Amgen recently filed its Reply brief to the Supreme Court in Amgen v. Sanofi. While a conventional proportion of Amgen's Reply is directed to arguments Respondent Sanofi made in its brief, at trial, and before the Federal Circuit, significant portions of the Reply brief are focused on the Federal Circuit's grounds for its decision, which may be wise for at least two reasons. The first, of course is the seeming anathema the Supreme Court has had for Federal Circuit jurisprudence for over a decade, reversing almost all that circuit court's decisions in whole or part. (It will be understood that the likelihood of reversal of Federal Circuit decisions may be higher than other circuit courts because the court's exclusive jurisdiction over patent matters means the only situations analogous to "circuit splits" arise from disagreements between the Federal Circuit judges, tempered by whatever deference lingers for that court's "special expertise" noted in Warner-Jenkinson Co. v. Hilton Davis Chemical Co.) The other reason is that Amgen's argument in its principal brief is that the Federal Circuit has crafted a new test for enablement, which it termed the "full scope" test, that is contrary to the plain meaning of the statute (as well as Supreme Court precedent) because the new test mandates that each and every species in a claimed genus be set forth rather than that the disclosure merely teach the skilled worker how to make and use the claimed invention.
In its argument, Amgen asserts that the defects of the Federal Circuit's "full scope" enablement test include that it depends on "the cumulative time and effort" needed to determine and disclose the full scope of what is claimed (which for a genus claim would encompass all or substantially all of the species falling within the scope of the genus) (which Amgen asserts that the Solicitor General contends is "a test [that] has no foundation in the text [of § 112] or [the] Court's precedent"). The proper enablement test, Amgen asserts, is the statutory "make and use" test, which "requires a practical inquiry into what skilled artisans can do using the patent's disclosures." And if the disclosure is "sufficiently definite to guide those skilled in the art to" the "successful application" of "the invention," Amgen contends, the patent is enabled, citing Minerals Separation, Ltd. v. Hyde, 242 U.S. 261, 271 (1916). Amgen further argues that the number of embodiments (or species in a genus) is irrelevant to this standard (although as a practical matter it seems this argument cannot be an absolute one). "The Federal Circuit's opinion [that Amgen's claims are not enabled] plainly turns on the effort required to identify and make all or nearly all the antibodies within Amgen's patent claims," Amgen asserts, which is a "heightened" enablement standard beyond the statutory requirements. The Federal Circuit's purported new test "consider[s] the quantity of experimentation that would be required to make and use, not only the limited number of embodiments that the patent discloses, but also the full scope of the claim" Amgen contends (emphasis in brief). The amount of experimentation (how many potential species fall within the scope of a genus) needed can by itself amount to undue experimentation under the Federal Circuit's test, Amgen contends. The basis for the Federal Circuit's decision below was the "substantial time and effort" necessary for this identification of all these antibodies (undue experimentation would be required to practice the full scope of these claims) and the consequence of this new interpretation of enablement law, in Amgen's view, is that the Federal Circuit reached its non-enablement conclusion based on the "millions of candidate antibodies" the claims encompass. As a result "'it would be necessary to first generate and then screen each candidate antibody' to determine whether those antibodies fell within the claims" to satisfy the Federal Circuit's test. Illustrating both this meaning of the Federal Circuit's opinion and the potential danger it raises, Amgen cites the decision in Human Power of N Co. v. Heartbeet Ltd., No. PGR2021-00110, 2022 WL 683124, at *9 (P.T.A.B. Feb. 25, 2022), and the district court's decision in Baxalta Inc. v. Genentech, Inc., 579 F. Supp. 3d 595, 616 (D. Del. 2022), and Astra-Zeneca AB v. Mylan Pharms. Inc., No. 1:18CV193, 2022 WL 16857400, at *8 (N.D. W. Va. Nov. 9, 2022), which have applied the Federal Circuit's decision below to impose this increased disclosure requirement in finding claims non-enabled.
The Federal Circuit erred, according to Amgen, because that court "varied the Wands factors by asking whether there would be 'undue experimentation in identifying,' from 'millions of ' potential embodiments, those that 'satisfy' the genus's 'requirement[s],' i.e., the effort to "reach the full scope." Amgen asks the Supreme Court to "restore the statutory standard" of "make and use" which requirement is a clear, practical test, so that "the patentee [is required] to describe his invention so that others may construct and use it after the expiration of the patent," citing Schriber-Schroth Co. v. Cleveland Tr. Co., 305 U.S. 47, 57 (1938). The fact that a claim encompasses numerous embodiments that may be different from one another does not render the claim nonenabled according to Cont'l Paper Bag Co. v. E. Paper Bag Co., 210 U.S. 405, 418-419 (1908), Amgen argues. And rather than the Federal Circuit's new bright line rule what should be used is a flexible inquiry "that takes into account the nature of the claimed invention and the field in which it arises," citing the Solicitor General's amicus brief in this regard.
One portion of Amgen's brief illustrates the thorny and contrary considerations the Supreme Court must grapple with:
Amgen agrees patents must reasonably enable the entire scope of the claim. Where an invention has many embodiments, the patent enables the invention's "full scope" if skilled artisans can "reasonabl[y]" make and use "variation[s]," without gaps in subject matter skilled artisans would consider significant with "regard to [the patent's] subject-matter." Minerals Separation, 242 U.S. at 270. If the patent's instructions are "sufficiently definite to guide" skilled artisans to "successful application" across the scope of the "invention," the claims are enabled. Id. at 271. In such cases, the "enabling disclosure" is 'commensurate with the scope of the claims.'" But cataloging every embodiment is a task for "the mechanic, not the inventor." 2 W. Robinson, The Law of Patents for Useful Inventions §485 (1890) (other citations omitted).
Amgen supports this calculus with Supreme Court and other precedent, including Mowry v. Whitney, 81 U.S. (14 Wall.) 620 (1872) (methods for manufacturing railway wheels not required to recite "the specific timing and temperature parameters for every type of wheel skilled artisans might make"); Wood v. Underhill, 46 U.S. (5 How.) 1 (1846) (while exceptions arose in formula for mixing clay and coal dust to make bricks, those exceptions did not amount to non-enablement where "artisans would be able to apply the formula to most types of clay."); and Carver v. Braintree Mfg. Co., 5 F. Cas. 235, 237 (C.C.D. Mass. 1843) (claims enabled where a "skillful mechanic" could use the specification to make a rib for many kinds of cotton).
Addressing the asserted necessity for the "full scope" test, that without it there would be no reasoned way to prevent overbroad genus claiming, Amgen argues that "[t]here are various ways defendants can disprove enablement" including where a skilled artisan cannot make the invention at all, Beidler v. United States, 253 U.S. 447, 453 (1920); where there is a "distinct category of embodiments within the claim's scope that cannot be made following the patent's disclosures, Auto. Techs. Int'l, Inc. v. BMW of N. Am., Inc., 501 F.3d 1274, 1285 (Fed. Cir. 2007); or in "needle-in-a-haystack" situations involving "labor-intensive hunt for working embodiments that exceeds what skilled artisans typically do," Consol. Elec. Light Co. v. McKeesport Light Co., 159 U.S. 465, 472-475 (1895); Béné v. Jeantet, 129 U.S. 683, 684-686 (1889); and Holland Furniture Co. v. Perkins Glue Co., 277 U.S. 245, 250-251, 257 (1928).
The brief then discusses the impact of inoperative embodiments in a genus claim on enablement, asserting that such emodiments within a genus do not alone invalidate a claim, citing Atlas Powder Co. v. E.I. du Pont De Nemours & Co., 750 F.2d 1569, 1576 (Fed. Cir. 1984). Amgen illustrates the impact (or lack of it) of inoperative embodiments in Consolidated Electric, wherein the defendant prevailed by presenting clear-and-convincing evidence that months of experimentation, with thousands of materials, was necessary to find one fiber (bamboo) that would function as claimed. 159 U.S. at 472-474. Amgen further states in this regard that the Federal Circuit's decision was based on a factual error: rather than being limited to 26 expressly disclosed antibodies its specification disclosed 384 antibodies (reminiscent of the situation in Wands where the Patent Office considered demonstration of 9 positive antibodies from amongst 143 produced to be nonenabled while the Federal Circuit considered these 9 antibodies in the context of 16 tested). The "millions" of potential candidates forming another (albeit related) basis for the Federal Circuit's decision could be produced by conservative substitution Amgen notes, but counters that the patent teaches these variants. What its specification doesn't do, Amgen concedes, is teach which ones would be functional but according to Amgen there is no "needle in the haystack" problem with their specification because there was no evidence that any conservative substitution produced an inoperative embodiment. Amgen contends that these considerations are only relevant under the full scope test rather than being a reasoned basis for it (and Amgen does not mention that this test bypasses the principle from Wands that even extensive experimentation does not negate enablement if it is routine, well understood in the art, and operative embodiments can be readily and reliably identified).
Another target for Amgen's arguments is the government: Amgen contends that while the Solicitor General's (SG) brief pays lip service to a "reasonableness" standard based on Minerals Separations it then deviates from that standard, "from the record and the rules of procedure and evidence." Amgen cites the government's reliance in its brief on the limited number (26) of expressly exemplified antibodies as a basis for the claims being overbroad, and that the SG asks the Court to infer that accordingly Amgen didn't make additional antibodies because it did not know how to do so (which of course Amgen counters with the 384 antibodies recited in the specification). And with regard to the procedural deficiencies Amgen asserts against the SG's brief, the Reply brief cites the presumption of validity of a granted patent that must be overcome by clear and convincing evidence to the contrary. This quote from Minerals Separations is usefully cited in supporting this argument: "A patent's teachings may "guide" skilled artisans to the "successful application" of a nearly "infinite[ ]" number of embodiments beyond those disclosed in the specification" as well as from Smith v. Snow, 294 U.S. 1, 11 (1935), that "it is "not necessary to . . . describe in the specification[ ] all possible forms in which the claimed principle may be reduced to practice."
One (perhaps novel) government argument identified by Amgen is the requirement that when a patent claims a genus, an enabling disclosure requires "some general quality" described in structural terms that distinguishes the claimed products from others outside the scope of the claimed genus, a rubric derived from Consol. Elec., 159 U.S. at 475 and Holland Furniture, 277 U.S. at 256). The error here, according to Amgen, is that "[w]hile describing a structure may be one way to enable a genus, the Court has never held it is the only way, regardless of the nature of the art," citing examples like Neilson v. Harford (1841) 151 Eng. Rep. 1266, 1273-1274 (Exch.), and more recently Diamond v. Chakrabarty, 447 U.S. 303, 305-306 (1980).
And with regard to Amgen's purported "roadmap" for how to produce functional antibodies having the required specificity, Amgen asserts that "[f]ar from requiring skilled artisans to 'retrace' Amgen's "research steps," the patents allow them to start where Amgen's research ended," which of course is the benefit disclosure in a patent confers on the public, Amgen enumerating eight specific benefits. Amgen refutes the assertion that the roadmap or its other teachings amount to "trial and error" methods, stating rather that these are screening methods understood by the skilled workers.
Most of Amgen's arguments specifically directed to Sanofi's Respondent's brief (besides the rhetorical) are directed to disputes about the record below, whether Sanofi's positions have changed between what was argued before the Federal Circuit and in its Supreme Court briefing and that the level of disclosure in Amgen's competitors' patents (including Sanofi's) are no more enabling under the full scope test than in Amgen's patent-in-suit.
The brief concludes with reference to proper patent policy, where Amgen argues that the Federal Circuit's full-scope test is not consistent with the patent bargain, citing Pfaff v. Wells Elecs., Inc., 525 U.S. 55, 63 (1998). The brief cites the analysis by the late Professor Karshtedt and colleagues, The Death of the Genus Claim, 35 Harv. J.L. & Tech. 1, 1 (2021), regarding the need for genus claiming in biotechnology and antibody inventions and the paucity of protection afforded to antibody claims limited to specific antibody sequences. Amgen further argues that using the roadmap provided by Amgen's patents permits competitions to make their own antibodies based on this disclosure. Amgen also reminds the Court that the doctrine of equivalents is insufficient to protect claims to specifically disclosed antibodies because "[c]ourts . . . have refused to apply 'the equivalence argument' because it is feared that this 'would read the amino acid sequence limitation out of' the claim and 'effectively expand' the claim's scope to encompass any 'antibody that has the [claimed] effect,'" citing Teva Pharms. Int'l GmbH v. Eli Lilly & Co., No. 18-cv-12029, 2022 WL 4824318, at *19 (D. Mass. Oct. 3, 2022). Under these circumstances, Amgen warns that "[n]o company would invest billions in researching and developing new antibodies in hopes that the doctrine of equivalents will prevent competitors from evading an amino-acid-sequence claim."
Addressing again the weakness (in Amgen's view) of arguments regarding the danger of overbroad genus claims Amgen cites Westinghouse v. Boyden Power Brake Co., 170 U.S. 537, 568 (1898), for the principle that a species (here, an antibody) that "'has so far changed the principle of the' patent claim can be held not to fall within its enabled scope 'even if it "literally" falls within the class [of antibodies])'", and to the argument that broad genus claims harm innovation Amgen argues that improvements (in the form of particular species) incentivizes innovation (i.e., that patents incentivize the search for a "better mousetrap" rather than preclude that search).
Finally, Amgen argues that this case is not the case presented in O'Reilly v. Morse that claimed "'the use of . . . electro-magnetism, however developed, for making or printing intelligible characters . . . at any distances,'" on the basis that there are alternatives (such as siRNA, or antibodies that bind outside the sweet spot and small molecule inhibitors) for the antibodies disclosed in its specification. As a consequence, Amgen argues:
That is the type of innovation the patent system should promote—encouraging companies to develop diverse therapies, not minor variations of antibodies performing the same function in the same way. Upholding genus claims for pioneering inventions accelerates the progress of science as inventors seek their own breakthroughs instead of developing "me too" products . . . ."
The Supreme Court is scheduled to hear oral argument from the parties and the Solicitor General on March 27th; in the meantime posts on the positions of various amici will appear here.
