Introduction
The Federal Circuit’s decision in Wyeth LLC v. AstraZeneca Pharmaceuticals LP, Case No. 2024-2325 (Fed. Cir. July 9, 2026), affirms that a patent claiming daily administration of a therapeutically effective drug dosage to a patient must teach more than laboratory activity and broad projected dosage ranges. The specification must provide sufficient guidance for a skilled artisan to identify a workable patient dosage without undue experimentation.
The unanimous panel affirmed judgment as a matter of law invalidating claims directed to treating drug-resistant non-small cell lung cancer with irreversible epidermal growth factor receptor inhibitors. Although a jury had upheld the patents and awarded Wyeth $107.5 million, the district court concluded that the evidence could not support the verdict because the patents did not explain how to translate in vitro activity into a therapeutically effective daily dose for human patients.
The decision is significant because it addresses the boundary between patent enablement and the regulatory process. A method-of-treatment patent ordinarily need not establish the level of safety and efficacy required for Food and Drug Administration approval. But where the claims themselves require administration of a therapeutic dosage to a patient, the patent must enable that claimed use. Broad statements that clinicians can determine an appropriate dose will not substitute for an enabling disclosure.
The Claimed Cancer-Treatment Methods
Wyeth’s patents concerned methods of treating non-small cell lung cancer that had become resistant to gefitinib or erlotinib. Those drugs inhibit EGFR signaling, which can promote cancer-cell growth, but some tumors eventually develop resistance.
The asserted claims addressed that problem by administering an “irreversible” EGFR inhibitor. Unlike reversible inhibitors, these compounds form covalent bonds with specified cysteine residues in EGFR or a related receptor.
The claims did not merely cover compounds capable of inhibiting EGFR. They required daily administration to a patient of a “unit dosage” of an irreversible inhibitor. The specification expressly defined a unit dosage as a predetermined quantity of active material calculated to produce the desired therapeutic effect.
That language became decisive. As Charles Gideon Korrell observes, the patents claimed a practical treatment regimen, not merely a promising biochemical mechanism.
The specification identified three compounds as examples and described in vitro experiments showing increased killing of lung-cancer cells. It also supplied broad dosage ranges, including a general range of approximately 0.5 to 1,000 milligrams per kilogram of body weight and a projected total daily dosage of approximately 1 to 1,000 milligrams.
But the patents did not provide a working example of any claimed compound administered to a human patient at a dosage that produced the claimed therapeutic effect. Instead, the specification stated that the effective dose would depend on the patient, the compound, the method of administration, the severity of the disease, and the judgment of the practitioner.
The Jury Verdict and Post-Trial Invalidity Ruling
Wyeth sued AstraZeneca, alleging that AstraZeneca induced infringement through the marketing and sale of Tagrisso, or osimertinib, an irreversible EGFR inhibitor.
After a five-day trial, the jury found the asserted claims valid and infringed and awarded Wyeth $107.5 million. AstraZeneca renewed its motion for judgment as a matter of law, arguing that the claims were not enabled under 35 U.S.C. § 112(a).
The district court granted the motion. It emphasized that the patents disclosed no working patient-dosage examples and that unrebutted evidence showed that some disclosed dosage levels would exceed the maximum tolerated human dose. The court concluded that the patents supplied only a starting point for further research and left practitioners to determine through iterative testing whether any particular compound could produce a therapeutic effect at a dosage suitable for daily administration.
Patient Administration Was a Meaningful Claim Limitation
On appeal, Wyeth argued that the district court had effectively changed its claim construction after trial by importing unclaimed requirements of clinical safety and efficacy.
The Federal Circuit rejected that characterization. The court agreed that the claims did not require FDA approval or compliance with formal clinical endpoints. But it also refused to treat “administering daily to the patient” as surplusage.
Under Phillips v. AWH Corp., claim language must be interpreted in the context of the claims, specification, and prosecution history. The preferred construction gives meaning to each claim term and aligns with the patent’s description of the invention.
That point proved decisive on appeal. Wyeth’s argument effectively sought to minimize the significance of the already-established construction of “unit dosage” by treating patient administration as little more than a statement of intended use. The Federal Circuit rejected that approach because the specification expressly defined “unit dosage” as an amount calculated to produce the desired therapeutic effect. Once that definition became part of the claim construction, the therapeutic patient-dosing requirement could not be disregarded during the enablement analysis. The court therefore evaluated whether the specification actually taught skilled artisans how to achieve the claimed clinical result.
The court therefore drew a careful distinction. It did not hold that the patents had to prove an FDA-acceptable safety profile. It held that a dosage incapable of practical daily administration to a patient could not satisfy the claims as construed.
According to Charles Gideon Korrell, this distinction is likely to recur in pharmaceutical cases. Courts may decline to import general safety and efficacy requirements into treatment claims while still requiring patentees to enable the clinical limitations they chose to claim.
The Specification Did Not Enable the Full Scope of the Claims
The Federal Circuit applied the Supreme Court’s full-scope enablement principle from Amgen Inc. v. Sanofi: the more a patentee claims, the more the specification must enable.
The Federal Circuit grounded its analysis in the Supreme Court’s recent decision in Amgen Inc. v. Sanofi, which reaffirmed that a patent must enable skilled artisans to make and use the full scope of the claimed invention without undue experimentation. That principle was particularly significant here because Wyeth did not claim only the three compounds described in the specification. Instead, the claims broadly encompassed irreversible EGFR inhibitors capable of satisfying the claimed functional limitations while requiring daily administration of a therapeutically effective dosage to a patient. The broader the scope of such functional claims, the greater the specification’s obligation to teach skilled artisans how to identify operative embodiments throughout that scope.
The court relied heavily on Baxalta Inc. v. Genentech, Inc., which requires the specification to teach skilled artisans to make and use the full scope of the claimed invention without undue experimentation. A specification may permit some routine testing, but it cannot leave the central work of identifying operative embodiments to the reader.
Read together, Amgen, Baxalta, and Idenix establish a common doctrinal framework. A patentee may claim a broad functional genus only if the specification provides meaningful guidance for identifying substantially all embodiments falling within that genus. Merely identifying a desired biological result and leaving skilled artisans to discover which compounds, dosages, or treatment regimens actually achieve that result is precisely the type of undue experimentation that the enablement requirement prohibits.
The patents failed that standard for several related reasons.
First, they disclosed only in vitro activity. The court cited Amgen, Inc. v. Chugai Pharmaceutical Co. for the proposition that in vitro data may not establish an in vivo limitation. The relevant question was not whether the compounds killed cancer cells in a laboratory. It was whether a skilled artisan could determine a daily dosage that produced a therapeutic effect in a patient.
Second, the broad dosage ranges lacked practical guidance. The patents did not explain how the ranges were derived, how to select a dosage for a particular inhibitor, or how to account for differences among compounds and patients.
Third, the record showed that disclosed therapeutic concentrations for at least two of the three exemplary compounds exceeded tolerated human exposures. That evidence demonstrated that the broad ranges contained substantial inoperative territory.
Fourth, the specification expressly acknowledged that dosing depended on numerous individualized variables.
Indeed, the court viewed those statements as evidence that the inventors themselves had not solved the central problem presented by the claims. Rather than describing a methodology for identifying therapeutically effective daily dosages, the patents deferred that determination to the judgment of future practitioners. That approach may acknowledge the realities of clinical medicine, but it does not satisfy § 112(a) when the claims themselves require administration of a therapeutically effective dosage to a patient.
The Limits of Reliance on Skilled-Artisan Knowledge
Wyeth also relied on the knowledge of skilled artisans, arguing that experienced clinicians and drug developers would understand how to determine appropriate dosing.
The court acknowledged that a specification need not repeat everything already known in the art. But it relied on Idenix Pharmaceuticals LLC v. Gilead Sciences Inc. and Genentech, Inc. v. Novo Nordisk A/S to explain that skilled-artisan knowledge supplements an enabling disclosure; it does not replace one.
The missing information here concerned the claimed therapeutic use itself. The specification did not provide a method for translating laboratory potency into patient dosing, identifying tolerable therapeutic concentrations, or screening the claimed class for compounds that could satisfy the daily-administration requirement.
Charles Gideon Korrell notes that this aspect of the decision reinforces an increasingly consistent line of Federal Circuit precedent. Idenix teaches that routine experimentation does not become enabling simply because skilled artisans know how to perform it. Where a patent provides little guidance regarding which members of a broad claimed class are likely to succeed, requiring researchers to screen potentially countless candidates becomes undue experimentation. That principle extends well beyond pharmaceutical dosing and applies equally to biotechnology, chemistry, and other unpredictable arts in which functional claiming often exceeds the accompanying technical disclosure.
[Internal link: Otsuka Pharmaceutical v. Hetero Labs]
Relationship to Prior Treatment-Method Precedent
Wyeth relied on United Therapeutics Corp. v. Liquidia Technologies, Inc., where the Federal Circuit declined to import safety and efficacy limitations into a method-of-treatment claim that did not contain them.
Wyeth argued that United Therapeutics foreclosed the district court’s reasoning because that decision cautioned against importing unclaimed regulatory requirements into method-of-treatment claims. The Federal Circuit agreed with that general principle but explained that it did not control this case. Unlike United Therapeutics, the asserted claims here expressly required daily administration of a therapeutic unit dosage to a patient, making patient-level dosing an actual claim limitation rather than an external regulatory consideration.
The court did not retreat from United Therapeutics. Instead, it distinguished between regulatory requirements and express claim limitations. A patent need not necessarily establish that a treatment will meet FDA standards. But where the claim requires a therapeutic unit dosage administered daily to a patient, the patent must teach how to achieve that result.
The opinion also resembles Wyeth & Cordis Corp. v. Abbott Laboratories, where the Federal Circuit invalidated broad claims in an unpredictable field because the specification provided only a starting point and required extensive screening to identify working embodiments.
The common principle is not that pharmaceutical patents always require human clinical data. The court expressly recognized that treatment claims may be enabled without such data. The problem was the combination of broad functional claiming, absent patient-dosage examples, unreliable projected ranges, and substantial experimentation needed to locate workable embodiments.
Practical Implications
The decision has immediate implications for pharmaceutical patent drafting and litigation.
The decision also cautions against relying on broad prophetic dosage ranges unsupported by corresponding technical guidance. While patent applicants frequently include estimated dosage ranges before clinical testing is complete, those ranges may later undermine enablement if the specification does not explain how they were derived or how skilled artisans can determine which dosages are therapeutically effective for the claimed compounds. In some circumstances, an overly broad range may become affirmative evidence that the patent encompasses substantial inoperative embodiments.
Applicants claiming treatment regimens should ensure that the specification explains how disclosed preclinical results support the claimed route, frequency, and dosage. General dosage ranges may carry little weight when they are not tied to particular compounds, pharmacologic characteristics, or therapeutic outcomes.
Prosecution amendments also deserve careful attention. Narrowing a claim to overcome prior art may introduce a limitation that the original specification does not adequately support. A daily patient-dosage limitation may distinguish the prior art, but it also changes what the patent must enable.
The opinion likewise underscores the importance of developing an evidentiary record connecting preclinical data to the claimed therapeutic regimen. Patent prosecutors should consider including technical explanations describing how laboratory potency, pharmacokinetic data, animal studies, or other evidence support the claimed dosage and administration schedule. Such disclosures may later prove critical in demonstrating that the inventors possessed more than a research plan and instead taught skilled artisans how to practice the full scope of the claimed invention.
For litigators, the decision illustrates how enablement can defeat a jury verdict at the judgment-as-a-matter-of-law stage. Evidence from the patentee’s own inventors and experts concerning toxicity, maximum tolerated doses, and the gap between in vitro and human exposure may establish that no reasonable jury could find the claims enabled.
The case also reinforces the importance of claim construction in enablement disputes. Words such as “patient,” “therapeutic,” “daily,” and “dosage” may appear ordinary, but together they can impose substantial technical requirements.
The procedural posture also deserves attention. The Federal Circuit affirmed judgment as a matter of law notwithstanding a jury verdict upholding validity, concluding that the enablement evidence permitted only one reasonable conclusion. The decision therefore illustrates that, although enablement is often fact-intensive, a sufficiently one-sided evidentiary record may justify setting aside a jury’s verdict after trial.
Charles Gideon Korrell believes the decision’s broader lesson is straightforward: a patentee cannot use narrow, clinically meaningful language to obtain allowance and then treat that same language as insignificant when enablement is challenged.
Conclusion
Wyeth v. AstraZeneca does not impose a categorical clinical-testing requirement on method-of-treatment patents. It instead applies a familiar exchange at the center of patent law. Claims may reach beyond the disclosed examples, but the specification must provide enough teaching to practice the claimed invention without undue experimentation.
Wyeth claimed daily administration of a therapeutically effective dosage to a patient across a broad class of irreversible EGFR inhibitors. Its patents disclosed laboratory activity, projected ranges, and generalized statements about practitioner judgment. They did not teach how to identify a workable patient dosage. The Federal Circuit therefore concluded that the claims exceeded the enabling disclosure.
Key Takeaways
- A method-of-treatment claim requiring a therapeutic daily dosage in a patient must enable that patient-level use, even when FDA approval is not claimed.
- In vitro activity does not necessarily enable an in vivo therapeutic limitation.
- Broad dosage ranges are insufficient when the patent does not explain how to select operative dosages for particular compounds.
- Skilled-artisan knowledge may supplement a specification but cannot supply the principal missing teaching needed to practice the claimed invention.
- Claim limitations added to overcome prior art may create corresponding enablement obligations that the original disclosure cannot satisfy.
- Express claim language requiring therapeutic administration or patient dosing expands the scope of what the specification must enable under 35 U.S.C. § 112(a).