Background and Key Holding
On April 16, in Teva Pharmaceuticals International GmbH v. Eli Lilly and Company, the Federal Circuit reversed a district court invalidity ruling and reinstated a jury verdict for Teva. The court held that method-of-treatment claims directed to using humanized anti-CGRP antagonist antibodies to treat headache satisfied the written description and enablement requirements of 35 U.S.C. § 112. The decision draws a clear line between claims that cover a biological agent itself and claims that use a known agent for a specific therapeutic purpose, and applies a meaningfully lighter disclosure burden to the latter.
CGRP is a protein that, when it binds to certain receptors, causes blood vessels to expand, which is a process associated with headache. Anti-CGRP antagonist antibodies block that activity. These antibodies occur naturally in mice, and a technique called humanization converts mouse antibodies into a form the human immune system will tolerate.
Teva holds three patents claiming methods of treating headache by administering a humanized anti-CGRP antagonist antibody. The specifications describe several murine (mouse) anti-CGRP antagonist antibodies and standard humanization methods, but disclose only a single humanized antibody, G1, the active ingredient in Teva’s Ajovy product. Lilly’s Emgality product uses a different humanized anti-CGRP antagonist antibody for the same purpose. A jury found that Lilly willfully infringed Teva’s patents and failed to prove the claims invalid for lack of written description or enablement. The district court then granted Lilly judgment as a matter of law, invalidating the claims on both grounds. Teva appealed.
PTAB Admissions Strengthened the Jury Verdict
Before reaching the merits, the Federal Circuit paused on a set of admissions Lilly had made in earlier Patent Trial and Appeal Board (PTAB) proceedings. Lilly had filed inter partes review (IPR) petitions challenging both Teva’s headache patents and a separate group of Teva antibody composition patents. The PTAB invalidated the antibody composition claims but declined to invalidate the headache patent claims, i.e., the very claims at issue here. In arguing for invalidity of the antibody composition patents, Lilly argued to the PTAB that anti-CGRP antagonist antibodies were “well known” in the art, that the prior art was “replete with exemplary disclosures” of such antibodies, and that humanization “was a well-established and routine procedure” by the priority date. Those admissions, made when it suited Lilly’s interests in a different proceeding, came back as substantial evidence supporting the jury’s verdict in this one, alongside independent expert testimony at trial reaching the same conclusions.
Written Description Analysis
For genus claims, a specification must disclose either a representative number of species or structural features common to the members of the genus so that a skilled artisan can visualize or recognize the members. The district court held that disclosing only G1, i.e., a single humanized antibody, was not enough. The Federal Circuit disagreed, and the reason why matters for how practitioners think about claim drafting.
The court drew on a line of decisions holding that the written-description analysis looks different when the claimed genus is not itself the invention, but rather a known tool used in service of a different inventive concept. Here, the invention was the therapeutic use of humanized anti-CGRP antagonist antibodies to treat headache, not the antibodies themselves. Because anti-CGRP antagonist antibodies and methods of making them were well known in the art, and because humanization was routine, the specification’s disclosure of multiple murine versions plus a single humanized example, read against that backdrop of established prior art, the court found that this was sufficient. As the court put it, murine antibodies were just a routine (and disclosed) humanization step away from being humanized versions.
Lilly argued that the absence of specific amino acid sequences in the specification was fatal. The court was unpersuaded. Evidence showed that skilled artisans could determine an antibody’s sequence using standard sequencing techniques once they had the antibody, and anti-CGRP antagonist antibodies were commercially available before the priority date. Lilly also urged the court to read its earlier AbbVie v. Janssen decision as requiring disclosure of an antibody structurally similar to the accused product. The court declined to go that far. The court countered that AbbVie established one contextual factor, but it did not create a bright-line rule applicable regardless of the circumstances.
Enablement Analysis
Lilly’s enablement argument tracked the Supreme Court’s 2023 decision in Amgen Inc. v. Sanofi: the genus of humanized anti-CGRP antagonist antibodies is enormous, the specification provides no reliable way to predict in advance which candidates will antagonize CGRP, and the practical work of screening thousands of candidates would constitute undue experimentation.
The Federal Circuit was not persuaded, and the distinction it drew is the key takeaway. In Amgen, the patentee claimed the antibodies themselves, for any and all purposes, and the specification left others with the research task of finding more. Here, Teva claimed only the use of humanized anti-CGRP antagonist antibodies to treat headache. Reframing the question accordingly: the relevant research assignment was not identifying which candidate molecules antagonize CGRP (that was already established in the prior art) but determining which humanized anti-CGRP antagonist antibodies treat headache. The court found that assignment was already complete in that the specification made clear that all such antibodies work for that purpose. There was no open research problem left for others to solve.
The court distinguished its earlier decision in Idenix Pharmaceuticals LLC v. Gilead Sciences Inc. on the same logic. In Idenix, the evidence did not support a finding that all members of the claimed genus treated hepatitis C, so extensive screening remained necessary. Here, the jury supportably found that all humanized anti-CGRP antagonist antibodies treat headache, eliminating any analogous screening burden.
Practical Takeaways for Patent Drafting
The Teva v. Eli Lilly decision offers a roadmap for drafting and defending method-of-treatment claims after Amgen.
In this decision, the court held that method-of-treatment claims that use a known biological agent for a specific therapeutic purpose carry a lighter disclosure burden than claims to the agent itself, but only if the agent is genuinely well known in the art. The court was careful not to overstate this, and did not create a blanket rule that method claims are easier to enable than composition claims. But where the prior art supplies the genus and the specification answers the therapeutic question directly, this decision provides that can be sufficient. It is noted that the specification of the Teva patents affirmatively stated that all humanized anti-CGRP antagonist antibodies achieve the claimed therapeutic outcome. For practitioners drafting method-of-treatment claims in the biotech space, this decision would support making that showing expressly and to back it up with data where available.
This decision underscores the importance of being deliberate about what is conceded in PTAB proceedings. In particular, Lilly’s admissions about the state of the art, made to win an IPR on obviousness grounds, became the evidentiary foundation for the jury’s verdict against it in district court. In other words, arguments asserted before the PTAB, or in any parallel proceeding, have a way of traveling, and those managing a multi-front dispute should think carefully about how each position plays across every forum where it might surface.
The harder question the court did not fully answer is what happens when a claimed method genus is large and therapeutic utility is not universal. That is, where the record cannot support a finding that every member of the class works. Future defendants will press that gap. It may be that district courts will read this decision narrowly and will continue to scrutinize whether the “all members work” finding is genuinely supported by the evidence, rather than treating it as a formality. For patent owners, in view of this decision, it can be important to build the record at the time of filing. Expert declarations, prior art citations, and explicit specification statements linking all genus members to the claimed therapeutic effect may be the difference between a valid claim and an invalid one.
For more information, please contact Fitch Even partner David M. Kogan, author of this alert.
Fitch Even IP Alert®
David M. Kogan
David M. Kogan has strong capabilities in both patent prosecution and litigation, combined with substantial experience in prosecuting reexamination proceedings for patents that are being litigated. His skills and experience integrate with a wider focus on counseling large, midsized, and start-up companies on worldwide IP portfolio management, strategic development, and licensing.