ARCHER, Circuit Judge.

Hormone Research Foundation and Hoffmann-LaRoche, Inc. (collectively HRF) appeal from the summary judgment of the United States District Court for the Northern District of California, Hormone Research Found. v. Genentech, Inc., 708 F. Supp. 1096, 8 USPQ2d 1377 (N.D.Cal. 1988), holding (1) that Genentech, Inc., Genentech Development Corporation, and Genentech Clinical Partners, Ltd. (collectively Genentech) did not infringe claims 3, 11, 12, 17, 20, and 25 of U.S. Patent No. 3,853,833 ('833 patent), and (2) that claims 1, 3, 11, 12, 17, 18, 20, and 25 of that patent, which constitute all the claims in the suit, are invalid under 35 U.S.C. § 112, first paragraph, for lack of enablement. Genentech cross-appeals from the denial of attorney fees and costs under 35 U.S.C. § 285. We affirm-in-part, vacate-in-part and remand.

Background

The facts underlying this appeal are set forth in the district court's opinion, id., and familiarity with that opinion is presumed. The facts may be summarized as follows.

In April, 1971, Dr. Chao Hao Li filed a patent application describing the production, via a process known as "solid phase peptide synthesis," of a synthetic, 188 amino acid sequence exhibiting growth promoting properties. According to the application, Dr. Li indicated that he had identified the structure of the natural human growth hormone (HGH). However, in a continuation-in-part application that eventually matured into the '833 patent, Dr. Li conceded error in that regard and proposed a different version of the sequence of HGH, which is shown in Figure 2 of the '833 patent appended hereto. It is a 190 amino acid sequence.

It was later discovered and is admitted by HRF that the structure in Figure 2 of the '833 patent actually differs from natural HGH in two respects:

(1) HGH has one additional amino acid in the chain. (Specifically, glutamine is included at a position after the 68th amino acid.); and

(2) HGH has slightly different amino acids in positions 73, 106, and 108. (Specifically, there is glutamic acid instead of glutamine at position 73, aspartic acid instead of asparagine at 106, and asparagine instead of aspartic acid at 108.)

Hormone Research Foundation is the current owner of the '833 patent and Hoffman-LaRoche has an exclusive license thereunder. Genentech produces, by a recombinant DNA method, a human growth hormone product it calls Protropin. The structure of Protropin differs from the structure identified in Figure 2 of the '833 patent in that Protropin contains 192 amino acids instead of 190. Specifically, Protropin has an additional methionine at the amino end of the sequence and, like natural HGH, it has an additional glutamine after position 68 and slightly different proteins in the positions corresponding to position 73 (glutamic acid rather than glutamine), 106 (aspartic acid instead of asparagine) and 108 (asparagine instead of aspartic acid).

Genentech has also produced Protropin II, which apparently has not yet been approved for marketing by the Federal Drug Administration. The parties do not dispute that Protropin II has a structure identical to that of natural HGH.

Believing that Genentech's activities relating to Protropin and Protropin II are in violation of its rights under the '833 patent, HRF filed suit against Genentech seeking both injunctive and monetary relief. The eight claims of the '833 patent asserted by HRF are:

1. A method of producing synthetic human pituitary growth hormone which comprises:

a. forming an unbridged polypeptide chain of amino acid residues in the sequence of natural human pituitary growth hormone;

b. generating sulfhydryl groups on the cysteine residues in said polypeptide chain;

c. oxidizing said sulfhydryl groups under conditions effective to form disulfide bridges between said cysteine residues, thereby forming two intramolecular rings in the polypeptide chain.

3. A method of producing a substance having growth-promoting activity which comprises:

a. forming an unbridged polypeptide chain of amino acid residues in a sequence corresponding to FIG. 2 or 3 of the accompanying drawing;

b. generating sulfhydryl groups on the cysteine residues in said polypeptide chain;

c. oxidizing said sulfhydryl groups under conditions effective to form disulfide bridges between said cysteine residues, thereby forming two intramolecular rings in the polypeptide chain.

11. A method of producing a substance having growth-promoting activity which comprises:

a. forming an unbridged polypeptide chain of amino acid residues in a sequence corresponding to (i) the portion of the sequence of FIG. 2 of the drawing from positions 86 to 190 or (ii) the said portion (i) combined with any fraction of the remaining portion from position 85 to position 1;

b. generating sulfhydryl groups on the cysteine residues in said polypeptide chain;

c. oxidizing said sulfhydryl groups under conditions effective to form a disulfide bridge between said cysteine residues across positions 181 and 188; thereby forming an intra-molecular ring in the polypeptide chain.

12. A composition of matter consisting essentially of a synthetic, biologically active substance which has a structure corresponding to FIG. 2 of the accompanying drawing.

17. A composition of matter consisting essentially of a synthetic, biologically active substance which has a structure corresponding to (a) the portion of the structure of FIG. 2 of the drawings from positions 86 to 190 or (b) the said portion (a) combined with any fraction of the remaining portion from position 85 to position 1.

18. A composition of matter produced in accordance with the method of claim 1.

20. A composition of matter produced in accordance with the method of claim 3.

25. A composition of matter produced in accordance with the method of claim 11.

Before the district court, Genentech filed three motions for summary judgment. In its first motion, Genentech alleged that it had not infringed claims 3, 11, 12, 17, 20, and 25 of the '833 patent (the Figure 2 claims) because its Protropin products do not contain a compound having the structure of Figure 2. In its second motion, Genentech alleged that its products do not infringe claims 1, 3, 11, 18, 20, and 25 because they are produced by a recombinant DNA technique which differs from the solid-phase peptide process covered by these claims. In its final motion, Genentech asserted that the '833 patent claims at issue were invalid because the specification failed to satisfy the enablement requirement of 35 U.S.C. § 112, first paragraph. The district court granted Genentech's first and third motions, but denied the second.

With regard to Genentech's first motion, the district court found that the Figure 2 claims were not literally infringed because Genentech's products do not "correspond" to the structure of Figure 2. The district court also concluded that prosecution history estoppel precludes HRF from recovering for infringement under the doctrine of equivalents. On the enablement motion, the district court held that the specification would not have enabled a person skilled in the art to make and use either HGH (claims 1 and 18) or the structure of Figure 2 (claims 3, 11, 12, 17, 20, and 25) because, in its view, the synthesis process disclosed in the specification would not produce the identical polypeptide sequences called for in these claims.

Genentech also filed a motion for attorney fees and costs under 35 U.S.C. § 285, which the district court denied on the basis that Genentech had not shown that such an award was warranted.

In this appeal, HRF contends that:

(1) The district court erred in concluding on summary judgment that there was no literal infringement of claims 3, 11, 12, 17, 20, and 25 of the '833 patent.

(2) The district court erred in concluding on summary judgment that prosecution history estoppel precludes a finding of infringement of the Figure 2 claims under the doctrine of equivalents.

(3) The district court erred in concluding on summary judgment that claims 1, 3, 11, 12, 17, 18, 20, and 25 of the '833 patent were invalid under 35 U.S.C. § 112, first paragraph, for lack of enablement.

In its cross appeal, Genentech asserts that:

(4) The district court erred in denying it an award of attorney fees and costs under 35 U.S.C. § 285.

OPINION

I. Summary Judgment

Under Rule 56(c) of the Federal Rules of Civil Procedure, summary judgment "shall be rendered if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show there is no genuine issue of material fact and the moving party is entitled to a judgment as a matter of law." See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247, 106 S.Ct. 2505, 2509, 91 L.Ed.2d 202 (1986). "In reviewing a district court's grant of summary judgment in a patent case, this court must . . . determine for itself whether the evidence is genuinely conflicting on the material issues of fact and, if not, whether the movant is entitled to judgment on those facts." Avia Group Int'l, Inc. v. L.A. Gear Cal., 853 F.2d 1557, 1561, 7 USPQ2d 1548, 1551 (Fed. Cir. 1988).

We conclude that the district court properly granted summary judgment to resolve literal infringement where the controlling issue was solely one of law. On the other appealed issues, however, summary judgment was inappropriate.

II. Literal Infringement

The determination of whether a patent claim has been literally infringed involves two inquiries: whether the claims have been properly interpreted to determine their scope, and whether each limitation of the properly construed claims is found in the accused product or process. See ZMI v. Cardiac Resuscitator Corp., 844 F.2d 1576, 1578, 6 USPQ2d 1557, 1559 (Fed. Cir. 1988). The first of these is a legal question and the second is factual. Id.; Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861, 866, 228 USPQ 90, 93 (Fed. Cir. 1988).

In the present case there is no dispute regarding either the chemical structure or the conformation of the accused products; neither have the identical structure and conformation of Figure 2. Since the district court interpreted the Figure 2 claims as limited to a material having a structure and conformation identical to that depicted in Figure 2 of the patent, it held that Genentech's products do not literally infringe the claims. Consequently, the issue of whether Genentech's products literally infringe the '833 claims at issue is dependent solely on whether the district court's claim interpretation was legally correct. See Loctite, 781 F.2d at 865-66, 228 USPQ at 92. Our review of this legal question is de novo.

Claim interpretation involves a review of the specification, the prosecution history, the claims (including unasserted as well as asserted claims), and, if necessary, other extrinsic evidence, such as expert testimony. See Senmed, Inc. v. Richard-Allan Medical Indus., 888 F.2d 815, 818-19, 12 USPQ2d 1508, 1511-12 (Fed. Cir. 1989); see also Fromson v. Advance Offset Plate, Inc., 720 F.2d 1565, 1569-70, 219 USPQ 1137, 1140-41 (Fed. Cir. 1983). Should there exist a genuine dispute concerning any underlying factual matters, summary judgment on the infringement issue would be inappropriate. Fed.R.Civ.Proc. 56(c); see generally Uniroyal, Inc. v. Rudkin-Wiley Corp., 837 F.2d 1044, 1054, 5 USPQ2d 1434, 1441 (Fed. Cir. 1988).

In the present case, the district court interpreted the term "corresponding," as used in the Figure 2 claims to refer to the polypeptide sequence depicted in Figure 2 of the '833 patent and determined that the claims should be limited to the identical amino acid sequence and conformation there depicted. The court based that determination on its view of the most applicable dictionary definition of "corresponding," without discussing or acknowledging how that term was used in the specification and prosecution history.

HRF asserts that the district court's interpretation of the claims was in error not only because the specification and prosecution history were seemingly ignored but also because the court focused on the wrong dictionary definition of the disputed term. HRF contends that the claims should be interpreted broadly to encompass materials, such as Genentech's Protropin products, which have a structure and conformation "similar" to that shown in Figure 2 of the '833 patent.

Although the district court's methodology in interpreting the claims may have been incorrect, we are not persuaded that the district court reached an erroneous claim construction. It is a well-established axiom in patent law that a patentee is free to be his or her own lexicographer, see Fromson, 720 F.2d at 1569, 219 USPQ at 1140, and thus may use terms in a manner contrary to or inconsistent with one or more of their ordinary meanings. For this reason, an analysis of the specification and prosecution history is important to proper claim construction. Here, however, the court's determination that the claim phrases "a sequence corresponding to Figure 2" and "a structure corresponding to Fig. 2" required identity in all respects to that drawing is fully consistent with both the specification and the prosecution history.

In the specification, the patentee stated:

The molecular structure illustrated in FIG. 1 was thought heretofore in the art to correspond to that of natural HGH, but as indicated earlier hereinabove, applicant has discovered that this is not correct. The correct structure of HGH has been determined by applicant and is illustrated in FIG. 2. It consists of a polypeptide chain of 190 amino acid residues, as compared to 188 amino acid residues in the polypeptide chain in FIG. 1. . . . .

[W]hile there is general similarity between the overall conformations and structures illustrated in FIGS. 1 and 2, there are differences in the number and sequence of amino acid residues and in the size of the layer of the two intramolecular rings present in both structures.

'833 patent specification, col. 4, 1. 55-col. 5, 1. 12 (emphasis added). This strongly suggests that the terms "correspond" and "similar" were not intended by the patentee to have the same meaning and that the former reflected true identity.

More convincing, however, are the statements made during the prosecution of the application for the '833 patent. There patentee described claim 12, which uses the phrase "a structure corresponding to Fig. 2," as "directed solely to the compound of Fig. 2," "specific to the structure of Fig. 2," "limited to the structure shown in the drawing" and "specific to the chemical formula of Fig. 2." Each of these descriptions pointedly evinces the patentee's use of the term "corresponding" in the narrow sense of requiring identity rather than "similarity."

In sum, we conclude that the district court correctly held, albeit on the basis of an incomplete analysis, that the term "corresponding" as used in the disputed phrases of the '833 patent claims limits all of the Figure 2 claims to a material having the exact structure and conformation shown in Figure 2 of the '833 patent. Stratoflex, Inc. v. Aeroquip Corp., 713 F.2d 1530, 1540, 218 USPQ 871, 880 (Fed. Cir. 1983) ("We sit to review judgments, not opinions."). HRF's arguments to the contrary are little more than a post-hoc attempt to redefine its claims during litigation. This it cannot do. See Senmed, 888 F.2d at 819 n. 8, 12 USPQ2d at 1512 n. 8. Accordingly, that part of the district court's judgment that Genentech's products do not literally infringe claims 3, 11, 12, 17, 20, and 25 of the '833 patent is affirmed.

III. Infringement under the Doctrine of Equivalents

Infringement under the doctrine of equivalents is an equitable doctrine intended, "in situations where there is no literal infringement but liability is nevertheless appropriate[,] to prevent what is in essence a pirating of the patentee's invention." Loctite, 781 F.2d at 870, 228 USPQ at 96. Under the doctrine of equivalents, the accused products and processes may infringe the '833 patent claims if they perform substantially the same function in substantially the same way to give substantially the same result. Id. The doctrine of equivalents, however, is not a tool for redrafting the claims of a patent. Senmed, 888 F.2d at 818, 12 USPQ2d at 1511.

Prosecution history estoppel is a judicially accepted limitation to the doctrine of equivalents. Under that limitation, a patentee cannot "recapture through equivalence certain coverage given up [by argument or amendment] during prosecution." Loctite, 781 F.2d at 870, 228 USPQ at 96. That is not to say, however, that, whenever a limiting amendment or argument is made during prosecution, the patentee loses all coverage between what the claims literally cover and what they would have covered prior to the amendment or argument. Such an approach was rejected in Hughes Aircraft Co. v. United States, 717 F.2d 1351, 1362-63, 219 USPQ 473, 481 (Fed. Cir. 1983). Instead, "[d]epending on the nature and purpose of an amendment, it may have a limiting effect within a spectrum ranging from great to small to zero." Id.

"[W]henever the doctrine is evoked, `a close examination must be made as to, not only what was surrendered, but also the reason for such a surrender.'" Loctite, 781 F.2d at 871, 228 USPQ at 96 (quoting Bayer Aktiengesellschaft v. Duphar Int'l Research, 738 F.2d 1237, 1243, 222 USPQ 649, 653 (Fed. Cir. 1984)). Thus, the scope of estoppel can depend on factual questions regarding the prosecution history, which may be disputed and preclude a disposition of the issue on summary judgment. That is the situation here. Cf. Howes v. Medical Components, Inc., 814 F.2d 638, 646, 2 USPQ2d 1271, 1275-76 (Fed. Cir. 1987) ("Summary judgment should not have been granted" where there are "too many unresolved fact issues to properly construe the scope of [the] claims."); Palumbo v. Don-Joy Co., 762 F.2d 969, 976, 226 USPQ 5, 10 (Fed. Cir. 1985) ("[I]f ambiguity is thought to surround the prosecution history in this case, that could give rise to a question of fact underlying the legal question of claim construction.").

In the present case, the district court did not decide the issue of whether the accused products or processes were equivalent to the claimed invention. Instead, it held that recovery under the doctrine of equivalents was precluded by prosecution history estoppel. The relevant prosecution history pertains to application claim 18 (A-claim 18), which ultimately issued as claim 15 of the '833 patent. Even though A-claim 18 is not at issue in the case, its prosecution history formed the basis of the estoppel found by the district court. A-claim 18 reads:

18. A composition according to claim 12 without the disulfide bridges shown in the drawing and substituted instead with tetra-S-carbamidomethyl groups on the cysteine residues thereof.

In his first action on the application, the patent examiner rejected A-claim 18 (which at that time covered Figures 1, 2 and 3) under 35 U.S.C. § 102(b) as anticipated by the Bewley, et al. article (Bewley or reference U). Bewley discloses natural HGH substituted in the same manner as the composition described in A-claim 18, i.e., where the disulfide bridges depicted in the Figure 2 structure of the '833 patent are substituted with tetra-S-carbamidomethyl groups. The applicant (now patentee) responded by stressing that Bewley did not disclose how to make synthetic HGH, the starting material needed to make the claimed invention:

The reference U discloses a reduced carbamidomethylated HGH. However, claim 18 is dependent from claim 12 which calls for "A composition of matter consisting essentially of a synthetic, biologically active substance. . . ." The reference U shows no way of producing the basic compound required as shown in Figs. 1, 2, or 3 which is necessary starting material for the carbamidomethylated derivative.

Still not convinced that A-claim 18 was patentable, the examiner again rejected it as anticipated by Bewley. He stated that "[t]he reference uses native hormone as starting material and in rejecting the instant claim the preamble `synthetic . . . substance . . .' is given no patentable weight."

In response to this continued rejection, the applicant stated:

The Examiner has rejected claim 18 under 35 U.S.C. § 102(b) as anticipated by Bewley et al (U). Claim 18 is dependent from allowed claim 12, which until this amendment was directed to the compounds of Fig. 1, 2 or 3. Presently, claim 12 is directed solely to the compound of Fig. 2. Newly presented claim 29 is directed solely to the compound of Fig. 3. There is no claim directed solely to the compound of Fig. 1. New claim 32 is dependent from claim 29 but is otherwise like claim 18.

The Bewley et al. (U) reference discloses the carbamidomethylated reduced HGH — or so the reference asserts — the reference does not, however, disclose the chemical structure of the asserted HGH, which is the starting compound. Claim 18 . . . depend[s] . . . from claim[s] 12 . . ., which [is] specific to the structure of Fig. 2. . . . The claims are therefore limited to the structures shown in the drawings and are not directed broadly to HGH or its derivates. No such structure is disclosed in the reference U. As such, the reference does not show the basic compound which must be reduced to achieve the derivative claimed in these claims. Indeed, the authors of reference U, while they refer to HGH and were in fact working with natural pituitaries, clearly did not know the structure. One of the authors of the article U was Dr. Li, the inventor in the present case. He, and one of the other authors, Jonathan S. Dixon, were also co-authors of the later publication T which asserts that the structure of Fig. 1 is in fact HGH. As stated in the specification herein at page 2, lines 9-11, the published formula of Fig. 1 does not in fact correspond to natural human pituitary growth hormone. This was learned only subsequent to the date of publication T. Accordingly, the publication U, earlier in time, cannot anticipate the formulas of the present figures since they were not then known. Inasmuch as these product claims are specific to the chemical formula of Fig. 2 or Fig. 3, and the reference does not show the same, and, further, that the record demonstrates the formula was not in fact known, these claims are clearly allowable. (Emphasis added).

The examiner then allowed A-claim 18.

In the district court's view, the underlined portion of the above excerpt was key to its estoppel holding. The opinion states:

The applicant has thus secured the patentability of A-claim 18 by arguing that the structural limitation in A-claim 12 was directed narrowly to Fig. 2 rather than broadly to HGH or its derivatives. The argument was successfully made to disclaim coverage of a substance, disclosed in the Bewley reference, where starting material had a structure identical to the structure of natural HGH. Plaintiffs now argue, however, that the same structural limitation in A-claim 12, which issued without further amendment as patent claim 12, should be expanded by the doctrine of equivalents to include a substance having a structure identical to that of natural HGH (or, in the case of Protropin, natural HGH with an additional terminal methionine) but not identical to Fig. 2. The doctrine of prosecution history estoppel prevents plaintiffs from recovering through the doctrine of equivalents subject matter surrendered during prosecution of the patent.

708 F. Supp. at 1105-06. To the district court, therefore, what transpired during prosecution of the '833 patent was believed to be clear. The district court interpreted Dr. Li's argument that the claims are "limited to the structures shown in the drawings and are not directed broadly to HGH or its derivatives" to mean that the pending claims were confined to the specific structure of Figure 2 and that the Bewley reference was not anticipatory because it depicted the different, although very close, structure of natural HGH. If it were clear that this was the intended meaning of Dr. Li's arguments, the district court's determination that the prosecution history precludes HRF from recovering under the doctrine of equivalents would be correct.

There are, however, other possible interpretations of Dr. Li's prosecution arguments. HRF suggests that because the starting material for producing the carbamidomethylated derivative of A-claim 18 was the synthetic hormone of Figure 2 (believed at that time by Dr. Li to be the same structure as natural HGH), whereas the starting material disclosed by Bewley was natural HGH, the patentee surrendered only carbamidomethylated substitutes derived from natural HGH. This interpretation, unlike the one accepted by the district court, is consistent with the view expressed throughout the '833 specification that the structure of Figure 2 and natural HGH were one and the same.

Another plausible interpretation of Dr. Li's argument is based upon his statements to the examiner that "[n]o such structure is disclosed in the reference U" and "[a]s such, the reference does not show the basic compound which must be released to achieve the derivative claimed in these claims." He further argued that "publication U . . . cannot anticipate the formulas of the present figures since they were not then known." Drawing all inferences in favor of HRF, which summary judgment requires, see, e.g., Avia Group Int'l, Inc. v. L.A. Gear Cal., 853 F.2d 1557, 1560, 7 USPQ2d 1548, 1550 (Fed. Cir. 1988), Dr. Li may have been suggesting that Bewley was not anticipatory of the rejected claims because it was not enabling under 35 U.S.C. § 112, since it did not disclose a structure of, or how to make, HGH. See Akzo N.V. v. United States Int'l Trade Comm'n, 808 F.2d 1471, 1479, 1 USPQ2d 1241, 1245 (Fed. Cir. 1986).

Accordingly, we cannot say that the meaning of Dr. Li's prosecution arguments is without ambiguity and that there are not unresolved factual questions regarding the intent of those arguments. The other possible interpretations of Dr. Li's prosecution arguments are very relevant to the proper application of prosecution history estoppel as a limitation to infringement under the doctrine of equivalents. The existence of disputed factual questions regarding the intent and meaning of the prosecution arguments precludes summary judgment.

Accordingly, we vacate that part of the judgment which held that the claims are not infringed under the doctrine of equivalents and remand the case to the district court to determine the intent and effect of the arguments made during the prosecution of A-claim 18.

IV. Enablement

The district court's summary judgment that claims 1, 3, 11, 12, 17, 18, 20, and 25 are invalid for lack of an enabling disclosure, see 35 U.S.C. § 112, 1st para., was also improvidently granted. The court granted Genentech's motion because, in its view, the solid phase peptide synthesis process disclosed in the specification would not have been sufficient to produce materials as lengthy as the claimed polypeptide sequence or in a pure form and having the potency of natural HGH.

"To be enabling under § 112, a patent specification must disclose sufficient information to enable those skilled in the art to make and use the claimed invention." Spectra-Physics v. Coherent, 827 F.2d 1524, 1533, 3 USPQ2d 1737, 1743 (Fed. Cir. 1987). "Although enablement is ultimately a question of law, this court has recognized that there may be underlying factual issues involved." Id. ( citing Moleculon Research Corp. v. CBS, Inc., 793 F.2d 1261, 1268, 229 USPQ 805, 810 (Fed. Cir. 1986), and Quaker City Gear Works, Inc. v. Skil Corp., 747 F.2d 1446, 1453-54, 223 USPQ 1161, 1166 (Fed. Cir. 1984)).

After reviewing the record and the district court's lengthy analysis of the evidence submitted by Genentech, we are convinced that the enablement question in this case should not have been resolved summarily. Although the district court indicated that Genentech produced considerable evidence tending to show that the disclosed sequencing method could not have yielded polypeptide sequences as lengthy as that of Figure 2 (190 amino acids), other evidence in the record raises a genuine issue about this material fact. The '833 specification itself discloses that Dr. Li's claimed method had produced the material depicted in Figure 1 of the '833 patent and that such a material exhibited lactogenic activity. Evidence tending to support this assertion can be found in several of the journal articles in the record. For example, see C.H. Li and D. Yamashiro, "The Synthesis of a Protein Possessing Growth-Promoting and Lactogenic Activities," J. Amer. Chem. Soc., 92(26), 7608-9 (1970); H.D. Niall, "The Chemistry of the Human Lactogenic Hormones," p. 14; Prolactin and Carcinogenesis, Proceedings of the Fourth Tenovus Workshop, Cardiff March 1972, Alpha Omega Alpha Publishing, Cardiff, Wales, U.K. (August 1972); C.H. Li, "Human Pituitary Growth Hormone," Clinical Orthopedics and Related Research, 89, 123 (Nov-Dec 1972). While this evidence may not be sufficient to sustain an ultimate holding of enablement, it does, giving all inferences to HRF, see Fed.R.Civ.Proc. 56, preclude a summary judgment that the '833 claims are invalid on the basis that sequences of the required length could not be formed by the disclosed method.

In addition, we are unconvinced that Genentech "is entitled to a judgment as a matter of law." Fed.R.Civ.Proc. 56(c). Whether the specification of the '833 patent is enabling with respect to the purity and/or potency of the claimed materials is dependent on the resolution of legal and factual issues not addressed by the district court. Although the district court applied the principles set forth in In re Fisher, 427 F.2d 833, 166 USPQ 18 (CCPA 1970), it did not in its analysis consider the effect of In re Hogan, 559 F.2d 595, 194 USPQ 527 (CCPA 1977), see also United States Steel Corp. v. Phillips Petroleum Co., 865 F.2d 1247, 1251, 9 USPQ2d 1461, 1464 (Fed. Cir. 1989), on the resolution of the enablement issues. In In re Hogan, the court stated:

Appellants disclosed, as the only then existing way to make such a polymer, a method of making the crystalline form. To now say that appellants should have disclosed in 1953 [original filing date] the amorphous form which on this record did not exist until 1962, would be to impose an impossible burden on inventors and thus on the patent system. There cannot, in an effective patent system, be such a burden placed on the right to broad claims. To restrict appellants to the crystalline from disclosed, under such circumstances, would be a poor way to stimulate invention, and particularly to discourage its early disclosure.

559 F.2d at 601-06, 194 USPQ at 533-37. Accord United States Steel, 865 F.2d at 1249-52, 9 USPQ2d at 1463-66.

The opinion in In re Hogan specifically commented on In re Fisher:

[In Fisher] this court set forth the basic considerations respecting enablement and the potential for domination of future developments, describing the effect of predictability factors upon those considerations. We adhere to what was there said concerning the high level of predictability in mechanical or electrical environments and the lower level of predictability expected in chemical reactions and physiological activity. With respect to the erroneous use of a later state of the art in determining enablement, however, we make no distinction between fields of invention.

559 F.2d at 606, 194 USPQ at 537-38. Merely because purer and more potent forms of the Figure 2 compound might be produced using later-discovered technology does not necessarily mean that the '833 patent specification did not provide sufficient enabling disclosures as of the filing date of the application.

It is unclear whether the high degree of potency and purity contemplated by the district court's analysis of enablement was influenced by the potency and purity obtainable through recombinant DNA methodology. Moreover, it is unclear from the record before us whether that technology existed at the time the application was filed. Further factual development as to the state of the art at the date of the application, and consideration of In re Hogan by the district court, is required for this court to review these enablement issues.

That part of the summary judgment holding the asserted claims of the '833 patent to be invalid is, therefore, vacated.

V. Attorney Fees

The district court denied Genentech's motion for attorney fees and costs. As the then prevailing party, Genentech was entitled to make that motion, but in view of our decision it does not now have that status. Accordingly, Genentech's cross-appeal for attorney fees under 35 U.S.C. § 285 is dismissed, without prejudice. If Genentech prevails on remand, it may renew its motion at that time.

VI. Conclusion

We affirm the district court's judgment insofar as it determined that Genentech did not literally infringe the asserted claims. However, we conclude that there are genuine issues of material fact underlying the issues of prosecution history estoppel and enablement. We therefore vacate the district court's summary judgment of non-infringement of claims 3, 11, 12, 17, 20, and 25 and of invalidity of claims 1, 3, 11, 12, 17, 18, 20, and 25 and remand the case for further proceedings consistent with this opinion.

COSTS

Each party shall bear its own costs.

AFFIRMED-IN-PART, VACATED-IN-PART, AND REMANDED.

APPENDIX