TEVA PHARMACEUTICALS INTERNATIONAL GMBH v. ELI LILLY AND COMPANY

Authored by: Jeremy J. Gustrowsky

The Federal Circuit has declined to take a second look at one of its more consequential enablement decisions of the year, denying Eli Lilly and Company’s petitions for panel rehearing and rehearing en banc in its long-running dispute with Teva Pharmaceuticals over migraine antibody patents. The court invited a response from Teva, granted leave to a sizable group of amici to weigh in, and polled the active judges on whether to rehear the appeal. The poll failed. Judge Newman did not participate, and District Judge Andrews, sitting by designation, took part only in the decision on panel rehearing. Judge Dyk dissented from the denial.

The patents at issue are U.S. Patent No. 8,586,045 and U.S. Patent No. 9,884,907. Claim 30 of the ‘045 patent, treated as representative, covers a “method for reducing incidence of or treating headache in a human” by administering an effective amount of an anti-CGRP antagonist antibody that is a humanized monoclonal antibody. The panel had concluded that these method claims were enabled, even while assuming that the time and expense needed to make and humanize the full universe of anti-CGRP antagonist antibodies would amount to undue experimentation. The panel’s reasoning was that the claims do not cover the antibodies themselves, only their use for the narrower purpose of treating headache, and that the specification taught that all such antibodies work for that purpose.

Judge Dyk’s dissent takes direct aim at that distinction. As he reads the claim, it contains two functional requirements: the antibody must act as an anti-CGRP antagonist, and administering it must reduce or treat headache. In his view, the second requirement adds nothing meaningful, because the panel itself acknowledged that every antibody in the genus works to treat headache. All of the difficulty, and all of the experimentation, lies in figuring out which antibodies have the necessary antagonist properties. The record showed that identifying and humanizing such antibodies took months of benchtop and animal testing and tens of thousands of dollars per antibody. A claim that demands undue experimentation, he wrote, is simply not enabled, citing the court’s longstanding decisions in In re Wands and Wyeth & Cordis Corp. v. Abbott Laboratories.

The dissent’s larger concern is the relationship between this decision and the Supreme Court’s 2023 ruling in Amgen Inc. v. Sanofi, which held that a patent must enable the full scope of the invention as defined by its claims and that a patentee cannot satisfy that obligation by handing skilled artisans a screening protocol. The panel acknowledged that a claim to the antibody genus itself would have resembled Amgen. Judge Dyk responded that where the compounds have no realistic use other than the claimed treatment, the practical reach of the method claim is the same as a compound claim, and the recited antibody remains a claim limitation that must be enabled either way. Under the panel’s approach, he suggested, Amgen could have sidestepped the Supreme Court’s ruling by claiming a cholesterol-lowering method using its PCSK9-blocking antibodies rather than the antibodies themselves.

The amicus briefs supplied much of the ammunition for that argument. Johnson & Johnson and Nagra USA told the court the decision creates an “end-run around Amgen” and has already generated practitioner guidance on using creative claim drafting to avoid its disclosure requirements. Merck Sharp & Dohme and Ipsen Biopharmaceuticals warned that the ruling risks foreclosing entire fields of independent scientific development. Amgen and Sanofi argued that the differing outcomes between the two cases will create uncertainty for the Patent and Trademark Office, for future courts, and for industry. Judge Dyk pointed out the unusual posture of these filings, noting that pharmaceutical amici are not typically the ones asking for stricter patentability standards.

For now, the panel decision stands as controlling law. Applicants and litigators in the biologics space should expect increased attention to method-of-treatment claims that recite functionally defined antibodies, and increased scrutiny of whether the practical scope of such a claim really is narrower than a claim to the compound. The dissent flags a genuine tension that the court did not resolve, and it is the kind of tension that tends to resurface, whether in a future panel decision, a petition to the Supreme Court, or examination practice at the Patent Office.