H. Lundbeck A/S v. Lupin Ltd.

Case Name: H. Lundbeck A/S v. Lupin Ltd., Civ. No. 18-88-LPS (D. Del. Oct. 5, 2021) (Stark, J.)

Drug Product and Patent(s)-in-Suit: Trintellix® (vortioxetine hydrobromide); U.S. Patents Nos. 7,144,884 (“the ’884 patent”), 8,476,279 (“the ’279 patent”), 8,722,684 (“the ’684 patent”), 9,861,630 (“the ’630 patent”), 9,101,626 (“the ’626 patent”), 9,090,575 (“the ’575 patent”), 9,278,096 (“the ’096 patent”), and 9,125,910 (“the ’910 patent”)

Nature of the Case and Issue(s) Presented: Trintellix was a tablet formulation indicated for the treatment of major depressive disorder (“MDD”) in adults. The ’884 and ’279 patents were the “Compound Patents”; the ’684 and ’630 patents were the “Crystalline Form Patents”; the ’626 and ’575 patents were the “Process Patents”; the ’096 patent was the “Sexual Dysfunction Patent”; and the ’910 patent was the “Cognitive Impairment Patent.” Plaintiffs sued sixteen sets of defendants. Six sets of defendants proceeded to trial: (i) Alembic; (ii) Lupin; (iii) Macleods; (iv) Sandoz; (v) Sigmapharm; and (vi) Zydus. The district court found that Plaintiffs had not proven that Defendants, with the exception of Lupin, infringed the asserted claims, and that Defendants had not proven that the patents-in-suit were invalid.

Why Defendants, except for Lupin (with Respect to the ’626 patent), Prevailed: The Compound Patents were not invalid. Plaintiffs argued that Defendants’ arguments failed as a matter of law because their analysis began with a “lead motif” model—analyzing the teachings from six pieces of prior art—and did not begin with a lead molecule analysis. The court declined to per se reject the “lead motif” analysis, but nevertheless found that Defendants failed to justify the numerous modifications required by a POSA to reach the claimed compound.

Defendants did not infringe the Crystalline Form Patents. The court found that “the overwhelming abundance of careful XRPD images developed expressly for purposes of determining infringement detect[ed] no more than one of the multiple characteristic peaks.” In particular, because the claims identified the particular crystalline form by the presence of multiple characteristic peaks, evidence of the presence of just a single (even discriminatory) peak was not sufficient to demonstrate infringement. Further, the court rejected Plaintiffs’ argument that Defendants’ ANDA products—that are manufactured using amorphous API—will, over the course of time, convert to crystalline forms.

Zydus did not invalidate the Process Patents. The court found that Zydus did not address why a POSA would have been motivated to combine the seven references relied upon to produce the alleged obviousness combination.

Plaintiffs proved Lupin would infringe the ’626 Process Patent. After adopting Plaintiffs’ proposed construction of “reacting,” the court found that Lupin’s process for manufacturing vortioxetine Hbr met all the limitations of the asserted claim.

Defendants did not infringe the Sexual Dysfunction Patents. Defendants’ proposed labels “carved out” comparative information for the treatment of emergent sexual dysfunction. As a result, Defendants’ labeling would not encourage, recommend, or promote the practice of switching MDD patients, who had previously experienced adverse sexual side effects while taking one of the four classes of antidepressants, to vortioxetine. Specifically, Plaintiffs failed to show that clinicians would understand that Section 6.1, which includes the heading “Voluntarily Reported Adverse Reactions of Sexual Dysfunction,” would encourage doctors to prescribe vortioxetine according to the claimed method.

Defendants failed to prove invalidity of the Sexual Dysfunction Patents. Plaintiffs were not entitled to the claimed 2007 effective filing date, but were able to claim priority to the 2008 applications, meaning that the claims of the asserted patent were not invalid for lack of written description or anticipation/obviousness.

Plaintiffs did not prove infringement of the Cognitive Impairment Patent. This patent was directed to a method of treating cognitive impairment in a patient diagnosed with MDD. The parties disputed how often clinicians treat cognitive impairment when prescribing vortioxetine, but the Court was persuaded by Defendants’ evidence that this happens on less than a majority of occasions.

Defendants did not prove that the Cognitive Impairment Patent was invalid. Defendants acknowledged that their anticipation/obviousness theory had merit only if the Court were persuaded by Plaintiffs’ infringement theory. As a result, this argument failed.