`
`No. 2022-1595
`
`United States Court of Appeals
`for the Federal Circuit
`
`
`GENENTECH, INC., INTERMUNE, INC.,
`Plaintiffs-Appellants
`
`v.
`
`SANDOZ INC., LEK PHARMACEUTICALS, D.D.,
`Defendants-Appellees
`
`
`
`
`
`
`
`
`Appeal from the United States District Court for the District of Delaware in
`C.A. No. 1:19-cv-00078-RGA, Judge Richard G. Andrews
`
`REPLY BRIEF FOR PLAINTIFFS-APPELLANTS GENENTECH, INC.
`AND INTERMUNE, INC.
`
`
`
`Kira A. Davis
`DURIE TANGRI LLP
`953 East 3rd Street
`Los Angeles, CA 90013
`Telephone: 213-992-4499
`Facsimile: 415-236-6300
`
`July 5, 2022
`
`
`
`Daralyn J. Durie
`Vera Ranieri
`DURIE TANGRI LLP
`217 Leidesdorff Street
`San Francisco, CA 94111
`Telephone: 415-362-6666
`Facsimile: 415-236-6300
`
`
`Counsel for Plaintiffs-Appellants
`Genentech, Inc. and InterMune, Inc.
`
`
`
`
`
`
`
`
`
`
`
`
`Case: 22-1595 Document: 25 Page: 2 Filed: 07/05/2022
`
`
`
`TABLE OF CONTENTS
`
`Page
`
`II.
`
`
`TABLE OF CONTENTS ............................................................................................ i
`TABLE OF AUTHORITIES ................................................................................... iii
`ARGUMENT ............................................................................................................. 2
`The District Court Erred in Finding the LFT Patents Not
`I.
`Infringed ................................................................................................ 2
`The district court’s legal error is not entitled to
`A.
`deferential review ........................................................................ 2
`The district court did not apply the correct legal standard ......... 4
`B.
`C. Applying the correct standard, Sandoz’s Label induces
`infringement ................................................................................ 8
`The District Court Erred in Finding the DDI Patents Not
`Infringed .............................................................................................. 11
`A.
`Sandoz’s Label recommends infringement ............................... 11
`B. Where the Label recommends infringement, that is
`dispositive ................................................................................. 12
`There is no requirement for real-world evidence to
`1.
`show direct infringement ................................................ 12
`An ANDA application resolves the infringement
`question when it provides a clear recommendation
`to infringe ........................................................................ 13
`Sandoz misreads Vanda .................................................. 16
`The district court erred in requiring Plaintiffs to
`provide evidence beyond the Label ................................ 17
`Reversal without remand is appropriate ................................... 20
`
`2.
`
`3.
`4.
`
`C.
`
`i
`
`
`
`Case: 22-1595 Document: 25 Page: 3 Filed: 07/05/2022
`
`
`
`TABLE OF CONTENTS (Cont’d)
`
`
`
`Page
`
`III. The District Court Erred in Finding the LFT Patents Invalid ............. 21
`The district court mis-applied the law of inherency and
`A.
`Sandoz cannot retroactively cure that legal defect ................... 22
`The district court erred in its evaluation of the Pirespa
`Label .......................................................................................... 24
`The district court opinion cannot support invalidating
`claims requiring re-escalation to a full dose ............................. 27
`CONCLUSION ........................................................................................................ 30
`CERTIFICATE OF COMPLIANCE ....................................................................... 31
`
`
`
`B.
`
`C.
`
`
`
`ii
`
`
`
`Case: 22-1595 Document: 25 Page: 4 Filed: 07/05/2022
`
`
`
`TABLE OF AUTHORITIES
`
`
`Cases
`AstraZeneca LP v. Apotex, Inc.,
`633 F.3d 1042 (Fed. Cir. 2010) .................................................................... 10, 20
`
`Page(s)
`
`In re Cyclobenzaprine Hydroxychloride Extended-Release Capsule
`Pat. Litig.,
`676 F. 3d 1063 (Fed. Cir. 2012) ......................................................................... 24
`Dynamic Drinkware, LLC v. Nat’l Graphics, Inc.,
`800 F.3d 1375 (Fed. Cir. 2015) .......................................................................... 18
`Eli Lilly & Co. v. Teva Parenteral Meds., Inc.,
`845 F.3d 1357 (Fed. Cir. 2017) ................................................................ 8, 15, 16
`Ferring B.V. v. Watson Lab’ys, Inc.-Fla.,
`764 F.3d 1401 (Fed. Cir. 2014) .............................................................. 14, 15, 17
`Glaxo, Inc. v. Novopharm, Ltd.,
`110 F.3d 1562 (Fed. Cir. 1997) ...................................................................passim
`GlaxoSmithKline LLC v. Teva Pharms. USA, Inc.,
`7 F.4th 1320 (Fed. Cir. 2021), reh’g and reh’g en banc denied, 25
`F.4th 949 (Fed. Cir. 2022) ................................................................................ 4, 7
`Grunenthal GMBH v. Alkem Lab’ys Ltd.,
`919 F.3d 1333 (Fed. Cir. 2019) .................................................................. 6, 7, 21
`HZNP Medicines LLC v. Actavis Lab’ys UT, Inc.,
`940 F.3d 680 (Fed. Cir. 2019) .......................................................................... 5, 6
`Karlin Tech., Inc. v. Surgical Dynamics, Inc.,
`177 F.3d 968 (Fed. Cir. 1999) ............................................................................ 20
`Microsoft Corp. v. i4i Ltd. P’ship,
`564 U.S. 91 (2011) .............................................................................................. 18
`Myco Indus., Inc. v. BlephEx, LLC,
`955 F.3d 1 (Fed. Cir. 2020) ................................................................................ 19
`
`iii
`
`
`
`Case: 22-1595 Document: 25 Page: 5 Filed: 07/05/2022
`
`
`
`TABLE OF AUTHORITIES (Cont’d)
`
`Page(s)
`
`Panduit Corp. v. Dennison Mfg. Co.,
`810 F.2d 1561 (Fed. Cir. 1987) .......................................................................... 24
`Par Pharm., Inc. v. Hospira, Inc.,
`835 F. App’x 578 (Fed. Cir. 2020) ..................................................................... 14
`Soverain Software LLC v. Newegg Inc.,
`705 F.3d 1333 (Fed. Cir. 2013), amended on reh’g, 728 F.3d 1332
`(Fed. Cir. 2013) ................................................................................................... 25
`Sunovion Pharms., Inc. v. Teva Pharms. USA, Inc.,
`731 F.3d 1271 (Fed. Cir. 2013) ...................................................................passim
`Takeda Pharms. U.S.A., Inc. v. W.-Ward Pharm. Corp.,
`785 F.3d 625 (Fed. Cir. 2015) ...................................................................... 15, 16
`Tech. Licensing Corp. v. Videotek, Inc.,
`545 F.3d 1316 (Fed. Cir. 2008) .................................................................... 18, 19
`Upjohn Co. v. Mova Pharm. Corp.,
`225 F.3d 1306 (Fed. Cir. 2000) .......................................................................... 29
`Vanda Pharms. Inc. v. W.-Ward Pharms. Int’l Ltd.,
`887 F.3d 1117 (Fed. Cir. 2018) ...................................................................passim
`Vanda Pharms. Inc. v. Roxane Lab’ys, Inc.,
`203 F. Supp. 3d 412 (D. Del. 2016) .............................................................. 13, 16
`Warner-Lambert Co. v. Apotex Corp.,
`316 F.3d 1348 (Fed. Cir. 2003) ............................................................................ 8
`Warner-Lambert Co. v. Teva Pharms. USA, Inc.,
`418 F.3d 1326 (Fed. Cir. 2005) .......................................................................... 20
`YBM Magnex, Inc. v. ITC,
`145 F.3d 1317 (Fed. Cir. 1998), overruled on other grounds by
`Johnson & Johnston Assoc. Inc. v. R.E. Serv. Co., Inc., 285 F.3d
`1046 (Fed. Cir. 2002) (en banc) ............................................................................ 3
`
`iv
`
`
`
`Case: 22-1595 Document: 25 Page: 6 Filed: 07/05/2022
`
`
`
`TABLE OF AUTHORITIES (Cont’d)
`
`Yee v. City of Escondido,
`503 U.S. 519 (1992) ............................................................................................ 19
`Statutes
`35 U.S.C. § 271(e)(2)(A) ......................................................................................... 20
`
`Page(s)
`
`
`
`v
`
`
`
`Case: 22-1595 Document: 25 Page: 7 Filed: 07/05/2022
`
`
`
`An ANDA applicant induces infringement when it includes label language
`
`that recommends, encourages, or promotes infringement. In reaching a contrary
`
`conclusion, the district court mis-applied controlling law.
`
`First, Sandoz’s Label induces infringement of the LFT Patents by including
`
`instructions to infringe. In adopting Sandoz’s argument that the Label cannot
`
`recommend, encourage, or promote infringement if it also proposes a non-
`
`infringing path, the district court erred as a matter of law. When the correct legal
`
`standard is applied to the undisputed contents of Sandoz’s Label, Sandoz induces
`
`infringement.
`
`Second, Sandoz’s Label also induces infringement of the DDI Patents.
`
`Sandoz escaped liability by arguing that Plaintiffs needed to come forward with
`
`real-world evidence of infringement. The district court faulted Plaintiffs because
`
`“[t]here was no evidence at trial of any patient receiving pirfenidone after being
`
`prescribed fluvoxamine, or of any patient taking [the two drugs] concurrently.”
`
`Appx33. This Court, however, has previously held where the label instructs,
`
`recommends, or encourages infringement, that is the end of the inquiry. In treating
`
`Sandoz’s Label as merely one piece of evidence relevant to the likelihood of future
`
`infringement—and not a particularly important one at that—the district court
`
`committed legal error. Sandoz’s other arguments aimed at avoiding reversal are
`
`likewise legally incorrect.
`
`1
`
`
`
`Case: 22-1595 Document: 25 Page: 8 Filed: 07/05/2022
`
`
`
`The district court also erred in finding the LFT Patents obvious. Sandoz
`
`fails to defend several of the district court’s findings, choosing instead to mis-
`
`characterize both what the district court found and what Plaintiffs argued. The
`
`record, however, reveals that the district court committed reversible legal error in
`
`reaching the conclusion that the LFT Patents were obvious.
`
`ARGUMENT
`
`I.
`
`The District Court Erred in Finding the LFT Patents Not Infringed
`A. The district court’s legal error is not entitled to deferential review
`
`Sandoz does not challenge the finding that Plaintiffs proved direct
`
`infringement with respect to the LFT Patents. Response Brief (“Opp’n”) 26. Nor
`
`does Sandoz dispute that where inducement is based on an ANDA-applicant’s
`
`label, the test is whether that label “encourages, promotes, or recommends”
`
`infringement. Compare Plaintiffs’ Opening Brief (“Br.”) 30, with Opp’n 26-27.
`
`There is also no dispute as to the contents of the Label. Appx13 (“The parties
`
`agree that Sandoz’s label . . . includes instructions for patients exhibiting Grade 2
`
`elevations in ALT and/or AST.”). Sandoz does not dispute that the language in the
`
`patent claims appears in the Label, Appx7171 (236:9-14), and that “[f]our of the
`
`five dose modification options provided in the Asymptomatic Section [of the
`
`Label] are covered by the Asserted Claims,” Appx14. Finally, the district court
`
`also found, and Sandoz does not dispute, that “some physicians will infringe by
`
`2
`
`
`
`Case: 22-1595 Document: 25 Page: 9 Filed: 07/05/2022
`
`
`
`implementing a dose modification covered by one of the Asserted Claims . . . .”
`
`Appx18.
`
`That leaves one question: does the undisputed language in the Label meet
`
`the legal test for encouragement, recommendation, or promotion? It does, as
`
`shown below. That is a legal question, and the district court’s failure to apply the
`
`law correctly is a legal error, entitled to no deference on appeal.
`
`Sandoz tries to turn that legal issue into a dispute of fact. In doing so,
`
`however, Sandoz mis-characterizes Plaintiffs’ arguments. Plaintiffs do not, for
`
`example, dispute that some of the relevant Label language is phrased in the passive
`
`voice while other language is more directive. Appx15. What Plaintiffs dispute is
`
`the legal effect of that language in ascertaining induced infringement. See infra
`
`Section I.B. That is a dispute of law.
`
`That conclusion is confirmed by the district court’s opinion, which is
`
`focused on interpreting various cases and not resolving disputed facts. “The
`
`meaning or interpretation of precedent is a question of law[.]” YBM Magnex, Inc.
`
`v. ITC, 145 F.3d 1317, 1320 (Fed. Cir. 1998), overruled on other grounds by
`
`Johnson & Johnston Assoc. Inc. v. R.E. Serv. Co., Inc., 285 F.3d 1046 (Fed. Cir.
`
`2002) (en banc). The district court’s discussion of the LFT Patents includes
`
`practically no citation to any record evidence other than the Label itself. See
`
`Appx12-18. The district court simply applied (incorrect) legal principles to the
`
`3
`
`
`
`Case: 22-1595 Document: 25 Page: 10 Filed: 07/05/2022
`
`
`
`undisputed contents of the Label. Appx15-18. The body of the opinion1 cites to
`
`non-Label evidence on one occasion, Appx18, and that evidence supports Plaintiffs
`
`because it confirms that the Label instructs physicians to infringe.2
`
`GlaxoSmithKline LLC v. Teva Pharmaceuticals USA, Inc., 7 F.4th 1320
`
`(Fed. Cir. 2021), reh’g and reh’g en banc denied, 25 F.4th 949 (Fed. Cir. 2022)
`
`(“GlaxoSmithKline” herein; in Opening Brief, “Glaxo”), is not to the contrary.
`
`That case involved a dispute as whether label language met the claim limitations
`
`despite Teva’s carve out—the “fact question” was “whether the post-MI LVD
`
`indication instructs a physician to prescribe carvedilol,” and whether that was
`
`“within the scope of the claims.” Id. at 1328, 1330-31. This case presents no
`
`similar fact question.
`
`B.
`
`The district court did not apply the correct legal standard
`
`Turning to the merits, the district court applied incorrect legal standards.
`
`First, it was legal error for the district court to require “active-voice verbs”
`
`to find induced infringement. Passive verbs can be recommendations, even where
`
`they appear in “clear contrast” with more “directive” language. Appx15. That is
`
`
`1 The district court also cited record evidence in two footnotes, see Appx14 n.5 &
`Appx17 n.6, but Plaintiffs do not challenge those factual findings here.
`2 The district court found that “evidence presented at trial confirmed that some
`physicians will infringe by implementing a dose modification covered by one of
`the Asserted Claims . . . .” Appx18. “Implement” generally means to put a plan
`into effect, i.e., to put the infringing plan in the Label into effect.
`
`4
`
`
`
`Case: 22-1595 Document: 25 Page: 11 Filed: 07/05/2022
`
`
`
`confirmed by Vanda Pharmaceuticals Inc. v. West-Ward Pharmaceuticals
`
`International Ltd., 887 F.3d 1117 (Fed. Cir. 2018) (“Vanda”). The asserted claim
`
`required “performing . . . a genotyping assay” to determine whether a patient had a
`
`poor metabolizer genotype, and if so, administering a half dose. Id. at 1121. The
`
`label in Vanda included directive language that mandated the dose reduction, but
`
`only passive-voice non-mandatory language regarding the genotyping test required
`
`by the patent claims: poor metabolizers “should have their dose reduced by one-
`
`half” and “[Genotyping tests] are available to identify” poor metabolizers. Id. at
`
`1131 (alterations in original) (emphases added). This Court held that these
`
`passive-voice references were sufficient to show inducement. Id. The district
`
`court erred in letting Sandoz off the hook for infringement because some of the
`
`Label’s recommendations were phrased in the passive voice.
`
`Second, the district court also erred in relying on HZNP Medicines LLC v.
`
`Actavis Laboratories UT, Inc., 940 F.3d 680 (Fed. Cir. 2019), to conclude that
`
`Sandoz’s Label does not “encourage” infringement. Appx16-18. The district court
`
`interpreted HNZP as holding that where a label does not require infringement, and
`
`what the label discusses is “broader” than infringement, the label does not
`
`encourage infringement because there is no instruction to infringe. Appx17. That
`
`general rule does not flow from HNZP. Critically, the reason that the label in
`
`HZNP did not encourage infringement is because it merely said “if” a patient
`
`5
`
`
`
`Case: 22-1595 Document: 25 Page: 12 Filed: 07/05/2022
`
`
`
`wanted to perform a first infringing step, “then” the patient should take another
`
`potentially infringing step. HZNP, 940 F.3d at 702. There was no instruction to
`
`perform the first step at all.
`
`The Label here is different. Unlike in HZNP, the Label is not agnostic as to
`
`the “if” step. Patients do not choose whether to experience the “if” step of an
`
`asymptomatic Grade 2 elevation—it is an inevitable consequence of treatment for
`
`some patients. And for those patients, Sandoz’s Label provides four explicit
`
`options, each of which infringes. Appx14 (listing the explicit options); see also
`
`Appx13 (“Sandoz’s label . . . includes treatment instructions for patients exhibiting
`
`Grade 2 elevations in ALT and/or AST.”).
`
`Third, the district court apparently concluded that the law requires that
`
`infringing methods must be recommended over all non-infringing ones for there to
`
`be induced infringement. See Appx15 (“while the Asymptomatic Section lists
`
`some dose modifications covered by the Asserted Claims as potential treatment
`
`options, it does not affirmatively recommend any of them.”). That was a legal
`
`error. Sandoz relies on Grunenthal GMBH v. Alkem Laboratories Ltd., 919 F.3d
`
`1333 (Fed. Cir. 2019), but in that case the label at issue did not “implicitly or
`
`explicitly encourage or instruct users to take action that would inevitably lead to
`
`the [infringing use].” Id. at 1340 (emphasis added). Here, however, the Label
`
`indisputably includes explicit instructions for how to treat asymptomatic patients
`
`6
`
`
`
`Case: 22-1595 Document: 25 Page: 13 Filed: 07/05/2022
`
`
`
`with Grade 2 elevations in liver function tests in ways that, if followed, cause
`
`infringement. See Appx13 (no dispute that the Label includes treatment
`
`instructions for patients exhibiting Grade 2 elevations); Appx14 (discussing the
`
`Label’s instructions for the four infringing methods and calling them “options” for
`
`treatment). And the district court found as a factual matter that “the evidence
`
`presented at trial confirmed that some physicians will infringe by implementing a
`
`dose modification covered by one of the Asserted Claims [of the LFT
`
`Patents] . . . .” Appx18 (emphasis added).
`
`To be sure, the district court also found that the Label implicitly allows for a
`
`non-infringing method of treatment, i.e., discontinuing treatment. See Opp’n 28
`
`n.11 (citing Appx14). But then, at best for Sandoz and unlike in Grunenthal, the
`
`Label here instructs both infringing options (continued treatment) and a non-
`
`infringing option (permanent discontinuation). The Label still induces
`
`infringement when treatment is continued. As Grunenthal noted, “[t]he pertinent
`
`question is whether the proposed label instructs users to perform the patented
`
`method.” 919 F.3d at 1339 (quoting AstraZeneca LP v. Apotex, Inc., 633 F.3d
`
`1042, 1060 (Fed. Cir. 2010)). Sandoz’s Label does. Indeed, GlaxoSmithKline
`
`rejected Sandoz’s argument that a recommendation not to infringe “somehow
`
`obviates infringement” where, as here, there is also a recommendation to infringe.
`
`7 F.4th at 1329-30.
`
`7
`
`
`
`Case: 22-1595 Document: 25 Page: 14 Filed: 07/05/2022
`
`
`
`Fourth, the district court also committed legal error by relying in part on the
`
`lack of prevalence of the infringing activity to find no infringement. Appx17 n.6;
`
`see Eli Lilly & Co. v. Teva Parenteral Meds., Inc., 845 F.3d 1357, 1368 (Fed. Cir.
`
`2017) (“[W]e have not required evidence regarding the general prevalence of the
`
`induced activity.”). Sandoz observes that this Court has held that “where a product
`
`has substantial non-infringing uses . . . , intent to induce infringement cannot be
`
`inferred . . . .” Opp’n 34 (quoting Warner-Lambert Co. v. Apotex Corp., 316 F.3d
`
`1348, 1363-64 [sic, 1365] (Fed. Cir. 2003)). The district court quoted Warner-
`
`Lambert for this proposition, too. Appx17-18 n.6. But Warner-Lambert was a
`
`situation where the label contained no discussion of the infringing method at all
`
`because the drug at issue was not approved for the patented use. See 316 F.3d at
`
`1364-65. There, the fact that a product had non-infringing uses was relevant to
`
`assess intent to induce the infringing conduct. Here, however, the Label explicitly
`
`instructs how to infringe. Thus, the district court’s and Sandoz’s “reliance on
`
`Warner-Lambert, an off-label use case, is misplaced.” Vanda, 887 F.3d at 1132.
`
`C. Applying the correct standard, Sandoz’s Label induces
`infringement
`
`Sandoz’s Label warns that some patients may experience elevated liver
`
`enzymes and recommends that “[t]he full daily dosage may be maintained, if
`
`clinically appropriate, or reduced or interrupted (e.g., until liver chemistry tests are
`
`within normal limits) with subsequent re-titration to the full dosage as tolerated.”
`
`8
`
`
`
`Case: 22-1595 Document: 25 Page: 15 Filed: 07/05/2022
`
`
`
`Appx16750. There is no dispute that this instructs how to implement the
`
`infringing methods. Appx13, Appx14. There is also no dispute that those
`
`instructions will be followed some of the time. Appx18.
`
`And there is more. Sandoz’s Label encourages the reader to “[c]onsider
`
`temporary dosage reduction, treatment interruption, or discontinuation for
`
`management of adverse reactions.” Appx16749 (emphasis added). The infringing
`
`methods are also “promoted” over permanently discontinuing treatment, as it is the
`
`infringing methods that are explicitly and indisputably disclosed. Appx14.
`
`Sandoz argues that none of this is a recommendation, citing the testimony of
`
`its expert. Opp’n 28 (citing Appx7290 (355:8-12)). The following colloquy
`
`occurred at trial:
`
`Q. So now, if we turn to the asymptomatic grade 2
`patients, are there any recommendations in the label for
`that you just point out?
`A. So I think that those are the bullets that are
`underneath the -- the introduction there. Discontinuing
`confounding medications, excluding other causes,
`monitoring the patient closely, including repeating the
`liver chemistry tests, and then we have the options down
`below for therapy.
`Appx7289 (354:14-21) (emphases added). The expert did observe that none of
`
`these “options” was “recommended over the other” because different infringing
`
`methods were mentioned in the Label and “[o]nly one thing can be done” at a time.
`
`Appx7290-7291 (355:8-356:10). Read fairly, Sandoz’s expert agreed that the
`
`9
`
`
`
`Case: 22-1595 Document: 25 Page: 16 Filed: 07/05/2022
`
`
`
`Label gives a number of recommended options, any one of which could be selected
`
`at a given time.
`
`It is of no moment that doctors retain discretion to pick the most appropriate
`
`option. The Label states “modify the dosage as follows,” then explicitly instructs
`
`doctors how to modify the dosages in infringing ways. This is more directive than
`
`the label in Vanda, which merely noted that tests (which needed to be taken in
`
`order to infringe) “are available.” Vanda, 887 F.3d at 1131. It is also more
`
`directive than the label in AstraZeneca, where, as Sandoz acknowledges, the label
`
`advised that the downward-titration that sometimes led to infringement was
`
`“desirable,” not mandatory. Opp’n 33-34 (quoting AstraZeneca, 633 F.3d at
`
`1057).
`
`Moreover, discretion to choose among options does not negate the
`
`recommendation, encouragement, or promotion of infringement. As Sandoz’s
`
`expert confirmed, only one path can be followed at a time. Appx7290-7291
`
`(355:20-356:10). Sandoz intends that prescribers pick an infringing path some of
`
`the time, and so intends that there be infringement. As this court said in Vanda,
`
`“[e]ven if not every practitioner will prescribe an infringing dose, that the target
`
`dose range ‘instructs users to perform the patented method’ is sufficient to ‘provide
`
`evidence of [the accused infringer’s] affirmative intent to induce infringement.’”
`
`887 F.3d at 1132 (quoting AstraZeneca, 633 F.3d at 1060).
`
`10
`
`
`
`Case: 22-1595 Document: 25 Page: 17 Filed: 07/05/2022
`
`
`
`II. The District Court Erred in Finding the DDI Patents Not Infringed
`Sandoz’s insistence that it can market an ANDA product with a label that
`
`recommends infringement and yet not be liable for infringement is contrary to
`
`Federal Circuit law.
`
`A.
`
`Sandoz’s Label recommends infringement
`
`In a footnote, Sandoz now claims that there might be a dispute as to whether
`
`the language in its Label corresponds to the DDI Patent claims. Opp’n 46 n.16.
`
`The language corresponds to the claimed dose modifications—discontinuing
`
`fluvoxamine (claim 6, ’383 patent), or concurrently administering fluvoxamine and
`
`a specific reduced dose of pirfenidone (claims 3 & 9, ’002 patent), and so
`
`recommends infringement. See Br. 13, 16-17. From Sandoz’s Label:
`
`
`
`
`
`
`
`“Discontinue fluvoxamine prior to administration of pirfenidone or
`reduce to 267 mg three times a day.” Appx16749.
`
`“Use of fluvoxamine or other strong CYP1A2 inhibitors should be
`discontinued prior to administration of pirfenidone and avoided during
`pirfenidone treatment. In the event that fluvoxamine or other strong
`CYP1A2 inhibitors are the only drug of choice, dosage reductions are
`recommended.” Appx16753.
`
`“2.4 Dosage Modification due to Drug Interactions,” “Strong CYP1A2
`Inhibitors (e.g., fluvoxamine, enoxacin),” “Reduce pirfenidone tablets
`to 267 mg three times a day (801 mg/day).” Appx16751.
`
`“The contents of the label itself may permit the inference of specific intent to
`
`encourage, recommend, or promote infringement.” Vanda, 887 F.3d at 1129. As
`
`the above shows, Sandoz’s Label recommends what the asserted claims require.
`
`11
`
`
`
`Case: 22-1595 Document: 25 Page: 18 Filed: 07/05/2022
`
`
`
`This is not a case where the label is silent on infringement or requires expert
`
`analysis to explain how the label’s indications match up against the indications in
`
`the patent.
`
`B. Where the Label recommends infringement, that is dispositive
`1.
`
`There is no requirement for real-world evidence to show
`direct infringement
`
`The district court held that “[t]he presence of language that ‘encourages,
`
`recommends or promotes’ an infringing use on a proposed label, without any
`
`additional evidence showing such an infringing use will in fact occur, especially
`
`where there is evidence that an infringing use likely will not occur, is insufficient
`
`for a finding of induced infringement.” Appx32 (emphasis added). That is not the
`
`law, and such a rule would conflict with this Court’s prior cases. In Vanda and
`
`other decisions, this Court did not require real-world evidence for the patent owner
`
`in a Hatch-Waxman case to succeed on its claim of induced infringement (with its
`
`subsidiary direct infringement component) even in cases where there was
`
`evidence that infringement might never occur in the real world. See Vanda, 887
`
`F.3d at 1129-30; see also Br. 36-39.
`
`Sandoz says that this case presents a “different situation” because here
`
`“there is evidence that no doctor would perform the claimed methods.” Opp’n 39.
`
`But in Vanda, the district court determined that “as a practical matter” doctors
`
`might not infringe the asserted claims by conducting the required genotyping,
`
`12
`
`
`
`Case: 22-1595 Document: 25 Page: 19 Filed: 07/05/2022
`
`
`
`Vanda Pharms. Inc. v. Roxane Lab’ys, Inc., 203 F. Supp. 3d 412, 434 (D. Del.
`
`2016) (“Vanda Pharms. Inc.”), and this Court still did not resolve whether there
`
`was evidence of real world infringement in the record. Vanda, 887 F.3d at 1129.3
`
`2.
`
`An ANDA application resolves the infringement question
`when it provides a clear recommendation to infringe
`
`Sandoz has not established that real-world evidence is required where the
`
`Label explicitly recommends infringement. Off-label cases are different, see
`
`Vanda, 887 F.3d at 1132-33, but this is not an off-label case, and Sandoz’s own
`
`authorities confirm the importance of this distinction.
`
`Sandoz relies on Glaxo, Inc. v. Novopharm, Ltd., 110 F.3d 1562 (Fed. Cir.
`
`1997) and its progeny as holding that it is always appropriate to look at evidence
`
`outside the label. However, as this Court has explained, in Glaxo the Court
`
`“endorsed the district court’s reference to evidence including biobatch data and
`
`actual samples of the generic composition . . . as relevant to the infringement
`
`inquiry because the ANDA specification itself did not resolve the question of
`
`infringement in the first instance. . . . However, the converse must also be true:
`
`if an ANDA specification defines a compound such that it meets the limitations of
`
`an asserted claim, then there is almost never a genuine issue of material fact that
`
`
`3 West-Ward contended that the claims “require that a single physician administer
`iloperidone to both poor and non-poor CYP2D6 metabolizers . . . .” Vanda, 887
`F.3d at 1129 (emphasis added). If so, there would have been no evidence of real-
`world infringement in the record—a dispute this Court saw no need to address.
`
`13
`
`
`
`Case: 22-1595 Document: 25 Page: 20 Filed: 07/05/2022
`
`
`
`the claim is infringed.” Sunovion Pharms., Inc. v. Teva Pharms. USA, Inc., 731
`
`F.3d 1271, 1279-80 (Fed. Cir. 2013) (emphases added) (internal citation omitted).
`
`Thus, as this Court elaborated in Sunovion, in cases like Glaxo, where the
`
`ANDA application does not resolve the question of infringement, evidence other
`
`than the ANDA application can be relevant. However, evidence outside of the
`
`ANDA application “cannot override the conclusion that when a drug manufacturer
`
`seeks FDA approval to market a generic compound within the scope of a valid
`
`patent, it is an infringement as a matter of law.” Id. at 1280 (emphasis added).
`
`That includes infringement of patents like these, claiming a use of a drug. See Par
`
`Pharm., Inc. v. Hospira, Inc., 835 F. App’x 578, 585-86 (Fed. Cir. 2020). “Even
`
`where internal documents suggest that a generic product will not meet a claim
`
`limitation in practice, representations about the ANDA’s scope control the
`
`infringement analysis.” Id. at 586. Infringement results even where the generic
`
`expects or even promises that infringement will not happen. “Simply saying ‘But I
`
`won’t do it’ is not enough to avoid infringement.” Sunovion, 731 F.3d at 1280.
`
`Ferring, on which Sandoz also relies, confirms this point. “In some cases,
`
`the ANDA specification directly resolves the infringement question because it
`
`defines a proposed generic product in a manner that either meets the limitations of
`
`an asserted patent claim or is outside the scope of such a claim. In cases in which
`
`the ANDA specification does not resolve the infringement question in the first
`
`14
`
`
`
`Case: 22-1595 Document: 25 Page: 21 Filed: 07/05/2022
`
`
`
`instance, we have endorsed the district court’s reference to relevant evidence . . . .”
`
`Ferring B.V. v. Watson Lab’ys, Inc.-Fla., 764 F.3d 1401, 1408-09 (Fed. Cir. 2014)
`
`(internal citations omitted).
`
`In short, where, as here, the ANDA application resolves the question of
`
`infringement, it does so “as a matter of law.” Sunovion, 731 F.3d at 1280. Even if,
`
`contrary to the trial record, Sandoz had proved that infringement likely will never
`
`happen, that would be akin to “saying ‘But I won’t do it,’” which is legally
`
`insufficient. Id. “If it had no intent to infringe, [Sandoz] should not have
`
`requested, or should not accept, approval to market a product within the scope of
`
`the claim.” Id. at 1279.
`
`Neither Eli Lilly nor Takeda is to the contrary.
`
`In Eli Lilly, the court looked to evidence outside the ANDA label to address
`
`the distinct issue of divided infringement. See Opp’n 37. In particular, real-world
`
`evidence was considered to determine whether “physicians ‘condition’ pemetrexed
`
`treatment on the admini