The PTAB has continued the trend of pushing the -“antibody exception” to written description into an ever-smaller corner. Claims to methods of using antibodies that bind Siglec-15 to impair osteoclast differentiation and inhibit bone resorption were deprived of priority because the parent application failed to disclose the “antigenic regions useful for generating antibodies having the desired functional properties.” Consequently, the claims were anticipated by the cited reference under 35 U.S.C. §102(a), and Patentee’s other evidence of prior conception, diligence and reduction to practice was insufficient to antedate the reference. Daiichi Sankyo Co., Ltd. v. Alethia Biotherapeutics, Inc., IPR2015-00291 (Paper No. 75, June 14, 2016) (final written decision).
The parent application did not disclose any antibodies with the claimed functional properties. Specifically, the parent application “does not disclose any structural information regarding an antibody that binds Siglec-15 or any epitopes or unique antigenic regions useful for generating antibodies having the desired functional properties,” and the parent application “contained no working example of a Siglec-15 antibody having the desired biological properties.” However, it did disclose that (a) “inhibiting expression of Siglec-15 using a short hairpin RNA (shRNA) knockdown assay impaired formation of osteoclasts from precursor cells,” and (b) “Siglec-15 is capable of restoring the osteoclastogenesis capabilities of the model cell line.” Despite being “persuaded that both studies would have indicated to a person of ordinary skill in the art that inhibiting Siglec-15 can potentially impair formation of osteoclasts from precursor cells,” and acknowledging that it was routine to obtain antibodies that could specifically bind to any target antigen, the PTAB held that the parent application lacked enablement and written description for the patent claims.
In weighing the Wands factors, the PTAB found it significant that the application provided no guidance regarding “any epitopes or unique antigenic regions useful for generating antibodies having the desired functional properties,” although the PTAB admitted that “epitope mapping may not be required to screen antibodies” for function. There was a “high level” of knowledge and skill in the art, but “the prior art was at least somewhat unpredictable.” The PTAB thus decided that “practicing the methods of the challenged claims at the time of the filing of the [p]arent [a]pplication would have required excessive experimentation, even if routine.” Contrast In re Wands, 858 F.2d 731, 737 (Fed. Cir. 1988) (“’a considerable amount of experimentation is permissible, if it is merely routine’ [citing In re Angstadt, 537 F.2d 489 (CCPA 1976)]”).
The PTAB also concluded that the claims lacked adequate written support for a genus of antibodies having the claimed functional properties. While acknowledging the Federal Circuit’s holding in Noelle v. Lederman, 355 F.3d 1343, 1348 (Fed. Cir. 2004), that antibodies are fully described by disclosure of “a ‘fully characterized antigen,’ either by its structure, formula, chemical name, or physical properties” (sometimes called the “antibody exception”), the PTAB determined that Noelle was inapplicable because the “challenged claims contain functional claim language concerning the biological properties of the recited antibody.” As support for its position, the PTAB relied on the Federal Circuit’s decision in Centocor Ortho Biotech, Inc. v. Abbott Labs., 636 F.3d 1341 (Fed. Cir. 2011), where the court similarly distinguished Noelle. Only some of the antibodies that bind to the polypeptide would have the functional property of inhibiting osteoclast differentiation, and other antibodies would not. Thus, the PTAB concluded that the parent application “fails to sufficiently describe common structural information to show possession to [sic] the genus of antibodies recited in the challenged claims.”