In Re DDAVP Direct Purchaser Antitrust Litigation

06-5525-cv In re: DDAVP Direct Purchaser Antitrust Litigation 1 UNITED STATES COURT OF APPEALS 2 FOR THE SECOND CIRCUIT 3 4 August Term 2008 5 (Argued: September 15, 2008 Decided: October 16, 2009) 6 Docket No. 06-5525-cv 7 8 -----------------------------------------------------x 9 10 In Re: DDAVP DIRECT PURCHASER ANTITRUST LITIGATION 11 12 MEIJER, INC., MEIJER DISTRIBUTION, INC., on behalf of 13 themselves and all others similarly situated, 14 15 Plaintiffs-Appellants, 16 17 ROCHESTER DRUG CO-OPERATIVE, INC., LOUISIANA 18 WHOLESALE DRUG CO., INC., 19 20 Consolidated-Plaintiffs-Appellants, 21 22 -- v. -- 23 24 FERRING B.V., FERRING PHARMACEUTICALS, INC., AVENTIS 25 PHARMACEUTICALS, INC., 26 27 Defendants-Appellees, 28 29 -- v. -- 30 31 VISTA HEALTHPLAN, INC., on behalf of itself and all 32 others, PENNSYLVANIA EMPLOYEES BENEFIT TRUST FUND, on 33 behalf of itself and all others, PAINTERS DISTRICT 34 COUNCIL NO. 30 HEALTH AND WELFARE FUND, PHILADELPHIA 35 FEDERATION OF TEACHERS HEALTH AND WELFARE FUND, 36 37 Consolidated Plaintiffs. 38 39 -----------------------------------------------------x 40 41 B e f o r e : FEINBERG, WALKER, and LIVINGSTON, Circuit Judges. 42 Appeal by Plaintiffs Meijer, Inc., Meijer Distribution, 43 Inc., Rochester Drug Co-operative, Inc., and Louisiana Wholesale -1- 1 Drug Co., Inc., from the dismissal of their class action 2 antitrust complaint in the United States District Court for the 3 Southern District of New York (Charles L. Brieant, Judge) for 4 lack of standing and failure to state a claim upon which relief 5 can be granted. We hold that the district court erred in 6 dismissing the complaint, because the plaintiffs have antitrust 7 standing and have adequately stated a claim upon which relief can 8 be granted. 9 VACATED and REMANDED. 10 DAVID F. SORENSEN, (Daniel 11 Berger, Daniel C. Simons, on 12 the brief), Berger & Montague, 13 P.C., Philadelphia, PA, Bruce 14 Gerstein, Adam Steinfeld, 15 Garwin Gerstein & Fisher, LLP, 16 New York, NY, Linda P. 17 Nussbaum, Kaplan Fox & 18 Kilsheimer, LLP, New York, NY, 19 for Plaintiffs-Appellants. 20 21 DOUGLAS L. WALD, (William J. 22 Baer, Barbara H. Wootton, Jon 23 J. Nathan, on the brief), 24 Arnold & Porter LLP, 25 Washington, DC, for 26 Defendants-Appellees Ferring 27 B.V. and Ferring 28 Pharmaceuticals, Inc. 29 30 JULIA E. MCEVOY, (Donald B. 31 Ayer, John M. Majoras, Donald 32 Earl Childress III, 33 Christopher R. Farrell, on the 34 brief), Jones Day, Washington, 35 DC, for Defendant-Appellee 36 Aventis Pharmaceuticals, Inc. 37 38 JAMES J. FREDERICKS, (Thomas 39 O. Barnett, Assistant Attorney -2- 1 General, Gerald F. Masoudi, 2 Deputy Assistant Attorney 3 General, Robert B. Nicholson, 4 Attorney, on the brief), U.S. 5 Department of Justice, 6 Washington, DC, William 7 Blumental, General Counsel, 8 Federal Trade Commission, 9 Washington, DC, for Amici 10 Curiae United States and 11 Federal Trade Commission. 12 13 J. Douglas Richards, Pomerantz 14 Haudek Block Grossman & Gross 15 LLP, New York, NY, for Amici 16 Curiae the American Antitrust 17 Institute, AARP, the Consumer 18 Federation of America, 19 Consumers Union, and Families 20 USA. 21 22 Andrew W. Cuomo, Attorney 23 General of the State of New 24 York, Barbara D. Underwood, 25 Solicitor General, Benjamin N. 26 Gutman, Deputy Solicitor 27 General, Jay L. Himes, Chief, 28 Antitrust Bureau, Robert L. 29 Hubbard, Director of 30 Litigation, Antitrust Bureau, 31 of counsel, New York, NY, for 32 Amici Curiae the States of New 33 York, Alaska, Arizona, 34 Arkansas, California, 35 Colorado, Connecticut, 36 Delaware, Florida, Hawaii, 37 Idaho, Illinois, Iowa, Kansas, 38 Kentucky, Louisiana, Maine, 39 Maryland, Massachusetts, 40 Michigan, Minnesota, 41 Mississippi, Montana, New 42 Hampshire, New Jersey, New 43 Mexico, North Dakota, Ohio, 44 Oklahoma, Oregon, 45 Pennsylvania, South Carolina, 46 Tennessee, Texas, Utah, 47 Vermont, Washington, West 48 Virginia, Wisconsin, and -3- 1 Wyoming, and the District of 2 Columbia and Puerto Rico. 3 4 Margaret M. Zwisler, Charles 5 R. Price, Latham & Watkins 6 LLP, Washington, DC, Diane E. 7 Bieri, Pharmaceutical Research 8 and Manufacturers of America, 9 Washington, DC, Thomas 10 DiLenge, Biotechnology 11 Industry Organization, 12 Washington, DC, for Amici 13 Curiae the Pharmaceutical 14 Research and Manufacturers of 15 America and the Biotechnology 16 Industry Organization. 17 JOHN M. WALKER, JR., Circuit Judge: 18 This case presents a novel question of standing that lies at 19 the junction of antitrust and patent law. The plaintiffs, direct 20 purchasers of desmopressin acetate tablets (sold under the name 21 DDAVP), filed this class action in the Southern District of New 22 York (Charles L. Brieant, Judge) against the defendants Ferring 23 B.V., Ferring Pharmaceuticals (collectively, “Ferring”), and 24 Aventis Pharmaceuticals (“Aventis”), alleging that Ferring and 25 Aventis abused the patent system to unlawfully maintain a 26 monopoly over DDAVP. Ferring developed, patented, and 27 manufactures DDAVP, and Aventis holds FDA approval for DDAVP 28 tablets as well as a license from Ferring to market and sell the 29 drug. The plaintiffs alleged that Ferring and Aventis inflated 30 the price of DDAVP by suppressing generic competition for the 31 tablets in violation of the antitrust laws. The district court 32 dismissed the suit, concluding that the plaintiffs both lacked -4- 1 antitrust standing and had failed to state a claim upon which 2 relief could be granted. This appeal followed. 3 BACKGROUND 4 The facts that follow are either undisputed or, because they 5 are properly pled, presumed true for the purposes of this appeal 6 at the pleading stage. See Diaz v. Paterson, 547 F.3d 88, 91 (2d 7 Cir. 2008). 8 DDAVP is an antidiuretic prescription medication used to 9 manage a form of diabetes, excessive urination, excessive thirst, 10 and bed wetting. Ferring developed and manufactures DDAVP and 11 owns U.S. Patent No. 5,047,398 (the “‘398 patent”), which claims 12 the compound in tablet form. Ferring filed its application for 13 the ‘398 patent on December 17, 1985, and the Patent and Trade 14 Office (“PTO”) issued the patent on September 10, 1991. Ferring 15 granted an exclusive license to Aventis to market and sell the 16 patented tablets under the DDAVP name. In addition to its 17 license from Ferring, Aventis also holds an approved New Drug 18 Application (“NDA”) from the FDA for the tablets. A manufacturer 19 seeking to market a new drug must file an NDA to demonstrate the 20 drug’s safety and efficacy. 21 In 2002, Ferring filed a patent infringement suit against 22 Barr Laboratories, Inc. (“Barr”), which came before the same 23 district judge who later presided over this action. Earlier that 24 year, Barr had filed an Abbreviated New Drug Application (“ANDA”) -5- 1 for a generic version of the compound. An ANDA filing 2 accelerates the approval process for a generic drug by allowing 3 the manufacturer to rely on the safety and efficacy data provided 4 in the NDA for the drug’s branded counterpart. As part of the 5 ANDA, Barr filed a certification (called a “paragraph IV 6 certification”) stating that the ‘398 patent was invalid, 7 unenforceable, and/or would not be infringed by Barr’s generic 8 product. This filing triggered Ferring’s infringement action 9 against Barr pursuant to 35 U.S.C. § 271(e)(2). 10 Ferring’s suit failed. On summary judgment, the district 11 court found that the ‘398 patent, rather than having been 12 infringed by Barr, was unenforceable due to inequitable conduct 13 before the PTO by Ferring and its agents. Ferring B.V. v. Barr 14 Labs., Inc., No. 7:02-CV-9851, 2005 WL 437981, at *10 (S.D.N.Y. 15 Feb. 7, 2005). The Federal Circuit affirmed. See Ferring B.V. 16 v. Barr Labs., Inc. (“Ferring I”), 437 F.3d 1181 (Fed. Cir. 17 2006). 18 The inequitable conduct that crippled the ‘398 patent 19 occurred on the patent’s troubled path to approval. In November 20 1986, PTO examiners rejected certain of the ‘398 patent’s claims 21 as anticipated by or obvious from U.S. Patent No. 3,497,491 (the 22 “‘491 patent”), a patent exclusively licensed to Ferring that 23 subsequently expired in 1987. See id. at 1183-84. The Board of 24 Patent Appeals and Interferences (“BPAI”) affirmed the examiners’ -6- 1 rejection in September 1990, but on slightly different grounds: 2 relying only on the ‘491 patent, the BPAI concluded that the ‘491 3 patent rendered the ‘398 patent “obvious” when considered “in 4 light of” a 1973 article written by Ivan Vavra. Id. at 1184. In 5 November 1990, in an effort to persuade the examiners of the ‘398 6 patent’s novelty, the applicants, two Ferring employees who 7 assigned their prospective patent rights to Ferring, submitted 8 declarations from several scientists stating that the ‘491 patent 9 and Vavra article did not suggest the ‘398 patent. Id. at 1185. 10 It turned out, however, that four of the five declarants 11 previously had either “been employed or had received research 12 funds from Ferring,” facts that the submissions failed to 13 disclose to the PTO. See id. On the strength of these 14 declarations, the earlier rejection notwithstanding, the PTO 15 issued the ‘398 patent in September 1991. See id. at 1185, 1188- 16 89. The district judge found this non-disclosure to be 17 inequitable conduct in the Barr litigation, and determined the 18 patent to be unenforceable. See id. at 1194-95. 19 Upon de novo review, the Federal Circuit held both that the 20 undisclosed affiliations would have been material to the 21 examiners’ decision to issue the ‘398 patent, see id. at 1187-90, 22 and that the evidence suggested that the affiants’ previous 23 relationships with Ferring were “deliberately concealed,” id. at 24 1193. The Federal Circuit therefore concluded that the district -7- 1 court had not abused its discretion in finding inequitable 2 conduct, see id. at 1194-95, rendering the ‘398 patent 3 unenforceable as against Barr and all other parties. 4 Less than two months after the Federal Circuit’s February 5 2006 ruling, the direct purchaser plaintiffs filed the instant 6 suit. The plaintiffs argue that the defendants’ conduct, in 7 addition to making the ‘398 patent unenforceable, violated the 8 antitrust laws. They allege that defendants Ferring and Aventis 9 “engaged in an exclusionary scheme” that included (1) 10 “[p]rocuring the ‘398 patent by committing fraud and/or engaging 11 in inequitable conduct before the PTO,” (2) “[i]mproperly listing 12 the fraudulently obtained ‘398 patent in the [FDA’s] Orange 13 Book,” thereby enabling patent infringement claims against 14 potential competitors, (3) prosecuting sham infringement 15 litigation against generic competitors, and (4) “filing a sham 16 citizen petition to further delay FDA final approval of Barr’s 17 ANDA.” Compl. ¶ 144. The plaintiffs claim that the lack of 18 competing, generic versions of DDAVP injured them by forcing them 19 to pay monopolistic prices for the drug. 20 Ferring and Aventis jointly moved to dismiss the complaint 21 on the basis that, inter alia, the plaintiffs lacked standing to 22 assert their claimed antitrust violations. Aventis separately 23 moved to dismiss the case on the ground that the plaintiffs had 24 not sufficiently alleged misconduct against it. The district -8- 1 court granted both motions and dismissed the antitrust action. 2 See In re DDAVP Direct Purchaser Antitrust Litig., No. 05 Cv. 3 2237, slip op. at 15 (S.D.N.Y. Nov. 2, 2006). 4 The district court acknowledged that, while conduct in 5 obtaining and enforcing a patent is generally protected from 6 antitrust liability by the First Amendment, a patentee loses this 7 immunity and can incur antitrust liability for enforcing a patent 8 if the patent was obtained by fraud on the PTO. See id. at 5-6 9 (citing Walker Process Equip., Inc. v. Food Mach. & Chem. Corp., 10 382 U.S. 172, 173 (1965)). However, the district court held that 11 the plaintiffs failed to plead fraud on the PTO with sufficient 12 particularity, noting that fraud requires a greater showing of 13 culpability than the inequitable conduct that can render a patent 14 unenforceable. See id. at 7-8. The district court concluded 15 that, “[f]or this reason alone, granting the motions to dismiss 16 is appropriate.” Id. at 8. 17 However, “in the interest of completeness,” the district 18 court also considered the plaintiffs’ standing. Id. As an 19 initial matter, the district court noted the lack of binding 20 precedent “with regard to the specific issue of whether purchaser 21 plaintiffs like those in this case have standing to assert a 22 Walker Process claim.” Id. at 10-11. The district court then 23 held that the plaintiffs lacked antitrust standing for their 24 Walker Process claim because the ‘398 patent had not been -9- 1 enforced against them, and they were not competitors of Ferring 2 or Aventis. Id. at 11-12 (citing In re Ciprofloxacin 3 Hydrochloride Antitrust Litig., 363 F. Supp. 2d 514 (E.D.N.Y. 4 2005); Walgreen Co. v. Organon, Inc. (In re Remeron Antitrust 5 Litig.), 335 F. Supp. 2d 522 (D.N.J. 2004)). 6 With little discussion, the district court also rejected the 7 plaintiffs’ non-Walker Process claims — the Orange Book listing, 8 the sham infringement litigation, and the sham citizen’s petition 9 — on the basis that the defendants had not acted “in subjective 10 bad faith.” Id. at 13. Finally, the district court granted 11 Aventis’s separate motion to dismiss, concluding that the 12 plaintiffs had failed to sufficiently allege that Aventis was 13 complicit in Ferring’s fraud upon the PTO. Id. at 14-15. 14 The plaintiffs now appeal the district court’s decision in 15 its entirety. They argue that the district court erred in 16 dismissing the complaint and, in doing so, violated their due 17 process rights. The defendants moved in this court to transfer 18 the case on the basis that the Federal Circuit has exclusive 19 jurisdiction over this appeal. We reserved decision on that 20 motion. 21 DISCUSSION 22 I. Jurisdiction 23 We disagree that this appeal properly belongs in the Federal 24 Circuit. The Federal Circuit has exclusive jurisdiction over -10- 1 appeals where the district court’s jurisdiction is “based, in 2 whole or in part, on section 1338 of [title 28],” 28 U.S.C. § 3 1295(a)(1), which in turn gives district courts “original 4 jurisdiction of any civil action arising under any Act of 5 Congress relating to patents,” id. § 1338(a). Such jurisdiction 6 exists if a case “arises under” patent law, such that “a 7 well-pleaded complaint establishes either [1] that federal patent 8 law creates the cause of action or,” as is relevant here, “[2] 9 that the plaintiff’s right to relief necessarily depends on 10 resolution of a substantial question of federal patent law, in 11 that patent law is a necessary element of one of the well-pleaded 12 claims.” Christianson v. Colt Indus. Operating Corp., 486 U.S. 13 800, 808-09 (1988). However, “as long as there is at least one 14 alternative theory supporting the claim that does not rely on 15 patent law, there is no ‘arising under’ jurisdiction under 28 16 U.S.C. § 1338.” In re Tamoxifen Citrate Antitrust Litig., 466 17 F.3d 187, 199 (2d Cir. 2006). In such a scenario, because “there 18 are reasons completely unrelated to the provisions and purposes 19 of federal patent law why petitioners may or may not be entitled 20 to the relief they seek under their monopolization claim, the 21 claim does not arise under federal patent law.” Id. (quoting 22 Christianson, 486 U.S. at 812). 23 In this case, because the plaintiffs have filed an antitrust 24 suit, patent law does not create the cause of action. The -11- 1 Federal Circuit’s jurisdiction over this case, if it exists, must 2 rest upon the second part of the Christianson test: that “the 3 plaintiff’s right to relief necessarily depends on resolution of 4 a substantial question of federal patent law.” 486 U.S. at 809. 5 The plaintiffs argue that their right to relief can stand on any 6 one of four different theories: (1) the defendants committed 7 Walker Process fraud in obtaining and securing the ‘398 patent; 8 (2) the defendants listed the ‘398 patent in the FDA’s Orange 9 Book despite knowing the patent was fraudulently procured and 10 therefore invalid; (3) the defendants, knowing the ‘398 patent 11 was invalid, prosecuted sham patent infringement litigation 12 against generic competitors in order to delay FDA approval of the 13 competitors’ generic DDAVP equivalent; and (4) the defendants 14 filed a sham citizen petition, asking the FDA to require 15 additional testing of a generic DDAVP equivalent (thus delaying 16 its approval) despite knowing that such testing was unnecessary. 17 The first three of the plaintiffs’ four theories plainly 18 “depend[] on resolution of a substantial question of federal 19 patent law,” as they all turn on how the ‘398 patent was 20 procured. In the absence of any fraud, actions taken by Ferring 21 and Aventis in obtaining, maintaining, and enforcing the ‘398 22 patent would be exempt from antitrust liability. See Walker 23 Process, 382 U.S. at 177 & n.5; SCM Corp. v. Xerox Corp., 645 24 F.2d 1195, 1206 (2d Cir. 1981). The plaintiffs’ antitrust claim, -12- 1 under their first three theories, requires that the ‘398 patent 2 had been fraudulently procured. See Walker Process, 382 U.S. at 3 177 & n.5. In this context, fraud requires (1) a false 4 representation or deliberate omission of a fact material to 5 patentability, (2) made with the intent to deceive the patent 6 examiner, (3) on which the examiner justifiably relied in 7 granting the patent, (4) but for which misrepresentation or 8 deliberate omission the patent would not have been granted. See 9 C.R. Bard, Inc. v. M3 Sys., Inc., 157 F.3d 1340, 1364 (Fed. Cir. 10 1998). Materiality, justifiable reliance, and but-for causation 11 each implicate the issue of patentability, namely, whether the 12 patent would have been granted in the absence of the fraudulent 13 representations or omissions. Because the first three theories 14 turn on substantial questions of patent law, appellate 15 jurisdiction would lie exclusively with the Federal Circuit if 16 the plaintiffs’ success solely depended on one or more of these 17 theories. 18 The plaintiffs’ fourth theory, however, does not turn on a 19 substantial question of patent law. Under this theory, Ferring 20 violated the antitrust laws when it filed a sham citizen petition 21 with the FDA, requesting that the FDA require Barr to conduct 22 extra testing to establish that its generic product was 23 bioequivalent to DDAVP, thereby delaying approval of Barr’s ANDA. 24 Generally, under the Noerr-Pennington doctrine, citizen petitions -13- 1 are immune from antitrust liability in light of the First 2 Amendment. See E.R.R. Presidents Conference v. Noerr Motor 3 Freight, Inc., 365 U.S. 127, 136 (1961) (“[T]he Sherman Act does 4 not prohibit two or more persons from associating together in an 5 attempt to persuade the legislature or the executive to take 6 particular action with respect to a law that would produce a 7 restraint or a monopoly.”); United Mine Workers of Am. v. 8 Pennington, 381 U.S. 657, 670 (1965) (“Joint efforts to influence 9 public officials do not violate the antitrust laws even though 10 intended to eliminate competition.”); Primetime 24 Joint Venture 11 v. Nat’l Broad. Co., 219 F.3d 92, 99 (2d Cir. 2000) (explaining 12 that “concerted actions before courts and administrative 13 agencies” are generally shielded from the Sherman Act by “the 14 right to petition the legislature”) (citing Cal. Motor Transp. 15 Co. v. Trucking Unlimited, 404 U.S. 508, 510-11 (1972)). 16 Noerr-Pennington protection is not absolute, however. When 17 petitioning activity “ostensibly directed toward influencing 18 governmental action[] is a . . . sham to cover what is . . . 19 nothing more than an attempt to interfere directly with the 20 business relationships of a competitor[, then] the application of 21 the Sherman Act would be justified.” Noerr, 365 U.S. at 144. 22 This sham exception requires that the petition be “(i) 23 ‘objectively baseless,’ and (ii) ‘an attempt to interfere 24 directly with the business relationships of a competitor through -14- 1 the use of the governmental process — as opposed to the outcome 2 of that process — as an anticompetitive weapon.’” Primetime 24, 3 219 F.3d at 100-01 (quoting Prof’l Real Estate Investors, Inc. v. 4 Columbia Pictures Indus., Inc. (“PRE”), 508 U.S. 49, 60 (1993)). 5 The plaintiffs claim that the citizen petition was a sham 6 because its sole purpose was “to delay the entry of generic 7 competitors into the market.” Compl. ¶ 105. Whether the 8 petition was a sham is an issue independent of patent law; the 9 substance of Ferring’s citizen petition did not rest on the 10 issuance or validity of a patent. A single lawsuit can violate 11 antitrust law as long as it is both an objective and subjective 12 sham. BE&K Constr. Co. v. NLRB, 536 U.S. 516, 526 (2002); see 13 also Tamoxifen, 466 F.3d at 213 (citing PRE, 508 U.S. at 60). 14 Administrative petitions, while less susceptible than lawsuits to 15 the sham exception, still carry the potential for antitrust 16 liability. See Kottle v. Nw. Kidney Ctrs., 146 F.3d 1056, 1062 17 (9th Cir. 1998) (noting that “the exact scope of the sham 18 exception to the Noerr-Pennington doctrine has not always been 19 clear in the administrative context” but finding it applicable to 20 “a sufficiently circumscribed form of administrative authority” 21 that is not “essentially political”). 22 The fact that a single citizen petition may trigger the sham 23 exception does not end our jurisdictional inquiry. Even if 24 Ferring’s citizen petition was a sham, “[p]roof of a sham merely -15- 1 deprives the defendant of immunity; it does not relieve the 2 plaintiff of the obligation to establish all other elements of 3 his claim.” PRE, 508 U.S. at 61. Thus we must determine if any 4 of these elements turn on a question of patent law, in which case 5 the sham petition theory cannot be our basis for jurisdiction. 6 “[T]o state a claim for monopolization under Section 2 of 7 the Sherman Act, a plaintiff must establish ‘(1) the possession 8 of monopoly power in the relevant market and (2) the willful 9 acquisition or maintenance of that power as distinguished from 10 growth or development as a consequence of a superior product, 11 business acumen, or historic accident.’” PepsiCo, Inc. v. Coca- 12 Cola Co., 315 F.3d 101, 105 (2d Cir. 2002) (per curiam) (quoting 13 United States v. Grinnell Corp., 384 U.S. 563, 570-71 (1966)). 14 The plaintiffs also must demonstrate an antitrust injury and 15 damages. See Cordes & Co. Fin. Servs., Inc. v. A.G. Edwards & 16 Sons, Inc., 502 F.3d 91, 105 (2d Cir. 2007). 17 The defendants argue that, because the citizen petition was 18 filed over a year before the ‘398 patent was ruled unenforceable, 19 the plaintiffs cannot prove the necessary intent to monopolize 20 without showing that the patent was foreseeably unenforceable — 21 an issue that raises questions of patent law. However, the ‘398 22 patent became unenforceable almost five months before the FDA 23 rejected the citizen petition. During that time, the defendants 24 were free to “supplement, amend, or withdraw” the petition, 21 -16- 1 C.F.R. § 10.30(g), which at that point they knew to be based upon 2 an unenforceable patent. Even if the defendants’ intent when 3 filing the petition raises questions of patent law, their intent 4 in maintaining the petition after they lost the infringement 5 litigation does not. And because the defendants’ failure to 6 address the citizen petition after the ‘398 patent became 7 unenforceable could plausibly constitute a Sherman Act violation, 8 the citizen petition supports a patent-independent theory of 9 liability. See Bell Atl. Corp. v. Twombly, 550 U.S. 544, 566 10 (2007) (suggesting that either “action or inaction” could be 11 plausibly alleged as an antitrust violation). 12 The failed infringement suit also dooms the defendants’ 13 suggestion that the ‘398 patent’s status might impact the 14 potential for antitrust injury from the citizen’s petition. If 15 the ‘398 patent had been valid, Barr’s ANDA might have been 16 denied regardless of the citizen petition. But the ‘398 patent 17 had already been held unenforceable due to inequitable conduct, 18 and the ANDA denial was plainly not inevitable (as evidenced by 19 its ultimate approval). Thus, the plaintiffs’ antitrust claim 20 can stand on the citizen petition theory without raising 21 questions of patent law. 22 Finally, the defendants argue that jurisdiction properly 23 lies with the Federal Circuit because the plaintiffs’ citizen 24 petition theory is a minor part of the overall allegations. -17- 1 Indeed, the plaintiffs themselves see the petition as only a 2 single piece of a larger anticompetitive scheme of which the ‘398 3 patent is the linchpin. But the relief sought by the plaintiffs 4 is not tied inextricably to this larger scheme. The plaintiffs 5 simply ask for a judgment declaring the defendants’ actions to 6 have violated the Sherman Act. See Compl. 37-38. The question 7 of whether the Federal Circuit has jurisdiction “focuses on 8 claims, not theories, and just because an element that is 9 essential to a particular theory might be governed by federal 10 patent law does not mean that the entire monopolization claim 11 ‘arises under’ patent law.” Christianson, 486 U.S. at 811 12 (internal citations omitted). The defendants stress that the 13 plaintiffs’ patent-related theories are essential to the overall 14 relief the plaintiffs seek, because the citizen petition theory 15 covers a time period shorter than the overall allegations, but 16 this fact lacks jurisdictional significance. Focusing on claims, 17 not theories, we have jurisdiction as long as any one of the 18 theories can support the claim without raising substantial 19 questions of patent law. The citizen petition theory satisfies 20 this requirement, and therefore we have jurisdiction over this 21 appeal. 22 II. Antitrust Standing 23 The defendants argue that the plaintiffs lack standing to 24 pursue this action. We review questions of standing de novo. -18- 1 Comer v. Cisneros, 37 F.3d 775, 787 (2d Cir. 1994). In addition 2 to demonstrating Article III standing, an antitrust plaintiff 3 must also establish antitrust standing. See Paycom Billing 4 Servs., Inc. v. Mastercard Int’l, Inc., 467 F.3d 283, 290 (2d 5 Cir. 2006). We analyze antitrust standing under a two-part test: 6 a plaintiff must show (1) antitrust injury, which is “injury of 7 the type the antitrust laws were intended to prevent and that 8 flows from that which makes defendants’ acts unlawful,” 9 Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 10 (1977), and (2) that he is a proper plaintiff in light of four 11 “efficient enforcer” factors: 12 (1) the directness or indirectness of the asserted injury; 13 (2) the existence of an identifiable class of persons whose 14 self-interest would normally motivate them to vindicate the 15 public interest in antitrust enforcement; (3) the 16 speculativeness of the alleged injury; and (4) the 17 difficulty of identifying damages and apportioning them 18 among direct and indirect victims so as to avoid duplicative 19 recoveries. 20 21 Volvo N. Am. Corp. v. Men’s Int’l Prof’l Tennis Council, 857 F.2d 22 55, 66 (2d Cir. 1988) (citing Associated Gen. Contractors v. Cal. 23 State Council of Carpenters (“AGC”), 459 U.S. 519, 540-45 24 (1983)). 25 In this case, the plaintiffs are purchasers of the 26 defendants’ product who allege being forced to pay supra- 27 competitive prices as a result of the defendants’ anticompetitive 28 conduct. Such an injury plainly is “of the type the antitrust 29 laws were intended to prevent.” Brunswick, 429 U.S. at 489; see -19- 1 also AGC, 459 U.S. at 530 (“Congress was primarily interested in 2 creating an effective remedy for consumers who were forced to pay 3 excessive prices by the giant trusts and combinations that 4 dominated certain interstate markets.”). Although the 5 defendants’ conduct at issue targeted their competitors, such as 6 Barr, the plaintiffs’ claimed injury of higher prices was 7 “inextricably intertwined” with the conduct’s anti-competitive 8 effects and thus “flow[ed] from that which makes defendants’ acts 9 unlawful.” Blue Shield of Va. v. McCready, 457 U.S. 465, 484 10 (1982) (internal quotation marks omitted). Antitrust injury is 11 therefore present. 12 As for the “efficient enforcer” factors that bear on whether 13 the plaintiffs are “proper” antitrust plaintiffs, spelled out in 14 Volvo, each favors granting antitrust standing. With respect to 15 the first factor, directness of injury, even though the 16 plaintiffs’ injuries were derivative of the direct harm 17 experienced by the defendants’ competitors, harming competitors 18 was simply a means for the defendants to charge the plaintiffs 19 higher prices. See id. at 478-79; In re Warfarin Sodium 20 Antitrust Litig., 214 F.3d 395, 400-01 (3d Cir. 2000). This 21 factor supports the plaintiffs’ standing. 22 As to the second factor, motivation, the defendants argue 23 that their competitors are the parties most motivated to enforce 24 the antitrust laws, because the competitors were most directly -20- 1 impacted by the alleged anticompetitive behavior. They note that 2 we declined to find antitrust standing in Paycom in part because 3 the plaintiff there was “not an entity whose self-interest would 4 most ‘motivate [it] to vindicate the public interest in antitrust 5 enforcement.’” 467 F.3d at 294 (quoting AGC, 459 U.S. at 542) 6 (alteration in original). But this argument overlooks the fact 7 that the Paycom court asked if the plaintiff was an entity most 8 motivated by self-interest, not the entity most motivated by 9 self-interest. See id. The second factor simply looks for a 10 class of persons naturally motivated to enforce the antitrust 11 laws. “Inferiority” to other potential plaintiffs can be 12 relevant, but it is not dispositive. See Andrx Pharms., Inc. v. 13 Biovail Corp., Int’l, 256 F.3d 799, 816 (D.C. Cir. 2001). Even 14 if the competitors might be the most motivated, the plaintiffs 15 are also significantly motivated due to their “natural economic 16 self-interest” in paying the lowest price possible. See Daniel 17 v. Am. Bd. of Emergency Med., 428 F.3d 408, 444 (2d Cir. 2005) 18 (internal quotation marks omitted). 19 Moreover, the defendants’ competitors, unlike the 20 plaintiffs, would be seeking lost profits, not overcharges. Lost 21 profits are the difference between the competitive price and what 22 the competitors’ costs would have been, while overcharges are the 23 difference between the defendants’ supra-competitive price and 24 the competitive price. Denying the plaintiffs a remedy in favor -21- 1 of a suit by competitors would thus be “likely to leave a 2 significant antitrust violation undetected or unremedied.” AGC, 3 459 U.S. at 542; see also Andrx Pharms., 256 F.3d at 817 (noting 4 that lost profits and overcharges are distinct injuries). The 5 second factor supports standing. 6 Turning to speculativeness, the third factor, the defendants 7 argue that the plaintiffs’ allegations rest upon tenuous 8 assumptions about the beneficial effects of generic competition. 9 The assumptions are not as speculative as the defendants suggest. 10 That no other manufacturer would have obtained a patent on the 11 drug is a fair assumption, we think, given that “[t]he reluctance 12 of the PTO to issue the ‘398 patent was evident” in advance of 13 the defendants’ inequitable conduct. Ferring I, 437 F.3d at 14 1186. And that generic manufacturers would have decided to 15 compete for DDAVP sales is self-evident: manufacturers sought 16 approval for generic DDAVP when the ‘398 patent was still 17 enforceable. It may be difficult to account precisely for the 18 likely effects of generic competition, but we have little doubt 19 that those effects can be sufficiently estimated and measured 20 here. See Geneva Pharms. Tech. Corp. v. Barr Labs. Inc., 386 21 F.3d 485, 499 (2d Cir. 2004) (listing literature analyzing 22 generic drug competition). This is especially so when “[t]he 23 most elementary conceptions of justice and public policy require 24 that the wrongdoer shall bear the risk of the uncertainty which -22- 1 his own wrong has created.” Bigelow v. RKO Radio Pictures, 327 2 U.S. 251, 265 (1946). Like the first two factors, the third 3 factor supports the plaintiffs’ antitrust standing. 4 As for the fourth factor, the potential for duplicative 5 recovery, the difference between lost profits and overcharges is 6 again relevant. Even assuming some overlap between lost profits 7 and overcharges (as could occur if generic manufacturers charged 8 more than the competitive price), the two are conceptually 9 different measures that we think can be fairly apportioned in 10 order to avoid duplicative recoveries. See Andrx Pharms., 256 11 F.3d at 817. This factor also supports the plaintiffs’ antitrust 12 standing. 13 In sum, then, although the relative weight given to each 14 factor is imprecise, see, e.g., Daniel, 428 F.3d at 443, the 15 plaintiffs would be efficient enforcers under any formulation. 16 What complicates the standing question, however, is the 17 centrality of the alleged Walker Process fraud to the plaintiffs’ 18 case. Walker Process claims are based on a fraudulently obtained 19 patent, and are typically brought as counterclaims in patent 20 infringement suits: the plaintiff claims the defendant infringed 21 his patent, and the defendant responds that the patent was 22 invalid as fraudulently obtained, and that the plaintiff’s 23 enforcement efforts violate Walker Process. See Nobelpharma AB 24 v. Implant Innovations, Inc., 141 F.3d 1059, 1067 (Fed. Cir. -23- 1 1998). If a patent is valid, a Walker Process claim cannot 2 stand. 3 Outside of an infringement suit counterclaim, a patent’s 4 validity can be challenged only by a party (1) producing or 5 preparing to produce the patented product, and (2) being 6 threatened or reasonably likely to be threatened with an 7 infringement suit. See, e.g., Cordis Corp. v. Medtronic, Inc., 8 835 F.2d 859, 862 (Fed. Cir. 1987). As purchasers of DDAVP, the 9 plaintiffs do not satisfy these requirements and cannot directly 10 challenge the ‘398 patent’s validity. As the district court 11 noted, whether the plaintiffs have standing to bring their Walker 12 Process claim, when a court has yet to find the ‘398 patent 13 fraudulently obtained, is a question of first impression. 14 The defendants acknowledge that Walker Process standing 15 might be warranted for a purchaser when a patent has already been 16 held to have been fraudulently procured. But the defendants urge 17 us to hold that, when dealing with a patent not yet found to be 18 fraudulently obtained, a party has Walker Process standing only 19 if that party also has standing to challenge the patent’s 20 validity. They argue that giving Walker Process standing to the 21 plaintiffs, who cannot directly challenge the ‘398 patent’s 22 validity, could result in an avalanche of patent challenges, 23 because direct purchasers otherwise unable to challenge a 24 patent’s validity could do so simply by dressing their patent -24- 1 challenge with a Walker Process claim. It would be relatively 2 easy, the defendants argue, for these purchasers to allege an 3 antitrust injury, as patent protection inherently leads to supra- 4 competitive prices. See Zenith Radio Corp. v. Hazeltine 5 Research, Inc., 395 U.S. 100, 135 (1969) (“A patentee has the 6 exclusive right to manufacture, use, and sell his invention.”). 7 Given that Walker Process fraud converts this fundamental feature 8 of the patent system into a potential antitrust violation, the 9 defendants contend that finding purchaser standing could 10 significantly increase the costs of defending and enforcing 11 patents by greatly expanding the universe of potential 12 challenges. They insist that “[i]f the threat of treble damage 13 liability . . . were imbedded in the minds of potential patent 14 holders as a likely prospect . . . , the efficacy of the economic 15 incentives afforded by our patent system might be severely 16 diminished.” SCM Corp. v. Xerox Corp., 645 F.2d 1195, 1206 (2d 17 Cir. 1981). 18 Walker Process itself, of course, reflects a willingness to 19 let antitrust liability impact the patent system. However, the 20 defendants argue that Walker Process is the product of the 21 Supreme Court’s careful balancing of antitrust and patent 22 policies, a balance which should not be upset and under which 23 Walker Process plaintiffs must be independently able to first 24 prove the patent’s fraudulent procurement. Yet the language of -25- 1 Walker Process does not necessarily suggest such a limit: 2 While one of [the claim’s] elements is the fraudulent 3 procurement of a patent, the action does not directly seek 4 the patent’s annulment. The gist of Walker’s claim is that 5 since Food Machinery obtained its patent by fraud it cannot 6 enjoy the limited exception to the prohibitions of § 2 of 7 the Sherman Act, but must answer under that section . . . to 8 those injured by any monopolistic action taken under the 9 fraudulent patent claim. Nor can the interest in protecting 10 patentees from ‘innumerable vexatious suits’ be used to 11 frustate the assertion of rights conferred by the antitrust 12 laws. 13 14 382 U.S. at 176 (emphasis added). To be sure, the Walker Process 15 Court also noted that allowing antitrust recovery “accord[ed]” 16 with the “long-recognized procedures” that controlled how parties 17 could challenge a patent’s validity, 382 U.S. at 176-77, thereby 18 suggesting that the Court may not have envisioned expanding the 19 universe of potential patent challengers. 20 Nonetheless, we are reluctant to embrace the defendants’ 21 position because we are wary of creating the potential “to leave 22 a significant antitrust violation undetected or unremedied.” 23 AGC, 459 U.S. at 542. As the defendants would have it, direct 24 purchasers would be able to recover antitrust damages from a 25 fraudulent patentee only after that patentee first loses on a 26 fraudulent procurement claim. This asks too much of the generic 27 competitors and other potential patent challengers, who may not 28 have the strategic interest or the resources to start or win such 29 a battle, or who may be presented with strong incentives to 30 settle their challenge by patent holders seeking not only to -26- 1 preserve their patent’s enforceability, but also to avoid 2 potential Walker Process liability. See C. Scott Hemphill, 3 Paying for Delay: Pharmaceutical Patent Settlement as a 4 Regulatory Design Problem, 81 N.Y.U. L. Rev. 1553, 1616 (2006) 5 (noting how “an innovator has an especially strong incentive to 6 pay to neutralize . . . potential competition” when a generic 7 manufacturer first files an ANDA). 8 Although settlements between patent holders and generic 9 manufacturers that delay generic entry into the market may 10 themselves invite antitrust liability, a plaintiff must be able 11 to show the settled litigation to have been a sham in order to 12 succeed. See Tamoxifen, 466 F.3d at 208-09 (“In such a case, so 13 long as the patent litigation is neither a sham nor otherwise 14 baseless, the patent holder is seeking to arrive at a settlement 15 in order to protect that to which it is presumably entitled: a 16 lawful monopoly over the manufacture and distribution of the 17 patented product.”). A purchaser seeking to challenge the 18 settlement by showing the underlying infringement litigation to 19 be a sham would need to attach antitrust liability to the patent 20 enforcement efforts — a move that would raise the same standing 21 issues presented by this case. Thus, not only are there strong 22 potential settlement incentives, but these settlements could be 23 shielded from purchaser attack. 24 The difference between inequitable conduct and fraud also -27- 1 bears on the question. Inequitable conduct is a lesser showing 2 than fraud, but is sufficient to render a patent unenforceable. 3 A generic competitor interested simply in selling its product may 4 not value the higher showing of fraud enough to pursue it, 5 especially if the competitor’s antitrust damages would be minor 6 or difficult to prove. Again, relying on generic competitors to 7 lead the antitrust charge may ask too much of them. 8 On the other hand, we do not pass lightly over the 9 defendants’ objections to expanding the universe of patent 10 challengers. The risk of disturbing the incentives for 11 innovation dictates that we tread carefully. As a result, we 12 decline to decide whether purchaser plaintiffs per se have 13 standing to raise Walker Process claims. In this case, the 14 plaintiffs are challenging an already tarnished patent. We are 15 able to grant them antitrust standing without altering the 16 typical limits on who can start a challenge to a patent’s 17 validity. We therefore hold only that purchaser plaintiffs have 18 standing to raise Walker Process claims for patents that are 19 already unenforceable due to inequitable conduct. The district 20 court erred by concluding to the contrary. 21 III. Was the Antitrust Claim Adequately Pled? 22 Granting standing to the plaintiffs does not resolve this 23 appeal, because the district court also concluded that the 24 plaintiffs had failed to state a claim. “We review the district -28- 1 court’s dismissal of a complaint for failure to state a claim de 2 novo, accepting as true all facts alleged in the complaint and 3 drawing all inferences in favor of the plaintiff . . . .” 4 Faulkner v. Beer, 463 F.3d 130, 133 (2d Cir. 2006) (internal 5 quotation marks omitted). “[O]nly a complaint that states a 6 plausible claim for relief survives a motion to dismiss.” 7 Ashcroft v. Iqbal, 129 S. Ct. 1937, 1950 (2009). We believe that 8 the plaintiffs meet this standard for their antitrust claim under 9 each of their four theories. 10 Walker Process fraud, the plaintiffs’ first theory, 11 requires: 12 (1) a representation of a material fact, (2) the falsity of 13 that representation, (3) the intent to deceive or, at least, 14 a state of mind so reckless as to the consequences that it 15 is held to be the equivalent of intent (scienter), (4) a 16 justifiable reliance upon the misrepresentation by the party 17 deceived which induces him to act thereon, and (5) injury to 18 the party deceived as a result of his reliance on the 19 misrepresentation. 20 21 Nobelpharma, 141 F.3d at 1069-70. A fraudulent omission, which 22 “can be just as reprehensible as a fraudulent misrepresentation,” 23 can be sufficient to “support a finding of Walker Process fraud.” 24 Id. at 1070. 25 A party “alleging fraud or mistake . . . must state with 26 particularity the circumstances constituting fraud or mistake.” 27 Fed. R. Civ. P. 9(b). The plaintiffs argue that they have pled 28 each element with sufficient specificity. They alleged a series 29 of “highly material” omissions, without which “the ‘398 patent -29- 1 would not have issued.” Compl. ¶ 74. The Federal Circuit agreed 2 on the “high[] material[ity]” of the omissions when it found the 3 ‘398 patent unenforceable. Ferring I, 437 F.3d at 1194. The 4 Ferring I litigation also addressed the third element of intent, 5 as the district court found “clear and convincing evidence of an 6 intent to mislead the examiners.” Ferring B.V., 2005 WL 437981, 7 at *9. Reliance and injury, the fourth and fifth elements, are 8 straightforward here: the PTO was justified in relying on the 9 information the defendants provided, and injury is a “matter of 10 course whenever the other four elements are met.” Unitherm Food 11 Sys., Inc. v. Swift-Eckrich, Inc., 375 F.3d 1341, 1361 (Fed. Cir. 12 2004), rev’d on other grounds, 546 U.S. 394 (2006). Thus, the 13 plaintiffs contend the district court’s dismissal on the 14 pleadings was erroneous. 15 The defendants respond that the district judge’s rejection 16 of the plaintiffs’ claim must be affirmed because he was also the 17 district judge in the initial infringement litigation, in which 18 he held the ‘398 patent unenforceable. The defendants argue that 19 the judge’s involvement in both cases enabled him to validly 20 conclude that his previous findings could not support a claim of 21 fraudulent procurement in the instant case. This is a logical 22 non sequitur. The district judge could be correct in determining 23 that inequitable conduct occurred and yet mistaken that such 24 conduct did not amount to fraud. Moreover, the defendants’ -30- 1 argument ignores the distinction between findings and pleadings. 2 Even if the district judge was correct that the earlier record 3 did not show fraud, the record in this case could be different 4 following discovery. 5 The defendants contend that simply adding a conclusory 6 allegation of fraud to the previous findings is inadequate to 7 meet the plaintiffs’ obligation to “allege facts that give rise 8 to a strong inference of fraudulent intent.” Acito v. IMCERA 9 Group, Inc., 47 F.3d 47, 52 (2d Cir. 1995). We are, however, 10 “lenient in allowing scienter issues to withstand summary 11 judgment based on fairly tenuous inferences,” because such issues 12 are “appropriate for resolution by the trier of fact.” Press v. 13 Chem. Inv. Servs. Corp., 166 F.3d 529, 538 (2d Cir. 1999). The 14 same holds true for allowing such issues to survive motions to 15 dismiss. The district court found “an intent to deceive” in the 16 patent litigation. Ferring B.V., 2005 WL 437981, at *9. 17 Granting the plaintiffs all favorable inferences as we must on a 18 motion to dismiss, and given that the omissions at issue occurred 19 repeatedly over a period of years, this intent is sufficient to 20 plausibly support a finding of Walker Process fraud. 21 The defendants next argue that the plaintiffs must allege 22 evidence of intent distinct from the omission itself. While a 23 false or clearly misleading statement can permit an inference of 24 deceptive intent, a misrepresentation in the form of an omission -31- 1 is more likely to be innocent and cannot support Walker Process 2 fraud without “evidence of intent separable from the simple fact 3 of the omission.” Dippin’ Dots, Inc. v. Mosey, 476 F.3d 1337, 4 1347 (Fed. Cir. 2007). The issue in the initial infringement 5 litigation was inequitable conduct, not Walker Process fraud. 6 Moreover, the district court in that litigation correctly noted 7 that high materiality could overcome a lesser showing of intent. 8 Ferring B.V., 2005 WL 437981, at *9; see Brasseler, U.S.A. I, 9 L.P. v. Stryker Sales Corp., 267 F.3d 1370, 1380-81 (Fed. Cir. 10 2001). While such balancing is impermissible with Walker Process 11 claims, we think the plaintiffs’ allegations are nonetheless 12 sufficient. Dippin’ Dots concerned findings, not pleadings, see 13 476 F.3d at 1341-42; even if the district court’s findings in the 14 Ferring I litigation could not satisfy Dippin’ Dots, the 15 plaintiffs’ pleadings could plausibly lead to additional findings 16 that would satisfy Dippin’ Dots, which is all that is required at 17 this stage of the litigation. 18 The defendants additionally argue that the allegations of 19 materiality are insufficient. Specifically, they contend that 20 the plaintiffs do not dispute the ‘398 patent’s patentability on 21 the merits or claim that, but for the alleged fraud, no patent 22 could have been issued to anyone. For antitrust purposes, 23 whether a patent could be issued matters more than who would 24 possess it; if a patent could still “have been issued to -32- 1 someone,” its market power would still have been concentrated 2 (properly) in one party. Brunswick Corp. v. Riegel Textile 3 Corp., 752 F.2d 261, 265 (7th Cir. 1984). As a result, Walker 4 Process fraud must concern a material issue of patentability; 5 otherwise, a patent would have issued regardless of any fraud, 6 and potential plaintiffs would have suffered the same monopoly 7 effects (but legitimately). 8 Although the plaintiffs do not address patentability 9 directly in the complaint, the issue is implicit in their 10 allegations. The defendants’ allegedly fraudulent affidavits 11 were attempts to explain away prior art. The Federal Circuit 12 found them “absolutely critical” to the defendants’ overcoming 13 the patent application’s initial rejection. Ferring I, 437 F.3d 14 at 1189. Whether or not these declarations, if accompanied by 15 full disclosure, would have resulted in an enforceable patent is 16 debatable, but we think that, at the pleading stage, the fact of 17 non-disclosure is sufficient to properly allege materiality. 18 Overall, then, the plaintiffs have sufficiently alleged Walker 19 Process fraud to survive the defendants’ motion to dismiss on the 20 pleadings. 21 We likewise conclude that the sham litigation claim has been 22 adequately alleged. In order to state a claim for sham 23 litigation, the plaintiffs need to allege that “the litigation in 24 question is: (i) ‘objectively baseless,’ and (ii) ‘an attempt to -33- 1 interfere directly with the business relationships of a 2 competitor through the use of the governmental process . . . as 3 an anticompetitive weapon.’” Primetime 24 Joint Venture v. Nat’l 4 Broadcasting Co., 219 F.3d 92, 100-01 (2d Cir. 2000) (citing 5 Prof’l Real Estate Investors, Inc. v. Columbia Pictures Indus., 6 Inc., 508 U.S. 49, 60 (1993)). Based on the same facts alleged 7 to sustain a Walker Process claim, we find that in the 8 circumstances of this case, the plaintiffs’ allegations are also 9 sufficient to make out a sham litigation claim. The defendants 10 effectively concede as much. See Brief of Defendants-Appellees 11 Ferring B.V. and Ferring Pharms. at 38 (“[A] sham litigation 12 claim here not only requires proof that defendants defrauded the 13 PTO, but also that they knew their misconduct before the PTO had 14 rendered the patent invalid. . . . [Plaintiffs’] ‘sham’ 15 litigation allegation is thus substantively duplicative of their 16 patent fraud claim . . . .”). 17 The plaintiffs also may proceed on their Orange Book claim. 18 The defendants all but concede that the plaintiffs would have a 19 basis for contending the Orange Book listing was fraudulent if 20 the ‘398 patent is found to have been fraudulently procured. See 21 id. at 41 (“[T]he allegations regarding the Orange Book listing 22 are – like the sham litigation claims – purely derivative of the 23 underlying claim that the ‘398 patent was fraudulently 24 procured.”). Indeed, the Orange Book listing’s validity flows -34- 1 from the patent’s validity. Having determined that the Walker 2 Process and sham litigation theories are still in play, we 3 similarly conclude that the plaintiffs have adequately alleged 4 that the defendants improperly listed the ‘398 patent in the 5 FDA’s Orange Book. 6 The final theory is the plaintiffs’ citizen petition theory, 7 which, as we have explained, stands apart from the ‘398 patent’s 8 validity. The district court dismissed this theory on the basis 9 that it concerned petitioning activity protected by the First 10 Amendment. Id. To reach this conclusion, the district court 11 presumably reasoned that the plaintiffs could not plausibly show 12 the petition to be a sham, i.e., objectively and subjectively 13 baseless, a proposition with which we disagree. The FDA found 14 that the citizen petition “had no convincing evidence” and lacked 15 “any basis” for its arguments. Compl. ¶ 115 (internal quotation 16 marks and emphasis omitted). In the Ferring I litigation, the 17 district court suggested that the petition might have been 18 “nothing more than a hardball litigation tactic, motivated by a 19 desire to keep out competition for as long as possible after the 20 expiration of the patent and raise transactional costs for Barr.” 21 Ferring B.V., 2005 WL 437981, at *17. Together, these findings 22 indicate the plaintiffs could plausibly show the citizen petition 23 to have been a sham. 24 Finally, the defendants contend that the citizen petition -35- 1 cannot give rise to antitrust liability because it could not have 2 impacted the FDA’s decision, as the FDA ultimately rejected the 3 petition. But this ignores the possibility that the sham 4 petition caused a delay in generic competition, a possibility 5 reinforced by the fact that the FDA approved the generic drug on 6 the same day that it rejected the petition. See 21 C.F.R. § 7 10.35(d)(1) (enabling a stay of FDA action after the filing of a 8 petition). Whether the ‘398 patent was valid on the date the 9 petition was filed is immaterial to this theory’s success, 10 because the plaintiffs can plausibly show the patent to have been 11 fraudulently procured. It may turn out at trial that this 12 petition was not a sham, or that the FDA’s approval of the 13 generic drug was not delayed by the petition, but the possibility 14 that the petition was a sham, and that it impacted the FDA’s 15 decision, is sufficiently plausible to defeat the motion to 16 dismiss. 17 Overall, the plaintiffs have stated an antitrust claim upon 18 which relief may be granted. Based on the pleadings, each of 19 their four theories could plausibly succeed. The district court 20 erred by concluding to the contrary. 21 IV. Can Aventis Be Kept in the Case? 22 The district court also granted Aventis’s separate motion to 23 dismiss, concluding that the plaintiffs had not alleged fraud 24 with sufficient particularity to satisfy Rule 9 of the Federal -36- 1 Rules of Civil Procedure. “Although Rule 9(b) permits knowledge 2 to be averred generally, plaintiffs must still plead the events 3 which they claim give rise to an inference of knowledge.” 4 Devaney v. Chester, 813 F.2d 566, 568 (2d Cir. 1987). In a case 5 involving multiple defendants, plaintiffs must plead 6 circumstances providing a factual basis for scienter for each 7 defendant; guilt by association is impermissible. See id. This 8 can consist of allegations as to who “possessed . . . knowledge” 9 of the fraud, “when and how they obtained [that] knowledge,” or 10 even why they “should have known” of the fraud. Id. The 11 district court concluded that the plaintiffs fell short of this 12 standard. We disagree. 13 In the district court’s view, “[t]hat Aventis would pay to 14 license a patent which it knew to be unenforceable flies in the 15 face of reason.” In re DDAVP, slip op. at 15. However, we find 16 the plaintiffs’ allegations plausible, and sufficient to survive 17 a motion to dismiss on the pleadings. At the time Aventis filed 18 its NDA and listed DDAVP in the Orange Book, the ‘398 patent’s 19 validity was already in question with the patent having been 20 rejected twice, and the PTO having raised concerns of bias. See 21 Ferring I, 437 F.3d at 1190. Yet, the plaintiffs assert that 22 Aventis apparently made no effort to independently investigate 23 and attest to the validity of the ‘398 patent. Rule 9(b) 24 requires only the circumstances of fraud to be stated with -37- 1 particularity; knowledge itself can be alleged generally. 2 Especially considering the long-standing relationship between 3 Aventis and Ferring, the plaintiffs have adequately stated 4 circumstances that give rise to a plausible inference of 5 knowledge and liability. At this early stage, the plaintiffs 6 need only state a plausible claim of monopolization, and they 7 have alleged enough for their suit against Aventis to proceed. 8 Because the dismissal as to both Ferring and Aventis was in 9 error, we have no cause to address the plaintiffs’ claim that 10 their due process rights were violated. 11 12 CONCLUSION 13 For the foregoing reasons, we VACATE the district court’s 14 dismissal of the plaintiffs’ case and REMAND for further 15 proceedings. -38-