WYETH LLC v. ASTRAZENECA PHARMACEUTICALS LP
Authored by: Jeremy J. Gustrowsky
The Federal Circuit has affirmed a Delaware district court’s decision to overturn a jury verdict and invalidate two Wyeth cancer treatment patents for lack of enablement, erasing a $107.5 million damages award against AstraZeneca. The dispute centered on U.S. Patents 10,603,314 and 10,596,162, which claim methods of treating gefitinib and erlotinib resistant non-small cell lung cancer (NSCLC) using irreversible EGFR inhibitors that covalently bind to specific amino acid residues.
Wyeth had sued AstraZeneca in 2021, alleging that AstraZeneca’s blockbuster lung cancer drug Tagrisso (osimertinib) induced infringement of the asserted patents. After a five-day trial, the jury sided with Wyeth on both infringement and validity. But the district court granted judgment as a matter of law post-verdict, concluding that no reasonable jury could have found the claims enabled. The claims required daily administration of a “unit dosage” of an irreversible EGFR inhibitor calculated to produce a therapeutic effect in a patient.
On appeal, Wyeth argued that the district court improperly rewrote its claim construction post-verdict by importing clinical safety and efficacy requirements into the term “unit dosage.” The Federal Circuit rejected that characterization. While acknowledging that the claims do not require FDA-type clinical safety or efficacy standards, the court explained that the claim language demanded meaningful operative effect. The requirement that a “unit dosage” be “administered daily” to a “patient” contemplates a repeatable dosing regimen actually capable of producing a therapeutic effect in humans, not merely killing cancer cells in a test tube.
Turning to enablement, the court found the specification fatally deficient. The patents disclosed only three example compounds (EKB-569, HKI-357, and HKI-272) and described only in vitro experimentation, with no working examples of unit dosages administered to patients. Although the specification recited broad dosage ranges of “about 0.5 to about 1000 mg/kg” and “about 1 to 1000 mg” daily, these were described merely as “general” and “projected” ranges, with no guidance on how to derive them or select among them for a given compound.
Perhaps most damaging to Wyeth’s position, its own experts and inventors admitted at trial that the specification’s disclosed dosage ranges far exceeded the maximum tolerated dose in humans for at least two of the three exemplified compounds. Co-inventor Dr. Haber testified that drug concentrations in the patents were “five times higher” than what could be administered to patients. This evidence showed that the compounds identified as preferred embodiments were largely inoperative for the claimed daily patient administration, forcing skilled artisans to perform undue experimentation to identify workable dosages.
The Federal Circuit acknowledged the accepted practice of claiming methods of treatment without full clinical data, leaving safety and efficacy to the FDA. But the panel drew a critical distinction: these particular claims were limited to dosage forms administered to patients, yet the specification disclosed only a broad range including toxic doses and provided no actual working dosages for any claimed compound. The specification itself emphasized that determining precise amounts depended on multiple variables and the “judgment of the practitioner,” essentially conceding that the invention required extensive experimentation to practice.
The decision reinforces the principle from *Amgen v. Sanofi* that “the more one claims, the more one must enable,” and serves as a cautionary tale for pharmaceutical patentees relying primarily on in vitro data to support claims requiring in vivo administration. Because the court affirmed invalidity, it did not reach Wyeth’s separate appeal regarding pre-issuance damages under 35 U.S.C. § 154(d).