There are potentially enormous clinical benefits of using an individual’s genetic sequence to predict how it affects the metabolism or function of a drug (pharmacogenetics), to allow improved dosage and minimise adverse reactions. The body of evidence is already growing with regard to warfarin treatment, the world’s most popular anti-coagulant (see previous news).
A recent study published online in the New England Journal of Medicine investigated whether the benefit of the drug clopidogrel when compared to placebo was decreased in carriers of alleles known to affect the function of a cytochrome P450 gene involved in drug metabolism [Paré et al. (2010) NEJM doi:10.1056/NEJMoa1008410]. When added to aspirin, clopidogrel reduces the rate of major vascular events in patients with acute coronary syndromes and atrial fibrillation, but recent reports have suggested that certain variants of the CYP2C19 gene may reduce its efficacy by causing a loss-of-function.
The study reports the outcomes of two large, randomised, double-blind, placebo-controlled trails comparing clopidogrel with placebo (both with combination with aspirin). The first trial (CURE) involves patients with acute coronary syndromes and the second trial (ACTIVE A) involves patients with atrial fibrillation with at least one additional risk factor for stroke whilst not being eligible for warfarin treatment. In the CURE trial, 5,000 participants of European or Latin American ancestry were genotyped for three alleles rs4244285, rs4986893 and rs12248560 (so-called *2, *3 and *17 alleles of CYP2C19, respectively), with 2,500 randomly assigned to each treatment group. In the ACTIVE A trial, 1,200 participants of European ancestry were genotyped for the same three alleles with 600 assigned randomly to each treatment group. The genotyped alleles were used to classify patients into metaboliser phenotype categories with carriers of two copies of loss-of-function alleles (*2 or *3) classified as ‘poor metabolisers’ and carriers of two copies of the gain-of-function allele (*17) classified as ‘ultra [fast] metabolisers’.
In the CURE trial, patients on treatment that carried loss-of-function alleles had similar outcomes to non-carriers, with both groups experiencing reduced cardiovascular events compared to their respective placebo groups. Patients on treatment that carried the gain-of-function alleles experienced a larger reduction in cardiovascular events versus the non-carriers when compared to their respective placebo groups. In the ACTIVE A trial, the effect of clopidogrel in reducing the rate of cardiovascular events was similar across carriers and non-carriers of all the functional alleles genotyped when compared to their placebo groups.
Comment: This well-conducted trial starts to clarify the confusion over whether loss-of-function alleles of the CYP2C19 gene alter the efficacy of clopidogrel. Previous studies were hampered by potential confounding factors, something these randomised trials can overcome. However, it is important to note that reductions in the clopidogrel treatment group were consistently observed regardless of metaboliser phenotype when compared to the placebo, suggesting that pharmacogenetic testing of this gene is unlikely to be either valid or useful for this drug.