Two Federal Circuit decisions, three months apart, analyzed enablement of similarly broad method-of-treatment claims, but reached opposite outcomes. In April, the court reversed a JMOL of invalidity and reinstated a $176.5 million verdict in Teva Pharmaceuticals International GmbH v. Eli Lilly and Company. Three months later, in Wyeth LLC v. AstraZeneca Pharmaceuticals LP, the same court affirmed a JMOL wiping out a $107.5 million verdict on nearly the same legal theory. Both cases involved broad, functionally-drafted method-of-treatment claims tested against the enablement standard set by the Supreme Court in Amgen Inc. v. Sanofi. A side-by-side comparison offers lessons on the current state of enablement law.
Key Takeaways
- In Teva, the claimed genus of anti-CGRP antibodies was well-understood and routine to produce in the art, allowing a single representative species and established protocols to enable the full scope of the claims.
- In Wyeth, the claimed “unit dosage” requirement – a daily dose “calculated to produce the desired therapeutic effect” in a patient – necessarily contemplates a repeatable, practical dosing regimen, not merely a compound’s ability to inhibit EGFR or kill cancer cells in vitro.
- The Federal Circuit in Wyeth emphasized that the claims did not require FDA-level proof of safety or efficacy. Nevertheless, because the claims required daily administration of a therapeutic unit dosage to a patient, the specification still needs to teach how to arrive at such a dosage without undue experimentation.
- Neither decision required clinical safety or efficacy data to satisfy § 112(a). The Wyeth panel was explicit that the claims did not need FDA-level proof of safety; they needed enough disclosure to allow a skilled artisan to reach a workable, non-toxic dosage without undue experimentation.
- Read together, the two decisions indicate that functional claiming survives enablement scrutiny where the underlying art is mature and predictable, but requires the specification to do the translational work itself where in vivo behavior remains genuinely unpredictable.
The facts of each case explain why the Wyeth decision stands in sharp contrast to Teva. While both cases evaluated broad genus method-of-treatment claims, the divergent outcomes highlight how enablement turns on where the primary technical uncertainty resides. In Teva, the claims were directed to using a well-understood class of humanized anti-CGRP antibodies to treat headaches, where standard art-established protocols enabled skilled artisans to practice the full scope from a single working species. Conversely, in Wyeth, the claimed novelty and explicit limitations resided in the human dosing regimen itself, yet the patent left the critical task of translating in vitro activity into non-toxic human administration entirely to the reader’s trial-and-error experimentation.
Analyzing the contrast between Wyeth and Teva v. Eli Lilly reveals clear boundaries for method-of-treatment claim drafting:
In Teva v. Eli Lilly, the Court found that the claimed composition genus (anti-CGRP antibodies) was well-understood in the prior art, allowing a single representative species and background knowledge to enable the method. In Wyeth, even if irreversible EGFR inhibitors were conceptually known, the Court found that the dosing regimen in human patients was unpredictable. Therefore, disclosure of a broad range and reliance on the physician to determine the dose were not enabling.
Wyeth underscores that when a method claim requires daily administration to a human patient, in vitro cell-killing data alone is insufficient unless the patent provides a reliable dose translation mechanism or working human dosing models.
While disclosing some inoperative embodiments is not automatically fatal, disclosing broad ranges where preferred species exceed maximum tolerated human doses forces undue experimentation to identify functional dosages.
In each of these cases, the outcome turned on the key limitation that distinguished the claims from the prior art, i.e., dosage regimen/toxicity in Wyeth, and use of a class of antibodies for a specific indication in Teva. The record evidence was critical in this regard, i.e., undisputed evidence of toxicity in the disclosed dosage ranges in Wyeth versus Lilly’s prior assertions in inter partes review proceedings that anti-CGRP antagonist antibodies were well known and their use was routine in Teva.
| Feature | Wyeth v. AstraZeneca | Teva V. Eli Lilly |
|---|---|---|
| Claim Focus | Specific human dosing regimen (“unit dosage”). | Use of a genus of molecules (anti-CGRP antibodies) for a specific indication. |
| Primary Uncertainty | Translation of in vitro activity to in vivo human administration. | Production and binding mechanics of the molecules. |
| State of the Art | Human dosing for these inhibitors was unpredictable. | Anti-CGRP antibodies were well-known and routine to produce. |
| Disclosure Quality | Broad ranges; preferred embodiments were toxic/inoperative. | Single representative species enabled the full scope via established protocols. |
| Outcome | Patent Invalid. Permanently set aside a $107.5 million jury verdict. | Patent Not Invalid. Reinstated $176.5 million verdict. |
Background on Wyeth
The patents-in-suit, U.S. Patents 10,603,314 (‘314 patent) and 10,596,162 (‘162 patent), claimed methods of treating gefitinib- and erlotinib-resistant (“g/e resistant”) non-small cell lung cancer (NSCLC) using irreversible EGFR inhibitors — compounds that covalently bind cysteine 773 (or cysteine 805 of erb-B2) rather than forming the reversible, non-covalent bonds characteristic of first-generation drugs like gefitinib and erlotinib.[1] Representative claim 1 of the ‘314 patent required “administering daily to the patient” a “unit dosage” of the claimed inhibitor.[2]
The specification defined “unit dosage” as a “predetermined quantity of active material calculated to produce the desired therapeutic effect,” and identified only three exemplary compounds — EKB-569, HKI-357, and HKI-272 — supported solely by in vitro data.[3] It offered a “general” dosage range of 0.5 to 1000 mg/kg and a “projected” daily total of 1 to 1000 mg (preferably 2 to 500 mg), but stated the “[p]recise amounts… depend on the judgment of the practitioner.”[4]
A Delaware jury found the claims not invalid and found that AstraZeneca’s Tagrisso (osimertinib) induced infringement, awarding Wyeth $107.5 million.[5] The district court later overturned the jury’s verdict by granting AstraZeneca’s renewed motion for judgement as a matter of law, holding the claims invalid for lack of enablement. The court reasoned that the patents “provide[d] ‘only a starting point, a direction for future research'” and left skilled artisans to “an iterative, trial-and-error approach.”[6] (quoting ALZA Corp. v. Andrx Pharms., LLC, 603 F.3d 935, 939–41 (Fed. Cir. 2010)).
Claim Construction: The Significance of “Daily” and “To the Patient”
Wyeth’s first argument was that the district court injected FDA-style safety and efficacy requirements into the claims post-verdict. The panel disagreed. It held that the claim terms “administer[ed] daily” and “to the patient” must be given independent meaning under Merck & Co. v. Teva Pharms. USA, Inc., 395 F.3d 1364, 1372 (Fed. Cir. 2005) (holding that claim terms must be construed to avoid rendering any claim language superfluous), and that doing so “draws the claims beyond merely requiring ‘killing cancer cells,’ yet constrains them short of requiring full-blown FDA-type clinical safety standards.”[7] In other words, the claims require “a repeatable dosing regimen” “tailored for practical administration to a human patient, not merely any concentration capable of producing anti-cancer effects in vitro.”[8]
Enablement: In Vitro Data Doesn’t Enable an In Vivo Limitation
Applying Amgen Inc. v. Sanofi, 598 U.S. 594, 610 (2023) (“[t]he more one claims, the more one must enable”[9]), and Baxalta Inc. v. Genentech, Inc., 81 F.4th 1362, 1365 (Fed. Cir. 2023), the panel held the specification “leaves the determination of the claimed ‘unit dosage’ entirely to the knowledge of the skilled artisan,” which is insufficient standing alone.[10] (citing Idenix Pharms. LLC v. Gilead Scis. Inc., 941 F.3d 1149, 1159 (Fed. Cir. 2019)). It was undisputed that the specification contains no working example of a unit dosage “calculated to achieve a therapeutic effect and suitable for daily administration in human patients.”[11]
The record evidence hurt Wyeth further. Trial testimony — including from Wyeth’s own expert and a co-inventor — established that the disclosed dosage ranges for at least two of the three exemplary compounds exceeded the maximum tolerated dose in humans.[12] (citing J.A. 17397–98, J.A. 17549; 17563, J.A. 11804, 11807). Slip op. at 16–17. The court treated that as strong evidence of non-enablement under Atlas Powder Co. v. E.I. du Pont De Nemours & Co., 750 F.2d 1569, 1576–77 (Fed. Cir. 1984), which held that substantial inoperative embodiments can render claims not enabled when practicing the full scope requires undue experimentation. ([I]f the number of inoperative combinations become significant, and in effect forces one of ordinary skill in the art to experiment unduly in order to practice the claimed invention, the claims might indeed be invalid). Id.
Importantly, the panel was careful to frame this as an enablement gap, not a toxicity limitation read into the claims: “the district court did not hold that the claims require proof that every claimed dosage be non-toxic, non-lethal, or satisfy any freestanding FDA clinical safety criteria.”[13] Rather, the toxicity evidence mattered because it showed the specification gives “little guidance as to which disclosed dosages, if any, are suitable for daily administration to a patient across the functionally claimed class of compounds” — which is “precisely the type of undue experimentation that the enablement requirement forbids.” (citing ALZA Corp., 603 F.3d at 941, 943).[14]
The court concluded by emphasizing the distinction between FDA approval and patent enablement: patentees may generally claim a dosing method “without showing actual clinical data,” leaving safety and efficacy determinations to the FDA.[15] “The problem with these patents is that, perhaps because of close prior art, their claims are limited to dosage forms to be administered to patients, yet they disclosed only a broad range of doses some of which were shown to be toxic, and they disclosed no actual dosages for any compound within the scope of the claims.”[16] In short, the patents failed because the specification did not enable skilled artisans to practice the full scope of the claimed therapeutic dosing methods.
Practical Implications
The Federal Circuit’s decision offers several practical lessons for patent applicants prosecuting method-of-treatment claims.
First, claim narrowing can have enablement consequences. Wyeth added the “unit dosage” administered daily limitation to overcome the prior art. The Federal Circuit construed that added claim language in accordance with the specification’s express definition: “physically discrete units suitable as unitary dosage for the subject, each unit containing a predetermined quantity of active material calculated to produce the desired therapeutic effect” – and that same limitation ultimately proved fatal to enablement. Had the claims been directed more broadly to inhibition of EGFR or cancer-cell killing, the existing disclosure may have been sufficient to survive the enablement inquiry. But by claiming administering a daily therapeutic dosing regimen of a unit dosage to patients, however, the specification needed to teach skilled artisans how to practice that limitation without undue experimentation.
Second, relying on the knowledge of a person of ordinary skill cannot substitute for an enabling disclosure. The specification stated that “[t]he skilled artisan is aware of the effective dose for each patient,” but the Federal Circuit viewed that statement as evidence that the patent left the critical dosing determination to the reader. The court reiterated a familiar principle of enablement: while the knowledge of a skilled artisan may supplement a disclosure, it cannot supply the invention’s missing teachings.[17]
Third, patent applicants and challengers should carefully consider the examples in the specification identified as preferred embodiments – forms of the invention the patent presents as particularly desirable. Here, the asserted patents identified only three exemplary irreversible EGFR inhibitors, yet the trial evidence showed that at least two – and potentially all three – could not be administered at any dosage within the disclosed dosage ranges without exceeding maximum tolerated human doses. The preferred embodiments therefore undermined, rather than supported, enablement by showing that the specification did not teach how to practice the claimed invention across its full scope.
Finally, the panel insisted that the decision does not establish a new requirement that method-of-treatment patents include clinical safety or efficacy data. The panel reaffirmed that enablement is distinct from FDA approval and consistent with its prior decision in United Therapeutics Corp. v. Liquidia Techs., Inc., 74 F.4th 1360, 1369 (Fed. Cir. 2023) (holding that treating pulmonary hypertension did not import additional safety or efficacy requirements into the claims because those limitations were not recited). Wyeth instead illustrates the other side of that principle: when claims expressly require a particular dosing regimen, the specification must enable that regimen across the full scope of the claims. Here, evidence of inoperative or toxic embodiments mattered because it showed that the specification did not provide sufficient guidance to identify workable dosages without undue experimentation. The broader lesson extends beyond dosing: claim amendments made to overcome prior art can create new § 112 vulnerabilities, including enablement, written description, and indefiniteness. Wyeth underscores the importance of reassessing § 112 support whenever claims are amended to ensure the amended claims remain supported by the specification.
Takeaways for Practitioners
For patent prosecutors in the pharmaceutical and biotechnology sectors, Wyeth v. AstraZeneca establishes that in vitro assays and broad mechanistic disclosures cannot bridge a clinical gap when a method claim includes an in vivo administration limitation. Whenever a claim recites explicit dosing, whether through a numerical range or functional terminology like a “unit dosage” tied to a therapeutic effect, the specification must provide concrete dosing guidance, or explicit translational modeling, to enable a safe and effective human dosing regimen. While disclosing or claiming broad dosing ranges is predicted to maximize nominal claim scope, doing so introduces severe § 112 vulnerability if the disclosures and in vitro data fail to share nexus with the ultimately determined in vivo results in humans.
Conversely, Teva v. Eli Lilly illustrates that when claiming a method of using a genus of molecules or biologics whose production and binding mechanics are already well-known and routine in the art, patent applicants do not need to provide data for every member of the claimed genus. During litigation, objective statements by others and admissions by challengers acknowledging that the claimed class of compounds was well-known and routine to produce and use can be crucial evidence supporting enablement.
Disclaimer: The information contained in this posting does not, and is not intended to, constitute legal advice or express any opinion to be relied upon legally, for investment purposes or otherwise. If you would like to obtain legal advice relating to the subject matter addressed in this posting, please consult with us or your attorney. The information in this post is also based upon publicly available information, presents opinions, and does not represent in any way whatsoever the opinions or official positions of the entities or individuals referenced herein.
[1] Slip op. at 2–3.
[2] Slip op. at 3.
[3] Slip op. at 3–4, 14.
[4] Slip op. at 4.
[5] Slip op. at 5.
[6] Slip op. at 6–7
[7] Slip op. at 10–11.
[8] Id.
[9] Slip op. at 13
[10] Slip op. at 14
[11] Slip op. at 14.
[12] Slip op. at 15–16 (evidence showed that, for those two compounds, all therapeutically effective dosage levels across the disclosed range exceeded the maximally tolerated dose in humans).
[13] Slip op. at 18.
[14] Id.
[15] Slip op. at 19 (citing United Therapeutics Corp. v. Liquidia Techs., Inc., 74 F.4th 1360, 1369 (Fed. Cir. 2023)).
[16] Slip op. at 19.
[17] Slip op. at 14.