Key Points

  • The First Circuit held that a state-law claim faulting the label a manufacturer proposed in its biologics license application is preempted, because that application is a request for FDA approval rather than a unilateral act.
  • The court independently held that the FDA knew Aimovig's trials excluded patients with seizure and neurological histories, considered a cerebrovascular warning, and rejected it, so a request for Warner's proposed label would have been denied.
  • The court reversed the denial of leave to amend, holding that the Rehni, Liu, Sorby-Adams and Zhai articles plausibly qualify as newly acquired information supporting a post-approval label change under 21 C.F.R. § 601.12(f)(2).
  • The court declined to exclude animal studies that did not test Aimovig, noting that the FDA itself relied on such studies, and held that a study published before approval can be newly acquired if it was never submitted to the FDA.
  • Amgen did not rely on a 'clear evidence' defense that the FDA would have rejected the change anyway, and the court left that reserved defense unaddressed on remand.

A federal appeals court on Friday revived a mother's wrongful-death suit against Amgen Inc. over its migraine drug Aimovig. The court ruled that she may try to prove Amgen could have added a brain-injury warning to the drug's label on its own after regulators approved it.

A three-judge panel of the U.S. Court of Appeals for the First Circuit affirmed the dismissal of Elissa Warner's original complaint, which faulted the label Amgen proposed when it first sought Food and Drug Administration approval. Judge Kayatta, writing for a unanimous panel that included Judges Montecalvo and Lipez, reversed the trial court's refusal to let Warner amend her complaint with a narrower theory based on four scientific articles.

Under the decision, claims that a manufacturer should have proposed a different label to the FDA are preempted, meaning federal law displaces them. Claims that it should have strengthened the label after approval may proceed under the "changes being effected" rule, which lets a manufacturer add a warning based on "newly acquired information" without the agency's advance permission. The panel also refused to rule out, at the pleading stage, animal studies that never tested the drug itself.

According to the complaint, which the court accepted as true, Lucas Warner had a history of seizures and cerebrovascular disease. He took a single dose of Aimovig on June 17, 2018, at age 23, and within 48 hours suffered seizures and was placed in a coma. He was left with irreversible brain damage and died in November 2020 at 25. His death certificate states that "complications from Aimovig" "contribut[ed]" to his death.

Aimovig blocks the receptor for calcitonin gene-related peptide, or CGRP, a molecule that plays a role in migraines. Its label disclosed that the clinical trials excluded people who had suffered a stroke or similar events within the prior 12 months, but not that they excluded anyone who had ever had seizures or neurological disorders. Warner sued in state court, Amgen removed the case to federal court, and U.S. District Judge Julia E. Kobick dismissed it in February 2025.

On the original complaint, the panel applied precedent holding that state law is preempted when a manufacturer cannot comply with it without first obtaining an agency's approval. Kayatta wrote that "a manufacturer's ability to ask the FDA to include something in a label falls short of the unilateral power required to preclude an impossibility preemption defense." A claim like Warner's fails, the opinion said, "because a BLA submission is, by its nature, a request for agency approval," referring to the biologics license application Amgen filed for Aimovig.

The panel offered a second, independent ground. The FDA's review documents showed that the agency knew the trials excluded people with seizure disorders, and a reviewer in its Division of Neurology Products proposed a warning of "theoretical concern" for "patients with major . . . cerebrovascular disease." The FDA's Division of Risk Management rejected that proposal, finding "insufficient evidence to include th[e] theoretical risk in labeling." The label as approved contained no warnings, and Kayatta wrote that the label Warner wanted "would have been denied—and indeed, was effectively denied."

Warner raised the post-approval theory at a hearing on Amgen's motion to dismiss and later submitted nine articles. Kobick concluded that the articles did not "plausibly constitute newly acquired information" and found no "precedent for treating a single animal study not involving the manufacturer's drug as newly acquired information."

The appeals court held that four of the nine could qualify: a 2007 mouse study by Rehni, a 2011 rat study by Liu, a 2017 literature review by Sorby-Adams and a 2018 mouse study by Zhai. Read together, the court said, the studies suggest that CGRP helps protect the blood-brain barrier and brain cells after a stroke or other injury, and that blocking CGRP may weaken that protection.

"We are inclined to read the record as the district court did," Kayatta wrote. But without expert testimony, the opinion continued, "that inclination falls short of allowing us to conclude with certainty that the CBE procedure would have failed." Because Amgen raised preemption as an affirmative defense, the opinion said, "we ask only whether we can say with certitude that they do not" qualify as new information.

The FDA's review of "the world's literature" was limited to the risk that blocking CGRP could worsen ischemia, or loss of blood supply, by impairing blood-vessel dilation, the court said, and the record did not show that the four studies ever reached the agency. A study published before approval can still be newly acquired if it was never "previously submitted" to the FDA. The panel also noted that the FDA itself relied on dog and rat studies, some of which never administered a CGRP blocker. "We see no clear reason why we should certainly exclude animal studies related to CGRP in our assessment of newly acquired information if the FDA did not do the same," Kayatta wrote.

On causation, Amgen had argued that Warner must show a "plausible connection between Aimovig and an increased risk of seizures" in patients like her son. The panel agreed with Warner that she need only allege a link between the drug and worsened outcomes after brain injury. "[W]e are simply not certain that there is no causal relationship between Aimovig and the extent of the injuries Lucas experienced," Kayatta wrote.

Abbye R. K. Ognibene of Hagens Berman Sobol Shapiro LLP argued for Warner. Jessica L. Ellsworth of Hogan Lovells US LLP argued for Amgen.

Amgen can still defeat the revived claim by showing "clear evidence" that the FDA would have rejected the label change even if it had seen the studies. Amgen chose not to make that argument at this stage, and the panel did not consider it. The case returns to the district court, limited to Warner's claim that the four articles allowed Amgen to warn that Aimovig could reduce CGRP's protective effects in the brain after cerebral injury.