by Dennis Crouch
The Federal Circuit recently handed down its decision in Bayer Pharma Aktiengesellschaft v. Mylan Pharmaceuticals Inc., No. 2023-2434 (Fed. Cir. Sept. 23, 2025), focusing on the the claim limitation “clinically proven effective.” Bayer’s U.S. Patent No. 10,828,310, which claims methods for reducing cardiovascular events in certain patients by administering specific doses of rivaroxaban (2.5 mg twice daily) and aspirin (75-100 mg daily) “in amounts that are clinically proven effective.” Generic manufacturers Mylan, Teva, and Invagen successfully challenged the patent in IPR proceedings — finding the claim term non-limiting. The Federal Circuit agreed that the term did not provide patentable weight – but through a different analytical path.
Clinically Proven Effective: Contrasting Analytical Approaches
The PTAB concluded “clinically proven effective” was non-limiting, reasoning that because the claims already recited specific dosage amounts (2.5 mg rivaroxaban twice daily and 75-100 mg aspirin daily), the additional characterization of these amounts as “clinically proven effective” was merely redundant. Drawing on Bristol-Myers Squibb Co. v. Ben Venue Labs., Inc., 246 F.3d 1368 (Fed. Cir. 2001), the PTAB found the phrase constituted an “expression of intended result [that] essentially duplicates the dosage amounts recited in the claims.”
On appeal though, the Federal Circuit took a markedly different approach, declining to decide whether “clinically proven effective” was limiting at all. Instead, the court applied the functional relationship test from King Pharms, concluding that even if the phrase were limiting, it constituted a “functionally unrelated limitation” that could not rescue an otherwise anticipated method from unpatentability. The court reasoned that clinical proof of efficacy “in no way transforms the process of taking the drug[s]” at the specified amounts and frequencies, emphasizing that the “the actual method . . . is the same” regardless of subsequent clinical validation.
The Functional Relationship Test from King Pharmaceuticals In King Pharmaceuticals v. Eon Labs, 616 F.3d 1267 (Fed. Cir. 2010), the Federal Circuit established an important test for determining when additional claim limitations can make an otherwise anticipated method patentable. The case involved patents claiming methods of taking the muscle relaxant metaxalone with food to increase the drug's bioavailability - a practice that was already disclosed in prior art for reducing stomach upset. The patentee argued that adding a step of "informing" patients about the increased bioavailability made the claims novel. The Federal Circuit rejected this argument, holding that the relevant inquiry is whether the additional limitation has a "new and unobvious functional relationship" with the known method. Since informing patients about the drug's benefits "in no way transforms the process of taking the drug with food" and "the actual method . . . is the same" regardless of whether patients are informed, the informing step lacked the required functional relationship. This test prevents patentees from indefinitely extending patent protection over known methods by simply adding instructional, informational, or other functionally unrelated steps that don't change how the method actually works.
This analytical framework allowed the Federal Circuit to sidestep the inherency analysis that occupied substantial attention in the PTAB proceedings, instead focusing on whether any purported limitation bore a functional relationship to the claimed method itself. In particular, the PTAB FWD included a detailed inherency analysis and conclusion that the prior art teaching of administering 2.5 mg rivaroxaban twice daily with 100 mg aspirin once daily to patients — inherently disclosed “clinically proven effective” amounts. Bayer challenged this inherency finding, arguing that numerous factors could impact whether a dosing regimen proves clinically effective, including study design, patient populations, and trial conduct. On appeal though, the Federal Circuit bypassed this inherency debate entirely through its functional relationship analysis discussed above.
The Court’s “Clawback” Concern
The Federal Circuit’s decision reflects the Court’s policy concern about preventing patentees from clawing back otherwise anticipated methods from the public domain through post-hoc validation. Chief Judge Moore articulated this concern directly during oral argument, warning against allowing developers to “put [precise dosages] out there and then later come back and say, now I want to claim it even though it’s been in the public domain for a very long time” simply by adding descriptive language like “clinically proven effective.” To illustrate the absurdity of this approach, she posed a hypothetical claim limitation requiring amounts that had been “voted best new drug of 2026” – a descriptor that, like “clinically proven effective,” adds nothing meaningful to the actual method of treatment. The court recognized that allowing such clawback would enable “the indefinite patenting of known products [and methods] by the simple inclusion of novel, yet functionally unrelated limitations.” (Quoting King).
Note: The Federal Circuit did provide a partial win to the patentee – finding that the PTAB had erred in construing the limitation “first product.”
The “Super-Utility” Problem and Patent Gamesmanship
I am still pondering this decision and want to reserve judgment. On the one hand, I am uncomfortable with the Functional Relationship Test because it ultimately means that we ignore express claim terms. On the other hand, it seems necessary to avoid undue patent gamesmanship — particularly by permitting post-invention validation to confer patentability.
The validating limitation (clinically proven effective) appears akin to a super-utility limitation. And, maybe there is something there. Our standard utility requirement is a low bar, and maybe we should provide incentives to patentees who do the work to actually prove effectiveness. But, if we’re going to permit applicants to explicitly claim clinical effectiveness in their patent claims, we should demand something more substantive than our traditional utility showing. That means at least a requirement of robust clinical evidence.
But this approach faces a fundamental problem: it risks creating patents for the mere act of validation rather than true invention. It seems plainly wrong to allow a patent simply for conducting studies to prove that drug X effectively treats disease Y when prior art already disclosed using drug X for that very purpose. The clinical work, while valuable to society, doesn’t transform an anticipated method into a patentable one – it merely confirms what was already taught.
Perhaps the answer lies in requiring such clinical proof only when the effectiveness claim goes beyond what was reasonably expected from the prior art disclosure, or when the claimed dosing regimen represents a genuine departure from known approaches. This would preserve incentives for meaningful clinical development while preventing the indefinite extension of patent protection through post hoc validation of anticipated methods. The challenge, of course, is crafting doctrine that can meaningfully distinguish between these scenarios without creating unworkable complexity.
Allergan Provides a Potential Roadmap for Patentees
The Federal Circuit had to distinguish a fairly close decision, and the language here reveals some potential strategies for patent drafters. Allergan Sales, LLC v. Sandoz, Inc., 935 F.3d 1370 (Fed. Cir. 2019). In Allergan, the Federal Circuit held that “wherein” clauses specifying minimum safety and efficacy requirements were indeed material to patentability and thus limiting. The key factual distinction was that these clauses were “written in open format” allowing for additional ingredients like solvents, buffers, and preservatives to ensure safety and efficacy. On the other hand, the “clinically proven effective” limitation in Bayer is directly tied to the amount of rivaroxaban and aspirin, those amounts are particularly claimed (2.5 mg and 75-100 mg); and there was no evidence that “clinically effective” further limited the dosage in any way. This meant that the clinically proven effective limitation did not provide any functional limitations beyond other elements of the claims. The basic claim drafting idea here then is to ensure that clinically proven effective will have some genuine limitation on the method or product itself. Here, that might exist if the aspirin dosage was listed as 50-100, but then only proven effective for 75-100. This may be enough to establish a functional relationship.