The U.S. Court of Appeals for the Federal Circuit sided with a jury’s 2017 finding that Teva swayed doctors away from using GlaxoSmithKline’s drug Coreg (carvedilol) for a use of its drug that was still protected under patent law, in a split, 2-1 decision announced Thursday
A company can be liable for “inducing” infringement of a patent by convincing someone else to do so, often through advertising, instruction manuals, or, as discussed in this case, the wording it uses on a label.
In its precedential opinion, the majority said that though their ruling was overturned by the district judge, members of the jury had actually been reasonable in concluding “that Teva’s partial label was evidence Teva instructed physicians to use its carvedilol in an infringing way” and hitting Teva with $234 million in damages for GSK’s lost sales and another $1.5 million in royalties.
In her dissent, Justice Prost staunchly disagreed, saying “With reasoning sometimes labored, sometimes opaque, the majority strains to prop up a jury verdict that is unsupportable.”
Teva’s not-so-skinny label
In 2007, Teva launched its generic version of carvedilol with a “carved-out” or “skinny” label indicating its effectiveness at treating hypertension and left ventricular dysfunction after a heart attack and avoiding the indication for which GSK still held a patent — treatment of congestive heart failure (CHF) — or so they thought.
However, despite this “skinny label,” doctors could still prescribe Teva’s generic and pharmacies could automatically substitute it for any use, including treatment of CHF, thanks to section viii of the Hatch-Waxman Act.
Section viii permits a generic manufacturer to seek FDA approval to market a generic version of a drug for a non-patented use where the reference drug has multiple indications, not all of which are covered by a method-of-use patent.
If a generic applicant goes this route, it must propose labeling for its product that “carves out” the patented methods of use, only seeking approval for methods that either were never protected by patent or for whom protection has expired.
GSK first sued Teva in 2014, arguing that Teva had infringed its 2008 reissue patent, which listed a number of “follow-on” ways you could use the drug, including as treatment for CHF. Teva replied it had done no such thing, having successfully carved out the patented methods of use. The jury disagreed, finding Teva either knew or should have known that their labels and advertising would lead doctors to infringe.
After GSK won its jury trial, Teva had the case sent back to the district court for review. The district judge rejected the jury’s findings saying they were based on insufficient evidence that doctors had relied on the label to use the drug in disallowed ways, and that any infringement was instead due to non-Teva factors (e.g., prior knowledge of cardiologists regarding the uses for carvedilol). On appeal to the Federal Circuit, GSK argued the district court had overstepped, which the Federal Circuit found persuasive, reinstating the jury verdict in its October 2020 decision. Immediately, a chorus of concerned stakeholders raised their voices in concern that this ruling left Hatch-Waxman’s skinny labeling provisions toothless.
In February, the court accepted Teva’s request for a rehearing, leaving critics of its October 2020 decision hopeful that it would narrow its finding to protect the generic industry’s ability to use “skinny labels” to launch non-branded alternatives at risk, a pathway laid out in the Hatch-Waxman Act of 1984.
The Federal Circuit’s latest decision
Instead of narrowing its findings, the court responded to criticisms one by one, showing how each of the possible grounds the jury had considered for inducement of infringement could have been found compelling by the jury, and how the finding that Teva induced was supported by the evidence as a whole.
In doing so, it clarified one point that had generics companies losing sleep: whether claiming chemical equivalence by marketing a drug’s AB rating was grounds for inducement, in and of itself. To their collective relief, the court found it was not.
AB rating is a regulatory agency’s approval of a company’s scientific evidence that its drug is bioequivalent with its reference, and thus equally safe and effective. This rating is what makes it legally possible for a generic to be interchanged with its brand, bringing drug prices down for providers and patients alike.
The court found that, under this specific set of facts, where the label had not sufficiently carved out the patented use, an AB rating was evidence of infringement.
“Teva’s AB rated representations under these limited circumstances, when substantial evidence supports the jury’s presumed determination regarding the label’s contents, are further affirmative evidence supporting the jury’s inducement finding,” the court said.
In a footnote, it distanced these circumstances from a “true section viii carve-out” (i.e., a skinny-enough label, that did not reference any patent-infringing uses), saying Teva had instead used the AB rating to prop up its “partial label,” which the court said the jury was “free to credit” as evidence of induced patent infringement.
The court made significant mention of expert testimony from doctors about how — or whether — they read a generic drug’s label before prescribing it to patients.
The majority found testimony that Teva’s mention of its drug in relation to patients with cardiovascular disease compelling evidence that it induced use of the drug in patients with CHF, the one indication that was still patented. This despite the same doctors admitting they did not read labels before prescribing, as the dissent pointed out.
“The idea is the label doesn’t speak for itself,” Dmitry Karshtedt, a George Washington University professor, said in a phone interview. “The question becomes, how much can expert testimony move the needle? And here it did, making this a very fact-specific finding — or at least the court tries to pitch it that way.”
Photo: Gearstd, Getty Images