GlaxoSmithKline LLC v. Glenmark Pharms. Inc., USA

Case Name: GlaxoSmithKline LLC v. Glenmark Pharms. Inc., USA, 2017 U.S. Dist. LEXIS 82534 (D. Del. May 23, 2017) (Burke, M.J.)

Drug Product and Patent(s)-in-Suit: Coreg® (carvedilol); U.S. Patent No. RE40,000 (“the ’000 patent”)

Nature of the Case and Issue(s) Presented: GSK manufactures and sells the drug carvedilol under the trade name Coreg. Defendants all seek FDA approval to market generic versions of carvedilol. The court construed congestive heart failure (“CHF”) to mean “a condition that occurs as a result of impaired pumping capability of the heart and is associated with abnormal retention of water and sodium.” Before 1997, the only treatments approved by the FDA for heart failure were diuretics, certain angiotensin converting enzyme (“ACE”) inhibitors and digoxin/digitalis. Carvedilol has been a known beta-blocker since at least 1978. Beta-blockers were compounds that prevent stimulation of the adrenergic receptors responsible for increased heart rate and contractility, which could cause the heart to pump slower or with less force. Historically, beta blockers were contraindicated in the treatment of CHF because of the medical community’s “widely-held concern” that this type of drug would further reduce the diseased and/or damaged heart's ability to pump blood through the body.

In February 1995, clinical trials revealed a significant effect of carvedilol on survival in CHF patients. In November 1995, GSK sought FDA approval of carvedilol in combination with ACE inhibitors, digoxin or diuretics. In May 1997 the FDA ultimately approved carvedilol as the first beta blocker for the treatment of CHF, as an adjunctive therapy. Also in 1997, GSK sponsored another clinical trial in order to obtain FDA approval for use in patients who had recently experienced a heart attack and who had a left ventricular ejection fraction of [<=]40%, “Post-MI LVD.” In September 2002, GSK submitted a supplement to its NDA, seeking approval of the Post-MI LVD indication for Coreg. By 2003, Coreg received FDA approval for the following three indications: (i) heart failure; (ii) left ventricular dysfunction following myocardial infarction, or Post-MI LVD; and (iii) hypertension.

In June 1995, the inventors filed a patent application directed to a method of using carvedilol to decrease the risk of mortality caused by CHF, which issued as U.S. Patent No. 5,760,069 (“the ’069 patent”). In November 2003, GSK requested a reissue of the ’069 patent, which resulted in the ’000 patent. In February 2008, GSK replaced the ’069 patent in the Orange Book with the ’000 patent, and identified “[d]ecreasing [m]ortality [c]aused [b]y [c]ongestive [h]eart [f]ailure” as the method of use covered by the ’000 patent. The ’000 patent claims methods of decreasing mortality caused by CHF in a patient in need thereof by administering carvedilol in a manner recited in the claims.

In March 2002, Teva submitted its ANDA including a Paragraph IV certification against the ‘’069 patent. Teva believed that it was the first generic applicant that included a Paragraph IV certification in its ANDA, which could make it eligible for 180-day exclusivity against all other subsequent Paragraph IV filers. But a few months before Teva’s planned September 2007 launch, Teva learned that other generic companies had chosen not to challenge the validity of the ’069 patent, instead submitting section viii statements with their ANDAs, and receiving FDA approval for labels that included the hypertension and Post-MI LVD indications, but that omitted the heart failure indication. In July 2007, Teva decided to amend its ANDA to include a section viii statement in line with the other generics. On September 5, 2007, the FDA approved the ANDAs of fourteen generic companies, including Teva. At the time of launch, the labels for Teva’s and Glenmark’s generic versions of Coreg included the hypertension and Post-MI LVD indications, but did not include the heart failure indication (the “Skinny Label Period”). In May 2011, Teva amended its label to include the heart failure indication, and it has since used that full label. For a three-month period from June to August 2010, Glenmark’s label also included the separate heart failure indication (although according to Glenmark, this was the result of an error in which the heart failure indication was inadvertently added to its package insert in the course of making an unrelated ANDA amendment to add different bottle sizes).

Defendants moved for summary judgment of no induced infringement on the basis that GSK has failed to produce any evidence establishing that: (i) Defendants’ conduct actually caused physicians to administer carvedilol in an infringing manner; and (ii) Defendants acted with the specific intent of inducing physicians to administer carvedilol in an infringing manner during the Skinny Label Period. The court recommended denial of the motion.

Why GSK Prevailed: Firstly, Defendants relied on the Federal Circuit’s recent holding in Power Integrations, Inc. v. Fairchild Semiconductor Int’l, Inc., 843 F.3d 1315 (Fed. Cir. 2016) where the Court found that “finding of induced infringement requires actual inducement”—i.e., “successful communication between the alleged inducer and the third-party direct infringer.” Here, defendants argued that GSK had put forth no expert opinion nor other evidence that any doctor’s prescription of carvedilol for the allegedly infringing purpose was caused by any act of defendants. GSK made a few different arguments in response. Firstly, at least as to the period in which defendants’ products included the full label with the heart failure indication, GSK suggested that the Federal Circuit had set out a per se rule that if a generic manufacturer’s label includes the indication that was patented, that, in and of itself, was “sufficient to establish induced infringement.” Here, GSK filed its actions almost seven years after defendants launched their generic carvedilol products into the market. Thus, GSK’s induced infringement claims were not premised on a hypothetical, but instead, needed to be supported by sufficient evidence as to what actually happened during the relevant time period. GSK argued that: (i) under the law, it may prove actual inducement with circumstantial evidence directed to a class of direct infringers, and it had put forward sufficient evidence to create a genuine issue of material fact on this element. In order to meet its burden, GSL relied on: (i) defendants’ labels (both the full labels and the labels from the Skinny Label Period); and (ii) defendants’ touting of the AB rating of their generic carvedilol for the entire infringement period from 2008-2015 in their product catalogues, websites, price sheets, Monthly Prescribing Reference and Generic Product Reference Guide. On this record, and at this stage of the proceedings, the court found that GSK’s claims survived.

Next, the court addressed whether GSK had demonstrated that a genuine issue of material fact exists regarding whether defendants actively encouraged infringement during the Skinny Label Period. This was a similar argument that Teva had raised in its case at the pleadings stage, which the court earlier rejected. While it was true that since the pleadings stage, the facts had shown that over 80% of uses of defendants’ generic carvedilol were non-infringing uses, that was but one factor to consider. The existence of a substantial non-infringing use, however, did not preclude a finding of induced infringement. Moreover, the court again found that “the mere existence of a skinny label does not foreclose the possibility of infringement liability.” In light of recent Federal Circuit guidance (from Eli Lilly and Co. v. Teva Parenteral Meds., Inc., 845 F.3d 1357 (Fed. Cir. 2017)) that when a plaintiff’s induced infringement claim relies on the content of a defendant’s label, the court must assess the link between the use described on the defendant’s labeling and the patented use, the court found that there was a genuine dispute of material fact as to whether defendants’ skinny labels instructed or encouraged physicians to infringe the ’000 patent.

In addition to the language of Defendants’ skinny labels, plaintiffs also relied on other evidence that they claimed demonstrated defendants’ intent to induce infringement. In particular, plaintiffs pointed to documents evidencing Teva’s original plan to market and sell a product expressly labeled with the CHF indication, as well as Teva’s sudden pre-launch decision to remove that indication. Plaintiffs also pointed to defendants’ press releases issued at launch that compared defendants’ generics to Coreg, that referenced COREG's annual sales (amounts that included sales of the drug to be used in treating patients with heart failure). All of this, the court reasoned, may demonstrate defendants’ intent to induce such that summary judgment was not warranted.