GlaxoSmithKline LLC v. Teva Pharms. USA, Inc.

Case Name: GlaxoSmithKline LLC v. Teva Pharms. USA, Inc., Nos. 2018-1976, 2018-2023, 2001 WL 3412496 (Fed. Cir. Aug. 5, 2021) (Circuit Judges Moore, Newman, and Prost presiding; Opinion per curiam) (Appeal from D. Del., Stark, J.) 

Drug Product and Patent(s)-in-Suit: Coreg® (carvedilol); U.S. Patents Nos. 4,503,067 (“the ’067 patent”), 5,760,069 (“the ’069 patent”), RE40,000 (“the ’000 patent”)

Nature of Case and Issue(s) Presented: GSK marketed carvedilol under the name Coreg. The FDA approved carvedilol for three indications: (i) hypertension; (ii) congestive heart failure (CHF); and (iii) to reduce cardiovascular mortality in patients suffering from left ventricular dysfunction following a myocardial infarction (“post-MI LVD”). The first two indications were approved in 1997, and the post-MI LVD indication was approved in 2003.

In March 2002, Teva filed an ANDA, which included all three indications. Teva certified under Paragraph III against the ’067 patent, which claimed the carvedilol compound, and certified under Paragraph IV against the ’069 patent, alleging invalidity based on obviousness. GSK did not file suit against Teva on the ’069 patent and, instead, applied for a reissue patent. Teva received tentative approval to market its generic product for treatment of heart failure and hypertension in 2004, which would become effective when the ’067 patent expired in 2007.

In January 2008, the PTO issued the ’000 patent, which claims “a method of decreasing mortality caused by CHF by administering carvedilol with at least one other therapeutic agent.” GSK listed the ’000 patent in FDA’s Orange Book.

Before Teva launched its product, it certified to FDA that its label would not include an indication for “decreasing mortality caused by congestive heart failure.” GSK sued Teva, alleging that Teva induced infringement of the ’000 patent. Teva argued at trial that it could not have induced infringement, at least before 2011, because it had carved out the CHF indication from its label. Teva also argued that it could not have induced infringement at any time because it had not caused others to infringe the ’000 patent.

The district court instructed the jury to evaluate infringement during two different times: (i) the “partial label period” from 2008 to 2011 when Teva carved the CHF indication out of its label; and (ii) the “full label period” from 2011 to 2015 when the label contained all three indications, including the CHF indication. The jury found that the ’000 patent was not invalid, and that Teva induced infringement of claims 1-3 during the partial label period, and additionally claims 6-9 during the full label period. The jury assessed damages based on a combination of lost profits and reasonable royalty.

Teva moved for JMOL, which the district court granted, finding that substantial evidence did not support the induced infringement verdict because GSK did not prove that Teva directly caused physicians to prescribe Teva’s product to treat CHF. The district court also found that no reasonable juror could have found induced infringement based on the post-MI LVD indication. GSK appealed, arguing the verdict should be reinstated, and the Federal Circuit agreed. Teva then petitioned for en banc rehearing, which the Federal Circuit also interpreted as a petition for panel rehearing, arguing the Federal Circuit’s October 2, 2020 decision could be read broadly to impose liability on ANDA filers that use section viii carve-outs to avoid infringement. Amici were also concerned about the effect on section viii carve-outs. The Federal Circuit granted panel rehearing, vacated the October 2, 2020 judgment, and withdrew its October 2, 2020 opinions. The Federal Circuit explained that it granted panel rehearing to assuage the concerns of the Amici, to show that the facts of this case were clearly outside the boundaries of the law regarding section viii carve-outs. The Federal Circuit held that in this case, the substantial evidence supported the jury’s finding that the patented use was on Teva’s label at the relevant times and that Teva therefore had failed to carve out the patented indications.

Why GSK Prevailed: Concerning the partial-label period, the parties disputed whether Teva effectively carved out the patented methods from its label, thereby making it a “skinny label” under section viii. At trial, GSK presented evidence that doctors read drug labels and that the portion of Teva’s label directed to the post-MI LVD indication satisfied the “decreasing mortality caused by congestive heart failure” limitation of the ’000 patent. Teva’s label stated:

Carvedilol is indicated to reduce cardiovascular mortality in clinically stable patients who have survived the acute phase of a myocardial infarction and have a left ventricular ejection fraction of ≤ 40% (with or without symptomatic heart failure) (see CLINICAL STUDIES [14.1]).

Both GSK and Teva’s experts agreed that patients experiencing left ventricular ejection fraction of less than or equal to 40% with symptomatic heart failure would be diagnosed with CHF. GSK presented further evidence that Teva’s label satisfied the other claim elements, and its expert testified that the label disclosed dosages that satisfied the “administering a therapeutically acceptable amount of carvedilol” and administering “daily maintenance dosages” limitations. GSK’s expert further testified that the label contained descriptions of patients who were taking ACE inhibitors and diuretics, which satisfied the claim limitation of administering carvedilol in conjunction with one of those drugs. GSK’s expert also testified that Figure 1 in the label showed treatment for longer than six months, which satisfied the limitation requiring a maintenance period greater than six months. In addition to the instructions in Teva’s label, GSK presented evidence such as marketing materials, catalogs, press releases, and testimony from Teva’s own witnesses showing that Teva encouraged sales of cardevilol for CHF despite its purported carve-out.

Concerning the full label period, on May 11, 2011, Teva changed its label for its carvedilol product to include all three indications. GSK’s expert testified that Teva’s full label met all the claim limitations of the ’000 patent. Teva did not dispute the jury’s finding that the full label included all of the claim limitations. GSK also presented evidence that Teva put information on its website and in marketing materials encouraging the infringing use, and instructed doctors to read Teva’s label, which instructed the infringing use. The Federal Circuit therefore held that substantial evidence supported the jury’s infringement verdict.

Concerning, causation, Teva argued that it did not cause doctors to prescribe carvedilol, but instead doctors were prescribing carvedilol based on information they received from GSK. Additionally, it argued that, based on information in industry guidelines, treatises, and medical textbooks, doctors knew to use carvedilol to treat CHF before Teva launched its product. But the Federal Circuit relied on evidence discussed above. Further, the Federal Circuit found that it was fair for the jury to make the determination that when Teva distributed its product with labelling and marketing materials encouraging an infringing use, it actually induced doctors to infringe. The Federal Circuit explicitly did not decide whether the label alone was enough to prove causation.

Concerning damages, the jury assessed lost profits and reasonable royalty damages. Teva argued that the jury was improperly instructed and would have awarded no damages—or only reasonable royalty damages—with proper instructions. Teva argued that because there were other generic carvedilol products on the market, GSK must prove that, for every infringing sale Teva made, the direct infringer must have purchased Coreg instead of another generic carvedilol product. The district court declined to present that instruction, explaining that in a but-for world the infringing alternatives of other generic producers would not exist and therefore were not properly considered in the damages analysis. The Federal Circuit explained that Teva’s argument was at odds with “long-standing precedent that the presence of non-infringing alternatives precludes award of lost profits, but the present of other infringers does not.” The Federal Circuit therefore held that the district court properly instructed the jury and therefore sustained the damages verdict.

Finally, Judge Prost dissented on three grounds: (i) the majority weakened the intention-encouragement prong of inducement by eliminating the difference between merely describing versus encouraging an infringing use in a label; (ii) the majority eviscerated the causation prong of inducement; and (iii) the majority created more confusion surrounding section viii carve-out patented uses.

In her opinion, the skinny label itself, a press release from 2007, and a press release from 2004, were not enough for a reasonable jury to find either intent to encourage infringement or causation.

The skinny label omitted the CHF indication, which is what GSK told the FDA the ’000 patent covered. At most, the evidence the majority pointed to in the skinny label merely described an infringing use, which is not enough for inducement. GSK would have had to connect the dots between a doctor’s reading of the Teva label, piecing together the disparate portions of the label to arrive at a recommendation to use carvedilol to treat CHF according to the specific claimed method, and then making the prescribing decision. Judge Prost stated that GSK did not connect those dots.

As for the 2007 press release announcing the FDA approval for Teva’s product, it published before the ’000 patent issued but remained on Teva’s website subsequently. Judge Prost took issue with the majority’s holding because it created liability for inducement for just saying the product is a generic equivalent to the brand drug, which Her Honor labelled a “drastic holding,” as all ANDA products are generic equivalents. Judge Prost further points out that the majority never explained how a reasonable jury could have found that the 2007 press release could have caused doctors to infringe the ’000 patent. For similar reasons, Judge Prost also took issue with the 2004 press release.