NEJM Retracts Pivotal Trial of FDA-Approved Drug

The New England Journal of Medicine (NEJM) retracted the pivotal study supporting FDA’s approval of avacopan (Tavneos) for severe forms of anti-neutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis.

The journal said the two academic authors on the study — David Jayne, MD, and Peter Merkel, MD, MPH — requested the retraction after discovering that nine patients’ primary endpoint assessments had been re-adjudicated after database lock and after the phase III trial data were unblinded.

Those actions, revealed as part of an ongoing FDA investigation into conduct of the ADVOCATE trial, took place without the knowledge of Jayne and Merkel, according to NEJM.

“This was not disclosed in the article and is inconsistent with proper research conduct,” the retraction notice stated. “The editors therefore retract the article.”

The FDA called for avacopan to be withdrawn from the market earlier this year and asserted that “manipulated” data submitted by developer ChemoCentryx supported the approval. The agency also flagged risks associated with avacopan, including fatal cases of drug-induced liver injury and vanishing bile duct syndrome, a condition where the bile ducts are progressively destroyed.

Last week, a key European Medicines Agency committee recommended pulling the marketing authorization for the drug in Europe, saying that avacopan’s “benefits are no longer proven to outweigh its risks” given what the committee called breaches in good clinical practice.

Amgen has rebuffed calls to withdraw the drug — the company acquired the rights for avacopan after the oral complement 5a receptor inhibitor was approved in 2021 as an adjunctive treatment for adults with granulomatosis with polyangiitis and microscopic polyangiitis, two forms of severe ANCA-associated vasculitis.

ADVOCATE randomized patients to either avacopan or prednisone tapering with standard cyclophosphamide or rituximab therapy in each arm. The primary endpoint of sustained remission at week 52 was observed in 65.7% of the avacopan group and in 54.9% of the prednisone group, which was deemed significant for both noninferiority and superiority.

But the FDA said that after an initial analysis of the primary endpoint failed to achieve significance, unblinded ChemoCentryx personnel selected participants for readjudication and ultimately changed five patients treated with avacopan from “not in sustained remission” to “sustained remission,” resulting in a statistically significant superiority benefit over the control arm.

“Amgen takes scientific integrity seriously and respects the role of journals in upholding the peer review process,” a spokesperson said in response to the news of the retraction.

The company said it recruited the Duke Clinical Research Institute earlier this year to conduct an independent, fully blinded readjudication of ADVOCATE’s primary endpoint results.

“Those results will be shared with the FDA as part of our hearing submission due by July 29 and submitted for publication,” the spokesperson said.