Bayer v. Mylan: “Clinically Proven Effective” Claim Language Fails to Breathe Patentability into Method Claims

In a precedential decision, the Federal Circuit held the phrase “clinically proven effective” was a functionally unrelated limitation to the claimed method of administering rivaroxaban and aspirin, and failed to make the challenged claims patentable.¹ As a result, claims that were otherwise anticipated were not saved by the addition of the “clinically proven effective” language. Likewise, objective indicia of nonobviousness tied to the “clinically proven effective” claim language failed to a have nexus to the merits of the claimed invention and did not support nonobviousness.

In Bayer Pharma Aktiengesellschaft v. Mylan Pharmaceuticals Inc., No. 23-2434 (Fed. Cir. Sept. 23, 2025), the Federal Circuit affirmed-in-part, vacated-in-part, and remanded the Patent Trial and Appeal Board’s (Board) decision holding claims 1-2 of U.S. Patent No. 10, 828,310 unpatentable as anticipated and claims 1-8 unpatentable as obvious.

Bayer owns the ‘310 patent, which describes the results of the phase III clinical trial “COMPASS,” an evaluation of the efficacy and safety of administering rivaroxaban with and without aspirin for the risk reduction of cardiovascular events in patients with coronary artery disease (CAD) and/or peripheral artery disease (PAD).² Mylan and Teva (collectively, Appellees) filed inter partes review petitions, challenging the claims of the ‘310 patent.³

In this case, Appellant argued the Board erred in:

• (1) construing “clinically proven effective” as non-limiting and finding, in the alternative, that it is inherently anticipated;
• (2) construing “first product comprising rivaroxaban and aspirin” to encompass administration of rivaroxaban and aspirin as separate dosage forms;
• (3) failing to articulate why a skilled artisan would have combined Foley⁴ and Plosker⁵ with a reasonable expectation of success; and
• (4) failing to analyze whether clinical proof of efficacy was an unexpected result.⁶

First, the Court affirmed the Board’s finding that “clinically proven effective” is non-limiting, and concluded that even if the phrase were limiting, it would still be a functionally unrelated limitation that fails to make the challenged claims patentable.⁷ The Court relied on its reasoning in King Pharmaceuticals, Inc. v. Eon Labs, Inc.,⁸ where it held that an anticipated method of treatment is not made patentable by adding an instructional limitation that does not have a “functional relationship” with the claimed method. ⁹ The Court explained that adding “clinically proven effective” was no better than adding a limitation with an accolade like “Best Drug of 2026.”¹⁰ The Court distinguished Allergan Sales, LLC v. Sandoz, Inc.,¹¹ stating the limitation in this case did not further define the dosages administered.¹² In Allergan the claims provided “specif[ic] safety and efficacy benchmarks.”¹³ The Court reenforced that an additional “informing” instructional limitation without a new and unobvious functional relationship with the known method of treatment, was not patentable.¹⁴

Second, the Court agreed with Appellant that “first product comprising rivaroxaban and aspirin” is limited to a “single dosage form” that “include[s] or contain[s] both rivaroxaban and aspirin.”¹⁵ Based on the plain language, “first product” cannot encompass separate dosage forms.¹⁶ The Court remanded this issue for further consideration of Appellees’ obviousness arguments under the correct construction of the “first product” term.¹⁷

Third, the Court found the Board adequately articulated a reason why a skilled artisan would combine prior art references.¹⁸ A skilled artisan would have reason to use 75 mg and 81 mg aspirin daily instead of the 100 mg aspirin taught by Foley, because it is consistent with the dosage range taught by Plosker.¹⁹ The Court found no failure by the Board to articulate an adequately supported rationale for why a skilled artisan would have combined Foley and Plosker with a reasonable expectation of success.²⁰

Fourth, the Court rejected Appellant’s argument that the Board failed to analyze the secondary consideration of unexpected results.²¹ Secondary considerations are only relevant if there is a nexus between the merits of the claimed invention and the evidence.²² In this case, the Court held Appellant’s evidence that clinical proof of efficacy was unexpected had no connection to the “merits of the claimed invention” and thus “does not compel a holding of nonobviousness.”²³

The Court affirmed the Board’s judgment holding claims 1-4 unpatentable, vacated the judgment of unpatentability with respect to claims 5-8, and remanded for further proceedings.²⁴

Specific safety and efficacy benchmarks should be used to support patentability rather than ambiguous accolades like “clinically proven effective.”