Patent Lodge

The U.S. Supreme Court recently issued its decision in Amgen Inc. v. Sanofi; a decision long-awaited by patent practitioners. In its decision, the Court unanimously held that several of Amgen’s patent claims to a class of antibodies were invalid for lack of enablement. Relying on hundred-year-old precedent involving, for example, Samuel Morse’s electromagnetic telegraph and Thomas Edison’s incandescent light bulb, the Court held that “[t]he more one claims, the more one must enable.” Thus, for a claim to a class of compounds to be adequately enabled, the specification must teach a person of skill in the art how to make and use the entire class. According to the Court, Amgen’s patents failed to do so. At a high level, this decision maintains the status quo with respect to the standard for patent enablement, neither narrowing nor broadening it. Interestingly, the Supreme Court did not acknowledge the Federal Circuit’s oft-cited In re Wands enablement factors, instead solely recounting its own precedent.

Case Background

Amgen and Sanofi both worked to develop drugs to reduce cholesterol by binding to or otherwise inhibiting PCSK9—a protein that binds to LDL receptors and, in a nutshell, stops them from extracting LDL cholesterol from the bloodstream. Each company successfully developed and commercialized a unique antibody medication specific to PCSK9: Amgen’s Repatha and Sanofi’s Praluent.

Amgen sought (and received) broad “genus” claims covering this technology in 2014. While Amgen’s patents specifically identified 26 amino acid sequences for antibodies that would block or inhibit PCSK9, Amgen took it a step further and claimed “the entire genus” of antibodies that:

1. “bind to specific amino acid residues on PCSK9,” and
2. “block PCSK9 from binding to [LDL receptors].”

Amgen provided two alternative methods for arriving at the remainder of the genus–“the roadmap” or the “conservative substitution.”

The “roadmap” involves four “steps”:

1. generate antibodies;
2. test them to determine if they bind to PCSK9;
3. test them to determine if they bind to the “sweet spot”; and
4. test them to determine if they bind to the “sweet spot” and block PCSK9 from binding to LDL receptors.

The “conservative substitution” method also involves multiple steps –

1. start with an antibody known to hit all four requirements of the “roadmap”;
2. replace amino acids with other amino acids that are similar; and
3. test.

Amgen then sued Sanofi for infringement of these claims.

Sanofi’s lead defense (and the most relevant here) was lack of enablement for the genus claims.

Sanofi argued that the genus claims, which included “the roadmap” or the “conservative substitution” methods, required extensive experimentation and trial-and-error to arrive at viable antibodies, and thus failed to enable the claimed invention under 35 U.S.C. § 112. Importantly, Amgen’s disclosed methods could result in “potentially millions more antibodies” than the 26 that were disclosed by amino acid sequence.

At the District Court and the Federal Circuit, Sanofi’s enablement argument won out. Both courts held that Amgen’s broad genus claiming failed the enablement standard.

Supreme Court Holding

On appeal to the Supreme Court, Sanofi won again. The Court held that the district court and Federal Circuit were right, and Amgen’s claims were not enabled.

The Court explained that its enablement precedent, dating back to O’Reilly v. Morse (1853) and The Incandescent Lamp Patent (1989), adheres to the principle that a patentee cannot claim much but enable little: “the more a party claims, the broader the monopoly it demands, the more it must enable.” Thus, if a patent claims a class (orgenus) of products, the specification must enable a person of skill in the art to make and use the entire class without “painstaking experimentation.”

The Court did, however, state that this does not mean the specification “must describe with particularity how to make and use every single embodiment within a claimed class.” It may be sufficient for the specification to describe only a few example embodiments “if the specification also discloses ‘some general quality . . . running through’ the class that gives it a ‘peculiar fitness for the particular purpose.’” (citing Incandescent Lamp). The Court also allowed that the specification can leave the artisan to do a reasonable amount of experimentation or testing to practice the claims. As is always the case, the amount of disclosure required or experimentation allowable “depends,” including upon the size of the genus, the nature of the art and the knowledge of a skilled artisan. But the skilled artisan must have more to go on than “advice to engage in trial and error.” (internal quotation marks omitted).

Amgen attempted to claim an entire class of compounds by their function, namely antibodies that bind to the “sweet spot” of PCSK9 thereby inhibiting it from binding to LDL, while only describing 26 amino acid sequences in its specification. The two processes, the “roadmap” and “conservative substitution” did not save Amgen. According to the Court, these amounted to “little more than two research assignments” which forced scientists to conduct “painstaking experimentation” to see what worked. (citing Incandescent Lamp). The Court therefore held that Amgen’s specification did not enable the claims.

Takeaways

The main problem with Amgen’s claim to a broad class of antibodies was that the class was defined by its function—the ability to bind to or interfere with PCSK9. But Amgen’s specification did not give scientists enough information to choose candidate antibodies from the millions of options and therefore required scientists to engage in a great deal of experimentation and failure. “That is not enablement”—it is a “hunting license.”

Those seeking to patent genus claims should take this decision to heart. Experimentation to practice the claim is fine, but the specification should contain some guidance to direct the experimentation. Preferably, that guidance should be more than “go test a million things to see what sticks.” Identifying a common characteristic of the genus that is helpful to identifying species that work is key. Additionally, patents with broad genus claims should also include narrower claims to specific species as a backup.

For those seeking to invalidate a genus claim, little has changed. The Court went to great lengths to show that the enablement requirement applies to modern technology in the same way it applied to the technology of the 1800s. When attacking enablement, practitioners should continue to hit on the theme of the patent “bargain,” i.e., that in exchange for the limited term of protection afforded by a patent, the patentee must fully disclose the invention to the public.