Duke University owns US 7,056,712 (‘712), which claims methods of treating a metabolic disorder known as Pompe disease. In particular, ‘712 claims methods of treating Pompe disease using a recombinant human acid α-glucosidase (hGAA) that is produced using Chinese hamster ovary (CHO) cells. BioMarin petitioned for IPR of ‘712 on two grounds: (1) anticipation by US 7,351,410 (van Bree); and (2) obviousness over WO 97//5771 (Reuser) in view of van Hove et al. (1997). The PTAB found Claims 1–9, 12, 15, 20, & 21 anticipated, and held Claims 1–9, 11, 12, 15, & 18–21 obvious. Duke appealed and the Federal Circuit affirmed in part, and reversed in part in Duke Univ. v. Biomarin Pharm. Inc., 2016-1106 (Fed Cir. April 25, 2017).

Regarding anticipation, van Bree teaches administering hGAA to treat Pompe disease, and lists CHO cells as a production system (albeit a disfavored one) for manufacturing the hGAA. Anticipation is a question of fact, and therefore the Federal Circuit (CAFC) only reviews an anticipation finding for substantial evidence. The mention of CHO cells in van Bree — even as a disfavored option — was enough to satisfy the CAFC that anticipation is supported by substantial evidence. The CAFC affirmed the invalidation of Claims 1–8, 12, 15, 20, & 21 for anticipation.

Claim 9, however, was a special case. Claim 9 provides that “the human acid α-glucosidase is a precursor of recombinant human acid α-glucosidase that has been produced in Chinese hamster ovary cell cultures.” The PTAB construed this claim as allowing for mixtures of precursor and full-length hGAA, but the CAFC held that “the human acid α-glucosidase” meant all the hGAA had to be in precursor form. Because van Bree does not mention the precursor form of hGAA, the CAFC reversed the PTAB’s anticipation conclusion for this claim.

For any claim already found anticipated, the CAFC declined to review the PTAB’s obviousness. That meant that the CAFC only had to review the obviousness of Claims 9, 11, 18, & 19. Claims 11 & 18 are dependent claims that had not been argued separately, so in effect the CAFC only had to review obviousness of Claims 9 & 19.

As noted above, the CAFC disagreed with the PTAB’s construction of Claim 9, so the CAFC remanded for reconsideration of obviousness of Claim 9 in view of the new construction. Claim 19 requires that an “immunosuppressant [be] administered prior to any administration of” hGAA. BioMarin had submitted an expert declaration asserting that “[i]f there is a high incidence of patients developing high antibody titers, and immunosuppressant could be administered prophylactically …” (emphasis added). This was the evidence on which the PTAB had relied to show a reasonable expectation of success for Claim 19. The CAFC considered that the fact that the person of ordinary skill could do something is not enough to establish obviousness. Rather, testimony was necessary as to what the person of ordinary skill would do. The CAFC considered that BioMarin’s expert declaration was not substantial evidence necessary to support a reasonable expectation of success. Therefore, the holding of obviousness for Claim 19 was reversed.

The most important take-away from this case is that expert declarations should avoid talking about what the person of ordinary skill “could” do. Instead, declarations should speak of what the person of ordinary skill “would” do, or “did” do.