Board of Patent Appeals and Interferences
Patent and Trademark Office (P.T.O.)
*1 FIDDES ET AL.
BAIRD ET AL.
Interference No. 102,229
October 27, 1993
Final Hearing August 25, 1993
Basic Fibroblast Growth Factor
Application of John C. Fiddes and Judith A. Abraham, filed December 16, 1985, Serial No. 06/809,163. Accorded benefit of Serial No. 06/775,521 filed September 12, 1985.
Application of Andrew J. Baird, Frederick S. Esch, Denis Gospodarowicz, Peter Bohlen and Nicholas C. Ling, filed December 31, 1987, Serial No. 07/139,953. Accorded benefit of Serial Nos. 06/747,154 filed June 20, 1985, now Patent No. 4,956,455 issued September 4, 1990; 06/940,524 filed December 10, 1986, now Patent No. 4,785,079 issued November 15, 1988; and PCT/US 86/01318 filed June 18, 1986.
Before Goolkasian, Sofocleous and Downey
Administrative Patent Judges [FN1]
Administrative Patent Judge
This interference involves an application of Fiddes et al. (Fiddes), assigned to California Biotechnology, Inc., and an application of Baird et al. (Baird), assigned to Salk Institute for Biological Studies.
The subject matter of this interference relates to recombinant DNA molecules encoding basic fibroblast growth factors. Fibroblast growth factor (FGF) is a protein which is active in causing endothelial and fibroblast cells to grow and divide and useful in effecting accelerated healing of wounds, bone fractures, burn tissue and other trauma. The subject matter at issue is defined by count 2 which is as follows:
Fiddes' claims 43 to 46, 50 to 54 and 57 to 69 and Baird's claims 11 to 21 and 23 to 28 correspond to the count.
After rendering the Decision on Preliminary Motions, each party was assigned a testimony period. Both parties took testimony and filed briefs. During the intervening period between the setting of testimony times and final hearing, our reviewing court issued its decision of Amgen Inc. v. Chugai Pharmaceutical Co., 977 F.2d 1200, 18 USPQ2d 1016 (Fed.Cir.1991); Fiers v. Sugano, 984 F.2d 1164, 25 USPQ2d 1601 (Fed.Cir.1993); and In re Bell, 991 F.2d 781, 26 USPQA2d 1529 (Fed.Cir.1993). The parties' briefs address the Amgen decision. After the briefs were filed, a schedule was set for filing supplemental briefs to address the applicability of the Fiers decision. Both parties appeared, through counsel, at final hearing [FN2].
1. Whether Baird's claims 11 to 21, 23, 24, 26 and 27 corresponding to the count are unpatentable under 35 U.S.C. 102(b) and 103?
3. Whether an interference-in-fact exists?
*2 4. Whether Fiddes derived its invention from Baird?
Patentability of Certain Baird's Claims
We hold that Baird's claims 11 to 21, 23, 24, 26 and 27 are unpatentable under 35 U.S.C. 102(b) and 103.
By not arguing the merits of the prior art references relied on in the preliminary motion (Paper No. 19) for judgment, Baird has conceded that the claims are anticipated under 35 U.S.C. 102(b) by the references. Cf. Guglielmino v. Winkler, 11 USPQ2d 1389 (BPAI 1989), rv'd. on other grounds, slip op. at 17 USPQ2d 1175 (Fed.Cir.1990). However, it is Baird's position that the references are not available as prior art against him because he has antedated these references by the June 20, 1985 filing date of his earlier filed application Serial No. 06/747,154.
It is Fiddes' position that Baird is not entitled to the benefit of the June 20, 1985 filing date of application Serial No. 06/747,154 with respect to the invention of these claims, because the application does not contain a written description of recombinant DNA molecules consisting essentially of DNA sequences encoding the genus of mammalian basic FGFs and does not enable a person skilled in the art to make recombinant DNA molecules consisting essentially of DNA sequences encoding the genus of mammalian basic FGFs.
Baird's claimed invention is directed to a method of producing a mammalian basic FGF (bFGF) (claims 11 to 18), a microorganism transformed with an expression vehicle capable of expressing bFGF (claims 19 to 21), an isolated or synthetic, substantially pure DNA sequence encoding mammalian FGF (claims 23, 24 and 26) and a cloning vector including the DNA sequence encodingmammalian FGF (claim 27) [FN3].
Since the '154 application has issued as U.S. Patent No. 4,956,455, we will refer to the patent disclosure. The patent contains the following disclosure relevant to our inquiry:
The present invention is directed to fibroblast growth factor (FGF) produced by synthetic methods, which will substantially enhance the availability of mammalian FGF. [Col. 1, lines 7 to 10]
Although the above-identified patent applications [CIP applications whose benefit Baird has not sought or been accorded] describe methods of purifying FGF from mammalian tissue, such as bovine pituitary tissue, these procedures may be difficult to scale up to large scale production. [Col. 1, lines 50 to 53]
The present invention provides pure FGF which may be produced by synthetic methods, and if so produced should substantially enhance the availability of mammalian FGF. [Col. 1, lines 54 to 57]
The present invention provides pure bovine pituitary fibroblast growth factor (bpFGF) which may be synthesized using recombinant DNA techniques or other suitable techniques. [Col. 1, line 60 to Col. 2 line 1]
*3 The invention provides the first known pure mammalian FGF, particularly bpFGF, and the production thereof by synthetic methods. [Col. 2, lines 50 to 52]
The '455 patent, column 2, lines 60 to column 3, line 22, sets forth the 146 amino acid sequence for bovine pituitary FGF. Example 2 sets forth a theoretical DNA sequence which will encode bovine pituitary FGF. As an alternative to a theoretical DNA sequence, the patent, column 5, lines 15 to 30, teaches that cDNA corresponding to bovine pituitary FGF may be prepared as follows:
A cDNA library or an expression library is produced in a conventional manner by reverse transcription from messenger RNA (mRNA) from a bpFGF-producing cell line. To select clones containing bpFGF sequences, hybridization probes (preferably using mixed probes to accommodate the degeneracy of the genetic code), corresponding to portions of the FGF protein are produced and used to identify clones containing such sequences. Screening of the expression library with FGF antibodies may also be used, alone or in conjunction with hybridization probing, to identify or confirm the presence of bpFGF-encoding DNA sequences in DNA library clones. Such techniques are taught, for example in CSH [T. Manatis et al., Cold Spring Harbor Laboratory Manual, Cold Spring Harbor New York (1982)], supra.
An applicant may antedate prior art by relying on the benefit of a previously filed U.S. application to establish an effective date earlier than that of the reference. See 35 U.S.C. 120. To be entitled to the benefit of an earlier filed application, the application must disclose the claimed subject matter in compliance with 35 U.S.C. 112, first paragraph. In re Gosteli, 872 F.2d 1008, 10 USPQ2d 1614 (Fed.Cir.1989); and In re Scheiber, 587 F.2d 59, 199 USPQ 782 (CCPA 1978). § 112, first paragraph, requires that the application contain ...
a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains ... to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out the invention.
The written description requirement of 35 U.S.C. 112, first paragraph, requires that an application convey with reasonable clarity to those skilled in the art that, as of its filing date, applicant was in possession of the invention. Vas-Cath Inc. v. Mahurkar, 935 F.2d 1555, 19 USPQ2d 1111 (Fed.Cir.1991). In Fiers v. Sugano, supra, the court discussed a DNA invention from the standpoint of both conception and the written description requirement. The discussion is applicable to our written description inquiry. At 25 USPQ2d 1606, the court stated:
*4 An adequate description of a DNA requires more than a mere statement that it is part of the invention and reference to a potential method for isolating it; what is required is a description of the DNA itself.
If a conception of a DNA requires a specific definition, such as by structure, formula, chemical name, or physical properties, as we have held, then a description also requires that degree of specificity. To paraphrase the Board, one cannot describe what one has not conceived.
In our view, the '455 patent does not contain a written description for the broad class of mammalian FGFs. The patent teaches no amino acid or DNA sequences for any mammalian FGF other than bovine pituitary FGF. For bovine pituitary FGF the patent teaches both the amino acid sequence and a theoretical DNA sequence for the factor. The patent does not teach the naturally occurring gene encoding the factor and thus, Baird was not in possession of the naturally occurring gene encoding bovine pituitary FGF [FN4]. Nor was he in possession of either the amino acid sequence for any other mammalian FGF or any naturally occurring gene encoding any mammalian FGF other than bovine pituitary FGF. Thus Baird was not in possession of the genes encoding mammalian FGFs since in 1987, knowledge of the amino acid sequence of a protein coupled with the established relationship in the genetic code between a nucleic acid and the protein it encodes would not establish possession of the gene encoding that protein. In re Bell, supra.
During oral argument, counsel for Baird argued that the Bell decision is inapposite since the decision does not reflect the state of the art or the skill of the art as of the June 1985 filing date of the '154 application. Counsel contends that the decision is applicable to state of the art as of 1981, the purported filing of the Bell application. We do not agree. First, the Bell application was filed in the PTO in 1987 and not in 1981. Furthermore, Baird's witness, Dr. Constance L. Cepko, a Ph.D. biologist, testified that a great deal of gene research was conducted between 1984 and 1989 and that "the basic strategies that were employed were the same," i.e., the use of oligomers and screening with oligomers in 1989 were very close to the same techniques as used in 1985. See the Baird record at pages 308, 410 and 411 (BR 308, 410 and 411). Thus, Baird's premises as to the state and level of skill in the art are not valid.
Turning to Baird's description of the DNA sequence encoding FGF, we note that he did not postulate the correct sequence for the naturally occurring gene but rather a theoretical DNA sequence for the bovine pituitary FGF out of the myriad possibilities [FN5]. Dr. Dudley D. Moore, a Ph.D. molecular biologist testified that Baird's theoretical DNA sequence differed from the naturally occurring sequence in 106 out of the 438 possible nucleotides. See the Fiddes record at 1104 and 1107 (FR 1104 and 1107). Dr. Moore also testified that the third nucleotide in a codon is often the critical choice, and that the Baird theoretical sequence differs from the naturally occurring sequence in 51 out of 146 codons. FR 1120 and 1141 to 1142. Thus, Baird was not in possession of the naturally occurring gene for bovine pituitary FGF or any other gene for any mammalian FGF at the time of filing of the '154 application.
*5 Since we have decided that the Baird '154 application does not contain a written description for the broad class of mammalian FGFs, Baird is not entitled to the benefit of that application. Under these circumstances, the question of whether the application contains an enabling disclosure for its claims 11 to 21, 23, 24, 26 and 27 is moot. However, we have reviewed the parties' records in light of the arguments raised in the briefs and agree with Fiddes that the Baird '154 application does not contain an enabling disclosure for those claims. Thus on this record with the evidence before us, we do not agree with Baird that a person skilled in the art of molecular biology would, as of June 1985, have been enabled, without undue experimentation, to clone the native gene encoding bovine basic FGF (and other homologous mammalian basic FGFs) having available the Baird disclosure of the full 146 amino acid sequence of basic FGF.
For the foregoing reasons, we hold that Baird's claims 11 to 21, 23, 24, 26 and 27 are unpatentable.
Substitution of Count 2 for Count 1
We have reviewed the Decision on Preliminary Motions granting Baird's preliminary motion to redefine by substituting count 2 for count 1 and agree with Fiddes that count 2 is not a proper count for this interference.
The decision complained of substituted count 2 for count 1 because of the holding that Baird's claims 11 to 21, 23, 24, 26 and 27 were patentable, i.e., Baird was entitled to the benefit of the '154 application. The substitution of count 2 was therefore necessary to embrace Baird's broadest claims [FN6] directed to mammalian FGFs which claims were designated as corresponding to count 1.
Since we have held that these claims are not patentable, we would normally substitute a narrower count for count 2 of this interference so as to encompass the broadest patentable claim of each of the parties. See the Manual of Patent Examining Procedure (MPEP), § 2309.01, Fifth Edition. The count is the vehicle for determining priority of invention and a proper count embracing the common patentable subject matter is necessary in order to make the determination. This interference, however, is not the normal situation since the interference will not proceed to a priority determination. In this case, since we have held, infra, that an interference-in-fact does not exist, we need not make a priority determination. Thus, we need not substitute a narrower count.
We hold that an interference-in-fact does not exist.
The interference having been declared under 35 U.S.C. 135(a), it is presumed that an interference-in-fact exists and that each party's claims designated as corresponding to a count define the same patentable invention as the count. An "interference-in-fact" exists when at least one claim of a party which corresponds to a count and at least one claim of an opponent which corresponds to the count define the same patentable invention. 37 CFR 1.601(j). A party urging that an interference-in-fact does not exist has the burden to show that its claims corresponding to the count are directed to a separate patentable invention from each of its opponents' claims designated in the notice as corresponding to the count. Case v. CPC International, Inc., 730 F.2d 745, 221 USPQ 196 (Fed.Cir.1984), cert. den. 105 S.Ct. 223, 224 USPQ 736 (1984); Hsing v. Myers, 2 USPQ2d 1861 (BPAI 1987). The test for interference-in-fact is set forth in § 1.601(n), which provides that an invention "A" is a separate patentable invention with respect to invention "B" when invention "A" is new (35 U.S.C. 102) and non-obvious (35 U.S.C. 103) in view of invention "B", assuming invention "B" is prior art with respect to invention "A".
*6 We now turn to the parties' claims to determine whether Fiddes' claims 43 to 46, 50 to 54 and 57 to 69 are new and nonobvious over Baird's claims 11 to 21 and 23 to 28, assuming that Baird's claims are prior art with respect to Fiddes' claims.
The Fiddes claims are directed to a recombinant DNA molecule encoding human basic FGF, an expression vector containing the DNA, transformed cells or organisms containing the expression vector capable of expressing human FGF or a method of producing human basic FGF from the cells or organisms. On the other hand, the Baird claims are directed to a recombinant DNA molecule encoding mammalian FGFs and bovine FGF, methods of producing these FGFs and transformed cells or organisms containing the expression vector capable of expressing these FGFs.
We agree with Fiddes that his claims which are essentially directed to a DNA molecule encoding human basic FGF would not have been anticipated or rendered obvious by Baird's claims which are directed to a DNA molecule encoding mammalian and bovine basic FGF. Since the present record shows that DNA molecules encoding various mammalian FGF's have not been sequenced, it is highly speculative whether they would have the same DNA sequence as the native gene encoding human FGF, especially where the native gene encoding bovine FGF is chemically different from the native gene encoding human FGF. See Amgen at 18 USPQ2d 1023 where the Court found that the human EPO gene could not have been identified and isolated with a reasonable likelihood of success where neither its DNA nucleotide sequence nor its exact degree of homology with the monkey EPO gene was known at the time. Dr. Moore testified that the synthetic gene for bovine FGF described in the Baird application is not the same as the native gene for human FGF since there are 106 non-matching nucleotides between the bovine gene and the native gene for human FGF. He also testified that there are 22 nucleotide differences between the native bovine gene and the native human gene. Furthermore, one of ordinary skill in the art having possession of Baird's synthetic bovine gene, which was postulated by Baird on the basis of the established relationship in the genetic code between an nucleic acid and the protein it encodes, would not have found it obvious to prepare the native gene encoding human FGF as claimed by Fiddes. Knowledge of the amino acid sequence of a protein coupled with the established relationship in the genetic code between a nucleic acid and the protein it encodes would not normally render obvious the native gene encoding that protein. In re Bell, supra. Cf. Ex parte Hudson, 18 USPQ2d 1322 (BPAI 1990).
In Hudson, the Board found that one of ordinary skill in the art having knowledge of the amino acid sequence of porcine relaxin would have been motivated to construct a synthetic gene for biosynthesis of the protein and would have had a reasonable expectation to achieve its biosynthesis under the facts therein, i.e., the amino acid sequences of A and B chains of porcine relaxin were known, the prior art of record showed that there was a preprorelaxin in rats, which contained in addition to the A and B chains, a C chain which connected them (these three chains together constituting prorelaxin), a signal peptide (the four chains together constituting preprorelaxin) and a suggestion in the prior art that porcine relaxin would have a similar precursor. Since the record before us does not show that the prior art (assuming that Baird's claims were prior art) was aware of any homology between the native gene encoding bovine FGF and the native gene encoding human FGF, it would have been highly speculative for one of ordinary skill in the art to have a reasonable expectation of success in obtaining the native gene encoding human FGF by using the native gene encoding bovine FGF.
*7 For the foregoing reasons, we hold that an interference-in-fact does not exist.
Since an interference-in-fact does not exist, we need not consider the issue of whether Fiddes derived the invention of the count from Baird. This issue is a question of priority which we do not reach and is accordingly dismissed as moot.
It is adjudged that, on the present record, John C. Fiddes and Judith A. Abraham, the junior party, are entitled to a patent containing claims 43 to 46, 50 to 54 and 57 to 69, and Baird et al., the senior party, are not entitled to a patent containing claims 11 to 21, 23, 24, 26 and 27 and are entitled to a patent containing claims 25 and 28.
John T. Goolkasian
Mary F. Downey
Administrative Patent Judges
FN1. As of October 4, 1993, the Commissioner authorized each Examiner-in-Chief to use the title, "Administrative Patent Judge."
FN2. During oral argument, counsel for the parties argued the applicability of the three decisions to this case.
FN3. Baird's remaining claims 25 and 28 are directed to bovine FGF.
FN4. Possession of the naturally occurring gene is normally necessary in order to probe for the corresponding human gene. While it may be obvious to try and to probe a human genomic DNA library with a mammalian gene to isolate its corresponding human gene, it is a matter of speculation whether the human gene would be isolated with a reasonable expectation of success. Amgen, supra at 18 USPQ2d 1023.
FN5. Considering the degeneracy of the genetic code and the fact that bovine pituitary FGF contains 146 amino acids, the Baird claims embrace substantially more DNA sequences than the Bell claims. In Bell, the human insulin-like growth factor, which has fewer DNA sequences than bovine pituitary FGF, could be encoded by more than 1036 different nucleotide sequences.
FN6. During ex parte prosecution, the primary examiner held that these claims were unpatentable under 35 U.S.C. 112, first paragraph. Nonetheless, he properly designated them as corresponding to count 1, because they define the same patentable invention as the count. The "same patentable invention" requirement of 37 CFR 1.637(c)(3)(ii) concerns only the relationship between the count and the claims sought to be additionally designated; it does not concern general patentability. Maier v. Hanawa, 26 USPQ2d 1606 (Comm'r.1992).