A Potential Poison Pill for Dosage Claims

Footnotes for this article are available at the end of this page.

On August 14, 2024, the US District Court for the District of Delaware entered final judgment in Wyeth¹—a patent infringement case relating to pharmaceuticals for treating non-small cell lung cancer (“NSCLC”)—vacating a jury verdict that awarded $107.5 mm of damages in favor of Plaintiff. In the final judgment, the court granted Defendants’ motion for judgment as a matter of law (“JMOL”), holding the asserted claims of the patents-in-suit—U.S. Patent No. 10,603,314 (the “’314 Patent”) and U.S. Patent No. 10,596,162 (the “’162 Patent”)—invalid under 35 U.S.C. § 112 for lack of enablement and lack of written description because the claims encompassed compounds that may need to be administered to patients at toxic dosages to achieve therapeutic effects and the patents-in-suit provided no guidance on which compounds could be administered safely to patients nor how one of ordinary skill in art (“POSA”) could reliably screen between compounds with therapeutic effects at toxic versus non-toxic dosages. On September 12, 2024, Wyeth appealed final judgment to the Federal Circuit. If the final judgment is affirmed by the Federal Circuit, the rationale employed in Wyeth could lead to a significant paradigm shift for the drafting and litigation of applications covering formulations administered to subjects (e.g., therapeutics and nutritional supplements) going forward.

Case Summary

In September 2021, Wyeth sued AstraZeneca, alleging that AstraZeneca’s drug Tagrisso® induced infringement of claims 1, 3, and 9 of the ’314 Patent and claim 1 of the ’162 Patent (collectively the “Asserted Claims”).²

Claim 1 of the ’162 Patent is:

1. 1. A method of treating gefitinib and/or erlotinib resistant non-small cell lung cancer [“g/e resistant NSCLC”] having a T790M mutation in SEQ ID NO: 1 in a patient, comprising administering daily to the patient having gefitinib and/or erlotinib resistant non-small cell lung cancer having a T790M mutation in SEQ ID NO: 1 a pharmaceutical composition comprising a unit dosage of 2-500 mg of an irreversible EGFR inhibitor that covalently binds to cysteine 773 of the catalytic domain within the SEQ ID NO: 1 having a T790M mutation; wherein the irreversible EGFR inhibitor is not CL-387,785.

Early in litigation, the parties disputed numerous terms of the Asserted Claims. In relevant part, the court construed: (1) the term “g/e resistant NSCLC” consistent with its plain and ordinary meaning; (2) the term “irreversible EGFR inhibitor…” as “a compound that irreversibly inhibits EGFR and covalently binds to cysteine 773…”; and (3) the term “unit dosage” consistent with the specification’s definition (i.e., “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…”).

AstraZeneca moved for summary judgment, contesting, in relevant part, that the Asserted Claims were invalid under 35 U.S.C. § 112 for lack of enablement and written description.³ AstraZeneca argued that the patents-in-suit did not enable a POSA to practice the Asserted Claims: (1) with the full scope of irreversible EGFR inhibitors claimed (including larger compounds⁴ and compounds with different reactive agents outside of those disclosed in the specification⁵) because there was insufficient disclosure in the specification on how to make and use such compounds; and (2) to treat the full scope of g/e resistant NSCLCs claimed because a POSA would consider particular NSCLC genotypes as g/e resistant and those genotypes would not be treated with irreversible EGFR inhibitors. Wyeth countered with arguments that a POSA: (1) would recognize that larger molecules would not be permissive embodiments; (2) could routinely make and test compounds with different reactive agents; and (3) would not consider the particular genotypes suggested by AstraZeneca as a g/e resistant NSCLC. The court denied AstraZeneca’s motion, holding there were genuine factual disputes.

The case proceeded to a jury trial, which resulted in a unanimous verdict that: (1) AstraZeneca induced infringement of the Asserted Claims; (2) AstraZeneca did not willfully infringe the Asserted Claims; (3) the Asserted Claims were not invalid under 35 U.S.C. §§ 102, 103, 112; and (4) awarded $107.5 million in damages to Wyeth.

Post-trial, AstraZeneca moved for Renewed JMOL on no induced infringement, invalidity, and no damages. AstraZeneca reasserted its lack of enablement/written description arguments from summary judgment, but added an additional, eventually dispositive, argument that the Asserted Claims were not enabled (or lacked written description⁶) for treatment with a unit dosage.⁷ AstraZeneca argued that the patent disclosure failed to inform POSA, without an undue amount of experimentation, the unit dosage for each of the irreversible EGFR inhibitors encompassed by the Asserted Claims because: (1) the process of identifying a unit dosage for a single compound is generally burdensome on its own; and (2) despite the patents-in-suit disclosing permissible ranges for unit dosages, these ranges would not reduce the burden on a POSA because they span a broad range (1-1000 mg, preferably 2-500 mg). Purportedly, trial evidence (testimony and a study by an inventor on the patents-in-suit) demonstrated that irreversible EGFR inhibitors expressly listed in the specification were toxic within the disclosed ranges. Wyeth countered by arguing that a POSA would understand that those ranges achieved the therapeutic effect of killing cancer cells and AstraZeneca’s arguments regarding toxicity, safety, and efficacy were attempts to import additional limitation into the claims, which the court had rejected at claim construction.

The court noted that “the claims require only that the ‘unit dosage’ ‘produces the desired therapeutic effect’ in a patient.”⁸ But the court rejected Wyeth’s argument that the patents are enabled even though the method of treatment kills cancer cells because the Asserted Claims do not recite “a method for killing cancer cells,” nor “a method for treating g/e resistant NSCLC” — rather, the claims recite “a method for treating g/e resistant NSCLC in a patient in need thereof, comprising administering daily to the patient . . . a unit dosage . . . ” and therefore must enable a unit dosage that is both therapeutically effective and can be administered to a patient daily.⁹ The court dismissed AstraZeneca’s argument that the breadth of the permissible unit dosage ranges would require undue experimentation, but nonetheless concluded that AstraZeneca presented clear and convincing evidence that no reasonable jury could conclude that the patents-in-suit enabled a POSA to administer a unit dosage of any irreversible EGFR inhibitor encompassed by the Asserted Claims to a patient without undue experimentation because:

1. 1. The specification did not disclose any working examples of unit dosages administered to patients;¹⁰ and
   2. AstraZeneca presented unrebutted evidence that some dosages of compounds within the permissible unit dosage ranges disclosed in the patents-in-suit would be toxic to patients, and those dosages may be required to achieve a therapeutic effect.

The court concluded that: “[i]n scenarios where there is no non-toxic therapeutic range for a given compound, a POSA would not be able to practice the claimed method of treatment in the patient.”¹¹ In this case, the patents-in-suit did not teach which unit dosages of compounds encompassed by the Asserted Claims could be administered to a patient daily and which could not, nor did the patents-in-suit provide guidance that would aid a POSA in reliably screening between compounds with therapeutic effects at toxic versus non-toxic dosages. Consequently, a POSA would have to undertake experimentation, unassisted by the patents-in-suit, which renders the claims insufficient to meet the enablement requirement.¹²

Discussion

On its face, the Wyeth decision appears to contradict the Federal Circuit’s precedent in United Therapeutics,¹³ despite significant factual similarities. Similar to Wyeth, in United Therapeutics: (1) the relevant claim recited a similar “unit dosage” term (“a therapeutically effective single event dose . . . compris[ing] from 15 micrograms to 90 micrograms of treprostinil . . . ”);¹⁴ (2) this term was similarly construed to be tied to a therapeutic effect;¹⁵ and (3) there were concerns over safety of the dosages of the claimed compounds for the claimed conditions treated.¹⁶ In United Therapeutics, however, the Federal Circuit reached the opposite conclusion as Wyeth — holding the Asserted Claims adequately enabled and not lacking written description because, in relevant part:

[B]ecause safety and efficacy are not recited in the claims, we need not deal with Liquidia’s arguments [relating to safety and efficacy]. Disease-specific treatment requirements are matters for the FDA and medical practitioners. They are best suited to make these determinations because practitioners are informed by the findings of the regulatory agency to avoid treatment of patients who will not properly respond. And every claim to a method of treatment of an ailment has refinements.¹⁷ (citations omitted).

Indeed, this long-standing precedent has been viewed as placing safety concerns as the subject matter left to the Food and Drug Administration, whereas patents for treatment methods merely need to achieve the claimed outcome.

Conclusion

It is unclear whether the Federal Circuit will affirm the district court’s decision in Wyeth in light of its holding in United Therapeutics. While Wyeth proceeds through the appellate process, patent practitioners should mitigate risk of future patents being invalidated by taking the following precautions:

• Avoid drafting claims where the therapeutic agent is defined only by its function. While the claims in Wyeth and United Therapeutics share similarities, the Asserted Claims in Wyeth recited the therapeutic agent by its function (i.e., “an irreversible EGFR inhibitor”), whereas the claims in United Therapeutics recited the therapeutic agent by name (i.e., “treprostinil or a pharmaceutically acceptable salt thereof”) — a distinction that may have contributed to the different outcomes in the two cases.
• While it is generally recommended to draft applications to recite broad ranges of permissible dosages, if the safety profile of the claimed compounds are not known over the entire range disclosed in the specification at the time of filing, consider drafting claims to expressly recite numerical dosages. Doing so can help mitigate the risk of having a patent invalidated as a consequence of the specification being read into the claims during litigation, especially in cases where an upper limit on safety of the claimed compound is discovered after the filing date of the application.
• Prior to filing an application, consider advising clients to undertake pre-clinical studies to evaluate upper bounds for safety/toxicity of novel compounds that they are seeking to patent. While these studies may not provide the exact value for the upper bounds of safety/toxicity in humans, they can provide a rough estimate. This information can be used by practitioners to draft claims that can help mitigate the risk of invalidation.

For further discussion on these issues or other similar inquiries, please contact AGG Intellectual Property partner Matt Zapadka or AGG patent agent John Schneible.

 

[1] Wyeth LLC v. AstraZeneca Pharmaceuticals LP and AstraZeneca AB, Case No. 1:21-cv-01338-MFK (D. Del. Aug. 14, 2024).

[2] For purposes of this analysis, only claim 1 of the ’162 Patent will be discussed.

[3] AstraZeneca’s argument on lack of written description largely mirrored their lack of enablement argument. The court held that there were genuine issues of material fact on lack of written description for the same reasons as lack of enablement. For brevity, the lack of written description argument will not be discussed.

[4] The specification stated that the irreversible EGFR inhibitor may also be a larger compound, such as proteins, oligonucleotides, ribozymes, DNAzymes, glycoproteins, siRNAs, lipoproteins, and/or aptamers.

[5] The examples in the specification were exclusively drawn to Michael acceptors. AstraZeneca argued that epoxides and non-Michael acceptors would be permissive reactive agents.

[6] AstraZeneca’s argument on lack of written description largely mirrored their lack of enablement argument. For brevity, the lack of written description argument will not be discussed.

[7] The court declined to grant AstraZeneca’s JMOL on the lack of enablement/written description based on the arguments that were reasserted from summary judgment. As such, the subsequent will focus solely on the “unit dosage” argument.

[8] Final Judgment (D.I. 516) at 30.

[9] See id. at 30. (emphasis added).

[10] The court noted that while this is not dispositive, it is nonetheless a relevant factor in the enablement inquiry. (citing In re Wands 858 F.2d 731 (Fed. Cir. 1998).).

[11] Final Judgment (D.I. 516) at 35.

[12] See id. at 36-37.

[13] United Therapeutics Corporation v. Liquidia Technologies, Inc. 74 F. 4th 1360 (Fed. Cir. 2023).

[14] Id. at 1364.

[15] See id. at 1370.

[16] See ibid.

[17] Id. at 1371.