FDA reviewer says no to cangrelor for PCI


The medical team leader of the FDA’s Division of Cardiovascular and Renal Products, Thomas A Marciniak, states in a clinical review of cangrelor (The Medicines Company) that the drug should not be recommended for reducing thrombotic cardiovascular events during percutaneous coronary intervention (PCI) because the studies supposedly supporting cangrelor for this indication are flawed and do not show superiority or non-inferiority of a cangrelor regimen compared with a clopidogrel (Plavix; Bristol-Myers Squibb, Sanofi-Aventis) regimen or with standard of care.

Marciniak believes that the CHAMPION series of studies (which evaluated cangrelor in patients undergoing PCI) did not “show superiority or non-inferiority” of cangrelor for several reasons, including the fact that clopidogrel “was delayed inappropriately in all of the trials”. He explains: “The trials themselves provide evidence that earlier administration of clopidogrel was better both by cross-trial comparisons and by logistic regressions of the PHOENIX data. Clopidogrel was never consistently administered early enough such that we cannot even conclude that cangrelor is non-inferior to clopidogrel.” Also, Marciniak claims that in the CHAMPION-PHOENIX study, some of the supposed superiority of cangrelor may be explained by the use of a 600mg loading dose in the cangrelor arm as “use of a 300mg loading dose was only allowed in the clopidogrel arm”.

He adds that the “flaws” of the CHAMPION studies are not the only reason why cangrelor should not be approved for use in PCI, writing: “Both prasugrel (Efient; Daiichi-Sankyo, Eli Lilly) and ticagrelor (Brilique, AstraZeneca) have demonstrated superior efficacy to clopidogrel. We do not know whether cangrelor affects this superiority so we do not know whether to use cangrelor with or instead of prasugrel and ticagrelor—or prasugrel or ticagrelor instead of cangrelor. Because the superiority involves irreversible harm (ie. deaths and myocardial infarctions), we should understand the tradeoff of these agents.”

Marciniak concludes: “I recommend not approving cangrelor until another trial succeeds in correcting the flaws that I have documented in this review,” adding that further studies should also address the fact that the CHAMPION studies were conducted “unethicially” (in a parallel review, Mariniak notes, he outlines why believes the studies were unethical).

However, in the same clinical review, two others reviewers—Fred Senatore and Nhi Beasley—state that they believe cangrelor should be recommended for use in PCI. They say that this is because cangrelor met the primary efficacy endpoint of the PHOENIX study of reducing death, myocardial infarction, ischaemia driven revascularisation, and stent thrombosis compared with clopidogrel. They add that the benefits of cangrelor outweigh the risks using GUSTO, TIMI, and Bad Bleed criteria, stating “One might consider these bleeds to be of greater clinical import”. But, they do state that overall risk-benefit profile of cangrelor is “mixed”. They explain the risks of the drug outweighs the benefit in terms of non- coronary artery bypass grafting bleeding and ACUITY major bleed—they write that these bleeds might be considered to be of “less clinical consequence”.

The FDA is still in the process of reviewing cangrelor and, therefore, the critical remarks of Marciniak (or positive remarks of Senator and Beasley) do not necessarily reflect the final decision of the agency.