An investigation by The BMJ has cast fresh doubts on the landmark PLATO clinical trial that led to the worldwide approval of ticagrelor (Brilinta/Brilique), an anti-platelet drug manufactured by AstraZeneca. The investigation reveals potential issues in data reporting and accuracy that raise questions about the drug's efficacy and safety, especially when compared to its cheaper, off-patent rivals like clopidogrel.
Concerns Over Data Reliability
Peter Doshi, senior editor at The BMJ, obtained primary PLATO trial records and unpublished data through a freedom of information request, uncovering discrepancies in the number, causes, and dates of patient deaths. These inconsistencies raise concerns about the accuracy of reporting to the FDA and the possibility of unblinding during the trial.
The PLATO trial, published in the New England Journal of Medicine in 2009, involved over 18,000 patients across 43 countries. It initially reported that ticagrelor reduced the risk of vascular death, heart attack, or stroke compared to clopidogrel. However, a subgroup analysis revealed that US patients on ticagrelor experienced higher mortality rates, leading to the FDA's initial rejection of AstraZeneca's application.
FDA Approval Amidst Internal Objections
Despite these concerns, the FDA eventually approved ticagrelor in 2011. This decision came despite strong reservations from FDA medical officer Thomas Marciniak, who described AstraZeneca's resubmission as "the worst in my experience regarding completeness of the submissions and the sponsor responding completely and accurately to requests." Marciniak recommended against approval due to concerns over the basic reliability of PLATO trial data.
Failure to Replicate PLATO's Results
Since its approval, numerous studies have failed to replicate PLATO's positive findings, prompting experts to question the trial's validity and call for a reappraisal of guidelines recommending ticagrelor. Eric Bates at the University of Michigan, a co-author of the US guidelines, expressed increasing concern over the lack of consistently positive results in subsequent trials.
Monitoring and Data Integrity
Critics have also pointed out that ticagrelor performed worse at trial sites monitored by third-party contract research organizations (CROs) compared to those overseen directly by AstraZeneca. While PLATO co-chairs Robert Harrington and Lars Wallentin have dismissed any influence of monitoring organization on study outcomes, The BMJ found that the cited statistical analysis did not directly compare primary endpoint results between CRO-monitored and sponsor-monitored sites. Furthermore, the lead author of the statistical paper was AstraZeneca's former chief statistician, a fact not disclosed in the publication.
Unresolved Questions and Calls for Re-evaluation
The controversy surrounding PLATO has persisted for years, with critics like Victor Serebruany at Johns Hopkins University raising concerns about inconsistencies and anomalies in the data. Serebruany, who initially found the trial results impressive, guided a US Justice Department investigation into PLATO in 2013. While the investigation was later terminated, Serebruany believes that re-engagement from the Justice Department is the only way to resolve the lingering questions about data integrity.
With generic versions of ticagrelor soon expected in the US, the renewed scrutiny of the PLATO trial raises important questions about the evidence supporting its widespread use and the need for a thorough re-evaluation of its place in clinical guidelines.
