Trial Magazine

For decades, drug manufacturers have raised “impossibility preemption”—that they could not comply with both state and federal requirements—as a defense to avoid liability for state law failure-to-warn claims. Over the past decade, the U.S. Supreme Court has repeatedly provided litigants and lower courts with guidance over certain standards governing impossibility preemption.¹

But it also has left open a crucial issue that is often at the heart of the defense: the type of evidence a manufacturer may rely on to show that compliance with state law would be impossible without also violating federal law. Since Wyeth v. Levine was decided in 2009, drug manufacturers have pushed lower courts to rule that state failure-to-warn claims are preempted when federal law permitted a company to strengthen its drug’s warning label so long as the FDA hypothetically might have rejected that new warning later.

But after the Court’s 2019 ruling in Merck Sharp & Dohme Corp. v. Albrecht ² firmly shut the door on this “hypothetical” approach to impossibility preemption, lower courts have begun to follow suit by rejecting manufacturer attempts to rely on hypotheticals as evidence for their impossibility preemption claims. A close review of these cases should assist plaintiff attorneys facing similar efforts by manufacturers to foreclose state law claims based on impossibility preemption.

Where It Started

The Supreme Court first addressed a manufacturer’s evidentiary burden a decade ago in Wyeth. The plaintiff had sued a drug manufacturer for failing to add a warning to its anti-nausea drug Phenergan about the risk of developing gangrene.³ The manufacturer argued that it could not have adopted the warning because federal regulations prevented it from changing the drug’s label.

The Court held the failure-to-warn claim was not preempted, ruling that because the FDA’s regulatory process permitted manufacturers to add new warnings to their labels, Wyeth could have added a stronger warning to its label.⁴ In support of this conclusion, the Court pointed to the “Changes Being Effected” (CBE) regulation, which permitted manufacturers to “add or strengthen the labeling” of their drug as long as they originally created the drug and were in possession of “newly acquired information” that justified the label change.⁵ The Court briefly noted that even though FDA regulations permit drug companies to add new warnings to their labels, the FDA can later reject those changes.⁶ But, the Court said, that would not lead to preemption in a failure-to-warn case unless there was “clear evidence that the FDA would not have approved [the] change” in question.⁷

In the years since Wyeth, that observation sparked a novel breed of “hypothetical impossibility” preemption, in which drug manufacturers have advanced a theory that does not turn on actual impossibility—whether a drug manufacturer could have added particular warnings to its label. Instead it turns on hypothetical impossibility—whether the manufacturer was right not to include the warning because the FDA could have and would have rejected it.⁸

In Forst v. SmithKline Beecham Corp., for example, a manufacturer pointed to “the amount of interaction it had with [an] agency” and the agency’s repeated review of a product’s safety data without requiring any new warning as clear evidence of preemption.⁹

In other cases, manufacturers focused on agency rejections of citizen petitions as “clear evidence that the FDA would not have approved a change.”¹⁰ As these cases illustrate, clear evidence had, for some, become a license to construct elaborate counterfactual scenarios in which the whole range of an agency’s actions—sometimes spanning the course of decades—was mined for hints of regulatory intent.¹¹

Wyeth’s Clear Evidence Standard Cannot Be Met by Hypotheticals

Impossibility preemption, however, does not deal in hypotheticals. And in Merck, the Supreme Court squarely rejected manufacturer efforts to treat Wyeth as a license for uncabined speculation based on a wide—and ever-increasing—assortment of contextual clues. The Court made clear that the question at the heart of the impossibility preemption inquiry in failure-to-warn cases—“agency disapproval”—is a “tightly circumscribed legal analysis.”¹² To answer it, “the judge must simply ask himself or herself whether the relevant federal and state laws ‘irreconcilably conflict.’”¹³

In Merck, the Court reiterated that it is not sufficient for there to be a “‘possibility of impossibility.’”¹⁴ Instead, impossibility exists only when federal law actually prohibits adding warnings to a label that would comply with state law.¹⁵ And that remains true, the Court held, even when an agency still has the “authority to reject labeling changes” to a product.¹⁶ Regardless of an agency’s authority, the manufacturer bears the “ultimate responsibility for its label”; it cannot avoid “state laws that would penalize [it] for failing to warn consumers of the risks” associated with its product without clearly showing that compliance would in fact force it to violate federal law.¹⁷

The Merck Court provided straightforward guidance on how to perform an impossibility preemption inquiry. Although the Court chose not to define clear evidence in terms of evidentiary standards, it identified the only type of evidence that could count: “agency actions taken pursuant to congressionally delegated authority.”¹⁸ The Court further defined the specific forms of agency action that could trigger impossibility preemption as a disapproval of a warning:

• “by means of notice-and-comment rulemaking setting forth labeling standards”
• “by formally rejecting a warning label that would have been adequate under state law”  
• “by taking other agency action carrying the force of law.”¹⁹

In Merck, the Court also established a clear two-step framework for determining whether a manufacturer has met the “demanding defense” of impossibility preemption.²⁰ First, a court must determine whether a manufacturer “fully informed” an agency of a product’s risks; if it failed to do so, the inquiry stops there and no impossibility preemption can exist.²¹ Second, if (and only if) the agency was fully informed of a product’s risks, the manufacturer must then show that the agency, acting within the scope of its lawful authority, would not approve a change to a product’s label.²² Only under those circumstances can a court conclude that state law failure-to-warn claims are foreclosed.

The Aftermath

In the wake of Merck, defendants have continued to argue impossibility preemption based on hypotheticals. But courts confronting this argument have had little difficulty rejecting it by applying Merck’s two-part test.

Third Circuit. The recent decision in In re Avandia Marketing, Sales and Products Liability Litigation illustrates how this approach works in practice.²³ The manufacturer sought to bar state law failure-to-warn claims under an impossibility preemption defense by arguing, first, that it fully informed the agency about the product’s safety risks because it “provided all ‘material’ information” to the agency and, second, that the agency had “rejected the proposed warning.”²⁴

The Third Circuit rejected both claims, holding that the manufacturer had “failed to satisfy either prong of Merck’s two-prong test.”²⁵ The court explained that the manufacturer had not shown that it fully informed the agency “of the justification . . . for the warning required by state law”—the FDA found that the information provided was inadequate and had informed the manufacturer that it “needed to submit various data and information in order to address the deficiency.”²⁶

The Third Circuit also flatly dismissed the manufacturer’s argument that none of the additional requested information was “‘material’ to its proposed warning,” explaining that a manufacturer “is not the arbiter of which data and information is or is not ‘material’” to an agency’s decision to approve or reject a label change.²⁷ Instead, it is the agency that can make that determination.²⁸ And the question of whether an agency was fully informed must be “tethered in time” to the question of whether the agency “indeed rejected the proposed warning.”²⁹ Were it otherwise, “the ‘fully informed’ prong of the test espoused in Merck would be rendered superfluous.”³⁰

The court also rejected the manufacturer’s effort to show that the agency actually “rejected the proposed warning.”³¹ The manufacturer pointed the court to an agency letter stating that the manufacturer’s request for a label change was “not approvable.”³² But the court explained that the agency refused to approve the label “because the information presented” was “inadequate”—not because the agency “was unconvinced of the need for a strong warning.”³³ The court explained that, at most, the letter indicated that it was “possible” that the agency “could have rejected the label change after receiving the various data and information it requested.”³⁴ But, the court reiterated, “the possibility of impossibility is not enough.”³⁵

Seventh Circuit. Earlier this year, the Seventh Circuit used Merck’s two-part test to guide its impossibility preemption inquiries in Dolin v. GlaxoSmithKline LLC, signaling its view that Merck adopted a clear and demanding approach when it comes to impossibility preemption in the failure-to-warn context.³⁶ Because clear evidence is “evidence that shows the court that the drug manufacturer fully informed the FDA of the justifications for the warning required by state law and that the FDA, in turn, informed the drug manufacturer that the [agency] would not approve a change to the drug’s label to include that warning,” the Seventh Circuit appears to have recognized that to succeed on an impossibility defense a “manufacturer must have actually requested a change and . . . the FDA [must have] rejected it.”³⁷

And it likewise read Merck’s focus on actions taken pursuant to the FDA’s congressionally delegated authority as at least an implicit rejection of the argument that a manufacturer can rely on “less formal exchanges of correspondence” to establish impossibility.³⁸ 

Lower courts. District courts have also applied Merck’s two-part test to reject manufacturer arguments that turn on something less than an agency’s formal, concrete, and informed disapproval of a particular warning. In Crockett v. Luitpold Pharmaceuticals Inc., the manufacturers argued that the plaintiff’s failure-to-warn claim was preempted on the theory that if the manufacturers “had submitted a new label—with additional warnings—to the FDA, the FDA would have rejected the warning.”³⁹ The court explained that the manufacturers had failed to “point to any evidence” that they had “proposed a stronger warning to the FDA or that the FDA would have rejected a different warning label.”⁴⁰ As a result, under Merck, the manufacturers had not met their burden.⁴¹

Even after Merck, failure-to-warn claims can still face credible preemption arguments. In particular, a claim may be preempted when a plaintiff cannot plausibly demonstrate that under the CBE regulation, a manufacturer could have made a label change without prior FDA approval. For example, in Drescher v. Bracco Diagnostics Inc., a manufacturer established its preemption defense by showing that there was no newly acquired information that could have justified a label change.⁴² But assuming a manufacturer could have changed its label without FDA approval, the manufacturer can no longer rely on speculation about what the FDA might have done.

Going Forward

Given this emerging understanding of impossibility preemption after Merck, plaintiffs facing this defense should carefully focus their analysis and responses on several key principles.

First, to show that federal law prohibited a drug manufacturer from adding a warning that would satisfy state law, the manufacturer must demonstrate that it fully informed the FDA of the justifications for the warning. This requirement, of course, assumes that the manufacturer had or should have had newly acquired information justifying a label change. But, in that event, the manufacturer must produce evidence that the FDA possessed all relevant information. And, significantly, it is the FDA—not the manufacturer—that determines relevancy. Only the agency can decide what is material to its approval or rejection of a label change.⁴³

Second, a manufacturer must show that the FDA would not approve changing the drug’s label to include the relevant warning. And here, as the Court in Merck held, the only agency actions that count are those that fall within the FDA’s congressionally delegated authority.⁴⁴ A manufacturer must point to a specific FDA action rejecting the proposed warning that satisfies this standard. In Merck, the Court identified three permissible avenues:⁴⁵

• a notice-and-comment rulemaking setting forth labeling standards⁴⁶
• a formal rejection of a warning label that would have been adequate under state law⁴⁷
• other agency action carrying the force of law.⁴⁸

And while this list may not be exclusive, whatever means the FDA uses to exercise its authority must lie within the scope of the authority Congress has lawfully delegated. Gone are the days when a manufacturer could rely on informal communications from mid-level agency bureaucrats to argue for impossibility preemption.⁴⁹

Third, neither of these two requirements can be divorced from each other. In other words, a manufacturer cannot rely on information provided to an agency that came after the agency acted to reject a proposed warning. Instead, a manufacturer must show that the FDA was fully informed of the justifications for the proposed warning “at the time that the FDA rejected the proposed warning.”⁵⁰ As a result, if a manufacturer wishes to rely on a particular agency action as proof that the agency rejected its proposed label change, it must also demonstrate that the agency possessed all of the information it deemed necessary to decide whether to approve or reject the proposed warning at the time it took such action.

Going forward, plaintiff attorneys should look to Merck and to lower courts’ application of that decision to guide their responses to impossibility preemption defenses.

---

Matthew Wessler is a partner at Gupta Wessler in Washington, D.C., and can be reached at matt@guptawessler.com.

---

Notes

1. See PLIVA, Inc. v. Mensing, 564 U.S. 604, 624 (2011); Mut. Pharm. Co., Inc. v. Bartlett, 570 U.S. 472, 490 (2013); Wyeth v. Levine, 555 U.S. 555, 573 (2009).
2. 139 S. Ct. 668 (2019); see also Matthew Wessler, Merck Moving Forward, Trial 60 (Oct. 2019); Matthew Wessler, Merck and Impossibility Preemption, Trial 42 (Apr. 2019).
3. Wyeth, 555 U.S. at 571.
4. Id. at 568, 570–71.
5. 21 C.F.R. §314.70(c)(6)(iii) (2019). See, e.g., Drescher v. Bracco Diagnostics Inc., 2020 WL 699878, at *6 (D. Ariz. Jan. 31, 2020) (state law failure-to-warn claims are preempted when the plaintiffs could not plausibly show “that the manufacturers could have changed their labels under the CBE regulations”); see also Goodell v. Bayer Healthcare Pharm., Inc., 2019 WL 4771136, at *3 (D. Mass. Sept. 30, 2019); Klein v. Bayer Healthcare Pharm., Inc., 2019 WL 3945652, at *4 (D. Nev. Aug. 21, 2019); Sabol v. Bayer Healthcare Pharm., Inc., 439 F. Supp. 3d 13 (S.D.N.Y. 2020).
6. Wyeth, 555 U.S. at 571.
7. Id.
8. See, e.g., Forst v. SmithKline Beecham Corp., 639 F. Supp. 2d 948 (E.D. Wis. 2009); Dorsett v. Sandoz, 699 F. Supp. 2d 1142 (C.D. Cal. 2010); Lofton v. McNeil Consumer & Specialty Pharms., 682 F. Supp. 2d 662 (N.D. Tex. 2010); Hunt v. McNeil Consumer Healthcare, 6 F. Supp. 3d 694 (E.D. La. 2014); Koho v. Forest Labs., 17 F. Supp. 3d 1109 (W.D. Wash. 2014).
9. 639 F. Supp. 2d at 954.
10. Dorsett, 699 F. Supp. 2d at 1157; see also Lofton, 682 F. Supp. 2d at 677–78.
11. See, e.g., Mason v. SmithKline Beecham Corp., 596 F.3d 387, 392–96 (7th Cir. 2010).
12. Merck, 139 S. Ct. at 1680.
13. Id. at 1679.
14. Id. at 1678.
15. Id.
16. Id. at 1677.
17. Id.
18. Id. at 1679.
19. Id.
20. Id. at 1672, 1678.
21. See id. at 1678.
22. Id.
23. See 945 F.3d 749, 756 (3d Cir. 2019).
24. Id.
25. Id. at 758.
26. Id.
27. Id. at 759.
28. Id.
29. Id.
30. Id.
31. Id. at 759–60.
32. Id.
33. Id. at 759.
34. Id. at 760.
35. Id. (quoting Merck, 139 S. Ct. at 1678).
36. See Dolin v. GlaxoSmithKline LLC, 951 F.3d 882 (7th Cir. 2020).
37. Id. at 890.
38. Id.
39. 2020 WL 433367, at *7 (E.D. Pa. Jan. 28, 2020).
40. Id.
41. Id. at *8. See also In re Testosterone Replacement Therapy Prods. Litig., 2019 WL 7290560, at *5–7 (N.D. Ill. Dec. 30, 2019) (a manufacturer’s previous approval of a similar product label does not constitute “clear evidence,” which must show the court that “the drug manufacturer fully informed the FDA” of the proposed warning and that “the FDA, in turn, informed the drug manufacturer that the [agency] would not approve a change”).
42. Drescher, 2020 WL 699878.
43. Id.
44. Merck, 139 S. Ct. at 1671.
45. Id. at 1679.
46. See, e.g., 21 U.S.C. §355(d) (2018); 21 C.F.R. §§201.57, 314.105 (2019).
47. See, e.g., 21 C.F.R. §§314.110(a), 314.125(b)(6) (2019).
48. Cf., e.g., 21 U.S.C. §355(o)(4)(A) (2018).
49. See, e.g., Lofton, 682 F. Supp. 2d at 677–78.
50. In re Avandia, 945 F.3d at 759 (emphasis in original).