District Court: ANDA Label including Non-Infringing Uses is Not Sufficient to Induce Infringement | Akin Gump Strauss Hauer & Feld LLP
In a Hatch-Waxman case, the District Court for the District of New Jersey recently found that a generic label that included an allegedly infringing permissive use did not induce infringement where the label cautioned against that use and provided non-infringing alternatives. The district court additionally found no direct infringement where the patentee failed to present evidence that the claimed methods had ever been practiced or that infringement is likely to occur in the future.
The Court held a bench trial on the sole issue of induced infringement. Both patents-in-suit relate to methods of treating Cushing’s syndrome by coadministering mifepristone and a strong CY3PA inhibitor. The relevant claims are directed to methods of administering specific doses of the two molecules, after titrating down from a higher dose of mifepristone.
In post-trial briefing the parties disputed whether the patentee proved either direct infringement or the specific intent necessary to establish that the defendant induced infringement.
Regarding direct infringement, the district court found the lack of evidence demonstrating that anyone had ever practiced the claimed methods—including during the 10-year span since the branded drug was approved—was persuasive evidence that supported a finding of no direct infringement. The district court also considered whether infringement was likely to occur in the future. First, the district court found that direct infringement would not be likely to occur even if a physician consulted the generic label because the label cautioned against co-administration of mifepristone. Second, the dosage and sequence claim requirements were not likely to be met because, even when following the label, a physician would not infringe if the mifepristone dosage was below the claimed amount or if the strong CYP3A inhibitor was given to a patient who is already taking a lower dose of mifepristone. Third, the district court credited expert testimony indicating that doctors would be reluctant to co-administer mifepristone with a CY3PA inhibitor due to difficulties with dosing and the risk of serious side effects. Finally, the availability of at least one non-infringing alternative to mifepristone that does not pose the same risks made it unlikely that physicians would directly infringe the claims in the future. Based on all of the above findings, the court concluded that the patentee did not meet its burden to prove that direct infringement was likely.
Although the court held that direct infringement was not proven, it completed its analysis by examining whether the defendant specifically induced infringement by providing the proposed generic label. The court found that because the generic label only provided general instructions on how to administer the combination if a physician deems it medically necessary without specific guidance leading to the claimed dosages, it could not be said to encourage direct infringement. The court was thus unwilling to infer intent from mere guidance, particularly because the label did not outline benefits of the treatment, indicate circumstances where the treatment would be appropriate, or guarantee use of the claimed doses, and warned against co-administration of the drugs. These facts, coupled with the existence of non-infringing alternatives, led the court to conclude that the patentee failed to prove the defendant had the specific intent to induce infringement. The court also rejected the patentee’s argument that a failure to seek an alternative label supports an inference of intent, noting that “requiring an ANDA applicant to modify its product label to avoid direct infringement” would turn the test for inducement “on its head.”
Practice Tip: Where patent claims are directed to methods or dosages that are permissive according to a label, evidence of actual use may be particularly important even for claims of indirect infringement. Likewise, evidence reflecting the existence and prevalence of non-infringing alternatives can weigh heavily in determining whether a party possesses the specific intent to induce infringement.
