Lack of Evidence of a Clopidogrel – Statin Interaction in the CHARISMA Trial

Lack of Evidence of a Clopidogrel – Statin Interaction in the CHARISMA Trial J Am Coll Cardiol 2007;50:291-5 Jacqueline Saw, MD*, Danielle Brennan, MS†, Steven Steinhubl, MD‡, Deepak L. Bhatt, MD†, Koon-Hou Mak, MD§, Keith Fox, MB^, and Eric J. Topol, MD#for the CHARISMA Investigators *Vancouver, British Columbia, Canada; †Cleveland, Ohio; ‡Lexington, Kentucky; §Singapore; ^Edinburgh, Scotland, United Kingdom; and #La Jolla, California

Study Objective • To evaluate the potential impact of long-term concomitant administration of clopidogrel and statins in the CHARISMA study on long-term clinical event-rates

CHARISMA Trial Design Median 28 months Bhatt DL, et al. N Eng J Med 2006;354:1-12.

Methods (1) • We performed a secondary analysis evaluating the differential treatment effect (interaction) of clopidogrel versus placebo according to the type of statin administered: • CYP3A4-MET versus Non-CYP3A4-MET • Atorvastatin versus pravastatin • Analyses of the entire CHARISMA cohort, symptomatic cohort, & asymptomatic cohort

Methods (2) • Statin administration was non-randomized • Recorded at baseline and each follow-up visit • Our analysis was based upon baseline statin administered —2 major groups: • CYP3A4-MET: Statins predominantly metabolized by CYP3A4 (atorvastatin, lovastatin, simvastatin) • Non-CYP3A4-MET: Statins not predominantly metabolized by CYP3A4 (pravastatin, fluvastatin)

Statistical Analyses • Intention-to-treat population, 2-sided tests, 5% α • Chi-square to compare baseline variables • Cox proportional-hazards model to estimate the HR and 95% CI for the primary efficacy endpoint • Logistic regression model to compute OR and 95% CI for the primary safety endpoint • Interactions tested with Cox proportional-hazards model, incorporating terms for randomized treatment and treatment-by-statin, to assess if treatment effect differed for CYP3A4-MET vs. non-CYP3A4-MET, and atorvastatin vs. pravastatin • SAS software (8.2, SAS Institute, Cary, NC).

Baseline Characteristics

HR 1.02p=0.87 HR 0.93p=0.23 HR 0.78p=0.19 HR 0.80p=0.06 HR 0.72p=0.13 HR 0.87p=0.08 HR 0.89p=0.18 Primary Efficacy Endpoint

Interaction: CYP3A4-MET The interaction of the type of statin (CYP3A4-MET vs. non-CYP3A4-MET) and randomized treatment was not significant. p=0.69

Interaction: Atorvastatin/Pravastatin The interaction of atorvastatin versus pravastatin and randomized treatment was not significant. p=0.54

p<0.001 Statins Versus No Statins p<0.001

Symptomatic Patients In the symptomatic subgroup, the interaction of the type of statin used (CYP3A4-MET vs. Non-CYP3A4-MET) and the randomized treatment (clopidogrel vs. placebo) remained insignificant: p=0.18 Likewise, the interaction of atorvastatin versus pravastatin and randomized treatment was not significant: p=0.25

OR 1.29p=0.20 OR 1.24p=0.11 OR 1.19p=0.33 OR 1.19p=0.39 OR 1.14p=0.76 OR 0.87p=0.61 OR 1.04p=0.93 GUSTO Major Bleeding

Limitations • Retrospective post hoc analysis • Statin administration not randomized • Statin dose not recorded • Analysis based upon baseline statin use • Platelet aggregation or activation not assessed

Conclusions • Despite theoretical concerns and ex vivo testing suggesting a potential negative interaction with concomitant clopidogrel and CYP3A4-MET statin administration, there was no evidence of an interaction clinically in a large placebo-controlled trial with long-term follow-up • Our study is concordant with other clinical analyses, suggesting that clinicians need not choose statins on the basis of CYP3A4-metabolism, even when long-term clopidogrel co-administration is required

Lack of Evidence of a Clopidogrel – Statin Interaction in the CHARISMA Trial