In patients with coronary artery disease (CAD), antiplatelet therapy is the mainstay of secondary prevention. Antiplatelet therapy is effective in reducing stent thrombosis, periprocedural damage and ischaemic events among patients undergoing percutaneous coronary interventions (PCI), and reducing plaque disruption and ischaemic events in patients with chronic coronary syndrome and decreasing thrombus burden and flow obstruction in patients with acute coronary syndrome (ACS).
Owing to its favourable safety profile, clopidogrel has replaced the first-generation thienopyridine ticlopidine.1 Clopidogrel is a prodrug activated by several enzymes, including cytochrome 2C19 (CYP2C19), to become its active form.
Genetic Testing of CYP2C19
The Food and Drug Administration (FDA) issued a boxed warning in 2009 suggesting the CYP2C19 genotype should be considered before prescribing clopidogrel. A series of randomised controlled trials (RCTs) have aimed to investigate whether genotype or platelet function testing (PFT) in clopidogrel-treated patients correlates with clinical outcomes.2 The American Heart Association and American College of Cardiology (AHA/ACC) were the first organisations to respond to the FDA warning, arguing that there was insufficient evidence to support it.2
In 2010, Paré et al. genotyped patients with ACS from the CURE trial and concluded that the effect of clopidogrel compared with placebo was consistent, irrespective of CYP2C19 loss-of-function (LOF) carrier status (primary efficacy outcome among LOF carriers 8.0% with clopidogrel versus 11.6% with placebo; HR 0.69; 95% CI [0.49–0.98]; among LOF non-carriers: 9.5% with clopidogrel versus 13.0% with placebo; HR 0.72; 95% CI [0.59–0.87]; p=0.84 for interaction).3
Similarly, in 2011, a meta-analysis by Holmes et al. included the results of four large studies, each of which reported ≥200 cardiovascular events and found no significant association between CYP2C19 genotype and composite cardiovascular outcomes in patients treated with clopidogrel (having one or more allele associated with reduced CYP2C19 enzyme function *2 through *8 versus having *1/*1 (wild-type) or 1 or more *17; gain of function RR 0.97; 95% CI [0.86–1.09]).2
The fact that these analyses have failed to demonstrate a correlation between CYP2C19 genotype and clinical outcomes has left genotyping testing in relation to the use of clopidogrel controversial.2,3
Guided Selection and De-escalation
A number of trials comparing guided or unguided selection of antiplatelet strategies using PFT have been conducted to investigate the role of PFT in the clinical setting.
Over the course of almost 10 years, trials investigating PFT-guided antiplatelet strategies consistently yielded no superiority in reducing ischaemic events regardless of patient type, such as those with ACS or chronic coronary syndrome, with either escalation or de-escalation of P2Y12 inhibitors (Table 1).4 The TROPICAL-ACS study published in 2017 was the only positive trial, in which non-inferiority but not superiority was achieved compared to the standard arm.5 Similarly, subsequent RCTs involving genotype-guided antiplatelet strategy showed no significant correlation between CYP2C19 LOF and clinical outcomes (Table 2).4
In patients with acute MI (AMI) receiving dual antiplatelet therapy (DAPT), ischaemic risk is the highest during the first month after PCI and lowers thereafter, but bleeding risk remains high over time.6 These findings have led to the development of the stepwise de-escalation of DAPT with potent platelet inhibition in the acute phase and de-escalation to clopidogrel in the maintenance phase, and the role of genotyping or PFT in selecting patients for de-escalation has been assessed.
TAILOR-PCI, the largest genotype-guided de-escalation trial to date, enrolled 5,302 patients with ACS or stable CAD randomised to a genotype-guided arm (n=2,652) or a control arm (n=2,650), with a background therapy of aspirin (Table 2).7 In the genotype-guided arm, CYP2C19 LOF carriers received ticagrelor while non-carriers received clopidogrel. In the primary analysis of 1,849 CYP2C19 LOF carriers, the primary ischaemic endpoint did not reach statistical significance (4.0% with genotype-guided arm versus 5.9% with control arm; HR 0.66; 95% CI [0.43–1.02]; p=0.06).
TALOS-AMI, a large-scale RCT investigating unguided de-escalation, enrolled 2,697 East Asian patients with AMI receiving ticagrelor and aspirin in the first month after PCI (Table 2).6 Patients were randomly assigned to the de-escalation group with clopidogrel and aspirin (n=1,349) or the control group, who continued ticagrelor and aspirin (n=1,348). While no significant difference in ischaemic events was observed between the two groups at 12 months (2.1 versus 3.1%; HR 0.69; 95% CI [0.42–1.14]; p=0.15), the de-escalation group experienced significantly fewer net adverse clinical events compared to the control group (4.6 versus 8.2%; p<0.001 for non-inferiority; HR 0.55; 95% CI [0.40–0.76]; p=0.0001 for superiority).
There are several potential explanations for the lack of efficacy of either genotype or PFT-guided antiplatelet strategy in different trials. First, the ischaemic risk is highest in the first month after AMI, and the selection of P2Y12 inhibitor could affect outcomes regardless of CYP2C19 genotype.8
Second, in TAILOR-PCI, only 23% of participants were East Asians, while the TALOS-AMI study enrolled East Asians only. As observed in several clinical trials, East Asians have a lower ischaemic risk and a higher bleeding risk, a phenomenon known as the East Asian paradox.9 Therefore, the unguided de-escalation from ticagrelor and aspirin to clopidogrel and aspirin was associated with better outcomes in TALOS-AMI, while similar outcomes were observed between clopidogrel and ticagrelor groups in TAILOR-PCI, which mainly included Caucasians.
Various network meta-analyses have been conducted to compare guided versus unguided antiplatelet strategies, but have yielded differing results.
In 2022, Galli et al. published a meta-analysis of 15 RCTs with 61,898 patients comparing guided de-escalation and standard DAPT with potent P2Y12 inhibitors for 12 months.10 A guided de-escalation strategy was associated with fewer major adverse cardiovascular events (IRR: 0.80; 95% CI [0.65–0.98]) without a statistically significant increase in bleeding (IRR: 1.22; 95% CI: [0.96–1.55]). The authors concluded that guided de-escalation of P2Y12 inhibitor therapy is associated with the most favourable balance between safety and efficacy.10
In the same year, Kuno et al. published a meta-analysis including 19 RCTs with 69,746 patients that compared unguided de-escalation and guided selection of DAPT. In this analysis, unguided de-escalation decreased bleeding without increasing ischaemic events compared to guided selection of DAPT.8
The key difference between the two meta-analyses is that Galli et al. did not include TALOS-AMI because the data were not available at the time, which may potentially have contributed to the different results.
Kang et al. published an individual participant-level data meta-analysis in 2023 that included RCTs with >1,000 patients with ACS undergoing PCI. This meta-analysis included 10,133 patients from four RCTs, which were all considered to have a low-bias risk.11 This analysis showed that the cumulative incidence of the ischaemic endpoint was significantly lower in patients assigned to the de-escalation strategy, guided or unguided, than in those assigned to the standard strategy (2.3% versus 3.0%; HR 0.761; 95% CI [0.597–0.972]; log rank p=0.029). The cumulative incidence of bleeding was also significantly lower in the combined de-escalation strategy group (6.5% versus 9.1%; HR 0.701; 95% CI [0.606–0.811]; log rank p<0.001). Subgroup analyses revealed a significant interaction between guided versus unguided de-escalation and the bleeding endpoint (p for interaction=0.007), but not in the ischaemic endpoint (p for interaction=0.688).
With a lack of clinical advantages with genotyping or PFT in the context of both DAPT selection and de-escalation being demonstrated in RCTs and meta-analyses, the clinical value of genotyping or PFT in cardiology remains elusive.
Guideline and Consensus Recommendations
While the AHA/ACC have maintained their stance on genotype or PFT-guided approach since their joint statement in 2011, the position of the European Society of Cardiology (ESC) has evolved (Table 3).
The 2017 ESC guidelines, a focused update on DAPT in CAD, mentioned that ‘no randomized trial has ever demonstrated any clinical benefit’ of genotype or PFT-guided antiplatelet approach, and further recommended against routine PFT to adjust antiplatelet therapies before or after elective stenting (class III, level of evidence A).12
Later ESC guidelines recommended conducting an unguided or guided antiplatelet approach ‘based on clinical judgment’, citing the TROPICAL-ACS trial, which reached only non-inferiority.13,14
Most recently, the 2023 ESC guidelines for ACS management alluded to the stance of their committee.15 A recommendation on whether to guide antiplatelet therapy with genotype or PFT was removed and replaced with a Gaps in Evidence section, which concluded that the clinical benefit of guided antiplatelet approach ‘remains unclear’.15
Owing to the lack of positive prospective trials in Asian patients, the 2020 Asian Pacific Society of Cardiology consensus on the use of P2Y12 receptor antagonists in the Asia-Pacific region has recommended against the use of PFT or genotype-guided decision-making.16
Recently, a panel of international experts tried to fill this gap in the clinical guidelines. They suggested the use of genotyping or PFT in three clinical frameworks, while admitting that there are still major limitations, such as the cost-effectiveness of the tests, the necessity of using the tests while unguided de-escalation has already proved to improve clinical outcomes in RCT and the lack of evidence from large-scale RCTs or patient-level meta-analyses for the guidelines to support the routine selection of guided antiplatelet therapy.17
Reasons for the Failure of Guided Strategies
More than 10 years after the FDA warning, attempts to investigate genotyping or PFT-guided selection or a de-escalation approach in cardiology have resulted in more controversies than clarity.
Several factors led to the results observed. First, taking pharmacokinetics into consideration for accumulation of the active form of clopidogrel, once it has reached the steady state, its antiplatelet effect is still apparent even in CYP2C19 poor metabolisers carrying two LOF alleles as observed in previous analyses.18,19
Second, the cut-off values and timing for defining high platelet reactivity are neither standardised nor ethnicity specific. In 2021, Kim et al. published a position statement that included evaluation of the cut-off for P2Y12 reaction units (PRUs).9 Consistent with the East Asian paradox, clinical evidence suggests a higher PRU cut-off (252–289) for East Asian patients, in contrast to Caucasian patients with a PRU cut-off (208 or 235), as suggested by the position statement.9 Furthermore, as the ischemic risk varies over time, the optimal cut-off for platelet inhibition of various phases for secondary prevention remains in question.
Conclusion
From years of investigation, the available evidence has yielded only inconsistent or unclear correlation between genotype or PFT-guided approaches and clinical outcomes. This inconclusive evidence makes it difficult for clinical guidelines to provide a clear recommendation for their use in clinical setting. To focus on a single nucleotide polymorphism and disregard other metabolic pathways, epigenetics or pharmacokinetics, or solely to focus on phenotype and disregard the overall clinical profile of the patient, may not be the best approach to improve outcomes. Factors such as ethnic difference, comorbidity, bleeding risk, anatomy, complexity of procedure, type of stent placed, stenting technique and medication adherence could all play a role in a patient’s clinical outcomes.
Clinical Perspective
- Multiple randomised controlled trials have failed to demonstrate clear clinical benefits from antiplatelet therapy guided by either genotype (using CYP2C19 genetic testing) or platelet function testing.
- Current guidelines do not recommend routine use of genotype or platelet function testing to guide antiplatelet therapy owing to insufficient evidence demonstrating clinical benefit.
- Studies suggest that unguided de-escalation from potent P2Y12 inhibitors to clopidogrel after the acute phase can reduce bleeding risk without increasing ischaemic events, particularly in East Asian populations who exhibit the East Asian paradox; this is characterised by a lower ischaemic risk but higher bleeding risk compared to Caucasian populations.
- Clinicians should prioritise comprehensive patient assessment – including ethnicity, bleeding risk, comorbidities, procedural complexity and medication adherence – over single biomarker testing when making antiplatelet therapy decisions.