Federal Circuit Revisits Standard for Enablement of Antibody Claims

In Baxalta Incorporated v. Genentech, Inc., 2022-1461, the Federal Circuit affirmed the district court’s decision granting Genentech’s motion for summary judgment that claims 1-4, 9, and 20 of U.S. Patent No. 7,033,590 (“the ’590 patent”) are invalid for lack of enablement.

The ’590 patent is directed to resolving a known issue in treating Hemophilia A, a blood clotting disorder. Previously, to treat Hemophilia A, a patient was often intravenously administered Factor VIII. However, some patients cannot be effectively treated with Factor VIII because their bodies develop Factor VIII inhibitors. The ’590 patent sought to resolve this known issue with the treatment of Hemophilia A by providing an alternative means of treatment for patients who develop Factor VIII inhibitors. The alternative means of treating Hemophilia A involves a preparation comprising antibodies or antibody derivatives that bind to Factor IX/IXa to increase the procoagulant activity of Factor IXa. These antibodies allow Factor IXa to activate Factor X in the absence of Factor VIII/VIIIa. Representative claim 1 of the ’590 patent recites:

1.  An isolated antibody or antibody fragment thereof that binds Factor IX or Factor IXa and increases the procoagulant activity of Factor IXa.

Baxalta sued Genentech, Inc. for patent infringement, alleging that Genentech’s Hemlibra® (emicizumab) infringes the ’590 patent. Genentech moved for summary judgment of invalidity of claims 1-4, 19, and 20 of the ’590 patent for lack of enablement, which the district court granted. Baxalta appealed.

On appeal, Baxalta argued that summary judgment of invalidity for lack of enablement was improper because a skilled artisan could make and identify the new claimed antibodies using a routine hybridoma-and-screening process disclosed in the ’590 patent.  According to Baxalta, using this routine screening process to identify the claimed antibodies would not amount to undue experimentation.

The Federal Circuit disagreed with Baxalta, citing the recent Supreme Court case Amgen, Inc. v. Sanofi.[1] The Federal Circuit stated:

As the Supreme Court recently reaffirmed in Amgen Inc. v. Sanofi, “the specification must enable the full scope of the invention as defined by its claims,” allowing for “a reasonable amount of experimentation.” 598 U.S. 594, 610–12 (2023). In other words, “the specification of a patent must teach those skilled in the art how to make and use the full scope of the claimed invention without undue experimentation.” MagSil Corp. v. Hitachi Glob. Storage Techs., Inc., 687 F.3d 1377, 1380 (Fed. Cir. 2012) (internal quotation marks and citation omitted).[2]

Applying the “undue experimentation” standard, the Federal Circuit analogized the screening process identified by Baxalta to the technique in Amgen. In discussing Amgen, the Federal Circuit stated that “[t]he full scope of the claims covered potentially millions of antibodies, but the specification only disclosed the amino acid sequences of twenty-six antibodies that performed the two claimed functions,”[3]  and that “[t]o make and use the undisclosed antibodies, skilled artisans could either follow the ‘roadmap’ disclosed in the patent or employ a technique known as ‘conservative substitution.’”[4]  According to the Federal Circuit, “The Supreme Court held these methods ‘amount to little more than two research assignments’ and fail to enable the full scope of the claims.”[5] The Federal Circuit concluded that “[t]he facts of this case are materially indistinguishable from those in Amgen.”[6]. The court held that the claims of the ’590 patent potentially covered millions of antibodies, but the specification disclosed only eleven antibodies. The Federal Circuit concluded the specification of the ’590 patent merely disclosed a roadmap test that would direct a person of ordinary skill in the art to use a trial-and-error process. According to the Federal Circuit, nothing in the specification of the ’590 patent teaches how to identify the claimed antibodies, other than by repeating the trial-and-error process. Nothing in the specification of the ’590 patent would allow the person of ordinary skill in the art to predict which antibodies would perform the claimed functions.

Baxalta attempted to distinguish the ’590 patent from the patent invalidated in Amgen by alleging that the hybridoma-and-screening process predictably and reliably generates new claimed antibodies each time the process is performed. The Federal Circuit concluded that even accepting Baxalta’s argument as true, the hybridoma-and-screening process is still a “trial-and-error” approach because a skilled artisan must make candidate antibodies and screen them to determine which ones are covered by the patent. Such a trial-and-error approach, without more, constitutes undue experimentation.