Merck Sharp & Dohme Corp. v. Albrecht


Is a state law failure-to-warn claim preempted when the FDA rejects a drug manufacturer’s proposed warning, or should the jury consider whether the FDA would have approved a differently worded warning?

Oral argument: 
January 7, 2019

The Supreme Court will determine whether the Food and Drug Administration’s (“FDA”) prior rejection of a drug manufacturer’s proposed warning preempts a state law failure-to-warn claim against the same manufacturer. Merck Sharp & Dohme Corp. (“Merck”), a corporation that manufactures the drug Fosamax, argues that under the Supremacy Clause, a state law claim for failing to warn about a link between a drug and abnormal femoral fractures is precluded by the FDA’s rejection of a proposed warning about the fractures. Doris Albrecht, a consumer who took Fosamax and suffered atypical femoral fractures as a result, counters that the FDA’s rejection of the warning is not dispositive and that evidence showing that the FDA would have approved an alternative warning should be considered by a jury. The Third Circuit ruled that whether the FDA would have approved an alternative warning is a question of fact that should go to the jury. Merck is now appealing that decision in a case that will have implications for drug warnings, FDA reporting, and public health.

Questions as Framed for the Court by the Parties 

Whether a state-law failure-to-warn claim is pre-empted when the Food and Drug Administration rejected the drug manufacturer's proposal to warn about the risk after being provided with the relevant scientific data, or whether such a case must go to a jury for conjecture as to why the FDA rejected the proposed warning.


Before a drug manufacturer can permanently add warnings to a drug label, the Food and Drug Administration (“FDA”) must approve the addition. There are two different ways to implement this change. The first changes the label immediately and allows the FDA to review the change later. The second requires seeking approval from the FDA before making the change. The latter method is known as a Prior Approval Supplement (“PAS”).

Merck Sharp & Dohme Corp. (“Merck”) manufactures Fosamax, a drug that treats osteoporosis. In 2008, Merck submitted a PAS to the FDA that suggested adding information about atypical femoral fractures to the “Warnings & Precautions” and “Adverse Reactions” sections of the Fosamax label. Merck’s proposal involved language about stress fractures. The FDA emailed Merck, explaining that although they were not prepared to add language about stress fractures to the Warnings & Precautions section, they would approve a warning in the Adverse Reactions section regarding “low energy femoral shaft and subtrochanteric fractures.” The FDA sent an additional letter to Merck explaining that the warning’s use of the term “stress fracture” was problematic because it might not be associated with the atypical femoral fractures that other scientific studies describe.

In 2010, a FDA task force published a report explaining that they had found evidence of a link between drugs like Fosamax and abnormal femoral fractures even though the existing research did not indicate that this link was a causal one. Later that year, the FDA requested that Merck add language to the Fosamax label’s Warnings & Precautions section. The language would state that atypical femoral fractures were reported in patients taking medications like Fosamax. Merck proposed additional language, which would once again include information about stress fractures. The FDA rejected this suggestion, explaining that the term was inappropriate because doctors were unlikely to sufficiently appreciate the severity.

After the label change, over 1,000 patients who had taken Fosamax and who had suffered atypical femoral fractures sued Merck for failure-to-warn about the risk of atypical femoral fractures. One of these patients was Doris Albrecht.

The United States District Court for the District of New Jersey heard a test case which had become a multidistrict litigation. The district court addressed Merck's primary argument that FDA regulations preempted the state law failure-to-warn claim because there was clear evidence that the FDA would not have approved a label change. The district court agreed with Merck that the claim was preempted, citing the clear evidence standard from Supreme Court precedent. After this decision, Merck submitted an order to show cause. The order asked the other plaintiffs in the multidistrict litigation to explain why their cases should not also be preempted. After reviewing the plaintiff’s arguments, the district court concluded that all of the claims were preempted and should be dismissed.

Albrecht appealed to the United States Court of Appeals for the Third Circuit, arguing that the FDA rejected Merck's proposed label change because of the “stress fracture” language. She claimed that if Merck had made more specific references to atypical femoral fractures, the warning would have been approved. Merck, on the other hand, argued that the FDA would have rejected any suggested warning linking atypical femoral fractures with Fosamax use because there was not enough evidence that the two were connected.

The Third Circuit held that the clear evidence standard articulated by the Supreme Court in earlier cases established a standard of proof that Merck must meet. Furthermore, the Third Circuit concluded that whether the FDA would have ultimately rejected a label change is a question of fact that should be answered by the jury rather than the judge. The United States Supreme Court granted certiorari on June 28, 2018.



Merck argues that a manufacturer cannot be penalized for failing to warn about a disclosed risk when the FDA has rejected the manufacturer’s proposed warning about that risk. . Merck points to the Supremacy Clause, which states that where federal and state law conflict, federal law takes precedence. Merck asserts that such a conflict exists if it is “impossible for a private party to comply with both state and federal requirements.” Merck further asserts that because the FDA did not permit it to warn about the risk of atypical femoral fractures, penalizing Merck would be punishing them for not taking an action that was forbidden under federal law. Thus, Merck concludes that the conflict at issue falls within the scope of the Supremacy Clause, and that the state law claim should be preempted.

Merck cites existing Supreme Court case law to further support this position. In the past, Merck notes, the Court has held that if federal law had prevented a manufacturer from updating a label on one of its products, then a state law failure-to-warn claim would have been preempted. Merck asserts that the Court has also held that where it is “impossible” for a manufacturer to comply with both state and federal regulations, conjecture about ways in which the parties could act to reconcile the conflicting state and federal duties will not defeat a preemption defense. According to Merck, the case law also establishes that state law failure-to-warn claims are preempted if the manufacturer can prove that the FDA would have rejected a proposed warning. Merck posits that the FDA actually did reject Merck’s request to include a warning after Merck informed the FDA about the potential risks associated with the product. Merck therefore maintains that this rejection should have preemptive effect.

Albrecht contends that federal law does not typically preempt failure-to-warn claims arising under state law against prescription drug manufacturers. Albrecht asserts that in any preemption case, “the purpose of Congress is the ultimate touchstone.” Accordingly, Albrecht claims that when Congress passes a statute containing an explicit preemption clause, the state law covered by that clause is preempted. Albrecht further explains that Congress has never enacted a preemption clause concerning prescription drugs. According to Albrecht, the lack of an express preemption clause, coupled with Congress’ awareness of existing drug litigation in state courts, provides strong evidence that Congress did not intend for an FDA rejection of a proposed warning to preempt state law claims.

Albrecht also argues that Merck could have complied with both state and federal requirements. Albrecht asserts that nothing prohibited Merck from updating its FDA-approved label to meet state law warning requirements, after which it could have used the updated label without waiting for FDA approval. Additionally, Albrecht points to the Supreme Court’s statement in prior cases that the impossibility defense does not apply “absent clear evidence that the FDA would not have approved a change to [the] label.” Albrecht claims that Merck has offered no such clear evidence and that, to the contrary, the Food and Drug Administration Amendments Act confers discretionary authority on the FDA which it can use to require changes to warning labels. As such, Albrecht maintains that Merck’s duty to use adequate, up-to-date warning labels did not end with the FDA’s rejection of one proposed warning, and that this single rejection does not preempt state law claims resulting from Merck’s failure to warn.


Merck claims that the Third Circuit incorrectly treated the FDA’s rejection of the proposed warning as a single factor in determining whether Merck still had an obligation to rephrase its warning or to propose different warnings. In fact, Merck asserts that the FDA’s rejection was dispositive on this issue and that there is no need for a counterfactual analysis in light of the fact that the FDA handed down an actual decision. Merck contends that the Third Circuit’s finding that the FDA may have accepted an alternative wording of the proposed warning constitutes an impermissible “second-guess[ing]” of the FDA’s conduct. Merck posits that the Court has previously held that a presumption of regularity attaches to the actions of federal agencies, which should be presumed to have properly carried out their duties unless there is “clear evidence” suggesting otherwise. Merck maintains that no clear evidence exists in the present case. Additionally, Merck contends that this second-guessing constitutes the kind of conjecture that the Court prohibited in previous cases.

Merck further argues that if the Third Circuit is correct in holding that a counterfactual inquiry should be made into what the FDA might have done if presented with an alternative warning, the meaning and effect of what the FDA actually did is still a question of law for a judge— rather than a jury—to decide. Merck claims that questions about the meaning of an agency’s action are “classic questions of law,” and the Court has previously assessed the preemptive effect of an agency’s action as a question of law. Merck contends that the Third Circuit’s decision lays out an inappropriate standard of proof by requiring it to prove through clear and convincing evidence that the FDA would have rejected an alternate warning. Because such evidentiary standards do not apply to questions of law, Merck asserts that the Third Circuit’s conclusion is misguided.

Finally, Merck posits that even if the Court were to reject all of its prior arguments, it would still prevail under the Third Circuit’s framework because the FDA stated that it would not have authorized a proposed warning from Merck until October 2010. As such, Merck maintains that no jury could conclude that the FDA would have accepted any alternative warnings at an earlier time. Merck argues that to believe otherwise, a jury would have to conclude that the FDA either mislead the Court or misunderstood the reasons for its own actions, and that this kind of speculation runs contrary to the presumption of regularity that attaches to federal agency actions.

Albrecht counters that Merck never proposed an adequate warning addressing atypical femoral fractures to the FDA. Thus, Albrecht claims that the Third Circuit was correct in considering how the FDA might have responded to such a warning. Albrecht notes that Merck proposed a warning focusing on “stress fractures” rather than atypical femoral fractures, and the FDA exercised its discretionary authority to reject that warning. Albrecht contends that both the FDA and Merck have acknowledged the distinction between atypical femoral fractures and stress fractures and that instead of precluding a warning addressing atypical femoral fractures, the FDA’s response “invited” Merck to add such a warning. Accordingly, Albrecht argues that a counterfactual analysis is proper here because the warning that the FDA rejected did not actually address the risk that gave rise to the state law failure-to-warn claim.

Albrecht argues that the question of whether the FDA would have rejected an adequate warning is a question of fact rather than a question of law and is therefore appropriate for a jury to decide. Albrecht cites the traditional rule that juries decide factual or mixed-fact questions in common-law tort suits, and asserts that this rule is no less applicable to factual issues arising out of a preemption claim. Albrecht claims that the traditional rule applies despite the fact that the question at hand involves evaluating the actions of a federal agency. According to Albrecht, this is true because the question calls for analysis of what the agency might have done, not of what it actually did. Albrecht also maintains that federal drug regulations do not articulate a rationale for preventing factual questions relevant to drug regulation from going before a jury. Thus, Albrecht argues that the “clear evidence” standard for questions of fact is appropriate.

Finally, Albrecht contends that the statement by the government in its amicus brief that the FDA would not have approved a proposed warning until October 2010 is a legal argument, not a factual representation. Albrecht asserts that the government clearly indicated in its amicus brief which of its representations were factual and, in contrast, described the FDA’s claim concerning approval as an interpretation of the FDA’s response to Merck. Albrecht argues that government litigation briefs are not entitled to deference, as the Court has previously given only “some weight” to the government’s conclusions regarding an agency’s objectives and positions.



The Pharmaceutical Research and Manufacturers of America and Biotechnology Innovation Organization (“PhRMA and BIO”), in support of Merck, argue that affirming the Third Circuit’s decision will lead drug manufacturers to attach speculative warnings to their products that are not grounded in real scientific evidence. Furthermore, PhRMA and BIO contend that if the Court affirms the Third Circuit, drug manufacturers will feel substantial pressure to include overly long labels on their products that doctors will not read. PhRMA and BIO further assert that a ruling in favor of Albrecht will result in risks being overstated, which could distort the analysis that all doctors must perform when weighing a medication’s risks against its benefits.

The Cato Institute, on the other hand, in support of Albrecht, argues that reversing the Third Circuit will invite drug manufacturers to abuse the preemption test by incentivizing them to propose inadequate warnings designed to cause delay and avoid labeling their products with warnings that might discourage customers. Similarly, Jerome P. Kassirer et al., in support of Albrecht, argue that in situations where drug manufacturers have given the FDA a purposely inadequate warning, a decision in favor of Albrecht would allow failure-to-warn litigation to compensate for this information gap. Furthermore, Jerome P. Kassirer et al. assert that reversing the Third Circuit would grant drug manufacturers almost complete immunity through federal preemption such that they would have little incentive to provide adequate warnings for their drugs.


PhRMA and BIO contend that drug manufacturers will be inclined to submit multiple versions of the same warning to avoid any potential future litigation. Accordingly, PhRMA and BIO argue that affirming the Third Circuit’s decision would overwhelm the FDA with alternate versions of warnings. In the same vein, the Product Liability Advisory Council and the Chamber of Commerce posit that the Third Circuit’s approach motivates drug manufacturers to continually propose labeling changes to the FDA so that potential plaintiffs will not be able to defeat preemption by suggesting alternative warnings.

Jerome P. Kassirer et al. respond that the Third Circuit’s decision should be affirmed because the FDA needs assistance from tort litigation to help monitor pharmaceutical risk. Furthermore, Jerome P. Kassirer et al. assert that this increased litigation would help the FDA gather more accurate information after a drug has gone to market. Finally, Jerome P. Kassirer et al. maintain that a ruling for Albrecht would push drug manufacturers to report more accurate risk information to the FDA and work more closely with the agency in an effort to limit their own liability.


PhRMA and BIO argue that if the Court affirms the Third Circuit, the resulting increase in litigation within the drug manufacturing industry will discourage companies from bringing new drugs to market. Furthermore, PhRMA and BIO assert that it is unlikely that affirming the Third Circuit will help to uncover new safety risks or produce any public health benefits when the FDA has already rejected the need for a warning. Similarly, the Product Liability Advisory Council and the Chamber of Commerce argue that allowing failure-to-warn litigation to proliferate in state courts despite the high likelihood that the FDA would have rejected the proposed warnings will open manufacturers up to unpredictable litigation costs, making them less likely to invest in new drugs.

Jerome P. Kassirer et al. respond that affirming the Third Circuit will bolster public health because litigation can shed light on adverse effects when drug manufacturers attempt to downplay the health risks of their products. Furthermore, tort law professors John Goldberg and Benjamin Zipursky point out that state tort claims help to ensure that the drug warning labels accurately and clearly explain the relevant risks for a particular product so that the drug manufacturer can avoid failure-to-warn liability. Jerome P. Kassirer et al. add that if the Third Circuit’s decision is overturned, doctors will have less information to share with their patients, making it more difficult for them to select the drug that is best suited for each individual.

Edited by 

Additional Resources