The Supreme Court’s unanimous 2023 decision in Amgen v. Sanofi, 598 U.S. 594 (2023), reshaped enablement analysis for broad genus claims under 35 U.S.C. §112. There, Amgen’s evolocumab (Repatha®) patents claimed a genus of antibodies that bind to specific amino acid residues of PCSK9 protein and block its activity, which helps lower LDL cholesterol. While the specification disclosed 26 examples, the claims sought to cover all antibodies with that function. The court held that such broad claims require equally broad disclosure – “the more a party claims, the more it must enable.” The ruling confirms that functional genus claims must teach a POSA how to make and use substantially all embodiments without undue experimentation.

In the wake of Amgen, broad functional claims – especially ones covering potentially thousands or millions of undisclosed variants – have been scrutinized rigorously for sufficient disclosure. The Federal Circuit, in particular, has affirmed invalidity of several genus claims for lacking enablement or adequate written description support, while some patents have withstood these challenges where the disclosure was more tailored to the claim breadth. The courts have made a key distinction between a functional claim and structural ones, where less disclosure is required for structural claims. Below, we summarize key post-Amgen decisions, highlighting the written description and enablement issues, the courts’ analyses, and the implications for patent practice and would-be challengers. These cases illustrate how patent drafters and litigators must navigate the fine line between claim breadth and disclosure depth in the post-Amgen era.

Baxalta v. Genentech (Fed. Cir. 2023)

Here, the Federal Circuit affirmed summary judgment of invalidity for non-enablement. Baxalta’s emicizumab (Hemlibra®) for treating hemophilia by binding to Factor IX/IA and increasing procoagulant activity, was defined by functional claims. The court found that although Baxalta disclosed some examples and general methods (like hybridoma screening), the examples did not enable the full scope of the broad functional genus claims. Finding the facts “materially indistinguishable” from Amgen, the court noted that the disclosure amounted to “little more than two research assignments” and would require undue experimentation to identify antibodies across the claimed genus without further guidance.

Post-Amgen Patent Playbook: Section 112 Under the Microscope

Baxalta, like Amgen, confirms that broad antibody claims defined by function are likely unpatentable unless the specification provides either a large and varied set of working examples or predictive guidance ( e.g., common structural patterns) to cover the full scope. Patent drafters must examine functional claims that lack sufficient disclosure, as courts will require more than a few examples plus an invitation to perform screening.

Medytox v. Galderma (Fed. Cir. 2023)

In this post-grant review, Medytox’s patent claimed a botulinum toxin (similar to Botox®) composition with an efficacy limitation of a “responder rate of 50% or greater at 16 weeks.” The specification disclosed only three data points – 52%, 61% and 62%. Galderma argued, and the PTAB agreed, that “50% or greater” implied a range up to 100%, and that the patent failed to enable the full scope. The Federal Circuit affirmed, applying the Wands factors and finding no guidance for achieving higher responder rates. Because a POSA would need to engage in undue experimentation to reach the upper end of the claimed range, the claims were not enabled across their full breadth.

Though the claims were directed to composition (neither functional nor structural), the legal issue nonetheless was enablement. This illustrates that even for claims not explicitly drawn to a “functional” genus, broad numeric ranges or open-ended limits must be fully enabled. The specification should provide enough examples or reasoning to show that any value in that range can be achieved without undue experimentation. Patent practitioners should thus avoid claiming ranges that extend beyond the demonstrated or scientifically extrapolatable performance disclosed in the specification. Patent challengers should likewise review for Amgen like claims lacking sufficient disclosure.

Melinta Therapeutics v. Nexus Pharmaceuticals (N.D. Ill. 2024)

Here, the district court considered enablement and written description challenges to patents covering an intravenous antibiotic formulation of minocycline with magnesium, claimed to reduce injection-site hemolysis, i.e., red blood cell damage). The claims included specific limitations of pH (3.0-5.0) and osmolality (concentration) of (<500 mOsm/kg). Nexus argued the claims were invalid because the patent did not teach how to maintain solubility across the pH range or address the absence of a lower osmolality limit including zero. The court, however, upheld the claims, distinguishing the case from Medytox. Relying on expert testimony, the court found that a POSA would know how to achieve and maintain appropriate pH, osmolality, and volume through routine formulation techniques, making the claims both enabled and adequately described.

This case shows that broad claims can survive enablement challenges if a POSA can use common knowledge to reach workable embodiments ( e.g., a POSA would know that an intravenously administered solution cannot have zero osmolality). Courts may uphold claims where routine adjustments known to POSA can fill gaps in disclosure, especially if expert testimony supports it. Patent drafters should still provide reasoning for chosen ranges and note routine techniques. Generic challengers, in turn, should be ready to argue that unaddressed extremes effectively broaden the claim scope beyond what is enabled.

Purdue Pharma v. Collegium Pharm. (Fed. Cir. 2023)

Not all post-Amgen Section 112 disputes have focused solely on enablement – written description also remains a critical issue. Here, the Federal Circuit affirmed the PTAB’s invalidation of claims covering an abuse-deterrent opioid formulation. The claims required an “aversive agent,” and though Purdue’s patent listed polyglycolyzed glycerides (PGGs) in the specification, it described PGGs only as surfactants and never indicated they could serve as the aversive agent. The court found that listing alone, without more, failed to show that the inventors possessed the concept of using PGGs in the claimed functional role. Purdue’s own expert conceded that the specification did not disclose PGGs as aversive agents, and the court noted that other parts of the patent described surfactants as distinct from aversive agents. Thus, the patent lacked the necessary “blaze marks” for a POSA to use PGGs as aversive agent.

Purdue stresses that written description demands more than naming a compound; it must link the compound to its claimed function. Simply including an ingredient in a broad list does not establish that the inventor had possession of its specific use without more detailed disclosure or a clear knowledge of a POSA that the component generally serves that function. Boilerplate lists of ingredients in broad classes won’t suffice if the particular function or property is not explicitly described but claimed – patentees should identify each component’s function in the specification to lower the risk a written description failure. Challengers should emphasize the absence of “blaze marks” directing a POSA to the claimed use, and leverage expert admissions or an alternate teaching in the specification that treat components differently from their claimed function.

Exelixis v. MSN (D. Del. 2024)

This case contrasts with Amgen because it deals with structural genus claims under §112 and not functional ones. Exelixis’ patent claimed crystalline forms of cabozantinib (L)-malate, the active ingredient in Cabometyx®. MSN argued that the patents were invalid because the specification disclosed only two specific polymorphs (“N-1” and “N-2”) and did not describe the full range of possible crystalline forms. The court rejected this argument, holding that the genus was adequately described through the compound’s name, formula, and crystalline nature. Citing precedent, the court explained that, unlike functional claims, structural genus claims require disclosure only of the structural features shared across the genus – even if polymorphic forms differ in crystal arrangement and properties. A disclosure of a single representative species may even suffice. A POSA would be able to “visualize or recognize the members of the genus,” since the scope was bounded by the known compound and its crystalline form.

Exelixis shows that for structurally-defined claims, written description can be satisfied by disclosing a representative number of species or the defining characteristics of the genus. If all members of the genus share certain structural features (here, the same molecule and crystalline lattice structure), and those are disclosed, then the claim may meet the written description requirement even if not every variant is enumerated. For patent prosecutors, this means that when claiming a genus of chemical compounds, emphasizing common structure ( e.g., a core scaffold or specific formula) may support broader claims. For challengers, this means that failure to meet the §112 standard may require more than pointing to undisclosed variants. They can achieve this by showing that genus is not sufficiently bounded by the disclosure or that a POSA would not recognize undisclosed forms as falling within it.

Allergan USA v. MSN (Fed. Cir. 2024)

This case addresses written description in the context of optional claim elements. Allergan’s patent covered an abuse-deterrent tablet of eluxadoline (Viberzi®), with excipients including an optional “glidant and/or lubricant.” The district court noted that every example included a glidant ( e.g., colloidal silicon dioxide), and invalidated the claims for lack of written description, reasoning that a POSA would expect a glidant to be necessary for proper manufacturing and that the patent failed to show a glidant-free formulation would function effectively. The Federal Circuit reversed finding sufficient written description because the specification described two embodiments with “one or more” of several inert ingredients, including colloidal silica. This, the majority reasoned, implied that a glidant could be present or absent. The patent never stated a glidant was required and thus permitted combinations that could omit it – making the excipients optional. Judge Dyk dissented, emphasizing that every complete formulation used a glidant and that nothing in the disclosure affirmatively taught its omission.

The case illustrates that when claiming optional components, patentees must ensure the specification supports both the presence and absence of the element, though courts may infer such support from flexible language and contextual clues. Ideally, the patent should either include an embodiment lacking the optional element or contain language that makes clear the element isn’t essential. The Federal Circuit’s reversal here provides a precedent that courts may infer support for an optional element if the specification does not demand that element and includes it only as one of several interchangeable excipients. Nonetheless, best practice is to explicitly describe alternatives (both with and without the element) to avoid uncertainty. Challengers may pounce on the lack of clear disclosure of optional elements and see if the patent affirmatively suggests that the omitted element is indispensable.

In re: Entresto (Fed. Cir. 2025)

This case revisits written description and enablement issues in the long-running litigation over Novartis’s heart failure drug Entresto®, a combination of valsartan and sacubitril. The issue was whether a patent is invalid for failing to describe an embodiment that was later discovered but unknown at the time of filing. As of the 2002 priority date, the 1:1 complex of valsartan-sacubitril had not yet been discovered. The district court sided with generic challengers holding that the patent failed the written description requirement, because the inventors could not have possessed a complex not known in 2002. As for enablement, the court concluded that since it must be evaluated based on the state of the art at the time of filing, the patentee was not obligated to enable a complex that had not yet been discovered. The Federal Circuit reversed the written description ruling, explaining that the patent did not claim the later-discovered complex. Instead, it simply claimed the use of valsartan and sacubitril in combination – as a co-formulation or co-administration in separate tablets. Because written description is tied to the claimed invention, and the patent supported co-administration, the court found that the inventors had possession. The complex, according to the court, was a later improvement outside the claim scope and irrelevant to the §112 analysis.

This ruling reinforces a critical principle: written description and enablement are judged as of the filing date, and patents need not predict or describe later developments beyond the scope of the claims. The key takeaway for drafters is to claim only what was actually invented. Novartis may have properly claimed the combination of valsartan and sacubitril but not any specific complex they might later form. The later discovery of a complex between the two did not undermine the original disclosure, but Novartis also cannot retroactively claim that specific improvement. Strategically, patentees should avoid overreaching in their specifications and instead focus on clearly describing and claiming what they knew and possessed at the time of filing. Patent Challengers should thus focus on what the patent does claim and whether the disclosure supports that at the time of filing.