This post gets a “Part II” designation because the Fed. Cir. clarified the support necessary to find a negative limitation in the prior art (Almirall LLC v. Amneal Pharmaceuticals LLC, Appeal No. 2020-2331 (Fed. Cir., March 14, 2022)). In my earlier post on Novartis v. Accord Heathcare, 21 F.4th 1362 (Fed. Cir. 2022), the Fed. Cir. panel addressed the question of how much support was required in the specification for an applicant to add a negative limitation to a claim, to distinguish the prior art.

In this case,  Almirall owns U.S. Pat. No. 9,517,219 which claims a method for treating acne with dapsone formulations that contain a thickening agent, “A/SA” that is an acryloyl/sodium taurate copolymer, and which formulations do not contain adapalene, another anti-acne agent. The Fed. Cir. panel affirmed the PTAB finding that the claims were obvious in view of the art cited by Amneal. Claim 1 reads as follows:

1. A method for treating a dermatologic condition selected from the group consisting of acne vulgaris and rosacea comprising administering…a pharmaceutical composition comprising:

About 7.5% w/w dapsone;

About 30% w/w to about 40% w/w diethylene glycol monomethyl ether;

About 2% w/w to about 6% w/w of a polymeric viscosity builder comprising A/SA; and

Water;

Wherein the topical pharmaceutical composition does not comprise adapalene.

The primary reference, Garrett, discloses the use of dapsone to treat acne but doesn’t mention adapalene. Rather, it uses a different thickener, 0.5% to 4% carbomer. The secondary references such as Nadau-Fourcade disclose anti-acne formulations that can comprise carbomers or an A/SA thickener at 0.01% – 5%. The Bonacurla reference discloses a specific A/SA polymer that can be used at 0.5-5.0%. So the range of the A/SA thickener recited in claim 1 is partially overlapped by, i.e., the range in Nadau-Fourcade. The panel affirmed that the claims in the ‘219 patent were obvious, since the panel agreed with the Board that the POSA would be motivated to replace the carbomer thickener used by the primary reference with the A/SA thickeners disclosed by the prior at.  The panel agreed with the Board that a prima-facie case of obviousness does not require that a single reference such as Garrett disclose all claimed ranges: “In the absence of evidence indicating that there is something special or critical about the claimed range, an overlap suffices to show that the claimed range was disclosed in—and therefore obvious in light of—the prior art.”

Almirall argued that the Board failed to give any weight to the negative limitation in the claims, when it resolved the obviousness question. Relevant here, is that Almirall argued that more than a mere absence of adapalene from the primary reference is required to support an implicit negative limitation in the primary reference. Citing Novartis Pharms. Corp. v. Accord Healthcare, Inc., the panel found that the primary reference disclosed complete (self-contained?) formulations that would be read by the POSA to exclude adapalene. In other words, the panel agreed with Amneal that Garrett “effectively teaches the negative adapalene claim limitation…It was reasonable  for the Board  to find that, in the context of Garrett, a skilled artisan would recognize that [Garrett] discloses a complete formulation—excluding the possibility of an additional active ingredient.” Slip. op. at 12.

In Novartis, the question was whether or not a negative limitation that was added to the claims during prosecution was adequately supported when the excluded element was not mentioned in the specification. The panel in that case held that there was sufficient extrinsic evidence that the POSA would read the specification as excluding the step that was “limited out” of the claim.  In other words, the negative limitation, “no adapalene,” was implicit in the Barrett disclosure. The present case is therefore a mirror image of Novartis. While there was adequate support for the limitation in the ‘219 specification, Almirall needed the panel to find that the negative limitation was not present in the prior art, so it could use the negative limitation to distinguish Barrett. This seems to be a “What is sauce for the goose is sauce for the gander” opinion. Unfortunately for Almirall, its goose got cooked.