The Federal Circuit’s August 5, 2021 ruling in GlaxoSmithKline LLC v. Teva Pharm. USA, Inc. (link to issued opinion), is an important development in the law of “skinny” generic pharmaceutical labels and induced infringement claims. See GlaxoSmithKline LLC v. Teva Pharm. USA, Inc., 7 F.4th 1320 (Fed. Cir. 2021) (“GSK II”) (Google Scholar link). Read on for important aspects of the majority’s opinion and Judge Prost’s dissent.
Background on Section viii Carve Outs. ANDA filers can seek approval for generic products with labels that “carve out” patented indications. Those labels are commonly referred to as “skinny labels” or “section viii carve outs” (the applications require a statement under Section 505(j)(2)(A)(viii) of the FDCA, 21 U.S.C. § 355(j)(2)(A)(viii)).
FDA accepts a carved-out label if the omissions “do not render the proposed drug product less safe or effective than the listed drug for all remaining, nonprotected conditions of use.” 21 CFR 314.127(a)(7). Courts have also recognized the viability of this labeling practice. In 2003, for example, the Federal Circuit explained that a Hatch-Waxman infringement case is “limited to an analysis of whether what the generic drug maker is requesting authorization for in the ANDA would be an act of infringement if performed.” Warner–Lambert Co. v. Apotex Corp., 316 F.3d 1348, 1360, 1364–65 (Fed. Cir. 2003) (explaining that an ANDA applicant’s “‘paragraph IV certification’ . . . was effectively a statement of non-applicable use pursuant to 21 U.S.C. § 355(j)(2)(A)(viii)” and holding that “the request to make and sell a drug labeled with a permissible (non-infringing) use cannot reasonably be interpreted as an act of infringement (induced or otherwise) with respect to a patent on an unapproved use, as the ANDA does not induce anyone to perform the unapproved acts required to infringe”).
But since 2010, the Federal Circuit has issued several decisions upholding induced infringement claims against skinny labels for infringement of the carved-out uses. See, e.g., AstraZeneca LP v. Apotex, Inc., 633 F.3d 1042, 1060 (Fed. Cir. 2010) (finding induced infringement where “despite being aware of the infringement problem presented by the proposed label, Apotex nonetheless proceeded with its plans to distribute its generic drug product”); Eli Lilly & Co. v. Teva Parenteral Meds., Inc., 845 F.3d 1357, 1369 (Fed. Cir. 2017) (“[E]vidence that the product labeling that Defendants seek would inevitably lead some physicians to infringe establishes the requisite intent for inducement.”); Sanofi v. Watson Lab’y Inc., 875 F.3d 636, 645 (Fed. Cir. 2017) (finding induced infringement where the label “directs medical providers to information identifying the desired benefit for only patients with the patent-claimed risk factors” and “[t]here was considerable testimony that this label encourages . . . administration of the drug to those patients”).
First Panel Decision in GSK v. Teva. The August 5 decision in GSK II reconsidered the panel’s earlier decision from October 2020. The earlier ruling had reversed the District of Delaware’s grant of judgement as a matter of law (“JMOL”) in favor of Teva. GSK II, 7 F.4th at 1325–26.
Teva had marketed a generic version of GSK’s carvedilol product, Coreg®, since 2008, first with a carved-out “skinny label” and then with a complete label. Id. at 1324–25. Coreg® has three FDA-approved indications, two of which are off-patent. Id. at 1323. The third indication, treatment of congestive heart failure (“CHF”), was covered by method-of-use claims. Id. at 1324. Teva’s initial product label carved out the CHF indication but, in 2011, FDA instructed Teva to revise its label to match GSK’s label, which included the CHF indication. Id. Teva complied with FDA’s instructions and marketed the generic product with a complete label. Id. at 1325. GSK sued Teva in 2014 for induced infringement of the CHF method patent, which subsequently expired in 2015. Id. Under GSK’s infringement theory, both Teva’s carved-out and full labels induced physicians to prescribe the generic product for the patented CHF indication. Id. at 1327.
The district court (Stark, D. Del.) held a jury trial and the jury returned a verdict in favor of GSK for both the skinny-label (2008–2011) and full-label periods (2011–2015). Id. at 1325. It awarded $234M in damages. Id. at 1340. But Judge Stark granted Teva’s JMOL motion on inducement, holding that GSK had failed to provide sufficient evidence to establish that Teva caused direct infringement by physicians who had prescribed the generic product. Id. at 1325. GSK appealed.
In October 2020, the Federal Circuit reversed Judge Stark’s post-trial ruling. GlaxoSmithKline LLC v. Teva Pharm. USA Inc., 976 F.3d 1347, 1348 (Fed. Cir. 2020). The panel majority (Judge Newman joined by Judge Moore) held that the total evidence of record, including Teva’s labels and marketing materials, supported the jury’s verdict. Then-Chief Judge Prost dissented.
Although Teva sought en banc rehearing, the panel construed the petition “as also requesting panel rehearing” and granted the perceived request. GSK II, 7 F.4th at 1325–26. The same majority, this time per curiam, again reversed the JMOL ruling and reinstated the jury verdict. Judge Prost again dissented.
The majority acknowledged that its “prior decision could be read to upset the careful balance struck with section viii carve-outs.” Id. at 1326. It granted rehearing to clarify how the specific facts in the record supported the jury’s verdict, and contended that its “narrow, case-specific review of substantial evidence does not upset the careful balance struck by the Hatch-Waxman Act regarding section viii carve-outs.” Id.
The majority catalogued all the evidence presented at trial that could support the jury’s infringement verdict. In its view, substantial evidence showed that Teva’s carved-out label instructed doctors to perform the patented methods and thus “was not a skinny label.” Id. at 1328. The majority viewed it instead as an infringing “partial label.” Id. It focused on GSK’s expert testimony suggesting that Teva’s label, while omitting the CHF indication, nonetheless encouraged doctors to use the product for the patented CHF treatment. The label discussed a different indication, left-ventricular dysfunction after a myocardial infarction (“post-MI LVD”), and GSK’s expert explained that post-MI LVD “is intertwined with heart failure.” Id. Teva’s expert acknowledged that some post-MI LVD patients have CHF. That supported the verdict even though GSK’s FDA submissions did not identify the asserted CHF patent as related to the post-MI LVD indication. According to the majority, the jury was also entitled to weigh and discount that evidence given the limited scope of FDA’s review. Namely, FDA does not determine infringement, and FDA submissions do not substitute for an infringement analysis. Id. at 1332–33.
In addition, Teva’s marketing materials and references to its AB-rating could be viewed as promotion of the generic product for infringing use and thus affirmative evidence of inducement. In a footnote, however, the majority noted that it was not concluding “that an AB rating in a true section viii carve-out … would be evidence of inducement.” Id. n.7.
Finally, the majority explained that two Teva press releases supported the verdict. One release from 2004 said the generic product was indicated for treatment of heart failure, and the other from 2007 described the product as a “cardiovascular agent.” Id. at 1336. That term, GSK’s expert explained, conveyed to doctors that they could use the generic for all indications, including the covered CHF treatment. Although the press releases were published before the asserted patent issued, the jury heard evidence that the releases remained on Teva’s website until 2015 and thus could be used to support the verdict.
Turning to the full-label period, because the jury heard testimony about that label and that Teva instructed doctors to read and follow the label, the majority concluded that substantial evidence supported the finding of inducement. Id. at 1339.
On the issue of causation, Teva argued that doctors prescribed the generic product for CHF based on information received from GSK long before Teva’s generic launch, not based on anything Teva did. Id. at 1339. But the jury was instructed on causation and presumably rejected Teva’s factual argument. Id. at 1340. The evidence about Teva’s labels, promotional activities, and press releases provided substantial evidence to support the necessary causative link. Id.
Judge Prost again issued a lengthy and pointed dissent. She painstakingly evaluated the majority’s analyses of the record on intent and causation, especially as to the skinny-label period, and laid out the negative effects she believed the ruling will have on the pharmaceutical industry. According to Judge Prost, Teva “played by the rules” and came to market with a skinny label just as Congress had intended yet was found liable based on “thin to nonexistent” evidence of inducement. Id. at 1342.
She explained that “the majority further weaken[ed] the intentional-encouragement prong of inducement by effectively eliminating the demarcation between describing an infringing use and encouraging that use in a label,” and “eviscerate[ed] the causation prong of inducement.” Id. at 1343. In her view, “the majority create[d] confusion for generics, leaving them in the dark about what might expose them to liability,” upsetting “Congress’s design for enabling quick public access to generic versions of unpatented drugs with unpatented uses.” Id.