Making the Case for Tailored Selection of Antiplatelet Therapy

Graphical Abstract: A guided approach to escalation or de-escalation of antiplatelet therapy.

As we evolve toward an era of more personalized medicine, there exists appeal for a tailored approach toward antiplatelet therapy. It has long been known that there is significant interpatient variability in response to clopidogrel. Following a 300 mg loading dose of clopidogrel, ~30% of individuals are observed to have an attenuated pharmacodynamic response as assessed by platelet function testing.^[1] A fraction of the variability in response to clopidogrel appears to be explained by genetic inheritance of a reduced function CYP2C19 allele,^[2] a genetic variant that encodes an enzyme that is involved in both steps of the drug's bioconversion to its active metabolite. Carriers of a reduced function CYP2C19 allele have diminished pharmacodynamic and kinetic response to clopidogrel and are at increased risk of major adverse cardiovascular events (MACE) and stent thrombosis in the setting of percutaneous coronary intervention (PCI).^[3] In response to these findings, the United States Food & Drug Administration applied a warning to the package insert for clopidogrel to make clinicians aware that some individuals are poor metabolizers of the drug and noted that use of an alternative treatment strategy should be considered in this setting.

Despite substantial data to support these observations, the clinical use of platelet function and genotyping for patients being treated with clopidogrel remains infrequent. Some of the hesitation stems from the argument that randomized clinical trials and national guidelines now support the routine use of ticagrelor or prasugrel preferentially over clopidogrel in most patients after an acute coronary syndrome (ACS).^[4] However, the pivotal trials of prasugrel and ticagrelor in this setting demonstrated that a significant reduction in MACE came at the price of increased bleeding.^[5,6] In addition to concerns about bleeding, questions have also been raised as to whether the observed beneficial effects toward reduction in MACE with the more potent P2Y₁₂ inhibitors might have been attenuated if the trials had excluded clopidogrel hyporesponders.

There are also inherent limitations to a genotype- or platelet function-guided approach. Overall, <20% of the observed variability in response to clopidogrel is explained by the CYP2C19 genotype, so genotyping may not always provide a complete phenotypic picture.^[2] In contrast, intraindividual platelet function may fluctuate and requires that the patient be treated with the P2Y₁₂ inhibitor before treatment decisions can be made, during which time the patient may be at risk for bleeding or MACE. Limited data exist to explore the value of both strategies in tandem.

Contributing to the confusion, trials designed to demonstrate the utility of a platelet function- or genotype-guided approach for selection of antiplatelet therapy have yielded inconsistent results. Some of this heterogeneity is explained by challenges and variability in study design. If the applied intervention for a clopidogrel non-responder is to use prasugrel or ticagrelor, then one cannot be sure if a positive trial result is simply due to use of a more potent agent. These trials therefore require a control arm whereby genotyping or platelet function are not applied, and patients are universally treated with either clopidogrel or alternatively a more potent P2Y₁₂ inhibitor. Since only a fraction of patients in the experimental arm ultimately qualify for an escalation or de-escalation in antiplatelet therapy, the treatment received between randomized groups is often identical for a substantial proportion of patients. Therefore, a very large sample size is required to be sufficiently powered to show a benefit for the tailored approach.

In the largest study to date, TAILOR PCI tested a genotype-guided approach by employing 'escalation' of therapy to ticagrelor in carriers of a reduced function CYP2C19 allele, but the trial was unable to demonstrate a significant reduction in MACE in the loss-of-function carriers compared with routine use of clopidogrel despite a very favourable trend.^[7] In another large study, the POPular GENETICS trial evaluated a genotype-guided approach toward 'de-escalation' of more potent P2Y₁₂ inhibition to clopidogrel in patients who were not carriers of a reduced function CYP2C19 allele and showed that the strategy was non-inferior for MACE when compared to universal use of the more potent P2Y₁₂ inhibitors and with less bleeding.^[8] In TROPICAL ACS, a platelet function-guided strategy toward de-escalation of prasugrel to clopidogrel was again found to be non-inferior in regard to MACE compared with routine use of prasugrel for 12 months.^[9] Since many have argued that individual trials may have been underpowered, a meta-analysis that combined data from 11 randomized controlled trials and three observational studies comprising 20 743 patients concluded that guided selection of P2Y₁₂ inhibitor therapy was associated with a 22% reduction in MACE and a favourable trend toward less bleeding.^[10]

The study published by Galli and colleagues in this issue of the the European Heart Journal builds upon this prior meta-analysis from some of its shared co-authors by focusing primarily on the acute setting and employing a frequentist network meta-analysis approach.^[11] This allowed for comparisons across P2Y₁₂ inhibitor strategies tested in 15 ACS trials and represented >60 000 patients. Five of the 15 trials tested a strategy of guided selection (two platelet function and three genetic testing). Thus, as is often the case with network meta-analyses, comparisons between groups sometimes occurred when there were imbalances in the numbers of individuals and included both direct and indirect estimates. The authors determined that guided selection of a P2Y₁₂ inhibitor in the acute setting led to the most favourable balance between efficacy and safety, when compared with an unguided approach. In particular, a guided strategy appeared to reduce MACE without a clear increase in bleeding when compared with unguided use of clopidogrel.^[11] No firm conclusions were drawn regarding the relative value of a strategy of escalation vs. de-escalation of P2Y₁₂ inhibitor therapy or the use of a genotype- vs. platelet function-guided approach.

Although comparisons were also made between the unguided use of prasugrel or ticagrelor vs. clopidogrel, the conclusions from these comparisons should be interpreted with caution as there was marked heterogeneity in regard to study designs and populations. To that end, the analysis included specialized populations where the benefit of more potent P2Y₁₂ inhibition is less clear (e.g. use of prasugrel in the elderly or in medically managed patients).^[12,13] Moreover, it is not immediately intuitive why a guided approach with only selective use of ticagrelor and prasugrel might lead to a larger reduction in MACE than the universal use of the more potent P2Y₁₂ inhibitors. However, one could hypothesize that bleeding may prompt earlier discontinuation of therapy, which in turn could increase risk of thrombotic events.

Ultimately, it is unclear what further evidence would be required to reshape clinical practice and for the use of genotyping or platelet function testing to become routine. Many continue to debate whether all patients should be universally treated with the more potent P2Y₁₂ inhibitors post-ACS or whether unguided de-escalation after a period of time may be appropriate for many lower risk patients.^[14,15] The landscape in antiplatelet therapy is also continuing to shift as a growing number of clinicians consider shorter durations of dual antiplatelet therapy (DAPT). In particular, there is interest in a strategy of P2Y₁₂ inhibitor monotherapy which has been shown to markedly decrease bleeding compared with traditional DAPT, without a clear increase in MACE.^[16] To that end, a strategy of ticagrelor monotherapy could be a simpler approach than requiring genotype or platelet function to determine suitability for de-escalation to clopidogrel but would need to be further studied to better understand the relative efficacy and safety of ticagrelor monotherapy vs. the combination of clopidogrel with aspirin.

As physicians, we are accustomed to the use of surrogate metrics [including international normalized ratio (INR), LDL-cholesterol, or blood pressure] to select dosing and assess response to commonly used therapies. It is therefore somewhat perplexing that the use of platelet function or genetic testing has not become more mainstream despite knowledge of the marked variability that exists in the pharmacodynamic response to clopidogrel. Currently, physicians are encouraged to tailor duration of DAPT based on individual patient characteristics. By extension of logic and based on the clinical trial evidence, it would be appropriate for the medical community to also shift toward a more guided selection of antiplatelet therapy based on patients' perceived bleeding and ischaemic risk, as currently supported by some updated guideline documents.^[4] Ultimately, an era of more personalized medicine is helping clinical practice to evolve beyond the 'one size fits all' approach that has failed many in the past.