In that case, Amgen asserted the ’138 patent against Apotex based on Apotex’s aBLAs for biosimilar versions of Amgen’s Grastofil™ (filgrastim) and Lapelga™ (pegfilgrastim) products. Claim 1, the only independent claim, recites “a method of refolding a protein expressed in a non-mammalian expression system and present in a volume at a concentration of 2.0 g/L or greater comprising . . . (a) contacting the protein with a refold buffer . . . ; (b) incubating the refold mixture; and (c) isolating the protein from the refold mixture.” The district court construed the “refold mixture” to have a “high protein concentration . . . at or above about 1 g/L.”
During the patent dance, Apotex had sent Amgen two letters, one for filgrastim and one for pegfilgrastim, stating that it did not infringe the ’138 patent because the concentration of filgrastim or filgrastim critical intermediate in the refold buffer was limited to 0.9-1.4 g/L. At trial, Apotex presented expert testimony that the maximum concentration of “protein” in the refold mixture would actually be 0.708 g/L in its manufacturing process. Apotex’s expert further explained that the pre-litigation letters were inaccurate because the inclusion bodies in Apotex’s processes are not pure proteins, but instead a paste constituting about two-thirds water. Apotex also relied on two batch records providing the actual data from its process. These records showed that the protein concentration in the refold mixture never went above ~0.56 g/L. Based on this evidence, the district court found that Amgen had failed to meet its burden of proving direct infringement.
Amgen challenged the district court’s finding for three reasons:
- Apotex’s Pre-Litigation Letters. The district court had stated in its opinion that Apotex’s pre-litigation letters “are not probative on the issue of protein concentration.” On appeal, Amgen argued that the district court erred by not ascribing probative value to the letters as a matter of law. The Federal Circuit acknowledged that “a district court cannot ignore letters sent” during the patent dance, which constitute party admissions. However, the court found that in context the district court’s statement about the probative value of the letters should be interpreted as a finding of their relative weight, not their relevancy as a matter of law, especially in view of expert testimony that the letters were inaccurate. Thus, the district court’s finding regarding the probative value of the letters was not clear error.
- Construction of “Protein Concentration.” Amgen argued on appeal that “protein concentration” in the claims is interchangeable with “washed-inclusion-body concentration,” discussed in the specification. Because the specification repeatedly distinguishes between proteins and inclusion bodies, the Federal Circuit rejected this argument.
- Apotex’s aBLAs. Amgen argued that the district court’s non-infringement finding was based on too narrow a reading of Apotex’s aBLAs. In particular, Amgen argued that the Federal Circuit should follow Sunovion v. Teva, a Hatch-Waxman case in which the Federal Circuit found that “internal guidelines and a certification were insufficient to avoid a finding of infringement when the [ANDA] itself authorized the activity that would infringe.” Although the parties agreed that Sunovion’s analysis applies in the BPCIA context, the Federal Circuit distinguished the case on its facts and rejected Amgen’s argument that Apotex infringes because its aBLA allegedly left open the possibility of an infringing process. The Federal Circuit found that, unlike in Sunovion, Apotex’s applications do not “authoriz[e] processes that infringe,” and “indeed,  constrain the processes to non-infringing levels.”
Because the Federal Circuit affirmed the district court’s finding of non-infringement on the “refold mixture” limitation, it did not reach the district court’s other finding of non-infringement based on a limitation directed to “2mM or greater,” including the district court’s construction of that term.