Genetic influences on aspirin response in patients undergoing percutaneous coronary intervention

This editorial refers to ‘Association of the coronary artery disease risk gene GUCY1A3 with ischaemic events after coronary intervention’, by T. Kessler et al., pp. 1512–1518.

The last three decades have seen significant advances in interventional and pharmacologic therapy for coronary artery disease (CAD). Aspirin is the cornerstone of CAD treatment, both for primary prevention of cardiovascular events and in combination with P2Y12 inhibitors such as clopidogrel and ticagrelor for patients who undergo percutaneous coronary intervention (PCI). And yet, despite decades of experience and evidence supporting the use of aspirin, variability in outcomes in patients with CAD remains a challenge. We still do not have proven techniques to identify patients at low risk who can be treated with dual antiplatelet therapy (DAPT) for a short period of time and those who risk developing stent thrombosis despite adherence to DAPT.

While there is considerable literature investigating the possible mechanisms of high-on-aspirin platelet reactivity, true ‘aspirin resistance’, defined as an inability of aspirin to inhibit its target platelet cyclooxygenase (COX)-1, is rare. COX-1 is completely inhibited in over 99% of patients;¹ however, there are significant differences in off-target effects of aspirin via COX-1-independent mechanisms.² In contrast, for clopidogrel, the influence of several genes, notably ABCB1 and CYP2C19, has been extensively investigated³ and has been shown to confer a 76% increased risk of major adverse cardiac events and an almost four-fold risk of stent thrombosis.⁴ Clearly, a genetic test or series of tests which could identify patients in whom antiplatelet therapy is less effective would be extremely useful in guiding CAD and post-PCI therapy, but development of such strategies is still in its infancy.

Kessler et al.⁵ pursue the hypothesis that a known allele for CAD risk, GUCY1A3, modifies the platelet functional response to aspirin and clinical outcomes in patients undergoing PCI, the great majority of whom were treated with aspirin and clopidogrel. They test this hypothesis using two resources of genetic data: (i) the ISAR-ASPI registry of PCI patients with available arachidonic acid (AA) and ADP-induced impedance platelet aggregometry data and (ii) three cohorts of patients treated with PCI, including ISAR-ASPI, the clopidogrel arm of the PLATO trial, and the UCORBIO biobank. These patients were all followed for at least 30 days, allowing for analysis of differences in 30-day death, cardiovascular events, and stent thrombosis. The authors found that GUCY1A3 ‘high-risk’ allele homozygous patients had higher AA- but not ADP-induced platelet aggregation and were at 1.7-fold higher risk for ischaemic events following PCI, with no differences in bleeding. Based on these findings, the authors conclude that the high-risk allele is associated with a form of high on-treatment aggregation that also confers an increased risk of ischaemic events.

This study focused on a single genetic locus, GUCY1A3, which was identified in a family with extremely high rates of early-onset CAD.⁶ Further investigation demonstrated that GUCY1A3 is involved in nitric oxide signalling and that the changes in soluble guanylyl cyclase conferred by the ‘high-risk’ allele (which is intronic and lowers GUCY1A3 promoter activity) have significant implications for risk of myocardial infarction (MI) and make GUCY1A3 a promising target for cardiovascular therapeutics.⁷ The combination of several cohorts of over 6000 patients in the current study significantly increases the power to detect small differences in adverse cardiovascular events including stent thrombosis, which in the era of second-generation drug eluting stents during which the cohorts were studied is quite rare. In addition, the study identifies intermediate (laboratory platelet function) and clinical (cardiovascular death and stent thrombosis) endpoints to connect GUCY1A3 genotypes to physiological and clinically relevant outcomes.

Still, many questions remain about the utility of GUCY1A3 genotype in guiding antiplatelet therapy. First, the comparison of outcomes at 30 days after PCI leaves open the question of how GUCY1A3 high-risk genotypes might influence events out to 12 months, the usual length of DAPT, or after 12 months when aspirin monotherapy is the standard of care. Second, the study does not consider other gene variants, including CYP2C19 or PEAR1, which are known to influence aspirin and clopidogrel response.⁸ Third, variants in GUCY1A3 are also associated with blood pressure and the presence of coronary atherosclerosis which were not completely adjusted for in the current analysis. Lastly, whether the influence of high-risk GUCY1A3 alleles on aspirin response can be overcome by alternative P2Y12 inhibitors remains unknown. With this in mind, analysis of the ticagrelor arm of the PLATO trial, the prasugrel arm of the TRITON-TIMI-38 trial, and/or subsets of cohorts such as UCORBIO containing patients treated with ticagrelor or prasugrel would be of great interest.

Development of tests and strategies utilizing our knowledge of the genetic influences on response to antiplatelet therapies is still in its infancy. In the future, one can envision a package of genetic testing performed at the time of CAD diagnosis or just after PCI that could provide a truly personalized strategy for antiplatelet therapy (Figure 1). Many questions remain regarding the effectiveness of implementation strategies for such data. How can complex genetic information best be presented to clinicians and patients to influence decision making? If a high-risk genotype is discovered, how should the treatment strategy be altered—a change to a higher daily aspirin dose, to an alternative P2Y12 inhibitor, to a different stent platform, or if risk is sufficiently high with no alternative, to a strategy of medical management? How can we ensure that the racial and ethnic diversity, which were not well reflected in the current study, are better represented in future studies so that we have a more comprehensive understanding of the effects of a variant in diverse genetic backgrounds? Answering these questions will be a cornerstone of advancing the field of personalized cardiovascular medicine in the coming years and decades.

Figure 1.

Current practice and potential precision medicine approach for choosing a P1Y12 inhibitor. Identifying patients with genetic variants such as CYP2C19, ABCB1, and GUCY1A3 could guide personalized decision-making about medical treatment after MI, PCI, or CABG or in the setting of PAD. Platelet aggregation testing may help define relevance of genetic test results but it has not been shown to predict outcomes. ASA, aspirin; CABG, coronary artery bypass graft; CAD, coronary artery disease; DAPT, dual antiplatelet therapy; ISR, in-stent restenosis; IST, in-stent thrombosis; MACE, major adverse cardiovascular events; PCI, percutaneous coronary intervention; MI, myocardial infarction; NSTEMI, non-ST elevation myocardial infarction; PAD, peripheral arterial disease; STEMI, ST-elevation myocardial infarction.

Conflict of interest: none declared.

The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology.