Peripheral blood gene expression signatures which reflect smoking and aspirin exposure are associated with cardiovascular events

BACKGROUND: Cardiovascular disease and its sequelae are major causes of global mortality, and better methods are needed to identify patients at risk for future cardiovascular events. Gene expression analysis can inform on the molecular underpinnings of risk factors for cardiovascular events. Smoking and aspirin have known opposing effects on platelet reactivity and MACE, however their effects on each other and on MACE are not well described. METHODS: We measured peripheral blood gene expression levels of ITGA2B, which is upregulated by aspirin and correlates with platelet reactivity on aspirin, and a 5 gene validated smoking gene expression score (sGES) where higher expression correlates with smoking status, in participants from the previously reported PREDICT trial (NCT 00500617). The primary outcome was a composite of death, myocardial infarction, and stroke/TIA (MACE). We tested whether selected genes were associated with MACE risk using logistic regression. RESULTS: Gene expression levels were determined in 1581 subjects (50.5% female, mean age 60.66 +/−11.46, 18% self-reported smokers); 3.5% of subjects experienced MACE over 12 months follow-up. Elevated sGES and ITGA2B expression were each associated with MACE (odds ratios [OR] =1.16 [95% CI 1.10–1.31] and 1.42 [95% CI 1.00–1.97], respectively; p < 0.05). ITGA2B expression was inversely associated with self-reported smoking status and the sGES (p < 0.001). A logistic regression model combining sGES and ITGA2B showed better performance (AIC = 474.9) in classifying MACE subjects than either alone (AIC = 479.1, 478.2 respectively). CONCLUSION: Gene expression levels associated with smoking and aspirin are independently predictive of an increased risk of cardiovascular events. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s12920-017-0318-6) contains supplementary material, which is available to authorized users.