10 hours ago
BMJ Investigation Increases Concerns About Ticagrelor Trials
An investigation by The BMJ is raising fresh concerns about the clinical studies that supported the approval of the antiplatelet drug ticagrelor (Brilinta, AstraZeneca), almost 15 years after the medication was first approved and as generic versions are set to hit the market.
Peter Doshi, PhD, a senior editor at The BMJ , previously reported inconsistencies and omissions in data reporting from the 2009 PLATO study, published in The New England Journal of Medicine , which showed ticagrelor was superior to clopidogrel in treating acute coronary syndrome.
Now a follow-up investigation of two supporting studies published in Circulation , ONSET/OFFSET and RESPOND, has revealed primary endpoints were reported inaccurately, data were missing from the submission to the US Food and Drug Administration (FDA), and study centers may not have received adequate training.
Doshi said the results of his investigations call into question the drug's approval and suggested that it should be revisited.
'The FDA's approval in 2011 went against the evidence according to its reviewers, and now, my investigations into PLATO, ONSET/OFFSET, and RESPOND, suggest that even the data presented to the FDA and reported in The New England Journal of Medicine and Circulation , is not trustworthy,' he told Medscape Medical News.
The investigation identified several problems with data integrity in the two trials. The original primary endpoint results for RESPOND, which aimed to test whether ticagrelor could convert nonresponders to clopidogrel into responders, were statistically nonsignificant ( P = .157) but were subsequently reported in Circulation as significant ( P = .005) because of an undeclared change in its primary endpoint definition.
For ONSET/OFFSET, which reported ticagrelor provided faster and greater inhibition of platelets than clopidogrel, the investigators now claim several patients were excluded from the analysis. However, those who remained were identified as the 'intention-to-treat' population, implying all patients were included. Implausible data points were also included in the analysis of the primary endpoint but were first transformed through an unpublished data analysis, Doshi claimed.
Doshi also gained access to readouts from some of the platelet function test machines used in the trial. He found more than 60 of 282 readings were not present in the datasets submitted to the FDA, and the levels of platelet activity in those readings were significantly higher than those reported in Circulation .
Victor Serebruany, MD, from Johns Hopkins University in Baltimore, and one of the more high-profile critics of ticagrelor, told The BMJ the missing readings show 'there are episodes of skyrocketing rebound and profound platelet inhibition after ticagrelor, making patients prone to thrombosis or bleeding. If doctors had known what happened in these trials, they would never have started using ticagrelor.'
The investigation also revealed oddities around the authorship of the publications. One active trial investigator was never identified as a study author, while one author told The BMJ he was not involved in the trial.
The BMJ states that AstraZeneca, the journal Circulation , and many of the original investigators either declined to comment on the new claims or were unreachable.
Ticagrelor has been under fire since the beginning. The drug failed in its first bid for FDA approval and was the subject of an investigation by the US Department of Justice in 2013 at the urging of Serebruany. That investigation was closed in 2014 with no further action.
A review of several major trials of ticagrelor by Eric Bates, MD, professor of internal medicine at the University of Michigan in Ann Arbor, Michigan, and a co-author of the US guidelines that recommend ticagrelor, concluded 'the clinical conventional wisdom and clinical trial guideline support for…ticagrelor compared with clopidogrel may be overemphasized.' Bates is now calling for a review of ticagrelor's recommendation in guidelines, according to the earlier The BMJ report.
