Catherine GIBBONS, et al., Plaintiffs-Appellants, v. BRISTOL-MYERS SQUIBB COMPANY and Pfizer Inc., Defendants-Appellees.
Plaintiffs in this multi-district litigation appeal from judgments entered by the United States District Court for the Southern District of New York (Denise L. Cote, J.) dismissing their products liability claims for injuries allegedly caused by the drug Eliquis (apixaban). Specifically, Plaintiffs assert that the district court (1) incorrectly denied motions to remand forty-four of the sixty-four cases before it, and (2) wrongly concluded that Plaintiffs’ state law claims were preempted by the Food, Drug, and Cosmetics Act (“FDCA”). Because, as explained below, we agree that removal was proper and that Plaintiffs’ state law claims are preempted, we AFFIRM.
Defendants Bristol-Myers Squibb Co. (“BMS”) and Pfizer Inc. (“Pfizer”) are pharmaceutical companies that are incorporated in Delaware and maintain their principal places of business in New York. Together, Defendants manufacture and distribute Eliquis, a blood-thinning medication used to reduce the risk of stroke in patients with atrial fibrillation. As might be expected, Eliquis increases patients’ risk of bleeding. To that end, the drug, which was approved by the Food and Drug Administration in 2012, carries warnings about the risk of serious, and possibly fatal, bleeding events.
In 2015, plaintiffs nationwide began to bring products liability actions against Defendants, asserting that they or their decedents had suffered excessive bleeding after taking Eliquis that resulted in substantial injury (both physical and financial) or even death. Although the suits arose under the laws of several states, plaintiffs generally alleged that the injuries they or their decedents suffered were attributable to the improper design of Eliquis and the insufficient warning labels that accompanied the drug. Seventeen such suits made their way to the United States District Court for the Southern District of New York, where they were assigned to Judge Cote. The district court ordered the parties to identify a single bellwether case, and the parties selected Utts v. Bristol-Myers Squibb Co., No. 16-cv-5688, for that purpose. Defendants moved to dismiss the complaint in that exemplar action, and on December 23, 2016, the district court granted Defendants’ motion in part. See Utts v. Bristol-Myers Squibb Co., 226 F.Supp.3d 166, 189 (S.D.N.Y. 2016) (“Utts I”). The district court concluded that most of the claims in the Utts complaint, including the failure-to-warn claims, were preempted by the FDCA, and that the others simply did not meet the Rule 8 pleading threshold. Id. The district court dismissed the design defect allegations with prejudice, but granted leave to amend the remaining claims. Id. The Utts plaintiffs filed their first amended complaint on January 20, 2017.
Following the district court's dismissal of the original complaint in Utts, the Judicial Panel on Multidistrict Litigation transferred all Eliquis products liability actions pending in federal court to the Southern District of New York and assigned them to Judge Cote. See S.D.N.Y. Case. No. 17-md-2754 (DLC), Doc. No. 1. Judge Cote thereafter determined that the best procedure for the MDL would be to (1) allow the plaintiffs in Utts to amend their complaint again, (2) have the parties brief a motion to dismiss that complaint, and then, (3) following its ruling on the motion to dismiss the amended complaint in Utts, permit the parties either to (a) proceed to discovery in all cases (if the motion were denied) or (b) have the non-Utts plaintiffs show cause why their similar complaints should not also be dismissed on the grounds set out in the Utts opinion. The parties agreed.
The plaintiffs in Utts amended their complaint on February 24, 2017, setting out ten claims for: (1) manufacturing defect; (2) failure to warn; (3) strict liability; (4) negligence; (5) breach of express warranty; (6) breach of implied warranty; (7) fraudulent concealment; (8) negligent misrepresentation; (9) violation of the California consumer protection laws; and (10) loss of consortium. Defendants again moved to dismiss, and the district court again granted the motion, rejecting all of the Utts plaintiffs’ claims under Federal Rule of Civil Procedure 12(b)(6). See Utts v. Bristol-Myers Squibb Co., 251 F.Supp.3d 644, 684 (S.D.N.Y. 2017) (“Utts II”). Following the entry of Utts II, the district court gave the plaintiffs in the other pending MDL suits two weeks to (1) amend their complaints and (2) show cause why those complaints should not be dismissed in light of Utts II. Out of sixty-eight then-pending actions, just nineteen plaintiffs attempted to show cause why dismissal was not warranted. None was successful.
As it had earlier in the case, the district court set out its reasoning in an exemplar opinion – this time, Fortner v. Bristol-Myers Squibb Co., No. 17-cv-1562 (DLC), 2017 WL 3193928 (S.D.N.Y. July 26, 2017). In that decision, the district court concluded that (1) Plaintiffs’ negligence and strict liability claims were preempted by the FDCA to the extent they were based on a design defect theory, and to the extent they were based on a “failure to warn” theory, they were preempted by the FDCA and failed due to the label's adequacy as a matter of law; (2) the breach of warranty claims failed to state a claim; (3) the fraudulent concealment and negligent misrepresentation claims failed under Rule 9(b); and (4) the state law consumer protection claims were preempted and, in any event, failed to state a claim. Id. at *2–5. The district court then applied that reasoning to dismiss the other eighteen suits before it. Plaintiffs timely appealed those dismissals, which were consolidated in this Court on September 15, 2017.
However, at the time the district court dismissed the nineteen actions before it, not all Eliquis cases pending nationwide had become part of the MDL. Thus, at least thirty-three cases – all brought by the same counsel, Salim-Beasley, LLC – were pending in federal court in California awaiting transfer to the MDL at the time of Utts II. Following the district court's dismissal order in Utts II, the plaintiffs in these thirty-three California actions voluntarily dismissed their suits without prejudice and refiled them in Delaware state court. Two days later, Defendants – who had not yet been served with the Delaware complaints – removed the actions to the United States District Court for the District of Delaware and requested that they be transferred and consolidated into the MDL before Judge Cote. For their part, the plaintiffs in the thirty-three Delaware actions asked the District of Delaware to remand their cases to state court, arguing that because the only basis for federal court jurisdiction was diversity of citizenship, Defendants’ status as citizens of Delaware meant that removal was prohibited under 28 U.S.C. § 1441(b)(2). The District of Delaware denied the plaintiffs’ motions, see Young v. Bristol-Myers Squibb Co., No. 17-609-LPS, 2017 WL 2774735, at *3 (D. Del. June 27, 2017), and transferred the actions to Judge Cote, who eventually dismissed them with prejudice for the reasons articulated in Utts II. Plaintiffs have not appealed the dismissal of those actions.
Nevertheless, additional Eliquis plaintiffs – also represented by Salim-Beasley, LLC – adopted a new strategy whereby they filed a series of new suits in Delaware state court. When Defendants again removed the cases to federal court in Delaware before service, the new plaintiffs consented to the transfer of these removed actions to the MDL, and then asked Judge Cote to remand the suits to Delaware state court. See, e.g., Cheung v. Bristol-Myers Squibb Co., 282 F.Supp.3d 638, 641 (S.D.N.Y. 2017) (addressing four such actions). The district court denied Plaintiffs’ remand motions, see id. at 644, and then applied the earlier reasoning of Utts II and Fortner to dismiss all forty-five actions that came to it by this route (the “Transferred Actions”). The plaintiffs in the Transferred Actions timely appealed, and the forty-five Transferred Actions were consolidated on appeal with the nineteen Eliquis actions already pending before this Court, of which fifteen now remain.1
II. Standard of Review
“We review a district court's denial of a motion to remand de novo.” O'Donnell v. AXA Equitable Life Ins. Co., 887 F.3d 124, 128 (2d Cir. 2018). “In reviewing a denial of a motion to remand, ‘the defendant bears the burden of demonstrating the propriety of removal.’ ” Id. (quoting Cal. Pub. Emps.’ Ret. Sys. v. WorldCom, Inc., 368 F.3d 86, 100 (2d Cir. 2004) ). We also “review de novo a district court's decision to grant a motion under Federal Rule of Civil Procedure 12(b)(6).” Spinelli v. Nat'l Football League, 903 F.3d 185, 196 (2d Cir. 2018) (emphasis added).
Plaintiffs first challenge the district court's denial of their motions to remand the Transferred Actions. Specifically, Plaintiffs argue that because the only basis for federal court jurisdiction is diversity of citizenship,2 and because BMS and Pfizer were sued in the state courts of their home state (Delaware), removal was barred by the forum defendant rule, 28 U.S.C. § 1441(b)(2), and the cases therefore should be remanded to state court. We disagree.
Generally, any civil suit initiated in state court over which a district court would have had original jurisdiction “may be removed by ․ the defendants, to the district court of the United States for the district ․ embracing the place where such action is pending.” 28 U.S.C. § 1441(a). Section 1441 permits removal on the basis of either federal question jurisdiction or diversity of citizenship. See Marcus v. AT&T Corp., 138 F.3d 46, 52 (2d Cir. 1998). But where, as here, the only basis for federal subject-matter jurisdiction is diversity of citizenship under 28 U.S.C. § 1332, “the forum defendant rule applies.” Encompass Ins. Co. v. Stone Mansion Rest. Inc., 902 F.3d 147, 152 (3d Cir. 2018). Under that rule, which is set out at 28 U.S.C. § 1441(b)(2), a suit that is “otherwise removable solely on the basis of ․ [diversity of citizenship] may not be removed if any of the parties in interest properly joined and served as defendants is a citizen of the State in which such action is brought.”
In the usual case, application of the forum defendant rule is straightforward: a defendant is sued in a diversity action in the state courts of its home state, is served in accordance with state law, attempts to remove the case, and is rebuffed by a district court applying Section 1441(b)(2). See, e.g., Wilmington Tr., N.A. v. Pearson, No. 18-cv-4845 (PAC), 2018 WL 3918182, at *1 (S.D.N.Y. Aug. 16, 2018). Here, however, Defendants removed each of the Transferred Actions to federal court after the suit was filed in state court but before any Defendant was served. The district court, reasoning from the text of the statute, concluded that such removal was proper. Cheung, 282 F.Supp.3d at 641–42;3 see also Stan Winston Creatures, Inc. v. Toys ‘R’ Us, Inc., 314 F.Supp.2d 177, 180 (S.D.N.Y. 2003). Other district courts in this Circuit have reached the opposite conclusion. See, e.g., Torchlight Loan Servs., LLC. v. Column Fin., Inc., No. 12-cv-8579 (RWS), 2013 WL 3863887, at *2 (S.D.N.Y. July 24, 2013); In re IntraLinks Holdings, Inc. Derivative Litig., No. 11-cv-9636 (TPG), 2013 WL 929836, at *1–2 (S.D.N.Y. Mar. 11, 2013). Nevertheless, in resolving this split among district courts, we agree with the district court here that 28 U.S.C. § 1441(b)(2) is no barrier to the removal of the Transferred Actions.
“Every exercise in statutory construction must begin with the words of the text.” Saks v. Franklin Covey Co., 316 F.3d 337, 345 (2d Cir. 2003). As the Third Circuit – the only other Court of Appeals to address the propriety of pre-service removal by a defendant sued in its home state – recognized in Encompass Insurance, “the language of the forum defendant rule in section 1441(b)(2) is unambiguous.” 902 F.3d at 152. The statute plainly provides that an action may not be removed to federal court on the basis of diversity of citizenship once a home-state defendant has been “properly joined and served.” 28 U.S.C. § 1441(b)(2) (emphasis added). By its text, then, Section 1441(b)(2) is inapplicable until a home-state defendant has been served in accordance with state law; until then, a state court lawsuit is removable under Section 1441(a) so long as a federal district court can assume jurisdiction over the action.
In fact, Plaintiffs do not even attempt to argue that the text of Section 1441(b)(2) supports their position. Instead, Plaintiffs argue that the Court should depart from the plain meaning of Section 1441(b)(2) because applying the text of the statute (1) produces an absurd result and (2) will lead to non-uniform application of the removal statute depending on the provisions of state law. Neither argument is persuasive.
“It is, to be sure, well-established that ‘[a] statute should be interpreted in a way that avoids absurd results.’ ” Sec. Exch. Comm'n v. Rosenthal, 650 F.3d 156, 162 (2d Cir. 2011) (quoting United States v. Venturella, 391 F.3d 120, 126 (2d Cir. 2004) ). That being said, a statute is not “absurd” merely because it produces results that a court or litigant finds anomalous or perhaps unwise. To the contrary, courts should look beyond a statute's text under the canon against absurdity “only ‘where the result of applying the plain language would be, in a genuine sense, absurd, i.e., where it is quite impossible that Congress could have intended the result and where the alleged absurdity is so clear as to be obvious to most anyone.’ ” Catskill Mountains Chapter of Trout Unlimited, Inc. v. Envtl. Prot. Agency, 846 F.3d 492, 517 (2d Cir. 2017) (quoting Pub. Citizen v. U.S. Dep't of Justice, 491 U.S. 440, 470–71, 109 S.Ct. 2558, 105 L.Ed.2d 377 (1989) (Kennedy, J., concurring in the judgment)).
Plaintiffs argue that applying the plain text of Section 1441(b)(2) produces an absurd result in light of the overarching purpose of the removal statute, which is to allow an out-of-state defendant to escape prejudice in the state courts of the plaintiff's home state by ensuring that a fair federal tribunal is available. In light of this broad purpose, Plaintiffs frame the forum defendant rule as a carve-out, premised on the understanding that defendants are unlikely to be “home-towned” in their home state's courts. Plaintiffs then explain the inclusion of the phrase “properly joined and served” as Congress's further recognition that crafty plaintiffs might take advantage of the forum defendant rule to secure a state-court trial by naming an unnecessary home-state defendant against which they did not intend to proceed. Thus, Plaintiffs assert that it is absurd to allow a home-state defendant to use an exception meant to protect defendants from unfair bias (in the courts of a plaintiff's home state) and language designed to shield them from gamesmanship (in the form of fraudulent joinder) to remove a lawsuit to federal court.
Plaintiffs are, of course, correct about the general purposes of the removal statute. See Lively v. Wild Oats Mkts., Inc., 456 F.3d 933, 940 (9th Cir. 2006) (“Removal based on diversity jurisdiction is intended to protect out-of-state defendants from possible prejudices in state court.”). But while it might seem anomalous to permit a defendant sued in its home state to remove a diversity action, the language of the statute cannot be simply brushed aside. Allowing a defendant that has not been served to remove a lawsuit to federal court “does not contravene” Congress's intent to combat fraudulent joinder. Encompass Ins., 902 F.3d at 153. In fact, Congress may well have adopted the “properly joined and served” requirement in an attempt to both limit gamesmanship and provide a bright-line rule keyed on service, which is clearly more easily administered than a fact-specific inquiry into a plaintiff's intent or opportunity to actually serve a home-state defendant. See Cheung, 282 F.Supp.3d at 643. Absurdity, then, cannot justify a departure from the plain text of the statute.
Plaintiffs also urge us to look past the language of Section 1441(b)(2) to avoid “non-uniform application” of the forum defendant rule based on the vagaries of state law service requirements. Plaintiffs are correct that allowing home-state defendants to remove on the basis of diversity before they are served might mean that defendants sued in some states – those that require a delay between filing and service, like Delaware – will be able to remove diversity actions to federal court while defendants sued in others – those that permit a plaintiff to serve an action as soon as it is filed – will not. But state-by-state variation is not uncommon in federal litigation, including in the removal context, see, e.g., Murphy Bros., Inc. v. Michetti Pipe Stringing, Inc., 526 U.S. 344, 354–55, 119 S.Ct. 1322, 143 L.Ed.2d 448 (1999) (discussing state-specific variations in connection with the deadline to remove a suit to federal court), and it does not follow from the existence of variation that we must look beyond the plain text of Section 1441(b)(2).
Put simply, the result here – that a home-state defendant may in limited circumstances remove actions filed in state court on the basis of diversity of citizenship – is authorized by the text of Section 1441(b)(2) and is neither absurd nor fundamentally unfair. We therefore have no reason to depart from the statute's express language and must affirm the district court's denial of Plaintiffs’ motions to remand.
Plaintiffs also challenge the district court's dismissal of the remaining sixty-four suits – fifteen of which are now before us – on the grounds that their failure-to-warn claims are preempted by the FDCA. Because we agree with the district court that Plaintiffs’ negligence and strict liability claims, as alleged, are preempted, we affirm the dismissal of those claims.4
The federal government regulates the manufacture, labeling, and sale of pharmaceuticals pursuant to the FDCA. 21 U.S.C. § 301 et seq.; see also Merck KGaA v. Integra Lifesciences I, Ltd., 545 U.S. 193, 196, 125 S.Ct. 2372, 162 L.Ed.2d 160 (2005). To bring a drug to market, a manufacturer must file a new drug application, which must explain the drugmaker's tests and studies, demonstrate that the drug is “safe for use under the conditions prescribed,” and include proposed labeling language. 21 U.S.C. § 355(b)(1)(A), (b)(1)(F), (d). “The FDA's premarket approval of a new drug application includes the approval of the exact text in the proposed label.” Wyeth v. Levine, 555 U.S. 555, 568, 129 S.Ct. 1187, 173 L.Ed.2d 51 (2009).
The FDA can direct a pharmaceutical manufacturer to change a drug's label after it has entered the market, see 21 U.S.C. § 355(o)(4), but “manufacturers, not the FDA, bear primary responsibility for their drug labeling at all times,” Wyeth, 555 U.S. at 579, 129 S.Ct. 1187. Nevertheless, drug manufacturers are limited in their ability to unilaterally change the labels on their products. Specifically, to make a change on their own, a manufacturer must comply with the “changes being effected” (“CBE”) regulation, set forth at 21 C.F.R. § 314.70(c)(6)(iii). That regulation
allows drug manufacturers to change [a label] without the FDA's preapproval if the changes ‘add or strengthen a contraindication, warning, precaution, or adverse reaction,’ or ‘add or strengthen an instruction about dosing and administration that is intended to increase the safe usage of the drug product,’ in order to ‘reflect newly acquired information.’
Wyeth, 555 U.S. at 591, 129 S.Ct. 1187 (internal citations omitted) (quoting 21 C.F.R. § 314.70(c)(6)(iii)(A), and 21 C.F.R. § 314.70(c)(6)(iii)(C)). “Newly acquired information” can include either new data or new analyses of previously submitted data. See 21 C.F.R. § 314.3(b).
“The Supremacy Clause establishes that federal law ‘shall be the supreme Law of the Land ․ any Thing in the Constitution or Laws of any State to the Contrary notwithstanding.’ ” PLIVA, Inc. v. Mensing, 564 U.S. 604, 617, 131 S.Ct. 2567, 180 L.Ed.2d 580 (2011) (quoting U.S. Const. art. VI, cl. 2). Where federal and state law conflict – that is, where it is impossible for a party to follow both federal and state law – state law must give way. Id. Because manufacturers may unilaterally update a drug's label if the change complies with the CBE regulation, a state law failure-to-warn claim that depends on newly acquired information – information that Defendants could have added to their label without FDA approval – is not preempted. See Wyeth, 555 U.S. at 568–72, 129 S.Ct. 1187; see also In re Celexa & Lexapro Mktg. & Sales Practices Litig., 779 F.3d 34, 40–41 (1st Cir. 2015).
Following Wyeth, PLIVA, and a third case that addressed FDCA preemption of state law failure-to-warn claims in connection with generic drugs, Mutual Pharmaceutical Co., Inc. v. Bartlett, 570 U.S. 472, 133 S.Ct. 2466, 186 L.Ed.2d 607 (2013), the Courts of Appeals have synthesized the requirements to properly plead and then prove a state law failure-to-warn claim based on post-drug-release information. Thus, to state a claim for failure-to-warn that is not preempted by the FDCA, a plaintiff must plead “a labeling deficiency that [Defendants] could have corrected using the CBE regulation.” In re Celexa, 779 F.3d at 41; see also Dolin v. GlaxoSmithKline LLC, 901 F.3d 803, 812 (7th Cir. 2018). If the plaintiff meets that standard, the burden shifts to the party asserting a preemption defense to demonstrate that there is “ ‘clear evidence that the FDA would not have approved a change’ to the [prescription drug's] label.” In re Fosamax (Alendronate Sodium) Prods. Liab. Litig., 852 F.3d 268, 283 (3d Cir. 2017) (quoting Wyeth, 555 U.S. at 571, 129 S.Ct. 1187).
Plaintiffs’ claims here fail at the first step because, as the district court recognized, they consist of “conclusory and vague” allegations and do not plausibly allege the existence of newly acquired information that could have justified Defendants’ revising the Eliquis label through the CBE regulation. For example, the operative complaint in Fortner v. Bristol-Myers Squibb Co. & Pfizer Inc., which is representative of all the pleadings now before us, alleges that “[b]efore and after marketing Eliquis, [D]efendants became aware of many reports of serious hemorrhaging in users of [their] drugs” and that “[n]umerous ․ studies published after Eliquis’ approval in 2012 confirm the problematic bleeding events associated with Eliquis.” Joint Appendix 1345–46, ¶¶ 62, 66. However, for these “reports” and “studies” to constitute newly acquired information, as the term is defined in 21 C.F.R. § 314.3(b), they must have “reveal[ed] risks of a different type or greater severity or frequency than previously included in submissions to the FDA.” Id. As the district court observed, the Fortner complaint provides no basis upon which the court could conclude that the bleeding events covered by the alleged “reports” and “studies” presented a different type of risk than those the company had discussed with the FDA, or were more severe or more frequent than bleeding events that the government already knew about. See In re Celexa, 779 F.3d at 42–43. Accordingly, Plaintiffs’ complaints were properly dismissed.
On appeal, Plaintiffs attempt to make hay out of the district court's references to Utts II in its Fortner opinion. Specifically, Plaintiffs argue that the Utts complaint attached nine “reports, studies, and articles” that the district court relied on in concluding that the Utts plaintiffs had failed to allege “newly acquired information” within the meaning of the relevant regulations. Utts II, 251 F.Supp.3d at 663–72. But because the Fortner complaint removed all references to the reports, studies, and articles that were attached to the Utts complaint (compare S.D.N.Y. Case. No. 16-cv-5668 (DLC), Doc. No. 33 with Joint Appendix 1335–72), Plaintiffs now argue that the district court's reliance on Utts II in Fortner must have improperly considered the reports, studies, and articles, which Plaintiffs assert were not part of the pleadings in Fortner.
Plaintiffs misread Fortner. Although the district court explained that the allegations in the operative complaint in that case simply amounted to a less-detailed restatement of the Utts II allegations, it did not dismiss the Fortner complaint because of the insufficiency of the sources cited in the second amended Utts complaint. Instead, it reasoned that the Fortner complaint – and, by extension, the other complaints now before this court – did not provide enough information about the existence of newly-acquired information to meet the Rule 8 threshold of “a short and plain statement of the claim showing that the pleader is entitled to relief.” Fed. R. Civ. Proc. 8(a)(2). Because we agree that Plaintiffs’ complaints lack sufficient factual allegations to state a claim that is not preempted, we affirm the judgments below.
Moreover, had the district court erred in the manner Plaintiffs suggest (which it did not), we would still affirm, since – as discussed above – Plaintiffs’ complaints simply do not contain sufficient factual information to state a claim, and we “may affirm on any grounds for which there is a record sufficient to permit conclusions of law.” Mitchell v. City of New York, 841 F.3d 72, 77 (2d Cir. 2016) (quoting Holcomb v. Lykens, 337 F.3d 217, 223 (2d Cir. 2003) ).
For the foregoing reasons, we AFFIRM the district court's denial of the motions to remand the Transferred Actions and AFFIRM the district court's dismissal of the fifteen actions now before this Court.
1. The other forty-nine cases were voluntarily dismissed prior to oral argument. See Motion Order, No. 17-2638 (2d Cir. June 20, 2018), ECF No. 128.
2. The parties do not contest that the plaintiffs in all fifteen actions now before this Court are diverse from Defendants.
3. The District of Delaware reached the same conclusion – that the text of Section 1442(b)(2) was no barrier to pre-service removal by a home-state defendant – in denying the motion to remand the thirty-three California actions. Young, 2017 WL 2774735, at *2. However, those actions are not before this Court.
4. While Plaintiffs purport to contest the dismissal of their complaints in their entirety, they make no arguments regarding the district court's conclusion that their breach of warranty, fraud, or state consumer protection law claims were inadequately pled. For that reason, we decline to address those claims. See Norton v. Sam's Club, 145 F.3d 114, 117 (2d Cir. 1998) (“Issues not sufficiently argued in the briefs are considered waived and normally will not be addressed on appeal.”). And while Plaintiffs do challenge the district court's purported error in “sua sponte considering, ruling on, and granting motions that Appellees never made in their cases,” as well as its finding that the adequacy of Eliquis's label could be determined at this stage in the proceeding, we need not reach these arguments given our affirmance of the district court's conclusion as to preemption.
Richard J. Sullivan, Circuit Judge: