Derivation, Validation, and Prognostic Utility of a Prediction Rule for Nonresponse to Clopidogrel: The ABCD-GENE Score

Abstract

ObjectivesThe aim of this study was to develop a risk score integrating cytochrome P450 2C19 loss-of-function genotypes with clinical risk factors influencing clopidogrel response that would allow the identification with more precision of subjects at risk for high platelet reactivity (HPR) and adverse clinical outcomes. BackgroundClopidogrel is the most broadly used platelet P2Y12 inhibitor. However, a considerable number of patients achieve inadequate platelet inhibition, with persistent HPR, an established marker of increased thrombotic risk, underscoring the need for tools to help identify these subjects. Although carriers of loss-of-function alleles of the cytochrome P450 2C19 enzyme have reduced clopidogrel metabolism leading to increased rates of HPR and thrombotic complications, this explains only a fraction of the pharmacodynamic response to clopidogrel, and a number of clinical factors have also been shown to have contributing roles. MethodsThree prospective and independent studies were used to: 1) develop a risk score integrating genetic and clinical factors to identify patients with HPR while on clopidogrel; 2) investigate the external validity of the risk score; and 3) define clinical outcomes associated with the risk score in a cohort of patients with myocardial infarction treated with clopidogrel. ResultsA risk score ABCD-GENE (Age, Body Mass Index, Chronic Kidney Disease, Diabetes Mellitus, and Genotyping) was developed incorporating 5 independent predictors of HPR: 4 clinical (age >75 years, body mass index >30 kg/m2, chronic kidney disease [glomerular filtration rate <60 ml/min], and diabetes mellitus) and 1 genetic (cytochrome P450 2C19 loss-of-function alleles). The C-statistics for the score as an integer variable were 0.71 (95% confidence interval [CI]: 0.68 to 0.75) and 0.64 (95% CI: 0.60 to 0.67) in the pharmacodynamic derivation and validation cohorts, respectively. A cutoff score ≥10 was associated with the best sensitivity and specificity to identify HPR status. The C-statistics for the score were 0.67 (95% CI: 0.64 to 0.71) for all-cause death and 0.66 (95% CI: 0.63 to 0.69) for the composite of all-cause death, stroke, or myocardial infarction at 1 year. Using multiple models for adjustment, the ABCD-GENE score consistently and independently correlated with all-cause death, as well as with the composite of all-cause death, stroke, or myocardial infarction, both as a continuous variable and by using the cutoff of ≥10. The score did not predict bleeding. ConclusionsThe ABCD-GENE score is a simple tool to identify patients with HPR on clopidogrel and who are at increased risk for adverse ischemic events, including mortality, following an acute myocardial infarction. In patients with a high ABCD-GENE score, long-term oral P2Y12 inhibitors other than clopidogrel should be considered.