ANDA filer avoids infringing a method-of-treatment by “carving-out,” still caught by a preparation process patent

The Court of Appeals for the Federal Circuit (CAFC) recently issued their decision in H. Lundbeck A/S v. Lupin Ltd. This decision by the CAFC affirms the judgment of the U.S. District Court for the District of Delaware, finding non-infringement by the Defendants (Lupin, ANDA-filers) of two patents owned by some of the Plaintiffs directed to different medical uses of the drug “vortioxetine”. However, the CAFC also affirmed the district court’s finding of infringement of a broader method of manufacture patent owned by the Plaintiffs.

The defendants tried to “carve out” infringing activities, like a hole in a donut.

Some of the Plaintiffs hold an approved NDA for the branded drug Trintellix®, for the treatment of major depressive disorder (MDD) in adults. Trintellix® contains vortioxetine as the active ingredient. The Plaintiffs own U.S. Patent Nos. 9,278,096 and 9,125,910 (the ‘096 and ‘910 patents, expiring on March 21, 2032, and June 15, 2027, respectively). The ‘096 patent claims a method of treating certain diseases (depression, anxiety, abuse, and chronic pain) by administering vortioxetine in a specific patient subset. The ‘910 patent claims a method of treating cognitive impairment in a different patient subset diagnosed with depression, by administering the same drug. The Plaintiffs also brought into proceedings U.S. Patent No. 9,101,626, which claims a process for preparing vortioxetine or its salt by “reacting” certain compounds under specific conditions. As a patent drawn to a production process, the ‘626 patent does not appear in the Orange Book.

The Plaintiffs sued the Defendants for direct and indirect patent infringement based on the Defendants’ ANDAs filed with a Hatch-Waxman Section viii statement. The CAFC initially addressed the issue of whether the Defendants, by merely filing an ANDA, infringed the ‘096 and ‘910 patents . The ANDA-at-issue sought FDA approval to market vortioxetine for the single indication of treating MDD in adults. However, the CAFC upheld the district court’s decision that the ‘096 and ‘910 patents did not cover the single indication (skinny label) sought by the ANDA-at-issue. Nor had the Defendants’ ANDA filings induced infringement by facilitating third parties to carry out infringing acts. The Defendants were also cleared of carrying out contributory infringement i.e., through advertising.

Where the Defendants fell short however, was in the construction of the term “reacting” of the ‘626 process patent. The Defendants attempted to argue that the order of reacting the chemicals in the process claimed was important. The Plaintiffs, on the other hand, argued for the plain meaning of the term “reacting” to generally mean “changing of reactant(s) to product(s)”. Absent a clear indication from the prosecution history otherwise, the court took the broader meaning. The CAFC upheld the lower court’s decision that the Defendants infringed claim 12 of the ‘626 process patent.

 This decision highlights the importance of a well-rounded patent portfolio in the pharmaceutical world, including both Orange Book listed patents and production process patents. NDA filers should remember to consider such full portfolios for enforcement, and ANDA filers should remember to consider freedom to operate analysis that does not simply stop at the Orange Book. This decision also shows that a Section viii carveout to avoid infringement by excluding a patented medical indication may be successful, but ultimately may not be enough if broader process patents could impede the generic’s route to market.