Kiren-Amgen Inc v Board of Regents of University of Washington and Genetics Institute, Inc

[1995] APO 61

official notice

decision of a DEPUTY commissioner of patents

Application        :    No. 600650 in the name of Kiren-Amgen, Inc

Title:    Polypeptides of epo

Action:    Opposition by Board of Regents of University of Washington, and Genetics Institute, Inc

Decision:    Issued            . 

-invention relates to the DNA sequence encoding erythropoietin. Evidence shows that researchers consistently failed to identify the gene. Submissions that the invention was obvious based on commercial issues, and unfair advantages with regard to availability of the natural protein.

Claims novel and not obvious. Inventive step is assessed as against the hypothetical addressee. The actions of US participants in a race to clone the gene demonstrated that the necessary information was not obvious; evidence of Australian experts was ex post facto and did not address the real difficulties experienced by those trying to solve the problem.

Minor s.40 issues were identified, and applicant given an opportunity to amend.

patents act 1990

decision of a DEPUTY commissioner of patents

Re:Patent Application No. 600650 by Kiren-Amgen, Inc and Opposition by Board of Regents of University of Washington, and Genetics Institute, Inc

background

Patent application 600650 was filed under the Patent Cooperation Treaty on 11 Dec. 1984 as PCT/US84/02021, claiming priority from four US patent applications the earliest of which was dated 13 Dec. 1983. On 23 August 1990 the application was advertised accepted. Subsequently oppositions were filed by Board of Regents of University of Washington (hereafter referred to as Board of Regents), and Genetics Institute, Inc. (Genetics).

All parties provided substantial evidence in the opposition, including further evidence. A considerable portion of the evidence relates to proceedings in other jurisdictions; in particular, the corresponding proceedings before the European Patent Office, and those in the United States of America.

The matter was heard in Canberra over 28 to 31 March 1995. The applicant was represented by Dr. A Bennett of counsel, instructed by Mr I. Ernst (patent attorney of Shelston Waters, Sydney) and Mr G. Cox (patent attorney). The Board of Regents was represented by Dr J Emmerson of counsel, instructed by Dr V Santer (patent attorney of Griffith Hack & Co, Melbourne). Genetics was represented by Dr W Pickering (patent attorney of F B Rice & Co).

The invention relates to a substance called erythropoietin. The substance has biological importance in the treatment of inter alia anaemia - it being the protein that initiates the production of red blood cells in bone marrow. The evidence shows that the most effective current use of the substance is in patients suffering kidney failure. Production of erythropoietin is reduced or ceases with kidney disease, and treatment of such patients with erythropoietin dramatically improves their quality of life.

Apparently erythropoietin was one of a number of substances that were identified early in the biotechnology revolution as being especially desirable to produce through cloning. Naturally occurring erythropoietin is produced in extremely small quantities. Prior to this invention the amount of erythropoietin that had been obtained from natural sources was in microgram quantities, and was apparently inactive in vivo after purification. Consequently its potential medical efficacy was subject to conjecture. Nevertheless a number of organisations became engaged in what could best be described as a race to clone the protein. The activities of the participants in that race are discussed by the US District Court in Amgen Inc v Chugai Pharmaceutical Co Ltd 13 USPQ2d 1737. The following is a summary of pertinent points:

-CalTech

The search for the erythropoietin gene commenced in 1980 at the California Institute of Technology, where a Dr Hewick sequenced 26 amino acid residues at the N-terminal end, using a sample of naturally occurring erythropoietin. The sequence contained question marks at positions 3 and 7. With the benefit of hindsight, that sequence also contained an error at position 24.

This sequence was presented at a meeting of the American Society of Hematology in December 1981, by a Dr Goldwasser. In June 1983 the sequence was published in a paper by Sue & Sytowski. In hindsight, this sequence had errors at positions 7 and 24.

-Genetics Institute

Initial discussions about finding the erythropoietin gene commenced at Genetics Institute in 1981. In August 1982 a paper was prepared describing two cloning approaches that they would undertake. The first relied upon the available amino acid sequence of erythropoietin; the second involved obtaining additional erythropoietin and deriving additional sequence information.

During 1982 and early 1983, the research at Genetics Institute was based on the sequence obtained at CalTech; that is, on their first approach.

With respect to their second approach, they sought additional erythropoietin from a Dr Miyake in mid-1982, but they 'could not afford the terms demanded'. By May 1983 they sought extra erythropoietin from a Dr Sytkowski, but no agreement was reached. They found a source from a Dr Sherwood who had a cell line that was producing erythropoietin, but by December 1983 they had concluded that she was unable to produce enough. Dr Sherwood did pass on some erythropoietin obtained from Dr Goldwasser, but this was insufficient to obtain useful amino acid sequence information.

In April 1984 (after Amgen had cloned the gene) Genetics Institute received purified erythropoietin from Dr Miyake, derived sequence information from two tryptic fragments, screened a genomic library using two fully-degenerate probes, and found the erythropoietin gene. They subsequently cloned the gene and used it to construct a probe to screen a cDNA library constructed from fetal liver. They published the sequence of erythropoietin in the February 1985 issue of Nature.

-Biogen

Biogen began its erythropoietin project at the end of 1981, and continued until March 1985, investing some US$4M to $6M. They had two major strategies. The first made use of mRNA. The second was to obtain a source of the erythropoietin, sequence the protein, design probes corresponding to the sequence, and screen a cDNA library.

Prior to 1983, Biogen's knowledge of the sequence of erythropoietin came from the presentation by Dr Goldwasser in December 1981 (see above). Two Biogen scientists were at that meeting, and they copied the sequence as best they could - but in addition to the errors in the sequence presented, their copy included gaps. By the end of 1982 Biogen suspected there were errors in the sequence they were using.

Subsequently, when Biogen heard rumours that the sequence might be incorrect, Biogen attempted in 1983 to sequence erythropoietin they obtained from a Dr Zanjani - but without success. In late 1983 they obtained erythropoietin from a Dr Suyama from which they obtained N-terminal sequence information - but this had an error at position 24. They received a second shipment from Dr Suyama by February 1984 from which they obtained the sequence of the first 9 residues.

Biogen commenced screening a genomic library in early 1984. They succeeded in cloning the EPO gene in mid-1985, after their erythropoietin project had been officially terminated.

-Genentech

Genentech began work in 1981. Their project finished at the end of 1982 when they heard a rumour that Amgen had succeeded in cloning the gene. The rumour later proved to be untrue.

-Amgen

The present inventor, Mr Lin, joined Amgen in late 1981 some time after Amgen had commenced their erythropoietin project. At that time Amgen had limited and ambiguous sequence information on erythropoietin. They designed many probes to find the gene, but were unsuccessful. In hindsight, the reason was errors in the sequence. Initially, the probes were designed having regard to the probability of success, rather than using fully degenerate probes; by mid 1982 Lin was trying fully degenerate probes.

In June 1982 Amgen obtained 15 tryptic fragments of human erythropoietin from Dr Goldwasser. These were sequenced and probes designed, but Lin was unsuccessful in obtaining successful probes from these fragments. By mid 1983 Amgen was very frustrated with the project, and felt it was dead. However Lin persisted and obtained additional tryptic fragments in August 1983. Following sequencing of these fragments, Lin succeeded in isolating the erythropoietin gene from a genomic library using two sets of fully degenerate probes designed using the new sequence information.

Finally, Lin cloned the monkey cDNA sequence, and hybridized the human erythropoietin gene to monkey erythropoietin cDNA so that he could identify the introns in the human erythropoietin gene.

Against this background, the essence of the oppositions consist of three matters:

-is the invention claimed obvious;

-is the invention claimed novel having regard to one particular patent specification; and

-do the claims comply with s.40 of the Act?

The Specification

The description commences with a general overview of genetic manipulation. On page 9 it commences a discussion of erythropoietin and its attendant problems, and then commencing on page 18 provides a summary of the invention which parallels the claims. The major part of the 'detailed description' of the invention involves 10 examples which:

"are specifically directed to procedures carried out prior to identification of EPO encoding monkey cDNA clones and human genomic clones, to procedures resulting in such identification, and to the sequencing, development of expression systems and immunological verification of EPO expression in such systems."

In essence, the specification details:

-the identification of monkey EPO using a cDNA library;

-the identification of human EPO using a human Genomic library;

-the construction of vectors containing erythropoietin-encoding DNA;

-the development of mammalian host expression systems;

-manufactured genes encoding human erythropoietin and analogues thereof having preference codons for expression in certain hosts; and

-antibodies for identifying erythropoietin.

The specification ends with 56 claims. The independent claims are:

1. A purified and isolated polypeptide having the primary structural conformation and one or more of the biological properties of naturally-occurring erythropoietin and characterized by being the product of procaryotic or eucaryotic expression of an exogenous DNA sequence.

14.A purified and isolated DNA sequence encoding erythropoietin, said DNA sequence selected from the group consisting of:

(a)The DNA sequences set out in Tables 5 and 6 or their complementary strands; and

(b)DNA sequences which hybridize under stringent conditions to the DNA sequences defined in (a).

17.A purified or isolated DNA sequence consisting essentially of a DNA sequence encoding human erythropoietin.

18.A purified and isolated DNA sequence consisting essentially of a DNA sequence encoding monkey erythropoietin.

33.A DNA sequence coding for a polypeptide analogue of naturally-occurring erythropoietin.

34.A DNA sequence coding for [Phe15]hEPO, [Phe49]hEPO, [Phe145]hEPO, [His7]hEPO, [ASN2 des-Pro2 through Ile6]hEPO, [des-Thr163 through Arg166]hEPO or [delta27-55]hEPO.

39.A non-naturally occurring glycoprotein product of the expression in a non-human eucaryotic host cell of an exogenous sequence consisting essentially of a DNA sequence encoding human erythropoietin said product possessing the in vivo biological property of causing bone marrow cells to increase production of reticulocytes and red blood cells and having an average carbohydrate composition which differs from that of naturally occurring human erythropoietin .

45.A synthetic polypeptide having the amino acid sequence as set forth in Table V and having one or more of the in vivo or in vitro biological activities of naturally occurring monkey erythropoietin.

46.A synthetic polypeptide having the amino acid sequence as set forth in Table VI, other than a sequence of residues entirely within the sequence numbered 1 through 20, and having a biological property of naturally-occurring human erythropoietin.

48.An antibody substance characterized by immunoreactivity with erythropoietin and with a synthetic polypeptide having a primary structural conformation substantially duplicative of a continuous sequence of amino acid residues extant in a naturally-occurring erythropoietin except for any polypeptide comprising a sequence of amino acid residues entirely comprehended within sequences,

A-P-P-R-L-I-C-D-S-R-V-L-E-R-Y-L-L-E-A-K

55.A purified and isolated DNA sequence as set out in Table V or VI or the complementary strand of such a sequence.

I observe that these claims fall into four categories:

-claims limited by the DNA sequence encoding human or monkey erythropoietin (claims 14, 17, 18, 39, 45, 46 and 55)

-claims limited to variations of those sequences (claims 33, 34)

-claims that are not limited to human or monkey erythropoietin (claims 1, 33, and 48); and

-claims to an antibody recognising small parts of those sequences (claim  48).

CONSTRUCTION OF THE CLAIMS

The claims use a number of terms, which require some discussion.

erythropoietin. This is the word used to define substances covered by the invention. The specification details both human erythropoietin and monkey erythropoietin - making it clear that there is no basis in the specification to interpret erythropoietin as used in the specification to 'human' erythropoietin. Additionally:

-erythropoietin apparently has different amino acid sequences in different animals (compare Tables V and VI); and

-the properties of human erythropoietin are evidently dependent upon the specific glycosylation that occurs upon the expression of the protein by the cell - that is, the properties depend upon more than the mere amino acid sequence.

At the hearing I sought clarification of the scope of the term erythropoietin. After some considerable debate the applicants advised that the term erythropoietin, when unqualified, was 'shorthand for a polypeptide having the amino acid sequence of naturally-occurring erythropoietin'.

Having regard to the drafting of the specification I conclude that erythropoietin (unqualified) refers to a group of substances defined on the following basis:

1. naturally-occurring erythropoietins occur in animals and have the property of causing bone marrow cells to increase production of reticulocytes and red blood cells in animals - that is, they cause erythropoiesis;

1. the group contains:

a.naturally-occurring erythropoietins, not limited to any particular species of animal;

b.all polypeptides having the same amino acid sequence as naturally occurring erythropoietins; and

c.all polypeptides having an amino acid sequence which is an analogue of the amino acid sequence of naturally occurring erythropoietins.

primary structural conformation. This is a term used in the art to refer to the amino acid sequence of the polypeptide. By way of comparison, secondary structural conformation refers to the shape of the folded polypeptide.

purified and isolated. There is an element of tautology in this phrase, as isolation necessarily requires some degree of purification, and vice versa. Nevertheless I do not consider there to be a lack of clarity by use of the phrase. I simply take it to mean that the material in question has been removed from its host, and associated impurities reduced or eliminated.

Clarity Matters

During the hearing a number of minor matters regarding the claims arose; specifically:

-In claim 14, paragraph (a) refers to the complementary strands of the DNA sequences in Tables 5 and 6 as encoding erythropoietin - in fact the complementary strands hybridize to form a sequence encoding the erythropoietin, and rightly belong to paragraph (b) of claim 14 rather than paragraph (a).

-The applicants indicated claim 17 should read "A purified AND isolated ...", as in claim 18.

-The applicants acknowledged a lack of clarity in claim 46, asserting that it should read "A synthetic polypeptide having AN amino acid sequence ..." to remove an internal inconsistency in the claim. However, while this suggested change addresses the internal inconsistency in the claim, it seems to me that it likely introduces a lack of fair basis, there being no ready means to determine which fragments have the relevant biological properties.

These matters, although relatively minor in the context of this opposition, do constitute a lack of clarity in the claims and hence a lack of compliance with s.40 of the Act.

The opponents also argued that the term 'stringent conditions' in claim 14 was not clear, and that claims 40 and 41 were not clear through the absence of an upper limit to the specified production rate. I am not satisfied that there is any lack of clarity in these matters. The term 'stringent conditions' is well known in the art; and there is no reason why a range cannot be defined by one limit alone.

Obviousness

The prime question in this opposition is whether the invention claimed is obvious, although it was not a ground of opposition argued by Genetics. In addressing this question I am mindful that the courts have often stressed that this question is a jury-type question - that is, the question of obviousness requires an objective assessment of the facts of the particular case. And in considering the facts of the case care must be taken to avoid any ex-post facto analysis; see for example Minnesota Mining and Manufacturing Co. v. Beiersdorf (Australia) Ltd. (1980) 144 CLR 253 at pages 293-4 where Aiken J. quotes the words of Fletcher-Moulton L.J. in British Westinghouse Electric and Manufacturing Co. Ltd. v. Braulik ((1910) 27 RPC 209 at p.230):

"I confess that I view with suspicion arguments to the effect that a new combination, bringing with it new and important consequences in the shape of practical machines, is not an invention, because, when it has once been established, it is easy to show how it might be arrived at by starting from something known, and taking a series of apparently easy steps. This ex post facto analysis of invention is unfair to the inventors, and in my opinion it is not countenanced by English Patent Law."

The present application was filed well before the commencement of the Patents Act 1990; consequently I am must determine whether the invention is obvious having regard to the common general knowledge in Australia at the priority date (3M v Beiersdorf 144 C.L.R. 253). That is, my determination of obviousness must depend upon the evidence as regards to common general knowledge in Australia, and not evidence of activities outside of Australia.

At the hearing, the Board of Regents, and Amgen, made extensive submissions on the issue of obviousness. Most of those submissions involved analysing where a potential inventive step lay in the particular methodology used to clone erythropoietin. I do not intend to deal with all those submissions - but only those necessary to establish either the presence, or absence, of an inventive step.

Both parties have provided evidence from Australian experts, which is in basic conflict. In addressing the question of obviousness I must resolve this conflict. Significantly I note that the Australian declarants were not actively involved in the search for the erythropoietin gene. On the other hand there is much evidence on file of the activities overseas of the participants in the race to find the gene. A consideration of that evidence (albeit in respect of activities that took place overseas) will assist me in identifying the critical issues that actually faced the participants in the race. I can then resolve the conflict in the Australian evidence by having regard to the manner in which that evidence addresses those critical issues.

The Race to Clone Erythropoietin

The activities of the participants in the race are summarised in some detail in the decision of Amgen Inc v Chugai Pharmaceutical Co Ltd (supra) at pp 1745 to 1754 - see my summary at the commencement of this decision. This sets a scene where a large amount of time, money, and effort were being expended to clone the gene without success. It is obvious from this summary that the participants were all 'groping' for a solution to the cloning problems.

There are several issues evident from this summary:

1. It was clearly obvious to the participants in the race to try to clone erythropoietin;

1. the process of cloning was critically dependent upon knowledge of at least some significant part of the amino acid sequence of naturally occurring erythropoietin. Also, any errors in that sequence were likely to be fatal to the search;

1. naturally occurring erythropoietin was exceedingly difficult to obtain in significant amounts;

1. despite the scarce supply, a partial sequence containing the first 26 amino acids of the protein had been published;

1. searches based on the published sequence of erythropoietin failed; and

1. with the benefit of hindsight, both the published sequence, and independently derived sequences contained errors; but

1. the cause of failure in cloning the gene was not immediately ascribed to errors in the published amino acid sequence.

Significantly, the participants largely relied on the available sequence information. The cause of their failure to clone the gene could have been a range of factors including (but not necessarily) an incorrect amino acid sequence. It was only after considerable time and effort that they came to the conclusion that the available sequence(s) might contain errors. Further, having come to this conclusion, the researchers were clearly unable to correct those errors without deriving a new sequence for erythropoietin.

Although there was a shortage of erythropoietin, there was enough for several researchers to independently derive further partial sequences. These sequences were not significantly different from the published sequence, themselves contained errors, and did not correct the errors in the published sequence.

The Board of Regents argued that the lack of ready supply of erythropoietin prevented the determination of a reliable amino acid sequence. They further argued that Amgen succeeded in cloning the gene merely because they were able to secure a significant supply of erythropoietin from Dr Goldwasser and could therefore derive the correct amino acid sequence. They submitted this merely amounted to a commercial advantage - not an inventive step. This argument is superficially very attractive. However it is not supported by the actions of the participants in the race at the relevant time. For example, the Biogen scientists copied the CalTech sequence at the seminar in December 1981, and it wasn't until the end of 1982 before they 'suspected' the sequence might be in error. Even then they did not try to obtain erythropoietin to sequence until they heard rumours that the sequence might be incorrect. Then there is Genetics Institute, whose research in 1982 and early 1983 was mostly based on the published sequence of erythropoietin. Although they tried to acquire more erythropoietin to obtain additional amino acid sequence information, they 'could not afford the terms demanded'. If they really thought the published sequence was erroneous, why did they continue to rely upon it? Further, the decision that the erythropoietin was unaffordable was clearly a commercial decision based upon their expectations of what was needed to solve the problem at that time. The fact that they didn't follow that path indicates that at that time they did not consider the path would directly lead to success. It is also noteworthy that in April 1984, after Amgen had cloned the gene, Genetics Institute was suddenly able to afford the necessary erythropoietin and complete the cloning process.

It is clear to me from these actions that the participants did not have a clear understanding of why they were unable to clone the gene. The path of obtaining more erythropoietin was only one possibility. It was per se a desideratum, but in the context of its scarcity and the consequences arising therefrom, it was not the obvious path to follow. It is only with the benefit of hindsight - a classic ex post facto analysis - that the acquisition of erythropoietin at any cost is seen as crucial to solving the problem.

From these facts I conclude that, at the priority date, the amino acid sequence of erythropoietin was unknown at the level required to identify the erythropoietin gene. Further, to the level of knowledge required to identify the gene, the sequence was manifestly not obvious to the participants - as indicated by the failure of the participants to correct the errors in the known sequence, and the series of failures to correctly sequence the protein. Furthermore I consider this conclusion to have great weight as it is drawn from the actions of participants actually trying to solve the problem - not someone's opinions or conjectures of what a skilled person would have done.

It follows that the DNA sequence, whose identification depended upon knowledge of the erythropoietin sequence, was not obvious. It likewise follows that any products derived as a result of the knowledge of that gene were likewise not obvious.

The Australian Evidence

Against this factual background of the participants in the race to clone erythropoietin, I will now consider the evidence of the Australian declarants.

The applicant provided declarations by Messrs Goding, Firkin, Johnson, Gray, Horvath, Linnane, Schofield, and Mattick as evidence from experts in Australia. I shall not generally elaborate on their evidence, other than to observe that it is consistent with the actions of the participants in the race.

However, I do note the declarations by Dr. Firkin and Dr Johnson. They both declare that to the best of their knowledge, in 1983 there were only two groups engaged in actual research with erythropoietin in Australia, groups led by each of them respectively. Firkin declares that to the best of his knowledge no research was being conducted in Australia on the specific structure of erythropoietin; that prior to 1985 research organisations in the USA led the world in this field; that there was nothing in the literature that led him to consider cloning erythropoietin; that in 1983 he did not have the necessary skills to clone the gene; and there simply was insufficient information known by him at that time. Johnson similarly declares that to the best of his knowledge no research was being conducted in Australia on the specific structure of erythropoietin; and that prior to 1985 research organisations in the USA led the world in this field. He further declares that in the period 1982-1983 he considered cloning the erythropoietin gene, but considered there was insufficient information available in the published literature to propose a course of research with a reasonable likelihood of success. He also declares that he was aware that a number of US research laboratories possessed a partial amino acid sequence, and were actively engaged in research towards cloning the gene, and accordingly focussed his attention on different matters.

I observe that Firkin and Johnson were apparently the Australian researchers working closest to the present invention. There is nothing in their declarations which is inconsistent with the actions of the participants in the race, and they are attesting to the inventiveness of the invention.

For the Board of Regents, the principal declaration is by Dr. Hudson. Dr Hudson covers in depth the development of recombinant DNA technology, both generally and in Australia. He concludes that it was obvious in December 1983 to attempt to clone erythropoietin. As regards the sequencing of erythropoietin and the design of probes, he declares as follows:

In para 22:"... Indeed, isolation, characterization, and enrichment of mRNA for EPO had been described. I do not consider that it would have been essential that the sequence of the protein should be known, since this could be determined concurrently with the purification, using conventional techniques in use in Australia at the priority date."

In para 27:"... Kirin-Amgen allege in the specification that the sequencing of the EPO molecule presented particular difficulties. I consider that well-known methods were readily available to overcome these problems, and I consider that there were many people in Australia at that time who were certainly capable of sequencing the molecule.

"Sequencing of the EPO protein in order to design oligonucleotide probes would have been well within the capability of many protein chemists in Australia at the priority date. The amount of protein available was not limiting, and the protein could have been purified using known techniques (the method of Miyake et al (1977), referred in the specification at page 16. An alternative, simpler purification method is described by .....)"

In para 32, sub-para 26:     "... Thus, although fully redundant probes are "impractical", in the sense discussed above, with perseverance and the use of sequences containing the minimum number of redundancies which can be located, I am confident that erythropoietin could have been cloned using oligonucleotides predicted from Sue's N-terminal protein sequence."

In para 33:"... I have no doubt that a fully redundant set of oligonucleotide probes could have been made in Australia at that time."

In my view Hudson's declaration suffers from being an ex post facto analysis of the situation; it overlooks the real difficulties that faced those trying to solve the problem. In particular, he has failed to demonstrate that the problems discussed above that were experienced by the participants in the race would not have applied in Australia. There is no reason for me to doubt that the various assertions raised by Hudson as regards the skills of Australian researchers applied equally to the participants. None of those assertions indicate to me that a skilled person in Australia at the relevant date would have had any relevant advantage as compared to the participants in the race. And there is nothing in Hudson's declaration that indicates the level of knowledge in Australia at the priority date was greater than the knowledge inherent in the actions of the participants in the race.

The Board of Regents provided a number of other declarations. Hertzog does little more than state agreement with Hudson's declaration, and provides 'brief answers' to a questionnaire associated with proceedings before the European Patent Office. Wettenhall, Mercer and Cutler discuss their view of relevant expertise in Australia at the relevant time. But none of them adequately address the real difficulties that faced those trying to solve the problem beyond ex post facto assertions.

Accordingly I prefer the evidence of the applicant over that of the Board of Regents. I conclude that before the priority date, the amino acid sequence of erythropoietin was not known in Australia at a level sufficient to identify the erythropoietin gene; more particularly, the required knowledge was not obvious in Australia. Accordingly, the DNA sequence for erythropoietin was not obvious, and products derived from a knowledge of that sequence were similarly not obvious. Accordingly I find that the ground of obviousness has not been made out with respect to any of the inventions claimed.

To conclude this discussion of obviousness, there are two issues on which I need to make observations.

Availability of naturally-occurring erythropoietin

It was suggested that the principle reason Amgen was successful in identifying the erythropoietin gene was because, as was allegorically put to me, they had a 'fairy godmother' in Dr Goldwasser. It was suggested that Dr Goldwasser was selective in supplying erythropoietin; that Amgen had an unfair advantage because he provided them with more erythropoietin that others; and that the principle reason for Amgen's success was their access to erythropoietin in quantity via Goldwasser.

Parenthetically, the source of naturally occurring erythropoietin was urine of anaemic patients. Part of the reason for low availability of the protein was that medical treatment for anaemia in the USA and Europe limited the degree of anaemia permissible in patients. However in Japan the permissible level of anaemia was lower, resulting in a 100-fold increase in the level of erythropoietin present in the urine. A Dr Miyake had access to this urine, and supplied erythropoietin from it to Goldwasser for purification. It would seem that Goldwasser's total supply of erythropoietin was obtained in 1976 from urine collected by Miyake in Japan; he held a relatively substantial amount of it; but he only released it in small quantities.

If a person deliberately removed from public availability things which were 'essential' for an invention to be made, I do not think that (or a related) person could claim inventiveness by reason of the artificially created state of scarcity of the material. But that is not what occurred here. It is apparent that Goldwasser was fortunate in obtaining a sizable stock of erythropoietin. But there is nothing in the evidence that is at all suggestive that Goldwasser collected 'available' erythropoietin and removed it from general availability. Indeed the situation seems to have been the exact converse - erythropoietin was generally scarce; Goldwasser had isolated a substantial quantity; and he made it publicly available (albeit subject to conditions, and only in limited quantities) when it otherwise would not have been. His actions would seem to be consistent with a person who has created a valuable asset seeking to dispose of it carefully; there is absolutely nothing to suggest that he sought to prevent (directly or indirectly) persons obtaining erythropoietin by any other means.

Consequently I do not think Goldwasser having a large stock of material has any bearing on the fact that the material was scarce and not readily available in large amounts. That scarcity was part of the common general knowledge at that time; it was not an artificially created scarcity (in the sense of it not previously being scarce); and it was the environment in which the notional person skilled in the art was in fact working.

The Genentech decision

Extensive submissions were made on the basis of the decision of the UK Court of Appeal in Genentech [1989] RPC 147. I have also read the more recent decision of Biogen Inc v Medeva plc [1995] RPC 25, which follows the decision in Genentech. Despite these submissions, I have not found these cases to be of assistance. In both the applicant made very significant concessions which would seem to have had great bearing on the ultimate finding in regard to inventive step. Such concessions were not made in the present matter.

Novelty

The opponents argued that the invention claimed was not novel on the basis US patent specification 4377513 (Sugimoto). That specification describes a process for the production of human erythropoietin involving human lymphoblastoid cells capable of producing human erythropoietin. Those cells are created by fusing lymphoblastoids with human kidney cells.

The opponents argue that claim 1 was anticipated by Sugimoto. I disagree. Claim 1 requires the protein to be the product of the expression of an exogenous DNA sequence. Exogenous means 'having its origins external; derived externally' (Macquarie Dictionary). An exogenous sequence is a sequence that is of external origin to the cell. At one level, one could argue that all material in a fused cell was exogenous. However I think the term imports a connotation of an existing cell having some material incorporated from externally, there being a continuity of the basic cell properties; and I accept the evidence of Dr Gray (for the applicant) on this point. I do not think the term extends to the material present in a fused cell by reason of the fusion alone.

The same reasoning applies in respect of claim 39, which in addition is limited to a non-human host cell (which is not disclosed in Sugimoto). Additionally, Sugimoto does not disclose the DNA sequence for erythropoietin; therefore those claims directed to 'purified and isolated' sequences, or to analogues or changed sequences, are not disclosed. I am satisfied that none of the claims read on to the disclosure of Sugimoto, and consequently no objection of lack of novelty arises.

Manner of Manufacture

Although not directly argued, the issue of manner of new manufacture arises in the issue of whether the invention relates to a mere discovery - that is, the discovery of the DNA sequence encoding erythropoietin.

The present invention fundamentally relies upon the discovery of the DNA sequence encoding erythropoietin. In my view a claim directed to naturally occurring DNA characterised by specifying the DNA coding for a portion of that molecule would likely be claiming no more than a discovery per se and not be a manner of manufacture.

The present specification contains claims to DNA sequences, in two categories:

-Claims 14, 17, 18, and 55, which claim a 'purified and isolated' sequence limited to that specified in Tables V or VI, or limited to being 'essentially' the sequence encoding erythropoietin. These claims are directed to a molecule which is a fragment of the full chromosome. They do not claim the naturally occurring chromosome, or any other naturally occurring entity. By being directed to a purified and isolated DNA sequence they claim 'an artificially created state of affairs'.

-Claim 33 claims

"A DNA sequence coding for a polypeptide analogue of naturally-occurring erythropoietin."

and claim 34 claims

"A DNA sequence coding for ....",

specifying human erythropoietin with a range of substitutions or deletions.

Both claims include within their scope a full length chromosome containing the relevant sequence, because they are not restricted to a 'purified and isolated' sequence. However both claims are directed to molecules which have been deliberately changed from the naturally occurring form - that is, they are directed to artificially created states of affairs.

I also observe that an objection of manner of manufacture might arise if the claims were directed to a mere chemical curiosity; but that is plainly not the case with this invention.

Accordingly I am satisfied that an objection of manner of manufacture does not apply to any of the claims.

Fair Basis

The law on fair basis was recently considered by the full Federal Court in CCOM v Jiejing (1995) AIPC 91-079. Key aspects from that decision are:

-the Mond Nickel test is relevant, but not conclusive - it is not a substitute for the application of the terms of the legislation;

-the term 'fair' does not involve taking into account the fairness in a broad sense of the applicant's conduct; 'what is required to be fair is the basis which one document affords for a claim in the other'; and

-what is required is "a real and reasonably clear disclosure".

Thus, as was said by Barwick C.J. in Olin Corporation v Super Cartridge Co. Pty Ltd (1977) 51 A.L.J.R. 525 at 537:

The question whether the claim is fairly based is not to be resolved, in my opinion, by considering whether a monopoly in the product would be an undue reward for the disclosure. Rather, the question is a narrow one, namely whether the claim to the product being new, useful, and inventive, that is to say, the claim as expressed, travels beyond the matter disclosed in the specification."

The attack on fair basis is essentially that the claims should be limited to the specific sequences of tables 5 and 6, to methods for producing those sequences, and to variants of those sequences specifically referred to.

Tables 5 and 6 disclose the specific DNA sequences for human and monkey erythropoietins. The specification also teaches that as a result of the invention DNA hybridization processes can be used to locate the erythropoietin gene position "in the human, monkey and other mammalian species chromosomal map". This statement is made presumably in reliance upon the well-known substantial homology of gene sequences across species. Further, it is well known that there is a redundancy in the DNA coding for any particular amino acid; and that one can make conservative substitutions for amino acids without making significant changes in the properties of the protein. It seems to me that, properly considered, the discovery of a natural DNA sequence is tantamount to the discovery of a class of compounds - which class would be readily understood by a person skilled in the art. Accordingly I am satisfied that there is sufficient teaching to provide a fair basis to claims to erythropoietin unlimited either by species or specific structure. (And I would also observe that if it was subsequently found that a particular variation of the sequence gave rise to new and surprising results, the law of selections would apply.)

The opponents also argue that an essential feature of operative erythropoietin is the glycosylation that occurs in its expression from a host cell; that the glycosylation pattern is not known in detail, but is dependent upon the host cell; that the applicant has only show utility in certain host cell(s); and that therefore they should be limited to the use of the sequence in specific host cells. I do not agree. I do not think there is any basis to limit the present claims to specific host cells under the guise of fair basis. The applicant has identified the DNA sequence, and apparently provided a best method of performance with the sequence being incorporated into specific host cells. It seems to me to have been common general knowledge at the relevant date that, having isolated a sequence one could appropriately insert it into any suitable host cell and gain some expression of the protein. It is well known that the glycosylation can vary with the host cell expressing the sequence. It is further known that the glycosylation pattern may, or may not, affect activity - but the activity of the expressed protein is at most an issue of utility, which is not available in these opposition proceedings. Further, if subsequently someone establishes that expression is greatly enhanced if the sequence is cloned into a particular cell line, that is properly an issue of selection - not of fair basing. Accordingly I consider the claims to be fairly based in this respect also.

Claim 1 characterises the substance claimed by reference to it having 'one or more of the biological properties ...'. The characterisation is not limited to the properties identified in the specification. The specification discloses certain biological properties of naturally occurring erythropoietin, but it does so by way of exemplification rather than a specific list of known properties. I am not convinced that the exemplification in the specification can reasonably be said to be an exhaustive list of all the biological properties or activities of erythropoietin. Consequently, to the extent that claim 1 characterises the substance by reference to unlimited biological properties rather than those either specifically identified or otherwise well known at the priority date, it cannot be fairly based on the specification.

A related problem arises with respect to claims 45 and 46. These claims include within their scope synthetic fragments of erythropoietin characterised by having 'a biological property'. The specification identifies certain fragments having relevant biological activity. It does not provide a basis for identifying all such fragments, and it would take an undue burden of experimentation for a skilled worker to determine them. Neither does the specification disclose a class of fragments with similar activity or properties (with the consequence that selection law does not apply). Consequently these claims, being unrestricted to specific fragments, are not fairly based on the specification.

A similar problem occurs with claim 48, which is directed towards an 'antibody substance' which can bind to erythropoietin (including fragments but excluding a particular fragment) and dependant claims 49 and 50. The specification gives inadequate teaching about which fragments of erythropoietin are immunogenic, and it would take an undue burden of experimentation for a skilled worker to determine these. Consequently I consider that there is only a basis for claims to the specific antibodies referred to in the specification.

Finally, claim 3 (dependent upon claim 1) characterises the exogenous DNA sequence as a cDNA sequence. Similarly claim 22 qualifies claim 17 as being a cDNA sequence. Genetics argue that the applicant has not disclosed isolating human cDNA - only monkey cDNA - and consequently their claims should exclude human cDNA. The fundamental difference between a DNA sequence and a cDNA sequence is the presence/absence of introns. To obtain a DNA sequence free of introns a principal technique is to use a cDNA library constructed from the messenger RNA expressed by relevant cells. However it seems to me that if one can identify the introns in a DNA sequence, one has identified the sequence absent the introns - ie the cDNA sequence. It is in my view irrelevant whether a cDNA library was used to identify that sequence, and accordingly claim 3 is fairly based with respect to this issue.

Conclusion

I have found:

-the claims are novel, and involve an inventive step;

-claims 14, 17, and 46 do not comply with section 40 in that they involve some minor lack of clarity; and

-    claims 1, 45, 46, and 48 to 50 are not fairly based.

The defects can be readily overcome, and accordingly I allow the applicant 60 days from the date of this decision to propose appropriate amendments.

Costs

In actions before the Commissioner, costs usually follow the event. In this case the applicant was successful on the major issues. I do not think the s.40 findings are sufficiently significant to derogate from the normal position. Accordingly I award costs against the opponents.

In making this award, I note both oppositions were heard concurrently. I do not think it appropriate that the applicant claim their full costs associated with their attendance at the hearing from each of both opponents; I consider the applicant ought only to be entitled to one claim for such costs. Further, I consider those costs should be apportioned between the opponents having regard to the extent of their oppositions and their degree of involvement at the hearing. Accordingly, in respect of items 10, 11 and 12 of part 1, and part 2, of schedule 8 I direct that costs be taxed as if there were but one opposition; 80% of those costs being awarded against the Board of Regents, and 20% against Genetics.

D Herald
Deputy Commissioner of Patents

Patent attorneys for the applicant  :  Shelston Waters, Sydney

Patent attorneys for the Board of Regents :  Griffith Hack & Co, Melbourne

Patent attorneys for the Board of Genetics:  F B Rice & Co, Sydney