Takeda Pharm. Co. Ltd. v. Norwich Pharms., Inc.

Case Name: Takeda Pharm. Co. Ltd. v. Norwich Pharms., Inc., No. 20-8966 (SRC), 2022 WL 17959811 (D.N.J. Dec. 27, 2022) (Chesler, J.)

Drug Product and Patent(s)-in-Suit: Vyvanse® (1-lysine-d-amphetamine dimesylate); U.S. Patents Nos. 7,655,630 (“the ’630 patent”), 7,662,787 (“the ’787 patent”), 7,687,466 (“the ’466 patent”), 7,105,486 (“the ’486 patent”), 7,678,770 (“the ’770 patent”), 7,671,031 (“the ’031 patent”), 7,223,735 (“the ’735 patent”), 7,700,561 (“the ’561 patent”), and 7,671,030 (“the ’030 patent”)

Nature of the Case and Issue(s) Presented: Vyvanse is a central nervous system stimulant indicated for the treatment of attention deficit hyperactive disorder. Norwich stipulated to infringement and argued that the patents-in-suit were invalid for obviousness and lack of enablement. After a three-day bench trial, the court found that Norwich failed to prove that the asserted claims were invalid.

Why Takeda Prevailed: Norwich argued that a skilled artisan would have begun its obviousness analysis with the problem to solve: the abusability problem of d-amphetamine. The court agreed both with the identified problem and that a skilled artisan would have selected d-amphetamine in its lead compound analysis. Norwich’s obviousness analysis required two primary modifications: (i) conjugating d-amphetamine into a prodrug; and (ii) selecting l-lysine as the promoiety. Norwich argued that a skilled artisan would make both modifications to reduce the abusability of d-amphetamine. The court disagreed.

As to the prodrug step, Norwich argued that post-administration euphoria is the reason that d-amphetamine is abused. Next, Norwich showed that euphoria results from initial spiking of the plasma concentration curve of d-amphetamine and it is this euphoric effect that leads to abuse. Norwich then argued that a prodrug can change the plasma concentration curve to reduce initial spiking. In other words, according to Norwich, decreased initial spiking results in decreased euphoria, which in turn results in decreased abuse. The court agreed with Norwich’s proposition that a skilled artisan would have been motivated to minimize the euphoria side-effect that contributed to d-amphetamine’s abuse. But the court disagreed that a skilled artisan would have been motivated to use a prodrug to solve the d-amphetamine abuse problem. In particular, the court found that a skilled artisan would not have known the effect, if any, a prodrug would have on the plasma concentration curve for d-amphetamine. For that reason, Norwich could not prove that a skilled artisan would have had a reasonable expectation of success in using a prodrug to solve the d-amphetamine abuse problem. Further, the court found that a skilled artisan would not have been motivated to use a sustained release, prodrug approach to reduce the abuse of d-amphetamine. Indeed, the art showed that the abuse deterrence of extended-release formulations could be defeated by crushing. Prodrugs would not have defeated this form of abuse.

Norwich also failed to demonstrate that a skilled artisan would have selected l-lysine as the promoiety for the claimed prodrug. It two prior-art combinations to support the theory that a skilled artisan would have found it obvious to select l-lysine as the promoiety. But the court found that neither combination said anything about abuse of amphetamine APIs or reasoned that selection of 1-lysine as a promoiety would reduce abuse.

Additionally, the court found secondary indicia supported a finding of non-obviousness. In particular, the court found that Vyvanse was a commercial success and that Norwich’s use of the claimed salt, after failing to find an alternative salt outside the scope of the patents, supported a finding of copying.

Finally, Norwich argued that enablement should be gauged not by the disclosure in the claims and specification of the issued patent, but based on the disclosure in the specification of the application with the earliest priority date relied on by the patents-in-suit. The court disagreed and explained that enablement is gauged by the disclosures in the issued patent. The court further explained that methods of creating the claimed salts were well known in the art. For this reason, the patents-in-suit were not invalid for lack of enablement.