APOTEX INC v. ACORDA THERAPEUTICS INC

Reset A A Font size: Print

United States Court of Appeals,Second Circuit.

APOTEX INC., et al., Plaintiffs–Appellants, v. ACORDA THERAPEUTICS, INC., Defendant–Appellee.

Docket No. 14–4353–cv.

Decided: May 16, 2016

Before JACOBS, LIVINGSTON and DRONEY, Circuit Judges. Keith D. Parr (with Joseph N. Froehlich, Scott B. Feder, Hugh S. Balsam, James T. Peterka, and Andy J. Miller on the brief), Locke Lord LLP, Chicago, IL, for Appellants Apotex Incorporated & Apotex Corporation. John W. Nields, Jr. (with Jason C. Raofield and Colin P. Watson on the brief), Covington & Burling LLP, Washington, D.C. for Appellee Acorda Therapeutics, Inc.

The parties are rival manufacturers of tizanidine, a drug for treating spasticity. Plaintiffs Apotex Incorporated and Apotex Corporation (collectively “Apotex”) allege that defendant Acorda Therapeutics, Inc. (“Acorda”) (i) filed a sham citizen petition with the Food and Drug Administration (“FDA”) to hinder approval of Apotex's competing formulation in violation of Section Two of the Sherman Act, and (ii) violated the Lanham Act's proscription on false advertising. As relevant here, the competition between the manufacturers focused on the relative efficacy of tablets or capsules in controlling somnolence, one of the side effects of tizanidine.

The United States District Court for the Southern District of New York (Swain, J.) ruled that the simultaneous approval by the FDA of Apotex's drug application and its denial of Acorda's citizen petition (raising concerns about the application) was by itself insufficient to support a Sherman Act claim. After discovery, the district court (Torres, J.) granted summary judgment and dismissed all of Apotex's false advertising claims on the grounds that (with the exception of one graph) no representation was literally false or likely to mislead consumers; and that, as to that one graph, Apotex failed to show that the false depiction would meaningfully impact consumers' purchasing decisions.

Apotex appeals both rulings, arguing that precedent from this Court allows its antitrust claim to survive dismissal and that material issues of fact pertinent to Acorda's representations remain for a jury to decide. We affirm these rulings.

Although precedent supports an inference that a citizen petition is an anticompetitive weapon if it attacks a rival drug application and is denied the same day that the application is approved, that inference has been undercut by recent FDA guidance. As to false advertising, we agree with the district court that no reasonable jury could have found that Acorda made literally false or misleading representations in its advertisements, with the exception of a single representation that Apotex has failed to show affected decisions to purchase.

BACKGROUND

Tizanidine tablets are used to treat spasticity, a symptom of multiple sclerosis and Parkinson's disease. One of the tablets' most common side effects is somnolence: sleepiness or drowsiness. Tizanidine tablets were first marketed in the United States by Elan Pharmaceuticals, Inc. (“Elan”), under the trade name “Zanaflex.” (Elan later sold its rights to Acorda.)

Zanaflex tablets were initially approved for sale by the FDA on November 27, 1996. In October 2001, Elan submitted a New Drug Application (“NDA”) to the FDA seeking approval to market tizanidine in capsule form. During its review, the FDA concluded that the absorption of the drug was delayed when tizanidine capsules were taken with food (rather than without), and that the delay was associated with a mean 20 percent decrease in Cmax, the peak amount of the drug in a subject's bloodstream. More importantly, the FDA found that “[w]hen bioequivalence of the capsule relative to the tablet is examined under fed conditions [i.e., with food], there is a delay in absorption and the mean Cmax for the capsule is approximately 2/3 of the mean Cmax for the tablet (Figure 1).” Joint Appendix at 2230 (emphasis added). The FDA subsequently approved Elan's NDA on August 29, 2002.

The significance of this phenomenon, in plain terms, is that the faster the drug is absorbed, the more drowsy the patient may become, whereas the side effect may be reduced if absorption is slowed.

While the NDA was pending, Elan filed a patent application for methods of administering tizanidine capsules to reduce somnolence and Cmax. Less than a month after the FDA approved the NDA, Elan's patent issued as United States Patent No. 6,455,557 (“the ′557 patent”). Elan then received permission to market its newly approved tizanidine capsules under the trade name “Zanaflex Capsules.” In July 2004, Acorda acquired the rights to Zanaflex tablets and Zanaflex Capsules; in April 2005, it launched the sale of Zanaflex Capsules. Apotex had begun selling its generic tizanidine tablet product in 2004, and was one of about ten companies to do so. At the time Acorda began selling Zanaflex Capsules, Apotex's tizanidine tablet product had a five percent market share.

Zanaflex Capsules on the open market carried an FDA label pertaining both to the Capsules and the tablets. The preamble to this label invites doctors to distinguish between tablets and Capsules, and between the drug when taken with food and without:

PHARMACOKINETIC DIFFERENCES BETWEEN ZANAFLEX CAPSULES™ AND ZANAFLEX® TABLETS: ZANAFLEX CAPSULES™ ARE NOT BIOEQUIVALENT TO ZANAFLEX® TABLETS IN THE FED STATE. THE PRESCRIBER SHOULD BE THOROUGHLY FAMILIAR WITH THE COMPLEX EFFECTS OF FOOD ON TIZANIDINE PHARMACOKINETICS (see PHARMACOKINETICS and DOSAGE AND ADMINISTRATION).

Joint Appendix at 643. The Pharmacokinetics section of the label, under the subheading “Pharmacokinetic differences between Zanaflex Capsules™ and Zanaflex® Tablets,” advises (consistent with the FDA review of the Zanaflex Capsules NDA) that there is a 30 percent increase in Cmax when the tablets are administered with food, but that when the Capsules are administered with food, Cmax decreases by 20 percent. “Consequently, the mean Cmax for the [C]apsule when administered with food is approximately 2/3's the Cmax for the tablet when administered with food.” Id.

The label contains a graph that figures in a number of the false advertising claims. Referred to as “Figure 1,” the graph is titled “Mean Tizanidine Concentration vs. Time Profiles for Zanaflex Tablets and Capsules (2 x 4 mg) Under Fasted and Fed Conditions.” It displays the mean plasma tizanidine concentration at various hours from dosing. The peak for the curve representing tizanidine capsules (taken with food) is lower, and occurs later than the peak for the curve charting concentration over time for tizanidine tablets (taken with food).

The label then explains, in the Dosage and Administration section, that pharmacokinetic differences between the fed and fasted state may affect the frequency and onset of certain adverse events. (The text is in the margin.1 ) Somnolence is explicitly identified as one of these adverse events.2

In 2007, Apotex filed an Abbreviated New Drug Application (“ANDA”)—a filing that seeks generic drug approval for an existing licensed medication or approved drug—in order to sell generic tizanidine capsules which would provide competition to Acorda's Zanaflex Capsules. In the ANDA, Apotex certified that it was not encroaching on any validly claimed intellectual property rights because the ′557 patent was invalid. In predictable response, Acorda filed a patent-infringement suit in 2007. After a seven-day bench trial, the United States District Court for the District of New Jersey (Brown, J.) ruled in September 2011 that the ′557 patent was invalid. See Acorda Therapeutics Inc. v. Apotex Inc., No. 07–4937 (GEB–MCA), 2011 WL 4074116, at *27 (D.N.J. Sept.6, 2011).

Soon after that ruling, Acorda filed a citizen petition with the FDA raising problems with Apotex's ANDA. The citizen petition is a means afforded by the FDA for raising concerns about products the FDA reviews; any individual may file such a petition concerning scientific or legal issues before or while the product is on the market. See 21 C.F.R. § 10.30. Conceptually, citizen petitions provide an avenue for public input into the drug approval process; but the process has been abused by pharmaceutical companies that file meritless petitions intended to delay approvals sought by their competitors and inhibit competition. Acorda's citizen petition objected to (1) Apotex's statement that its product was bioequivalent to Reference Listed Drugs (RLDs) in the fed state; and (2) allegedly misleading or untrue statements in the proposed label for the Apotex ANDA.

The FDA denied Acorda's citizen petition on February 3, 2012. That same day, the FDA approved Apotex's ANDA. The Sherman Act claim relies principally on the FDA's simultaneous dispositions.

With the green light from the FDA, Apotex launched its product. Acorda countered with its own authorized generic version of Zanaflex Capsules. Apotex contends that in the course of Acorda's marketing: (1) its representatives misrepresented to doctors that Zanaflex Capsules reduced Cmax—in comparison with the tablets—and then improperly used reduction in Cmax as a proxy for a corresponding decrease in somnolence; and (2) Acorda distributed written promotional materials to the same effect. In total, Acorda's advertising efforts contributed to sales of more than $240 million attributable to its version of Zanaflex Capsules.

Apotex commenced this lawsuit in December 2011, amending its complaint in February 2012 to include the FDA's denial of Acorda's citizen petition. Acorda countered with a motion to dismiss, which the district court (Swain, J.) granted with respect to the Sherman Act claim and denied with respect to the Lanham Act claims. The district court observed that Apotex's Sherman Act claim relied purely on temporal proximity—the denial of Acorda's citizen petition on the same day the Apotex ANDA was approved—and concluded that was insufficient to state a claim in view of recent legislation making the requirements for delaying an ANDA application more stringent: “Congress'[s] explicit directive that ANDA processing should not ordinarily be delayed by a citizen petition, coupled with its narrowing for the grounds for any such delay and the statutory notice requirement, strongly undermines any inference that mere simultaneity of ANDA and citizen petition decisions is indicative of the delay of one by reason of pendency of the other.” Apotex Inc. et al., v. Acorda Therapeutics, Inc., No. 1:11–cv–8803 (S.D.N.Y. Feb. 7, 2013), Doc. 45 at 7 (Swain, J.) (“Apotex I ”). Apotex moved for leave to amend, but the motion was denied in the interest of judicial economy.

After the case was reassigned to Judge Torres, Acorda successfully moved for summary judgment on the Lanham Act claims. Judge Torres began the analysis with the principle that: “[i]n the context of pharmaceutical drugs, courts have generally rejected Lanham Act claims based on advertisements that merely repeat labeling information that has been approved by the FDA.” Apotex Inc. v. Acorda Therapeutics, Inc., No. 11–cv–8803(AT), 2014 WL 5462547, at *3 (S.D.N.Y. Oct. 23, 2014) (Torres, J.) (“Apotex II ”) (quoting Mylan Pharms., Inc. v. Proctor & Gamble Co., 443 F.Supp.2d 453, 460 (S.D.N.Y.2006)). The court held that because many of the allegedly false representations were “consistent with the product label,” Apotex failed to identify a genuine issue of material fact with respect to falsity of any of Acorda's representations, except one. Id. at *5. The remaining representation was Acorda's placement of Cmax data on Figure 1, as to which the district court found that “a reasonable juror could determine that the juxtaposition of this text and image communicates a literally false message.” Id. at *8. Nevertheless, the claim did not survive because Apotex failed to identify a material issue of fact as to the representation's bearing on decisions to purchase. See id. at *9.

Apotex appeals both decisions.

DISCUSSION

Apotex's Sherman Act claim was dispatched at the motion-to-dismiss stage, while its Lanham Act claims succumbed on summary judgment. “On a motion to dismiss, all factual allegations in the complaint are accepted as true and all inferences are drawn in the plaintiff's favor.” Littlejohn v. City of N.Y., 795 F.3d 297, 306 (2d Cir.2015). We review de novo a “district court's grant of a motion to dismiss under Rule 12(b)(6).” Id. And we “review a district court's denial of leave to amend for abuse of discretion.” Presbyterian Church of Sudan v. Talisman Energy, Inc., 582 F.3d 244, 267 (2d Cir.2009).

Summary judgment “is appropriate when, having resolved all ambiguities and permissible factual inferences in favor of the party against whom summary judgment is sought, there are no genuine issues of material fact in dispute and the movant is entitled to judgment as a matter of law.” Baez v. JetBlue Airways Corp., 793 F.3d 269, 273 (2d Cir.2015). “A genuine issue of material fact exists ‘if the evidence is such that a reasonable jury could return a verdict for the nonmoving party.’ ” Savino v. City of N.Y., 331 F.3d 63, 71 (2d Cir.2003) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986)). The grant of summary judgment in favor of Acorda is likewise subject to de novo review. See Baez, 793 F.3d at 273–74.

I

The claim that Acorda's filing of a sham citizen petition constitutes illegal monopolization arises under Section Two of the Sherman Act, which imposes liability on: “[e]very person who shall monopolize, or attempt to monopolize ․ any part of the trade or commerce among the several [s]tates․” 15 U.S.C. § 2. A complaint alleging that a competitor filed a sham citizen petition with the FDA states a claim under that section. See In re DDAVP Direct Purchaser Antitrust Litig., 585 F.3d 677, 694 (2d Cir.2009). For these purposes, a single sham petition may be analogized to a single sham litigation. “A single lawsuit can violate antitrust law as long as it is both an objective and subjective sham.” Id. at 686. “First, the lawsuit must be objectively baseless in the sense that no reasonable litigant could realistically expect success on the merits․ [S]econd ․, the court should focus on whether the baseless lawsuit conceals ‘an attempt to interfere directly with the business relationships of a competitor.’ ” Prof'l Real Estate Inv'rs, Inc. v. Columbia Pictures Indus., Inc., 508 U.S. 49, 60, 113 S.Ct. 1920, 123 L.Ed.2d 611 (1993) (emphasis in original) (quoting E.R.R. Presidents Conference v. Noerr Motor Freight, Inc., 365 U.S. 127, 144, 81 S.Ct. 523, 5 L.Ed.2d 464 (1961)). Because we conclude that Apotex has failed to show that Acorda's citizen petition was objectively baseless, we affirm on the first ground and do not reach the second.

As relevant here, the factual basis for the claim in DDAVP was “that the sham petition caused a delay in generic competition, a possibility reinforced by the fact that the FDA approved the generic drug on the same day that it rejected the petition.” 585 F.3d at 694. Given the simultaneous grant of the ANDA and denial of the citizen petition, Apotex argues that a straightforward application of DDAVP justifies reversal: “the possibility that the petition was a sham, and that it impacted the FDA's decision, is sufficiently plausible to defeat the motion to dismiss.” Id. at 695.

However, the FDA issued a new Guidance for Industry (“Guidance”) explaining how the FDA determines if a petition implicates an issue of public health, and how the FDA's decision-making process is affected by the simultaneous pendency of an ANDA application and a citizen petition dealing with the same drug. Although this case partially arises on a motion to dismiss, we may properly take judicial notice of this document (without converting Acorda's motion to dismiss into a motion for summary judgment) because the Guidance is publicly available and its accuracy cannot reasonably be questioned. See Fed.R.Evid. 201(b); Staehr v. Hartford Fin. Servs. Grp., Inc., 547 F.3d 406, 425 (2d Cir.2008) (“Although the general rule is that a district court may not look outside the complaint and the documents attached thereto in ruling on a Rule 12(b) motion to dismiss, we have acknowledged that the court ‘may also consider matters of which judicial notice may be taken.’ ” (quoting Kramer v. Time Warner Inc., 937 F.2d 767, 773 (2d Cir.1991))).3

The Guidance explains: “If a petition requests that the Agency take an action related to a specific aspect of a pending application, we will consider the review status of the affected application(s) in determining whether it would be appropriate for the Agency to respond to the request to take the action requested in the petition within the 180–day timeframe.” Joint Appendix at 530 (emphasis added). Such consideration is necessary because of the differing procedural rights of an ANDA applicant and the writer of a citizen petition. “[T]he Agency must give the [ANDA] applicant notice of an opportunity for a hearing on whether the application is approvable, with a specific timeframe and process should the applicant request such a hearing.” Id. at 531. FDA decisions adjudicating citizen petitions, on the other hand, “are subject to immediate review by the courts. They therefore carry with them none of the procedural rights for the affected applicants that attach to a decision to deny approval of an application.” Id. To bring these procedural arrangements into sync, the FDA states that it is preferable not to issue a decision on a citizen petition until it issues a decision on the corresponding ANDA application. See id. (“If we were to respond substantively to a petitioner's request regarding the approvability of ․ the application as a whole, such [a] response could interfere with the statutory and regulatory scheme governing the review of applications and related procedural rights of applicants.”).

Although it remains conceivable, notwithstanding the Guidance, that a citizen petition might cause anticompetitive delay, the Guidance tends to undermine the inference (drawn in DDAVP and advocated now by Apotex) that when a citizen petition is denied simultaneously with the grant of an ANDA petition, the citizen petition was a sham and an anticompetitive weapon. The Guidance favors contemporaneous adjudications to safeguard the procedural rights of ANDA applicants such as Apotex.4

Apotex has pled no other facts from which it can plausibly be inferred that Acorda's petition was a sham. Apotex alleges that, in the patent trial, Acorda falsely mischaracterized testimony and scientific evidence relating to the bioequivalence of Apotex's product. But that allegation shows that Apotex disagreed with the arguments Acorda advanced in its citizen petition (which is hardly surprising), not that the Acorda citizen petition was a sham.5

The FDA letter rejecting Acorda's citizen petition reinforces this conclusion.6 Although Acorda's objections to Apotex's ANDA based on bioequivalence left the FDA unpersuaded, the grounds for rejection were: (1) Acorda's use of arithmetic mean values for Cmax instead of geometric mean values; and (2) Acorda's reliance on a cross-study comparison. Nowhere does the FDA find that Acorda misrepresented testimony in the Apotex–Acorda patent litigation. See Joint Appendix at 416–17 n. 5 (“The Agency is not a party to that proceeding, and statements made in private litigation are not directly relevant to FDA's statutory obligation to determine the approvability of Apotex's ANDA.”).

Similarly, with respect to Acorda's labeling argument, the FDA “disagree[d]” with Acorda that the Cmax of Apotex's product in the fed state was increased compared to the RLDs. Id. at 420. Apotex elides the distinction between arguments that fail to move the FDA and arguments that are false and objectively baseless. As the Supreme Court has cautioned, courts need to “ ‘resist the understandable temptation to engage in post hoc reasoning by concluding’ that an ultimately unsuccessful ‘action must have been unreasonable or without foundation.’ ” Real Estate Inv'rs, 508 U.S. at 60 n. 5 (quoting Christiansburg Garment Co. v. Equal Emp't Opportunity Comm'n, 434 U.S. 412, 421–22, 98 S.Ct. 694, 54 L.Ed.2d 648 (1978)). Accordingly, the objective component of Apotex's claim is lacking; so there is no need to evaluate Acorda's subjective motivations in filing the petition.

Apotex relies on In re Suboxone Antitrust Litig., 64 F.Supp.3d 665, 690–91 (E.D.Pa.2014), which is inapposite. The district court in Suboxone held that a Section Two claim based on the filing of a sham citizen petition survived dismissal because of the many indicia that the petition was objectively baseless. Among the indicia: the “FDA acknowledged in its ruling that it had no authority to grant much of [the] requested relief,” id. at 689; and the competitor that filed the citizen petition requested that “the FDA investigate why Suboxone tablets had been withdrawn from the market” even though the competitor “was continuing to sell the product at that time.” Id. at 690. It was therefore plausibly pled that the petition was “objectively baseless in that no reasonable litigant could have realistically expected success on the merits.” Id. Here, however, the only fact Apotex has pled other than the timing of the FDA's decision on Acorda's citizen petition is that Acorda's citizen petition was ultimately fruitless.7

In the alternative, Apotex argues that the district court's denial of leave to amend the antitrust claim was based on a misapplication of the rules set forth in the Southern District's Pilot Project Regarding Case Management Techniques for Complex Civil Cases (“Pilot Rules”). However, the district court denied the motion to amend in the interests of judicial economy, not because of Apotex's noncompliance with the Pilot Rules. Apotex sought to amend in order to cite the FDA's fourth annual report to Congress, issued in December 2012, detailing the ineffectiveness of the FDAAA. The district court found that this report was available to Apotex before its decision issued dismissing Apotex's claims in February 2013, that Acorda expressly relied on the FDAAA in its initial motion to dismiss Apotex's Sherman Act claim, and that Apotex brought the FDA annual report to the district court's attention only after Acorda's motion to dismiss was adjudicated. Because Apotex delayed seeking leave to include the FDA's fourth annual report until March 2013, and because the general findings of the report were largely immaterial to Apotex's specific claim, the district court concluded that the interests of judicial economy warranted denial of Apotex's motion. We have consistently afforded district courts latitude to deny leave to amend on this basis. See Ruffolo v. Oppenheimer & Co., 987 F.2d 129, 131 (2d Cir.1993) (“Where it appears that granting leave to amend is unlikely to be productive, however, it is not an abuse of discretion to deny leave to amend.”).

In sum, Apotex has not stated a claim under Section Two of the Sherman Act, and the district court did not abuse its discretion in denying Apotex leave to amend. The district court's decision dismissing Apotex's antitrust claim is affirmed.

II

The Lanham Act's prohibition on false advertising states:

Any person who, on or in connection with any goods or services ․ uses in commerce ․ any ․ false or misleading description of fact, or false or misleading representation of fact, which ․ in commercial advertising or promotion, misrepresents the nature, characteristics, qualities, or geographic origin of his or her or another person's goods, services, or commercial activities, shall be liable in a civil action by any person who believes that he or she is likely to be damaged by such act.

15 U.S.C. § 1125(a)(1). False advertising claims based on this proviso contain two components.

First (and obviously), a plaintiff bringing a false advertising claim must show falsity. There are two ways to do that. First, a “plaintiff can demonstrate that the challenged advertisement is literally false, i.e., false on its face.” Time Warner Cable, Inc. v. DIRECTV, Inc., 497 F.3d 144, 153 (2d Cir.2007). “[C]onsumer deception is presumed, and ‘the court may grant relief without reference to the advertisement's [actual] impact on the buying public.’ ” Id. (quoting Coca–Cola Co. v. Tropicana Prods., Inc., 690 F.2d 312, 317 (2d Cir.1982)). This inquiry requires evaluating “the message conveyed in full context.” Id. at 158 (quoting Castrol Inc. v. Pennzoil Co., 987 F.2d 939, 946 (3d Cir.1993)). “If the words or images, considered in context, necessarily imply a false message, the advertisement is literally false and no extrinsic evidence of consumer confusion is required.” Id. Importantly, however, “ ‘only an unambiguous message can be literally false,’ ”; “if the language or graphic is susceptible to more than one reasonable interpretation, the advertisement cannot be literally false.” Id. (quoting Novartis Consumer Health, Inc. v. Johnson & Johnson–Merck Pharms. Co., 290 F.3d 578, 587 (3d Cir.2002)). One kind of literally false claim is a claim of test-proven superiority. The premise is that the “defendant's ad[vertisement] explicitly or implicitly represents that tests or studies prove its product superior” and “plaintiff satisfies its burden by showing that the tests did not establish the proposition for which they were cited.” Castrol, Inc. v. Quaker State Corp., 977 F.2d 57, 63 (2d Cir.1992).

“Alternatively, a plaintiff can show that the advertisement, while not literally false, is nevertheless likely to mislead or confuse consumers.” Time Warner, 497 F.3d at 153. Such an implicit falsity claim requires “a comparison of the impression [left by the statement], rather than the statement [itself], with the truth.” Id. (quoting Schering Corp. v. Pfizer Inc., 189 F.3d 218, 229 (2d Cir.1999)). “[W]hereas ‘plaintiffs seeking to establish a literal falsehood must generally show the substance of what is conveyed, ․ a district court must rely on extrinsic evidence [of consumer deception or confusion] to support a finding of an implicitly false message.’ ” Id. (second alteration in original) (quoting Schering Corp., 189 F.3d at 229).

Falsity alone does not make a false advertising claim viable; “[u]nder either theory, the plaintiff must also demonstrate that the false or misleading representation involved an inherent or material quality of the product.” Id. n. 3. Such a “requirement is essentially one of materiality, a term explicitly used in other circuits.” S.C. Johnson & Son, Inc. v. Clorox Co., 241 F.3d 232, 238 (2d Cir.2001) (quoting Nat'l Basketball Ass'n v. Motorola, Inc., 105 F.3d 841, 855 (2d Cir.1997)). This Court has defined materiality as “likely to influence purchasing decisions,” a definition in harmony with other Circuits' use of the term. Nat'l Basketball Ass'n, 105 F.3d at 855 (quoting Am. Tel. & Tel. Co. v. Winback & Conserve Program, Inc., 42 F.3d 1421, 1428 n. 9 (3d Cir.1994)).8

A majority of Apotex's claims attack Acorda's representations as inconsistent with the FDA-approved label for Zanaflex Capsules. A prior question is whether representations that are wholly consistent with an FDA label are subject to Lanham Act liability. This Court has yet to so hold, although a number of district courts in this Circuit have sensibly adhered to this principle.9 We agree with these courts and now hold that representations commensurate with information in an FDA label generally cannot form the basis for Lanham Act liability.10

Such a rule reflects proper “deference to the expertise” of the FDA as the regulatory agency responsible for issuing the label by respecting the exhaustive process preceding the issuance of a label. Am. Home Prods. Corp. v. Johnson & Johnson, 672 F.Supp. 135, 144 (S.D.N.Y.1987) (Conner, J.); see also Smithkline Beecham, 1996 WL 280810, at *13. This principle rightfully insulates pharmaceutical companies from liability when they engage in First Amendment speech that is consistent with the directive of the regulatory body having oversight of product labels.

We “have been careful not to permit overextension of the Lanham Act to intrude on First Amendment values.” Groden v. Random House, Inc., 61 F.3d 1045, 1052 (2d Cir.1995). Accordingly, in order to avoid chilling speech that ought to be protected, Acorda's advertisements cannot form the basis for Apotex's claims to the extent they were in line with the FDA-approved label.

Apotex, however, goes further and contends that Acorda's advertisements exceeded the boundaries imposed by the FDA label. Each of Apotex's specific Lanham Act challenges is now considered.

1. Statements by Acorda's Sales Representatives

Apotex challenges statements, attributable to Acorda sales representatives, that patients taking Zanaflex Capsules with food enjoy the benefits of reduced Cmax, including dosing flexibility and diminished somnolence. Representative samples of these sales pitches include the following:

Upon learning of the ability to decrease somnolence ․ by using Zanaflex Capsules, the doctors agreed to give it a try.

Joint Appendix at 3072;

I explained to [the doctor] that the [C]apsule[ ] [is] really unique in that it counteracts a lot of the drowsiness when you dose it with food. He said he would give that a try and see how well it works for his patients.

Joint Appendix at 3093.

When Acorda learned that its representatives may have made unauthorized promotional claims for Zanaflex Capsules, Acorda's head of sales sent a memorandum to the sales team explicitly forbidding promotions that Zanaflex Capsules had fewer side effects and less sedation than the tablets.

Apotex specifically objects to representations that Zanaflex Capsules provide more flexibility than the Capsules' counterpart—the tablets—a point on which the FDA label is silent. However, statements that Zanaflex Capsules reduce Cmax when taken with food are fully consistent with the FDA label, as the district court correctly found, and Lanham Act liability therefore cannot attach to these statements. See Apotex II, 2014 WL 5462547, at *5. To the extent Apotex challenges statements about dosing flexibility, it misconstrues its burden. It is not enough to show that Acorda made representations absent from the FDA label; instead, Apotex must show that these comments were inconsistent with the FDA label in a manner sufficient to support a false advertising claim. See Procter & Gamble Co. v. Chesebrough–Pond's Inc., 747 F.2d 114, 119 (2d Cir.1984) (“[E]ach plaintiff bears the burden of showing that the challenged advertisement is false and misleading, not merely that it is unsubstantiated by acceptable tests or other proof.” (internal citations omitted)). Apotex has adduced no evidence that Acorda's promotion of dosing flexibility was either literally false or likely to cause consumer confusion.

Apotex next argues that Acorda's sales representatives used Cmax as a proxy for somnolence and improperly claimed that Zanaflex Capsules reduce it. Critically, Apotex relies on a test-proven superiority argument—that the representations relied on tests or studies allegedly proving the superiority of Zanaflex Capsules. The district court, however, properly rejected liability under the test-proven superiority theory.

The theory comes into play only when the “defendant's ad[vertisement] explicitly or implicitly represents that tests or studies prove its product superior․” Castrol, 977 F.2d at 63 (emphasis added). Apotex proffers no evidence that sales representatives referred to tests or studies when they discussed the potential of Zanaflex Capsules to reduce somnolence. At most, Acorda's representatives used Figure 1 (or a similar graph of pharmacokinetic results) as a tool to aid their reduced somnolence pitch. There is no record evidence that Acorda representatives used graphs of pharmacokinetic results to represent that Zanaflex Capsules reduced somnolence. Consider one of the statements at issue:

Talked to Dr. Corondan for the first time. I asked him about the most common complaint with Zanaflex tablets and he said the drowsiness and then we went to the graph and I discussed the [C]apsules (which is how I love a call to work out).

Joint Appendix at 1520. At no point in this passage is it explicitly stated or implied that Figure 1 necessarily shows that Zanaflex Capsules reduce somnolence; instead, the graph was used as a tool for further discussion. The district court's reasoning on this issue was sound: “[a]t most, the statements suggest that, due to pharmacokinetic differences between the products, Zanaflex [C]apsules cause less somnolence than Zanaflex tablets when taken with food. The statements do not, by contrast, ‘explicitly or implicitly represent[ ] that tests or studies prove’ that there is less somnolence with Zanaflex [C]apsules.” Apotex II, 2014 WL 5462547, at *7 (third alteration in original) (quoting Castrol, 977 F.2d at 63). Unable to identify a misrepresentation based on test-proven superiority, Apotex is left to find a genuine issue of material fact on falsity on some other theory—something it has failed to do. It is immaterial that no study has shown a reduction in somnolence associated with Zanaflex Capsules; Chesebrough makes clear that Apotex must show falsity, not merely uncertainty. Moreover, Acorda's somnolence representations find a harbor in the FDA label, which states that increased adverse events, such as somnolence, may occur when switching between the tablets and Capsules in the fed state. Reinforcing this conclusion, the FDA's medical review concluded that the most problematic situation in terms of exacerbating somnolence was switching from the Capsules to the tablets in the fed state. In other words, the FDA has given some support to Acorda's representations; more importantly, however, there is no evidence that the representations are false. There is no basis to disturb the grant of summary judgment relating to Acorda's sales representatives' statements.

2. Acorda's Promotional Materials

Supplementing representations by Acorda's sales team were written promotional materials heralding the benefits of Zanaflex Capsules in a manner Apotex believes was false and misleading. Apotex specifically attacks Acorda's “gatefold brochure,” a piece of advertising that Acorda disseminated in the thousands. The front cover announces: “Flexible Control in a Capsule” directly above two images of the sun and the moon with the words “DAY” and “NIGHT” printed underneath. The bottom of the front cover urges: “For Treatment of Spasticity When Relief is Most Important.” The second page contains Figure 1 juxtaposed with relevant Cmax data, along with additional text at the bottom of the graph. The text reads:

Effects and Adverse Events are Dose Related to Plasma Levels of Tizanidine.

• Significant pharmacokinetic changes including plasma level differences occur when administering Zanaflex Capsules or tablets with food.

• These (pharmacokinetic) differences can result in clinically important differences in effectiveness and adverse events.

Joint Appendix at 3327–28. See Appendix A.

Apotex asserts that the brochure is misleading as a whole because the sun and moon imagery, with the text underneath the graph on page 2 of the brochure, delivers the message that Zanaflex Capsules undoubtedly reduce Cmax and somnolence.

Although Acorda argues that it merely reprinted the graph from the FDA label in its advertisements (including the gatefold brochure) Apotex accuses Acorda of manipulation because the following text is superimposed on the graph: “30% INCREASE FOR TABLETS”; “20% DECREASE FOR CAPSULES.” Joint Appendix at 3258. Apotex argues that the graphic, as a whole, conveys a false message because Figure 1 depicts mean tizanidine concentration, which is the average drug concentration at different points in time, while the text relates to Cmax data, which is the maximum drug concentration at different time points and, by definition, different from the mean drug concentration. See Appendix B.

With the exception of the version of Figure 1 with superimposed text, Apotex fails to create a triable issue of fact as to the falsity of the brochure. Literal falsity cannot be shown because no unambiguous message is conveyed by the remainder of the brochure; Apotex argues that the presence of the sun and the moon implies that the drug is equally effective during the day and at night and thereby implies efficacy in combating somnolence. This conclusion is plausible, but it is not unambiguous, especially because the cover never mentions somnolence. Nor has Apotex shown extrinsic evidence of consumer confusion with respect to the sun-moon imagery. Apotex relies on internal marketing statements from Acorda focusing on reduced Cmax and somnolence; but Acorda's motivations for launching the gatefold brochure do not constitute extrinsic evidence as required.

Apotex contends that the district court erred by examining the brochure in isolation while ignoring the context of Acorda's launch letters and other documents detailing Acorda's marketing efforts. True, a “district court evaluating whether an advertisement is literally false ‘must analyze the message conveyed in full context.’ ” Time Warner, 497 F.3d at 158 (quoting Pennzoil Co., 987 F.2d at 946). But the relevant context of the advertisement is the overall message conveyed by the brochure. The district court was not required to consider external marketing documents. A review of the brochure in its entirety does not change the conclusion reached here: there is no unambiguous message that Zanaflex Capsules reduce somnolence nor has Apotex proffered evidence of consumer confusion on this point.

Acorda's use of the version of Figure 1 with superimposed text, seen on the second page of the gatefold brochure, raises a closer issue.

• Figure 1 shows the average concentration for a group of subjects over time after the drug is administered (i.e., the mean drug concentration); while

• Cmax is the maximum concentration of the drug that is reached in a subject, which varies from subject to subject and is not correlated with the time elapsed from administration of the drug.

Since Cmax values are not time-dependent, and Figure 1 displays mean drug concentration over time, Figure 1 cannot display mean CMax values. We therefore agree with the district court that “a reasonable juror could determine that the juxtaposition of this text and image communicates a literally false message.” Apotex II, 2014 WL 5462547, at *8. However, falsity is not enough; Apotex must also raise a factual issue as to materiality. “Under either theory, the plaintiff must also demonstrate that the false or misleading representation involved an inherent or material quality of the product.” Time Warner, 497 F.3d at 153 n. 3 (emphasis added). Apotex contends that it is not required to show that the relevant misrepresentation would have an effect on consumers' purchasing decisions. This argument ignores precedent from this Court which endorsed that definition of materiality—in line with the vast majority of our sister circuits. See Nat'l Basketball Ass'n, 105 F.3d at 855. Apotex counters that, when an advertisement is literally false, “consumer deception is presumed, and ‘the court may grant relief without reference to the advertisement's [actual] impact on the buying public.’ ” Time Warner, 497 F.3d at 153 (alteration in original) (quoting Coca–Cola Co., 690 F.2d at 317). The argument conflates falsity with materiality. Once literal falsity is proved, there is no requirement of extrinsic evidence showing consumer deception. But Apotex is not thereby relieved of the burden of showing materiality, which requires that the allegedly “false or misleading representation involved an inherent or material quality of the product,” id. n. 3—i.e., that the representation was “likely to influence purchasing decisions,” Nat'l Basketball Ass'n, 105 F.3d at 855 (quoting Am. Tel. & Tel., 42 F.3d at 1428 n. 9).

Applying the materiality standard, the district court concluded that, at most, Acorda “overstated the increase in mean tizanidine plasma concentration” but that this evidence ultimately “does not reveal anything about the impact on consumers' purchasing decisions.” Apotex II, 2014 WL 5462547, at *9. This conclusion was sound; the only plausible effect attributable to the misrepresentation in the graph was an exaggeration of the scale of the mean drug concentration curves, or an improper conflation of the mean Cmax with the highest mean drug concentration for a given treatment. However, there is no record evidence that this inaccuracy would dissuade consumers from purchasing Zanaflex Capsules. Certainly, Apotex has provided none. Apotex's showing on this point consists of generalized evidence that Acorda's increased sales of Zanaflex Capsules stemmed from its advertisement efforts. Apotex fails to make the necessary showing that the specific misrepresentation in the graphic—in any of Acorda's advertisements—was likely to influence consumers' purchasing decisions.

The remainder of Apotex's attacks on Acorda's promotional materials suffer from a common flaw: although Acorda unquestionably made statements that were not drawn from the FDA label, the thrust of Chesebrough is that this fact is insufficient to show falsity. A pharmaceutical company is entitled to make advertising statements outside the four corners of an FDA label so long as none of its representations is inconsistent with it.

In sum, the district court correctly granted summary judgment on all of Apotex's false advertising claims.

CONCLUSION

For the foregoing reasons, we AFFIRM the judgment of the district court.

FOOTNOTES

1.  “These pharmacokinetic differences may result in clinically significant differences when [1] switching administration of the tablet between the fed or fasted state, [2] switching administration of the capsule between the fed or fasted state, [and][3] switching between the tablet and capsule in the fed state․ These changes may result in increased adverse events or delayed/more rapid onset of activity, depending upon the nature of the switch. For this reason, the prescriber should be thoroughly familiar with the changes in kinetics associated with these different conditions․” Joint Appendix at 646.

2.  The FDA's Medical Review of the Zanaflex Capsules NDA observed: “The most problematic potential situation is that of patients switching from the capsule formulation to the tablet formulation in the fed state, where there is a risk of more severe side effects with excessive hypertension and somnolence.” Id. at 2102 (emphasis omitted).

3.  The document can be accessed at the following website: Food and Drug Administration, Citizen Petitions and Petitions for Stay of Action Subject to Section 505(q) of the Federal Food, Drug, and Cosmetic Act–Guidance for Industry (June 8, 2011), http://www.regulations.gov/# !documentDetail; D=FDA–2009–D–0008–0011.

4.  While DDAVP was pending on appeal, Congress promulgated the Food and Drug Administration Amendments Act (“FDAAA”), which provides in part that consideration of a citizen petition shall not delay FDA approval of an ANDA unless a “delay is necessary to protect the public health.” 21 U.S.C. § 355(q)(1)(A)(ii). If the FDA determines that delay is necessary, it must give notice to the applicant “not later than 30 days after making such [a] determination” along with a brief summary of the “specific substantive issues raised in the petition which form the basis of the determination.” Id. § (B). However, it is not evident that the FDAAA has curbed all abuses of the citizen petition process. See Michael Carrier & Daryl Wander, Citizen Petitions: An Empirical Study, 34 Cardozo L.Rev. 249, 283 (2012) (“[W]hat is clear is that the 2007 amendment has not been successful in achieving its stated purposes․ [T]o date, the amendment has not reduced the number of unsuccessful (in other words, denied or essentially denied) citizen petitions that appear to be filed to delay generic competition.”). We cannot foreclose the possibility that Sherman Act liability might be predicated on the time taken for the FDA to decide whether a citizen petition raises a legitimate public health concern. But, for reasons explained below, Apotex has not plausibly pled any such anticompetitive delay here.

5.  Apotex's assertions that Acorda falsely mischaracterized trial testimony and misrepresented Apotex's bioequivalence studies are legal contentions, not factual allegations, and therefore need not be credited. See Bell Atl. Corp. v. Twombly, 550 U.S. 544, 555, 127 S.Ct. 1955, 167 L.Ed.2d 929 (2007) (“[O]n a motion to dismiss, courts ‘are not bound to accept as true a legal conclusion couched as a factual allegation.’ ” (quoting Papasan v. Allain, 478 U.S. 265, 286, 106 S.Ct. 2932, 92 L.Ed.2d 209 (1986))).

6.  We may consider this document because the pleading “ ‘relies heavily upon its terms and effect,’ which renders the document ‘integral’ to the complaint.” Chambers v. Time Warner, Inc., 282 F.3d 147, 153 (2d Cir.2002) (quoting Int'l Audiotext Network, Inc. v. Am. Tel. & Tel. Co., 62 F.3d 69, 72 (2d Cir.1995)).

7.  Apotex's reliance on Tyco Healthcare Grp. LP v. Mut. Pharm. Co., 762 F.3d 1338 (Fed.Cir.2014) is misplaced. The Federal Circuit found a disputed issue of fact as to whether Tyco's citizen petition was objectively baseless because the FDA's response stated that the petition “relie[d] entirely on uncorroborated generalities and theoretical speculation” and “fail[ed] to provide any evidence at all” on a key issue. Id. at 1347. Furthermore, Mutual provided an expert who testified that the citizen petition was baseless. See id. In contrast, the FDA's letter to Acorda provided no such language indicating the citizen petition was wholly baseless, and Apotex did not plead any other facts that would indicate baselesness.

8.  See also Grubbs v. Sheakley Grp., Inc., 807 F.3d 785, 798 (6th Cir.2015) (defining “material” in the false advertising context as likely to “influence the deceived consumer's purchasing decisions”) (quoting Am. Council of Certified Podiatric Physicians & Surgeons v. Am. Bd. of Podiatric Surgery, Inc., 185 F.3d 606, 613 (6th Cir.1999)); Cashmere & Camel Hair Mfrs. Inst. v. Saks Fifth Ave., 284 F.3d 302, 311 (1st Cir.2002) (“The materiality component of a false advertising claim requires a plaintiff to prove that the defendant's deception is ‘likely to influence the purchasing decision.’ ”) (quoting Clorox Co. P.R. v. Proctor & Gamble Commercial Co., 228 F.3d 24, 33 n. 6 (1st Cir.2000)); ALPO Petfoods, Inc. v. Ralston Purina Co., 913 F.2d 958, 964 (D.C.Cir.1990) (noting that to succeed on a false advertising claim, a plaintiff must show that representations were “material in their effects on buying decisions”); Taquino v. Teledyne Monarch Rubber, 893 F.2d 1488, 1500 (5th Cir.1990) (“To succeed, it must be proved that ․ the deception is material, in that it is likely to influence the purchasing decision․”); 5 McCarthy on Trademarks and Unfair Competition § 27:35 (4th ed. 1996) (“Plaintiff must make some showing that the defendant's misrepresentation was ‘material’ in the sense that it would have some effect on consumers' purchasing decisions.”).

9.  See Mylan Pharms., 443 F.Supp.2d at 460 (Castel, J.) (“In the context of pharmaceutical drugs, courts have generally rejected Lanham Act claims based on advertisements that merely repeat labeling information that has been approved by the FDA.”); Smithkline Beecham Consumer Healthcare, L.P. v. Johnson & Johnson–Merck Consumer Pharms. Co., No. 95 Civ. 7011, 1996 WL 280810, at *13 (S.D.N.Y. May 24, 1996) (Baer, J.) (“[T]o enjoin SmithKline Beecham from claiming that TAGAMET HB works faster than PEPCID AC on the basis of package [labeling], would substitute this Court's discretion for that of the FDA in approving package [labeling] for over-the-counter medications.”).

10.  Lanham Act liability might arise if an advertisement uses information contained in an FDA-approved label that does not correspond substantially to the label, or otherwise renders the advertisement literally or implicitly false.

DENNIS JACOBS, Circuit Judge: