Lost in follow-up rates in TRACER, ATLAS ACS 2, TRITON and TRA 2P trials: challenging PLATO mortality rates

Abstract

Context: Extreme rates of vascular and all-cause mortality especially in the clopidogrel arm of the Platelet Inhibition and Patient Outcomes (PLATO) non-USA cohort raise concerns of data integrity, and call for independent verification of vital records in the national death registries. Four recent acute coronary syndrome (ACS) trials: Thrombin Receptor Antagonist for Clinical Event Reduction in Acute Coronary Syndrome (TRACER), Anti-Xa therapy to lower cardiovascular events in addition to standard therapy in subjects with acute coronary syndrome (ATLAS-ACS 2), Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel (TRITON), and the Thrombin Receptor Antagonist in Secondary Prevention of Atherothrombotic Ischemic Events (TRA 2P), provide a valuable opportunity to match lost in follow-up (LIFU) with mortality rates among similar ACS studies.

Objective: To compare the LIFU from PLATO, TRACER, ATLAS-ACS 2, TRITON-TIMI 38 and TRA 2P trials.

Results: The disturbingly high (8.9%-14.7%) LIFU in PLATO was no match to LIFU in TRACER (0.1%), ATLAS ACS 2 (0.3%), TRITON (0.1%) and TRA 2P (0.1%). In fact, such an astronomical (49-147 fold higher) PLATO LIFU rate should result in less mortality compared to the other ACS trials since no event can be reported or adjudicated if the patient has been lost. Adjusting LIFU rate revealed that vascular (5.55%) and all cause (6.05%) mortality in PLATO was even more disparate than in TRACER (3.2% and 4.9%), ATLAS-ACS 2 (4.1% and 4.5%), TRITON-TIMI 38 (2.4% and 3.2%) and TRA 2P (3.0% and 5.3%) control arms, respectfully. Moreover, the incomplete CV follow-up in the ATLAS ACS 2 trial was later revealed to be around 12%, which lead to the rejection of rivaroxaban for the treatment of ACS. PLATO's LIFU rate was just as high, if not higher, than seen in ATLAS ACS 2.

Conclusions: The chance to die in PLATO far exceeds the mortality risks observed in the clopidogrel arms of four recent ACS trials, which becomes especially evident after adjustment for LIFU rates among trials. Astronomical LIFU numbers are not likely to be responsible for the PLATO mortality paradox, providing additional justification for an independent audit of the PLATO presumably deceased clopidogrel cohort.