Pull up the European Commission’s August 2026 revocation notice for Tavneos (avacopan) and read the operative phrase: “serious breaches” of Good Clinical Practice in the pivotal Phase III ADVOCATE study. The EU did not revoke the marketing authorization because the drug stopped working. It revoked it because the CHMP concluded, in a June 26, 2026 recommendation formalized weeks later, that GCP violations in the trial that produced the approval data were severe enough to invalidate the benefit-risk assessment entirely. That is a structurally different kind of failure than a safety signal or an efficacy miss. That is a finding that the evidentiary foundation was compromised before regulators on either side of the Atlantic ever voted.
Tavneos was developed by ChemoCentryx, acquired by Amgen in 2022, with CSL Vifor holding European marketing rights. The FDA approved avacopan on October 7, 2021, as an add-on treatment for adults with severe active ANCA-associated vasculitis. The EMA followed on January 11, 2022. Both agencies reviewed the same ADVOCATE trial. Both cleared the drug. Now one of them has reversed course on GCP integrity grounds. Which raises a question that nobody in the rare-disease space wants to answer out loud: what did FDA know during its parallel review, and what did it choose not to act on?
The Failure Preceded the Filing
The ADVOCATE trial did not fail at approval. It failed during execution, and the failure was invisible long enough to reach two regulatory dossiers on two continents. That is the operational timeline that matters here. GCP breaches of the kind the CHMP is now characterizing as “serious” are not post-hoc discoveries of statistical irregularity. They are site-level events: protocol deviations that were either not caught by monitoring, not escalated to the sponsor, not documented in the clinical study report, or all three. Someone signed off on a monitoring report at some point during ADVOCATE’s enrollment window. Someone reviewed the data management outputs. Someone at the CRO, at ChemoCentryx’s clinical operations function, certified compliance before the NDA and MAA packages were assembled.
The EMA’s CHMP recommended revocation because it determined the benefits of Tavneos no longer outweighed its risks, explicitly citing GCP principle violations in ADVOCATE. That language, “no longer outweighed,” is regulatory hedging for a harder truth: the committee cannot trust the data well enough to maintain the authorization. When a rare-disease approval survives four years of commercial use and then collapses on data integrity rather than new clinical evidence, the monitoring infrastructure that was supposed to catch deviations in real time failed at every level it was designed to operate.
Under ICH E6(R2), the sponsor holds primary accountability for trial conduct regardless of how much operational authority has been delegated to a CRO. Section 5.2 is explicit: the sponsor cannot transfer its regulatory responsibilities. ChemoCentryx, and by extension Amgen after the 2022 acquisition closed, owns the ADVOCATE data quality problem. But the CRO that managed site oversight during the trial years has not been named publicly in any of the enforcement documentation released so far, and that absence is itself a data point about how sponsor-CRO accountability gets obscured when the indictment finally arrives.
A Parallel Review That Demands a Parallel Answer
Here is the structural problem the FDA now faces. The agency reviewed the ADVOCATE package in 2021 and issued approval. The EMA reviewed the same package in 2022 and issued approval. The EMA has now looked at ADVOCATE again, found serious GCP breaches, and reversed its authorization. Under normal scientific logic, if the breach is serious enough to collapse a European marketing authorization, it is serious enough to require a formal FDA response. Not a press statement. Not a label update. A BIMO investigation, a clinical hold assessment, or a public-facing explanation of why the FDA’s benefit-risk calculation survives the CHMP’s integrity finding intact.
The FDA has not yet published that explanation. That silence is not unusual in the short term, but it becomes a governance problem the longer it persists. The agency approved Tavneos under its standard NDA pathway, and American patients with ANCA-associated vasculitis have been taking this drug since late 2021, nearly five years of commercial exposure built on a trial that EU regulators have now formally repudiated. The FDA’s own BIMO program exists precisely for this scenario: to audit clinical data in already-approved products when integrity questions surface. Whether BIMO is actively reviewing ADVOCATE right now is not publicly known, and that opacity is the accountability gap this column is identifying.
Comparable enforcement history should make everyone uncomfortable about how long these reviews take. The EMA’s reversal process for Tavneos moved from CHMP recommendation on June 26, 2026, to European Commission revocation in early August 2026, roughly six weeks. That is a decisive regulatory cadence for a rare-disease authorization. The FDA’s track record on parallel integrity reviews after partner-agency revocations is considerably slower, and in several cases has produced no formal public action at all. Sponsors operating in rare disease have quietly learned that U.S. approvals have a structural resilience to foreign GCP findings that EU authorizations do not. That asymmetry encourages a risk calculus that watchdog readers should recognize immediately.
What the Rare-Disease Infrastructure Gets Wrong
Rare-disease trials operate under conditions that make GCP compliance harder to enforce and easier to obscure. ADVOCATE enrolled patients with severe ANCA-associated vasculitis, a population that is small, geographically dispersed, and clinically complex. Site volumes are low, monitoring visit frequencies are often compressed by patient access constraints, and the investigator pool is narrow enough that sponsors are reluctant to impose the kind of corrective pressure that would generate a paper trail. In that environment, “serious breaches” do not announce themselves in real time. They accumulate quietly across a handful of sites and surface only when a regulator decides to look hard enough.
The Fabhalta (iptacopan) approval for C3 glomerulopathy, granted by FDA on March 20, 2025, based on the Phase 3 APPEAR-C3G trial, illustrates that the complement inhibitor space is actively expanding into rare kidney indications under exactly these same small-population, high-complexity conditions. The Tavneos revocation is not a historical footnote for that pipeline. It is an operational warning about what happens when trial governance infrastructure does not scale to match the regulatory ambition of rare-disease development programs. Sponsors filing complement inhibitor data in 2025 and 2026 should be reading the CHMP’s ADVOCATE findings as a technical brief, not a news item.
For clinical operations executives, the concrete directive is this: every rare-disease program with a pivotal trial currently in execution needs a vendor-quality audit that goes beyond standard monitoring metrics. Source data verification rates and query resolution timelines do not catch protocol deviations that were never entered into the EDC. Build a structured deviation-classification review into your CRO’s quarterly governance reporting, require investigator attestation at the site level, and make sure your clinical study report certification process includes an explicit GCP compliance sign-off that names the responsible party at the CRO, not just the sponsor’s medical officer. That paper trail is the only thing that survives a CHMP review four years after your NDA clears.
For FDA-watchers, the signal to track is whether the agency issues any public communication, a BIMO report, a drug safety communication, or an AdCom referral, on the ADVOCATE integrity findings before the end of calendar year 2026. If six months pass after the EU revocation with no formal FDA response, that silence will tell us something important about how the agency weights a partner regulator’s GCP findings against its own approval record. The next test is already set: watch what FDA does, or does not do, before February 2027.
References
- FiercePharma — “Tavneos pivotal trial flagged for ‘serious breaches’ of protocol as EU regulators dissect market withdrawal decision”
- European Medicines Agency — “EMA recommends revoking marketing authorisation for Tavneos” (June 26, 2026 CHMP recommendation)
- Drugs.com — Tavneos (avacopan) FDA Approval History, October 7, 2021
- FDA — “FDA Approves First Treatment for Adults with Complement 3 Glomerulopathy” (Fabhalta/iptacopan, March 20, 2025)
Moe Alsumidaie, MBA, MSF, is founder and Chief Editor of Vanguard Publications, which publishes Clinical Trial Vanguard, Pharma Vanguard and BullScope, and Head of Research at CliniBiz. He has two decades in clinical trial operations and data science, with earlier roles at Genentech, Abbott Vascular and Stanford University Medical Center, and is a guest lecturer in clinical trial sciences at Rutgers University.
