Case Name: Janssen Pharms., Inc. v. Teva Pharms. USA, Inc., No. 18-cv-734, 2021 WL 5323737 (D.N.J. Nov. 16, 2021) (Cecchi, J.)
Drug Product and Patent(s)-in-Suit: Invega Sustenna® (paliperidone palmitate); U.S. Patent No. 9,439,906 (“the ’906 patent”)
Nature of Case and Issue(s) Presented: The ’906 patent claims specific dosing regimens of injectable paliperidone palmitate to treat schizophrenia and other psychotic disorders. Specifically, the ’906 patent discloses unique combinations of dose amounts, dosing schedules, injection sites, and formulation properties. Janssen sells injectable paliperidone palmitate under the brand name Invega Sustenna. In 2017, Teva notified Janssen that it had submitted an ANDA seeking FDA approval to manufacture, use and sell generic paliperidone palmitate extended-release injectable suspension products prior to the expiration of the ’906 patent. Janssen subsequently sued Teva for patent infringement. Teva stipulated to infringement. Therefore, the only question before the Court was the validity of the ’906 patent. At trial, Teva mounted three invalidity defenses: (i) obviousness; (ii) lack of written description; and (iii) indefiniteness. The court found that Teva failed to show by clear and convincing evidence that the ’906 patent was invalid under any of its theories.
Why Janssen Prevailed: With respect to obviousness, the court found claim 2 was not obvious because the closest prior art reference (called “the ’548 protocol”) and the ’906 patent differ in several material respects. Where the ’548 protocol discloses equal doses of paliperidone palmitate administered in the gluteal muscle on fixed treatment days, the ’906 patent contains regimens comprised of unequal doses, two of which must be administered in the deltoid muscle, and a broader dosing window for the second and monthly maintenance doses. The Court found that Teva failed to persuasively argue that a POSA would know to alter the equal doses in the gluteal muscle described in the ’548 protocol to arrive at the unequal loading doses in the deltoid claimed by the ’906 patent without knowing the results of Janssen’s internal clinical trials, several of which yielded unexpected results. With respect to claims 10 and 13, which are directed to dosing regimens for patients with impaired kidney functions, the prior art does teach reduced dosing regimens of antipsychotic drugs for patients with renal impairment. But the prior art generally teaches a flat 50-percent dose reduction of antipsychotic drugs for patients with mild renal impairment. Here, applying a 50 percent dose reduction from the maximum recommended dose disclosed in claim 2 would not yield the regimens disclosed in Claims 10 and 13. Finally, with respect to claims 20 and 21, which are directed to formulations of paliperidone palmitate, the court found that of the primary prior art references, only the ’548 protocol discloses a treatment initiation regimen for paliperidone palmitate. But the ’548 protocol neither specifies the characteristics of the formulation used in the study, nor any information about the formulation’s particle size. Furthermore, while the two other primary art references discuss particle size for paliperidone palmitate formulations, they do not suggest that that these formulations are appropriate for a treatment initiation regimen.
The court made several findings concerning objective indicia of non-obviousness. Prior to the invention of Invega Sustenna, the conventional wisdom for both antipsychotic drug dosing was to “start low and go slow.” The claimed dosing regimens in the ’906 patent run contrary to these prior art teachings because they use depot injections of high, rather than low, loading doses to initiate treatment. In addition, in multi-dose regimens, such as those covered by the representative claims, there are many possible combinations of dose amounts, schedule, and sites of administration. The ’548 protocol, which Defendant identifies as one of the closest prior art references, tested three different combinations of equal doses of paliperidone palmitate administered in the gluteal muscle and failed to properly initiate treatment in a rapid manner to aid with treatment adherence. By contrast, the ’906 patent regimens, which are comprised of high loading doses that must be administered in the deltoid muscles for the first two doses, succeeded. Therefore, the court found that in view of the prior art, the claimed invention led to unexpected results. Janssen also proffered evidence of commercial success, industry praise, skepticism, copying, and long-felt need related to the claimed dosing regimens sufficient to support a conclusion of non-obviousness.
With respect to written description, Teva argued that the specification failed to properly identify the type of renal-impaired patient (e.g. mild, moderately, or severely impaired) that can be treated with the claimed dosing regimens. The court found otherwise. Both parties’ experts agreed at trial that the ’906 patent contains several representative embodiments of dosing regimens for mildly renal-impaired patients that a physician could choose in their medical discretion. Further, both parties’ experts agreed that a physician would not treat a patient with moderate or severe renal impairment using Invega Sustenna. Thus, the court found that the asserted claims reasonably convey to those skilled in the art possession of the claimed invention.
Next, Teva raised two indefiniteness arguments. First, Teva argued that claims 20 and 21 fail to properly characterize the claimed particle size range given that average particle size can be measured and expressed in a number of different ways. Second, Teva argued that claims 10, 20, and 21 fail to properly characterize the term “aqueous nanoparticle suspension,” which lacks a typical meaning in the art. Citing the Federal Circuit, the court found that “the mere possibility of different results from different measurement techniques does not render a claim indefinite.” Here, both sides’ experts agreed that throughout the development of Invega Sustenna, the prosecution of the patent, and the development of Teva’s ANDA, there was no apparent confusion over how to measure or express average particle size. The court also found that “aqueous nanoparticle suspension” was not indefinite because the ’906 patent adequately defines the term because “it says, an aqueous formulation would preferably be a nanoparticle suspension of wherein the nanoparticles would fit in average sizes of less than 2000 nanometers to about 100 nanometers.”
