Inferiority of ticagrelor in the PHILO trial: Play of chance in East Asians or nightmare confirmation of PLATO-USA?

Abstract

Long-awaiting PHILO trial, conducted in 2011-2012 has been submitted and published late in 2015. In contrast to overall PLATO results, but similar to PLATO-US cohort, PHILO revealed numerical inferiority of ticagrelor with regard to death, myocardial infarction, stroke, and bleeding over clopidogrel. Hence, we comprehend the PHILO results in light of the PLATO-US evidence. To assess the PHILO (n=801) outcomes, applied statistics, and trial conduct, matching them with the PLATO-US (n=1413) patients. The Asian, predominantly Japanese ticagrelor patients had worsened outcomes even when compared to the negative American cohort with regard to death (OR=1.44 (PHILO) vs. 1.17 (PLATO-US); myocardial infarction (OR=1.63 vs. 1.38); and composite primary endpoint (OR=1.60 vs. 1.27); but not for stroke (OR=1.51 vs. 1.75). Moreover, in contrast to the trend in PLATO-US (OR=1.11; CI=0.84-1.48, p=0.46), PHILO revealed significant excess of PLATO-defined composite of major and minor bleeding events (OR=1.83; CI=1.27-2.63,p=0.0014). PLATO-defined "net clinical benefit" in PHILO was also significantly (OR=1.61; CI=1.11-2.34, p=0.0145) inferior for ticagrelor. The "number needed to harm" suggests that for every 29 PHILO patients treated for 12months with ticagrelor instead of clopidogrel, one will suffer an additional major bleeding event. Finally, unexplained premature closure of PHILO just after 200-210days mean drug exposure, and short 240days mean follow-up seems deliberate and concerning with regard to unblinding. Aside from some information censoring, early trial closure, and creative statistics, Asian ticagrelor patients did consistently worse in PHILO, than even in the negative PLATO-US cohort. Especially alarming is a significant bleeding inferiority justifying a necessity for a separate outcome-driven low-dose ticagrelor trial in Asian post-PCI patients. This strategy was successfully implemented with the low-dose prasugrel in Japan. Regulatory authorities in Asia may consider conducting an independent review of PHILO dataset to ensure adequate ticagrelor safety.

Keywords: Aspirin; Bleeding; Clinical trial; Clopidogrel; Follow-up; Outcomes; Ticagrelor; Trial conduct.