Induced Infringement: The Federal Circuit Addresses the Role of Skinny Labels in the Determination of Induced Infringement for ANDA Products

On October 2, 2020, the Federal Circuit issued a decision in GlaxoSmithKline LLC v. Teva Pharm. USA, Inc.1 that addressed whether a generic drug manufacturer induced infringement of a method of use patent when its product label for an AB-rated generic equivalent of a branded product initially omitted—but later added—an indication contained in the branded-product’s label that was claimed by the patent. Typically, a generic product is indicated for the same treatments as the branded product. However, in certain circumstances, a generic manufacturer may choose to omit an indication from its product label that the branded product is approved for in an attempt to avoid infringement of a method of use patent, which is known as a section viii “carve out” pursuant to 21 U.S.C. § 355(j)(2)(A)(viii).2 A generic product’s label with such a carve out may be referred to as a “skinny label.”3

In this case, Teva’s product label for its generic carvedilol product attempted to carve out an indication for treating congestive heart failure (“CHF”), which was one of three indications on the FDA-approved label for the branded product Coreg® (carvedilol), to avoid a method of use patent. At trial, despite the carve out, the jury found that Teva induced infringement of the method of use patent, found that the infringement was willful, and assessed lost profits and reasonable royalty damages.4 After the trial, the district court granted judgment as a matter of law (“JMOL”) of non-infringement, finding that the verdict was not supported by substantial evidence that would show that Teva “actually caused the physicians . . . to directly infringe” the patent by prescribing the product for CHF.5 On appeal, the Federal Circuit panel was divided on the induced infringement issue. The panel majority (Judges Newman and Moore) reversed the district court’s grant of JMOL, reinstated the jury verdict, and remanded the case to the district court.6 But there was a lengthy dissent (Chief Judge Prost) that opined that the majority’s “holding is counter to Congress’s intent and incorrectly concludes that the jury’s verdict was supported by substantial evidence.”7 The Federal Circuit panel also considered whether a jury instruction relating to damages under a lost profits analysis was proper despite instructing the jury that generic products sold by other manufacturers were not non-infringing substitutes. Without extensive analysis, the majority concluded that this instruction was in conformity with the law.8

a. Summary of Facts

GlaxoSmithKline (“GSK”) markets Coreg® to treat hypertension, left ventricle dysfunction following myocardial infarction, and CHF.9 GSK had a method of treatment patent, U.S. Patent No. 5,760,069 (“the ’069 patent”), listed in the FDA’s Orange Book for Coreg®.10 In March 2002, Teva filed an Abbreviated New Drug Application (“ANDA”) seeking approval to manufacture a generic version of Coreg®, which included a Paragraph IV certification that the ’069 patent is invalid, unenforceable, and not infringed.11 In November 2003, GSK filed an application to reissue the ’069 patent, which resulted in the patent-in-suit relevant to the appeal, U.S. Patent No. RE40,000 (“the ’000 patent”), issuing on January 8, 2008 with claims directed to a method of treating CHF.12

In 2004, Teva received tentative approval from the FDA for its ANDA to market carvedilol, a generic version of Coreg®, for all three of the branded product’s approved indications, i.e., hypertension, left ventricle dysfunction following myocardial infarction, and CHF.13 Teva announced its tentative approval with a press release, stating that its carvedilol product is the “AB-rated generic equivalent” of Coreg® and is “indicated for treatment of heart failure.”14 However, the final approval of Teva’s ANDA was delayed because a patent on the carvedilol compound was still in force and Teva had filed a Paragraph III certification that it would not launch a generic product until after the compound patent expired.15 Prior to receiving final approval, Teva amended its ANDA and product label to carve out the CHF indication that would later be claimed in the ’000 patent, thus resulting in Teva’s skinny label.16 Teva did not remove its tentative approval press release from its website.17

In 2007, the compound patent expired.18 Teva received final approval of its ANDA and launched its generic product with the carve out of the CHF indication.19 Both Teva and the FDA issued press releases announcing the approval of generic carvedilol.20 Teva’s press release apparently did not mention the CHF carve out, and stated that it had been “granted final approval” of its ANDA to market its generic version of Coreg®.21 The FDA’s press release announced the approval of “generic versions of Coreg (carvedilol)” and, despite the CHF carve out, stated that “Coreg is a widely used medication that is FDA-approved to treat” CHF and other indications.22 In 2011, the FDA required that Teva amend its carvedilol label so as to be “‘identical in content to the approved GSK Coreg® labeling’ . . . .”23 Teva did so. As the dissent noted, “[i]n marketing its generic carvedilol, Teva never stated that it was approved, or could be used, to treat CHF,” but in its product catalogs Teva listed carvedilol and “reported that the therapeutic equivalence rating was ‘AB’ and the ‘Brand’ was Coreg®.”24

In July 2014, GSK sued Teva for induced infringement of the ’000 patent.25 At this point in time, Teva’s carvedilol product had been on the market for several years. The jury found that Teva induced physicians to prescribe carvedilol, thus infringing the ’000 patent, and assessed damages.26 As noted above, the district court granted Teva’s JMOL of non-infringement after deciding that the jury’s finding of induced infringement was not supported by substantial evidence and lacked the requisite proof of causation.27 The district court decided that the “non-Teva factors were what caused the doctors to prescribe generic carvedilol for an infringing use.”28 The decision was appealed to the Federal Circuit.

b. The Majority Opinion

The majority explained that under the patent statute, “[w]hoever actively induces infringement of a patent shall be liable as an infringer.”29 The majority cited the Supreme Court’s Global-Tech decision, explaining “that copying of a patented product is evidence of inducing infringement.”30 Further, the majority quoted the Supreme Court’s MGM decision, “stating that ‘active steps . . . taken to encourage direct infringement, such as advertising an infringing use or instructing how to engage in an infringing use, show an affirmative intent that the product be used to infringe.’”31 The majority also stated that circumstantial evidence can be used to prove the intent element of induced infringement, noting that “‘we have affirmed induced infringement verdicts based on circumstantial evidence of inducement (e.g., advertisements, user manuals) directed to a class of direct infringers (e.g., customers, end users) without requiring hard proof that any individual third-party direct infringer was actually persuaded to infringe by that material.’”32 The majority also explained that “these principles have been applied to the circumstances of FDA-regulated products,” citing several cases concerning product labels.33

GSK argued that there was “substantial evidence” presented to the jury to “meet the legal requirements of active inducement of infringement.”34 Such evidence spanned Teva’s promotional materials, product catalogs, and press releases—many of which were available on Teva’s website for years—including, for example, a press release that referred to “Teva’s carvedilol as the ‘AB rated generic equivalent of GlaxoSmithKline’s Coreg® tablets.’”35 GSK’s physician expert testified that physicians are “‘completely reliant’ on information provided by the generic producers,” and that Teva had provided “‘lots of information . . . that indicated that . . . it was a complete replacement.’”36 The expert also testified that Teva’s catalog lists “Teva’s carvedilol tablets next to Coreg® tablets and uses the phrase ‘AB-rating,’ and that this would lead a doctor to believe that ‘they’re therapeutically interchangeable.’”37 Teva’s expert also testified that “if he just wrote Coreg on a prescription, the patient would get the generic unless he explicitly wrote ‘dispense as written’ or ‘DAW.’”38 When GSK’s physician expert was asked whether he had ever “‘come to believe that Teva’s generic carvedilol had not been approved for’” CHF, he testified that he “‘never knew it.’”39 In addition, GSK’s regulatory expert explained the AB-rating means that “‘if the generic drug is used in accordance with its label, you would expect it to have the same clinical effect’ as the brand drug.”40

Teva argued that it could not be liable for induced infringement because it “had deliberately omitted, or ‘carved out’ from its 2007 label, reference to [CHF].”41 However, a Teva witness testified that using the carve out provision to create a skinny label is “‘a legal strategy, not a commercial strategy,’” and confirmed that “Teva still expects to get sales where the doctor prescribed carvedilol for [CHF]” unless “‘the doctor feels strongly’” and writes “brand only.”42 Teva further argued that its 2004 and 2007 press releases should not be considered as evidence for its alleged induced infringement because the ’000 patent was not issued until January 8, 2008.43 However, the majority noted that “the evidence before the jury was that the 2007 press release remained on Teva’s website throughout the life” of the patent with the caption that “Teva Announces Approval and Shipment of Generic Coreg® Tablets.”44

The majority concluded that the “district court applied an incorrect legal standard” in granting Teva’s JMOL.45 The majority held that precedent “makes clear that when the provider of an identical product knows of and markets the same product for intended direct infringing activity, the criteria of induced infringement are met.”46 The majority found that “ample record evidence” had been presented to the jury, including press releases, promotional materials, product catalogs, and witness testimony for both sides, to support the jury’s finding that Teva induced infringement.47 Further, the majority found that the jury instruction “correctly” permitted a finding of induced infringement if Teva “‘continued to take an action that began before the ’000 patent issued, after the ’000 patent was issued on January 8, 2008, intending to cause the physicians to directly infringe by administering Teva’s carvedilol product.’”48

c. Dissent by Chief Judge Prost

Chief Judge Prost commenced her lengthy dissent by noting the long-standing practice of striving to achieve a “critical balance” between incentivizing innovation and preserving patent rights on the one hand, and public access to those rights upon expiration of the patent on the other.49 The dissent asserted that the majority’s decision misconstrued the facts and misapplied the law, which “undermines this balance by allowing a drug marketed for unpatented uses to give rise to liability for inducement . . . where causation has not been shown.”50 According to the dissent, Teva’s marketing of generic carvedilol tablets based upon its skinny label did not induce infringement by the physicians that prescribed the product.51 Rather, the dissent agreed with the district court’s “thoughtful and thorough opinion” concerning Teva’s JMOL, which in reversing the jury verdict “got it right” because there was insufficient evidence to establish “that Teva actually caused the doctors’ infringement” for either of Teva’s product labels (i.e., with and without the attempted carve out).52 The dissent stated that the majority’s holding is “no small matter: it nullifies Congress’s statutory provision for skinny labels—creating liability for inducement where there should be none.”53 To frame this critique of the majority’s opinion, the dissent first provided an overview of the Hatch-Waxman Act, noting that “[o]ne essential purpose was to ‘speed the introduction of low-cost generic drugs to the market,’”54 and stated that the majority’s holding “directly undermines Congress’s design.”55

The dissent continued the critique of the majority’s opinion by stating that “Teva did not launch its product with a label that was identical to GSK’s” and therefore, opined that the majority’s reliance on Vanda or Sanofi is misplaced as those cases concerned infringement of brand manufacturers’ patents based on ANDAs that “included a virtually identical label.”56 The dissent stated that when Teva launched its generic carvedilol tablets its label “did not suggest that it was approved to treat CHF at all, much less the ’000 patent’s narrow method of treating CHF,” and the only two indications in Teva’s label (hypertension and left ventricle dysfunction following myocardial infarction) “were not patented” and therefore, Teva’s skinny label did not infringe.57

Further, the dissent stated that the majority’s holding is “contrary to our caselaw.”58 The dissent first cited Warner-Lambert, in which the court held that a generic label may not be an act of inducement under § 271(b) under certain circumstances because a “‘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) . . . as the ANDA does not induce anyone to perform the unapproved acts required to infringe.’”59 The dissent also cited Takeda Pharmaceuticals, which found that a particular drug’s label did not induce infringement because “‘vague label language cannot be combined with speculation about how physicians may act to find inducement’ and held that to induce infringement of a patented method, a ‘label must encourage, recommend, or promote infringement.’”60 In the dissent’s view, Teva’s skinny label did none of these things and “did not even suggest the patented method; it said absolutely nothing about CHF.”61 The dissent stated that Teva exercising the carve-out mechanism provided by Congress “is not gamesmanship . . . [n]or is it infringement.”62 Finally, the dissent stated that Teva also can not be liable for induced infringement based upon an “expectation that ‘off-label’ sales of generic carvedilol would occur,” relying on Warner-Lambert (“[m]ere knowledge of possible infringement by others does not amount to inducement”) and Takeda Pharmaceuticals (the Hatch-Waxman Act “was designed to enable the sale of drugs for non-patented uses even though this would result in some off-label infringing uses”).63 The dissent focused a substantial amount of its review and analysis to assert that GSK failed to prove that Teva actually caused doctors to directly infringe the patented method either during the skinny or full label periods. The dissent concluded stating that “[w]ithout causation, GSK failed to prove inducement.”64

d. Conclusion

The GlaxoSmithKline LLC v. Teva Pharm. USA, Inc. decision offers guidance about asserting induced infringement claims for method of use patents in view of an ANDA with an attempted section viii carve out. Both the majority’s and dissent’s opinions applied very fact-based analyses, but they each did so with different interpretations of the law and how the facts should apply to such law. Given the tensions between the interpretations applied by the majority and the dissent, as well as the judges’ respective viewpoints about the import of the decision—the majority stated that the “implications of the dissent’s position are vast”65 and the dissent stated that the “[m]ajority’s holding that the content of Teva’s skinny label can itself establish inducement nullifies Congress’s provision for skinny labels”66 —it appears likely that this decision will be well-positioned as a candidate for en banc rehearing.

1GlaxoSmithKline LLC v. Teva Pharm. USA, Inc., No. 18-1976 (Fed. Cir. Oct. 2, 2020) (“Slip Op.”), available at see also --- F.3d ---, 2020 WL 5856868
2Under 21 U.S.C. § 355(j)(2)(A), an abbreviated application for a new drug (commonly referred to as an “ANDA”) “shall contain (i) information to show that the conditions of use prescribed, recommended, or suggested in the labeling proposed for the new drug have been previously approved for a [listed drug, i.e., the branded product],” and section viii (§ 355(j)(2)(A)(viii)) allows the ANDA-filer to include a statement that it is not seeking approval for a method of use that is claimed in a patent that is listed in the Orange Book for the branded product.
3Slip Op., Dissent at 7.
4Slip Op. at 7.
5Id. at 8 (emphasis in original).
6Id. at 3.
7Slip Op., Dissent at 4.
8Teva argued that the district court’s jury instruction concerning lost profits damages was incorrect because it “require[d] the jury to ignore the reality of the marketplace,” which consisted of multiple generic manufacturers of carvedilol “who had not been sued for infringement.” Slip Op. at 19. Teva had sought a jury instruction requiring that GSK prove that “for every infringing sale made by Teva, the direct infringer would have purchased the prescribed carvedilol as GSK’s Coreg® branded product, and not from another generic producer.” Slip Op. at 18. The majority concluded, however, that the “district court correctly instructed the jury that the availability of carvedilol from other generic producers is not a ‘non-infringing substitute.’” Slip Op. at 20. The dissent did not address this.
9See Coreg® Product Label and Full Prescribing Information, Revised Dec 2008, available at
10Slip Op. at 3-4.
11Id. at 4.
12Id. at 5.
13Slip Op., Dissent at 8-9.
15Slip Op., Dissent at 8-9.
16Id. at 9.
17Id. at 23.
18Slip Op. at 5.
19Slip Op., Dissent at 9.
21Slip Op. at 13, Dissent at 9.
22Slip Op., Dissent at 10.
23Slip Op. at 6 (modification in original incorporated).
24Slip Op., Dissent at 10 (emphasis in original).
25Slip Op. at 6.
26Id. at 7.
27Id. at 8.
29Id. at 10 (quoting 35 U.S.C. § 271(b)).
30Slip Op. at 11 (citing Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754, 770-71 (2011)).
31Id. (quoting MGM Studios Inc. v. Grokster, Ltd., 545 U.S. 913, 936 (2005)).
32Slip Op. at 11 (quoting Power Integrations, Inc. v. Fairchild Semiconductor, Int’l, Inc., 843 F.3d 1315, 1335 (Fed. Cir. 2016)).
33Slip Op. at 11-12 (citing Eli Lilly & Co. v. Teva Parenteral Meds., Inc., 845 F.3d 1357, 1369 (Fed. Cir. 2017); Sanofi v. Watson Labs. Inc., 875 F.3d 636, 645 (Fed. Cir. 2017); AstraZeneca LP v. Apotex, Inc., 633 F.3d 1042, 1060 (Fed. Cir. 2010); Mentor H/S, Inc. v. Med. Device All., Inc., 244 F.3d 1365, 1379 (Fed. Cir. 2001)).
34Slip Op. at 10.
35Id. at 12 (quoting Teva’s June 9, 2004 press release).
36Slip Op. at 13-14 (quoting trial transcript).
37Id. (modification in original, quoting trial transcript).
38Id. at 14 (quoting trial transcript).
39Id. at 14-15 (quoting trial transcript).
40Id. at 15 (quoting trial transcript).
41Slip Op. at 14.
42Id. (quoting trial transcript).
43Id. at 15.
45Id. at 16.
47Slip Op. at 16.
48Id. at 15 (quoting jury instructions).
49Slip Op., Dissent at 1.
50Id. at 1-2.
51Id. at 2-3.
52Id. at 2-3 (emphasis in original).
53Id. at 3.
54Slip Op., Dissent at 4 (citing Caraco Pharm. Labs., Ltd. v. Novo Nordisk, 566 U.S. 399, 405 (2012)).
55Id. at 7.
56Id. at 17-18 (citing Vanda Pharm., Inc. v. West-Ward Pharm. Int’l Ltd., 887 F.3d 1117, 1129 (Fed. Cir. 2018) and Sanofi v. Watson Labs. Inc., 875 F.3d 636, 646 (Fed. Cir. 2017)).
57Id. at 18 (emphasis in original).
58Id. at 19.
59Id. at 19 (citing Warner-Lambert Co. v. Apotex Corp., 316 F.3d 1348, 1364-65 (Fed. Cir. 2003)).
60Slip Op., Dissent at 19-20 (citing Takeda Pharm. U.S.A., Inc. v. West-Ward Pharm. Corp., 785 F.3d 625, 631-32 (Fed. Cir. 2015)).
61Id. at 20.
63Slip Op., Dissent at 20 (citing Warner-Lambert, 316 F.3d at 1364; Takeda Pharm., 785 F.3d at 631).
64Id. at 28.
65Slip Op. at 17.
66Slip Op., Dissent at 17.