt). In carrying out that exercise, he will have found EPO which has the very characteristics, namely O-glycosylation and fucosylation, which, on Roche's case, are meant to be unobvious.
- Before passing on from my conclusion that Claim 8 is obvious over Example 10, it is right to refer back to the four step process identified in Windsurfing International Inc. -v- Tabur Marine (Great Britain) Ltd [1985] RPC 59 at 73 to 74. The inventive concept in Claim 8 is said to be the production of rEPO which is O-glycosylated and fucosylated. Assuming the mantle of the appropriately skilled person armed with common general knowledge (the second step) I pass on to the third step, namely identifying the alleged difference between the prior art, Example 10, and the claimed invention, Claim 8. For the various reasons I have given, it appears to me that the analysis reported in the deleted matter in Example 10 does not represent any significant difference. First, it would not be believed by the skilled reader, and in any event the skilled reader would not have set out to make EPO which was not O-glycosylated or fucosylated, would not have been able to do so, and, in any event, in doing so would inevitably have made the very product which is claimed by Roche to be inventive. That conclusion renders the fourth step unnecessary, but, essentially for the same reasons, any difference of Claim 8 over Example 10 would have been obvious.
- As I have mentioned, quite apart from 605A, Amgen rely on Jacobs as rendering Claim 8 obvious. In summary form, Jacobs discloses the EPO-encoding sequence and refers to it producing recombinant EPO COS cells. Both Professor Cummings and Professor Proudfoot explained that, at least for anything other than transient expression, a CHO cell would either be the cell, or a cell, of choice. I also note that Professor Clausen accepted that CHO cells were on the shortlist of cell lines which one would expect to use. Further, Professor Wall thought that CHO cell lines were obvious in 1984 as a choice for effecting mammalian gene expression, and he was not challenged on this. In light of that, and in light of the arguments I have considered in relation to Example 10 rendering Claim 8 obvious, I think that Claim 8 is also obvious over Jacobs.
R. CONCLUSIONS
- In these circumstances, my conclusions are as follows:
(1) The attacks on the 605 patent fail, save that Claims 19 and 20 (and the Claims dependent thereon) are invalid on the ground of insufficiency;
(2) Subject to further arguments, I am prepared to uphold the 605 patent subject to Amgen deleting the insufficient Claims;
(3) TKT infringe Claim 26 of the 605 patent;
(4) Roche infringe Claims 26, 27, 28, 29, 30 and 31 of the 605 patent;
(5) The 678 patent is invalid on the grounds of obviousness and lack of novelty.
- This has been a difficult case involving a great deal of detailed technical evidence, a lot of documentary evidence (running to over 50 pretty full files) and much legal argument. I am grateful to all eight counsel for the assistance they gave me, orally and in writing, before, during and after the hearing. I also appreciate the way in which the solicitors involved prepared the evidence and organised matters during the hearing.
