EMA and FDA Advocate Removing Amgen’s Drug from the Market: How Unblinded Personnel Manipulated the ADVOCATE Trial

It’s my pleasure to welcome back Sahar Van Waalwijk, who covers an important topic following an unusual development. Enjoy! Timothée

---

Imagine a wine competition. Judges taste wines from unlabelled bottles and score them without knowing which winery produced them. This prevents famous brands from receiving more favourable scores simply because of their reputation. Now imagine that all the wines have been scored and the bottles have been “unblinded”, allowing the judges to see which winery produced each wine. But then, without telling anyone, they go back and revise some of the scores. This would be regarded as foul play, as it undermines the integrity of the competition and may disadvantage wineries whose products were originally rated more highly.

Manipulating study results

Now imagine if the same thing happened in a blinded clinical trial. A change in outcome assessment can alter the conclusions of a study and potentially affect which medicines are approved, funded, and prescribed to patients. In the worst case, patients may receive treatments that cause more harm than benefit.

Remarkably, this is exactly what the FDA alleges happened in the ADVOCATE trial of avacopan (Tavneos), a drug of Amgen. According to the FDA, information that came to light more than three years after approval showed that “unblinded study personnel manipulated the results of the pivotal clinical study so the drug looked effective when the original analysis did not support that conclusion. The applicant also did not disclose the original analysis to FDA, in violation of FDA regulations.”

The same conclusion was recently reached by the EMA, which had issued a positive opinion on the drug just one month after the FDA’s approval in 2021.

The EMA now concludes that “Advocate study was conducted in breach of good clinical practice (GCP)” and the data submitted for the registration “were found to be incorrect and misleading”.

Serious harm

Avacopan is authorised for treating adults with severe, active granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA), two rare inflammatory conditions of the blood vessels.

EMA writes however that this drug “is associated with an increased risk of drug-induced liver injury (DILI) and vanishing bile duct syndrome (VBDS, a rare condition where the small bile ducts inside the liver are gradually damaged and disappear over time), including cases with a fatal outcome…” Also FDA is increasingly concerned about the safety profile of avacopan.

Advocating for the retraction of ADVOCATE

Last week, the New England Journal of Medicine (NEJM) retracted the landmark ADVOCATE trial publication after two academic authors requested its withdrawal. According to the retraction notice, they were unaware that primary endpoint assessments for nine patients had been re-adjudicated after database lock and trial unblinding.

In the ADVOCATE study, patients with GPA or MPA were randomised to receive either avacopan or placebo, both in addition to standard treatment (either rituximab or cyclophosphamide followed by azathioprine). The placebo group also received a 20-week course of high-dose corticosteroids. According to the now-retracted ADVOCATE publication, avacopan was reported to be at least as effective as corticosteroids in inducing remission and to provide higher sustained remission rates at 52 weeks.

What happens next?

In the EU, EMA has recommended that the medicine’s marketing authorisation be revoked. If this recommendation is confirmed by the European Commission, avacopan will no longer be authorised in the EU. In the US, the drug will remain on the market until either the applicant decides to withdraw it or the FDA Commissioner orders its removal. If the applicant requests a hearing, the FDA Commissioner will first decide whether to hold one and will subsequently decide whether approval of avacopan should be withdrawn.

Trust is hard to earn, easy to lose

In a foul wine competition, consumers are not harmed much. In the worst case, they end up with a wine they do not like and do not get value for their money. The avacopan case is a different story. Here, many patients were exposed to serious side effects, and healthcare budgets are wasted. Yet both cases have something in common: a loss of trust.

Manipulating data affects trust in science itself. It is uncomfortable to think that cases like this may only occasionally come to light, raising the question of how many similar cases go unnoticed.

It also raises questions about the role of academic investigators participating in industry-sponsored trials. If a sponsor does not consider it important to disclose deviations or breaches of this nature, what is the purpose of co-authoring their publications?