October 25, 2016

Case Name: Vanda Pharm. Inc. v. Roxane Labs., Inc., Civ. No. 14-757-GMS (Consolidated), 2016 U.S. Dist. LEXIS 113521 (D. Del. Aug. 25, 2016) (Sleet, J.) 

Drug Product and Patent(s)-in-Suit: Fanapt® (iloperidone); U.S. Patents Nos. RE39,198 (“the ’198 patent”) and 8,586,610 (“the ’610 patent”)

Nature of the Case and Issue(s) Presented: The ’198 patent claims iloperidone, and the ’610 patent claims methods of treating patients suffering with schizophrenia using that drug. More specifically, the method claims require performing genotyping assays to identify poor metabolizers of iloperidone, and lowering these patients’ doses accordingly. Before the court were the following issues: (i) whether the asserted claims of the patents-in-suit were invalid for obviousness, subject-matter ineligibility, and inadequate written description; (ii) whether Roxane’s proposed ANDA product would induce infringement of the ’610 patent; (iii) whether Roxane’s proposed ANDA products would contributorily infringe the ’610 patent; and (iv) if successful, whether Vanda were entitled to a permanent injunction against Roxane.

Why Vanda prevailed: Concerning validity of the drug patent, the court found fault with the Roxane’s lead-compound analysis and rejected its obviousness defense. The court found that no one had, over the decades, followed the path Roxane’s experts argued was obvious, despite the long-felt need to find improved treatments for schizophrenia. One of Roxane’s lead compounds had extremely serious side effects, and the other had no known antipsychotic activity, so neither was a good starting point. Further, the court found the motivation to make required substitution in one of the lead compounds was hindsight.

The court also rejected Roxane’s arguments that the ’610 patent was obvious, finding that the level of clinical testing required, and the inherent unpredictability in the field, undercut Roxane’s arguments that metabolism of iloperidone was sufficiently understood at the time of the invention. The court was particularly impressed that Novartis had purchased the rights to iloperidone from Vanda, and then sold them back because of Novartis’ inability to determine safe dosing levels. Finding those safe levels, the court said, was the invention at issue. The court also identified as a secondary consideration of non-obviousness the fact that Novartis bought the rights back, paying a 40-fold increase over the previous sales price.

Roxane argued that the ’610 patent covered ineligible subject matter, specifically a law of nature applied in a routine and conventional way. The court rejected this argument. While it acknowledged that the asserted claims depend on a law of nature, the court concluded that the claims incorporate an additional step sufficient to make them valid. The court observed that in Mayo v. Prometheus, 132 S. Ct. 1289 (2012), the “determining” step that led to invalidation in that case occurred “through whatever process the doctor or the laboratory wishes to use.” In other words, the determining step there was routine. In this case, however, the court concluded that Roxane had not proven by clear and convincing evidence that the precise test and the discovered results were routine or conventional.

Next, the court rejected Roxane’s written-description argument that the claimed dose for poor metabolizers was unsupported, because the data supporting the dose reduction were not statistically significant. In rejecting this argument, the court found the data sufficient. The court reasoned that it was not necessary that the patent demonstrate an actual reduction to practice.

As to infringement, Roxanne conceded infringement of the ’198 patent. For the ’610 patent, Roxanne argued that no physician would perform both the genotyping steps and the dosing step, but Vanda proved direct infringement by presenting testimony and medical records of one physician who performed both steps. Accordingly, the court found that there would be direct infringement by the ANDA drugs. The court further concluded that Roxane’s proposed label induced infringement of the ’610 patent, because while it might be true that some physicians administered the drug without performing the genotyping step, the label still recommended that type of testing, and that the label was thus probative of intent to induce.

The court found that Vanda was entitled to a permanent injunction against Roxane’s proposed ANDA product prior to the expiration of the ’198 patent pursuant to 35 U.S.C. § 271(e)(4)(B). The court said that, because the ’610 patent issued only after the FDA approved Roxane’s ANDA submission, Vanda was not entitled to an injunction under 35 U.S.C. § 271(e)(4)(A). However, the court said that the four eBay factors weighed in favor of issuing an injunction pursuant to the court’s equitable powers, and that Vanda was entitled to a permanent injunction for the ’610 patent, too.

Roxane did, however, prevail on the issue of contributory infringement. The court said that the evidence presented at trial demonstrated that a physician could prescribe iloperidone without practicing the claims of the ’610 patent by not using a genotyping assay. Specifically, a physician could adjust the dosing by measuring metabolites in the blood. But plaintiffs, the court said, had not shown this non-infringing use to be not substantial.