The FDA has moved to pull Tavneos (avacopan) from the U.S. market, alleging that manipulated clinical trial data propped up the drug’s 2021 approval for a rare and potentially fatal blood vessel disease. In a formal withdrawal proposal issued in spring 2026, the agency’s Center for Drug Evaluation and Research (CDER) said unblinded study personnel altered results in the only trial used to justify the drug’s approval, making it appear effective when it may not have been.
For the roughly 40,000 to 80,000 Americans living with ANCA-associated vasculitis, a condition in which the immune system attacks small blood vessels and can destroy kidneys, lungs, and other organs within weeks, the announcement lands in already difficult territory. Treatment options are limited and carry serious side effects. Tavneos was approved as the first oral complement inhibitor for the disease, and some patients have been taking it for years.
Now the FDA says the evidence that justified that approval cannot be trusted.
What the FDA found
CDER’s withdrawal proposal rests on two legal grounds under the Federal Food, Drug, and Cosmetic Act. First, the agency says new information discovered more than three years after approval shows the drug application lacked substantial evidence of effectiveness. Second, CDER asserts the original new drug application (NDA 214487) contained untrue statements of material fact.
The alleged manipulation centers on the ADVOCATE trial, a Phase 3 study (NCT02994927) that enrolled 331 patients across 20 countries. According to the agency’s formal hearing notice, unblinded study personnel made changes to the trial database after it had been locked, altering endpoint results. Database lock is the point in a clinical trial when patient data is frozen specifically to prevent post-hoc tampering. The FDA says the changes made avacopan appear to meet efficacy thresholds it otherwise would not have cleared.
The margins were already thin. When the FDA originally approved Tavneos in October 2021, it cited a sustained remission rate of 65.7% at week 52 for avacopan versus 54.9% for the control group. A gap that narrow means even modest data manipulation could have been the difference between a successful and a failed trial. The FDA now says that reported gap cannot be relied upon.
Eight deaths and a growing safety signal
The integrity questions arrived alongside a worsening safety picture. Through October 2024, the FDA identified 76 cases of drug-induced liver injury with reasonable evidence of a causal link to Tavneos, according to a drug safety communication from the agency. Of those, 74 were classified as serious: 54 required hospitalization, 7 were confirmed by liver biopsy, and 8 patients died. Those liver injury findings fed directly into the agency’s decision that the risk-benefit balance no longer supported keeping the drug on the market.
The concerns have crossed the Atlantic. The European Medicines Agency (EMA) launched its own review after what it described as “emerging information” about the integrity of the ADVOCATE data. The EMA’s Committee for Medicinal Products for Human Use (CHMP) is conducting that assessment independently, though both regulators are examining the same underlying trial. No interim findings or timeline from the European review have been published. If the EMA reaches a different conclusion than the FDA, patients could face a split regulatory landscape, with the same drug pulled in one jurisdiction and still prescribed in another.
Questions the FDA has not yet answered
The agency’s hearing notice describes post-lock manipulation by unblinded personnel but does not release the detailed audit trail, raw data logs, or the names of those involved. Outside researchers cannot independently verify which endpoints were changed, by whom, or by how much. The public record contains the FDA’s conclusions but not the full forensic evidence behind them.
The timeline raises its own questions. CDER says it learned of the manipulation more than three years after granting approval. What triggered the investigation, and why it took that long, remains unexplained in the documents released so far. Notably, FDA reviewers had flagged concerns about trial design, endpoint definitions, and safety issues during a May 6, 2021 Arthritis Advisory Committee meeting, before the drug was approved. Whether those earlier red flags should have prompted closer scrutiny of the underlying data is a question the agency has not addressed.
CSL Vifor, the Swiss pharmaceutical company that markets Tavneos, has not issued a public statement responding to the withdrawal proposal as of May 2026. The company retains the right to request a formal hearing before the FDA can finalize the withdrawal. Whether CSL Vifor will contest the findings, offer an alternative explanation for the data irregularities, or propose steps short of full withdrawal remains unknown.
What patients and doctors face now
ANCA-associated vasculitis is not a disease with a deep bench of alternatives. The current standard of care relies on potent immunosuppressants, primarily rituximab or cyclophosphamide, combined with high-dose glucocorticoids that carry their own serious long-term risks, including infection, bone loss, and diabetes. Tavneos was specifically valued because it offered a way to reduce glucocorticoid exposure. Losing it narrows an already tight set of options.
Some patients may have experienced genuine improvement on the drug. But without trustworthy trial data, clinicians cannot confidently separate avacopan’s contribution from the effects of background therapies or natural disease fluctuation. They now must weigh the possibility that the drug is less effective than believed, or not effective at all, against documented liver toxicity that includes fatalities.
The Vasculitis Foundation, the largest U.S. patient advocacy organization for the disease, has historically urged expanded access to new therapies. How the group and treating rheumatologists respond to the withdrawal proposal will shape whether patients currently on Tavneos are transitioned off the drug immediately or managed through a more gradual process.
A single trial, a single point of failure
Tavneos was approved on the strength of one pivotal Phase 3 study. In areas of high unmet medical need, regulators sometimes accept a single strong trial rather than requiring the traditional two. That shortcut assumes the pivotal study is both methodologically sound and free from fraud. When that assumption collapses, there is no backup dataset to fall back on, and the entire evidentiary foundation for approval goes with it.
No independent academic reanalysis of the ADVOCATE trial data has been published. No whistleblower account or court filing has surfaced to describe from the inside how the alleged post-lock changes occurred. Until those external sources emerge, the public record is shaped almost entirely by the regulator’s account, which is detailed but not fully transparent about its own methods and triggers.
The FDA’s withdrawal proposal is not yet final. A formal hearing, if requested by CSL Vifor, could take months. The EMA’s parallel review adds another variable. But the agency’s willingness to pursue one of its most aggressive enforcement actions in recent memory signals a judgment that the data behind Tavneos are too compromised to support continued use, and that the harm to patients, both from a potentially ineffective drug and from documented liver toxicity, is too serious to wait.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
