The $3.7B Drug That Europe Just Said Was Built on a Lie
Europe's drug regulator just recommended revoking Amgen's Tavneos after investigators found the clinical trial that got it approved was built on manipulated data. The case is a $3.7 billion cautionary tale with implications for every biotech acquirer, investor, and drug developer in the industry.
Imagine buying a house for $3.7 billion, only to find out the inspection report was forged.
That's roughly what just happened to Amgen. Europe's drug regulator has recommended revoking the marketing authorization for Tavneos (avacopan), a treatment for a rare and brutal autoimmune disease called ANCA-associated vasculitis. The reason isn't a new side effect or a failed follow-up study. It's worse: the clinical trial data that got the drug approved in the first place were, in the regulator's words, "incorrect, misleading and no longer reliable."
This isn't a slap on the wrist. It's one of the rarest and most extreme actions a drug regulator can take. And it carries implications that stretch far beyond one drug.
The Trial That Changed After the Blindfold Came Off
Tavneos was approved in the U.S. in 2021 and in Europe in 2022, both based on a single pivotal study called ADVOCATE. The trial enrolled 331 patients with severe vasculitis (a condition where the immune system attacks blood vessels) and tested whether Tavneos could replace the high-dose steroids that are brutally effective but come with a laundry list of side effects.
The results looked good. Good enough for the FDA. Good enough for the European Commission. Good enough for the New England Journal of Medicine, which published the findings.
But an investigation later revealed something alarming. Staff at ChemoCentryx, the small biotech that originally developed the drug, altered results for nine participants after the initial blinded analysis. Five of those patients were in the Tavneos arm. Those changes weren't cosmetic; they were the difference between a study that worked and one that didn't. The manipulations flipped the trial from non-significant to statistically significant.
Think of it like a student changing answers on a test after peeking at the answer key. Except the stakes aren't a letter grade. They're patients' lives.
Europe Pulls the Plug
The European Medicines Agency's human medicines committee (known as the CHMP) reviewed the evidence and didn't mince words. At its June 2026 meeting, the committee concluded that the ADVOCATE study , the foundational rules that govern how clinical trials must be conducted.
Get tomorrow's biotech intelligence before your competitors.
Join thousands of biotech professionals who start their day with our free, daily briefing.
breached good clinical practice principles
The committee looked at post-approval data and additional analyses to see if there was any way to salvage the drug's case. There wasn't. The benefit-risk balance, regulators said, could no longer be confirmed. Their recommendation: pull the authorization entirely.
Patients in Europe are already being told no new prescriptions should be written. Current patients should switch to alternative treatments, which typically means going back to the steroid-heavy regimens Tavneos was supposed to improve upon.
The European Commission still needs to formally act on the recommendation, but the CHMP's opinion is the scientific verdict. It rarely gets overturned.
The FDA Isn't Far Behind
This isn't just a European problem. In January 2026, the FDA requested a voluntary withdrawal of Tavneos from the U.S. market, and what it found was equally damning. Beyond the data manipulation, the agency tracked 76 cases of drug-induced liver injury in Tavneos patients, with evidence suggesting a causal link. Some cases involved a rare condition called vanishing bile duct syndrome. Some ended in death.
With the efficacy evidence now in tatters and serious safety signals piling up, the FDA has proposed withdrawing Tavneos from the U.S. market. The EMA's recommendation is explicitly described as being "in line with" the FDA's position. Two of the world's most powerful regulators, on two different continents, reaching the same conclusion independently: this drug's foundation is cracked.
The New England Journal of Medicine has since retracted the original ADVOCATE publication. For a journal that publishes the most consequential clinical trials in medicine, that's an extraordinary and deeply embarrassing step.
Amgen Inherited a Mess
Here's where the story gets complicated. Amgen didn't run the ADVOCATE trial. ChemoCentryx did. Amgen acquired ChemoCentryx in 2022 for $3.7 billion, and Tavneos was a crown jewel of that deal. By the time the data integrity problems surfaced, the ink on the acquisition was long dry.
Nothing in the public record suggests Amgen itself manipulated any data. The misconduct traces back to ChemoCentryx staff. But regulators and investors don't really care about corporate genealogy; they care about who owns the asset now. And that's Amgen.
The company isn't going quietly. Amgen has engaged the Duke Clinical Research Institute to conduct an independent re-analysis of the Tavneos data, hoping to demonstrate that the drug still has a viable benefit profile. That analysis is expected to be submitted for an FDA hearing by late July 2026.
But analysts are skeptical. When regulators have declared a pivotal trial's data fundamentally unreliable because of GCP violations, no amount of re-analysis can undo the breach itself. You can re-examine the numbers, but you can't un-cheat the test.
A Rare Disease, an Even Rarer Regulatory Action
To understand how unusual this is, consider: EMA marketing authorization revocations driven by clinical data fraud are almost unheard of. The last major precedent was the GVK Biosciences scandal in 2015, where EMA reviewed over 1,000 affected product presentations and recommended suspending more than 700. That case involved a contract research organization fabricating bioequivalence data across dozens of generic drugs.
The Tavneos situation is different. This is a single branded drug, for a rare disease, approved on a single pivotal trial, where targeted data manipulation at the sponsor level altered the study's statistical conclusion. Industry commentators are calling it a first-of-its-kind precedent.
And that's what makes it so significant. Rare disease drugs often get approved on smaller, single-study evidence packages because, by definition, there aren't many patients to study. If regulators are now willing to reopen and revoke those approvals years later based on forensic data reviews, the entire risk calculus for rare disease drug development shifts.
What This Means for Everyone Else
The ripple effects here go well beyond Amgen's portfolio.
For pharma acquirers, this is a $3.7 billion cautionary tale. Companies buying late-stage or recently approved assets will now face pressure to conduct forensic data audits and GCP compliance reviews before closing deals, not just review top-line results and published papers. Expect contractual indemnities and escrow arrangements tied to data-integrity risks to become standard.
For small biotechs running pivotal trials, the scrutiny is about to intensify. Regulators and investors alike will look harder at companies where a single study carries binary, make-or-break value. The incentive to "optimize" results is greatest in exactly these settings, and everyone now knows it.
For drug regulators, this case validates a more aggressive posture on post-marketing oversight. Both the EMA and FDA have signaled that approval is not the finish line. If new information surfaces that undermines the data behind an approval, they will reopen the case. The EMA's new automated compliance monitoring system for safety reporting, launched in September 2025, fits squarely into this trend of tighter post-approval surveillance.
For investors, the lesson is painful but clear: a single pivotal trial with a marginal p-value is a fragile foundation. Programs with multiple independent, concordant studies will command a premium. Lone-trial approvals, especially in rare diseases where replication is hard, will carry a steeper risk discount.
The Uncomfortable Question
The ADVOCATE data were reviewed by the FDA, the EMA, and the New England Journal of Medicine. All three gave it a passing grade. The manipulation wasn't caught by any of them during the normal review process.
That raises an uncomfortable question: how many other approved drugs are sitting on similarly shaky foundations? The answer is almost certainly "very few." Outright data fabrication in pivotal trials is genuinely rare. But "very few" is not the same as "none," and this case proves that even the gold-standard gatekeepers can miss it.
Regulators are already responding. The EMA is piloting a voluntary proof-of-concept initiative for sponsors to submit raw, patient-level datasets in regulatory procedures, moving beyond reliance on summary tables and published papers. The agency is also developing a formal data quality framework for real-world evidence. These aren't reactive patches; they're structural changes designed to make the next ChemoCentryx-style manipulation much harder to pull off.
For now, though, the patients with ANCA-associated vasculitis are the ones paying the price. They're being switched off a drug they were told represented a meaningful advance over steroids, back to the very regimens that drug was designed to replace. Whether Tavneos actually helped them remains, officially, an open question.
And that might be the most unsettling part of all.
5 min read
Gene Editing Just Broke a Two-Year IPO Drought
For over two years, no gene-editing company dared to go public. Scribe Therapeutics just priced a $129 million IPO at the top of its range, with Sanofi buying in alongside public investors. The gene-editing IPO drought is officially over.
