Platelet function testing in atrial fibrillation patients undergoing percutaneous coronary intervention

Platelet function testing (PFT) could be a useful clinical tool to guide individualized antithrombotic treatment in patients with atrial fibrillation (AF) undergoing percutaneous coronary intervention (PCI). We aimed to investigate platelet reactivity (PR) in the context of a contemporary registry. “Real-world” data were retrieved from a nationwide, multicenter, observational study of AF patients on oral anticoagulants (OAC) undergoing PCI. Patients treated with a P2Y12 inhibitor, namely clopidogrel or ticagrelor, as part of double or triple antithrombotic therapy, were submitted to PFT before discharge and were followed up for 12 months. Out of 101 patients included in the study, 66 were submitted to PFT while on clopidogrel and 35 while on ticagrelor; PR was 162.9 ± 68 PRU and 46.02 ± 46 PRU, respectively (P < 0.001). High on-treatment PR (HTPR) was observed in 15 patients under clopidogrel (22.7%); 7 of them escalated to ticagrelor. Low on-treatment PR (LTPR) was found in 9 clopidogrel and 28 ticagrelor-treated patients (13.6% vs. 80%, P < 0.001), of whom only 1 de-escalated to clopidogrel. PR did not differ by OAC regimen. PFT results had no impact on aspirin prescription at discharge, while failed to predict significant bleeding events at follow up. Ticagrelor administration led to lower PR and lower incidence of HTPR in comparison with clopidogrel. Physicians’ behavior in response to knowledge of a patient’s PR was variable. Further studies are required to elucidate the role of PFT as a tool to guide individualized antithrombotic treatment in this clinical scenario. Supplementary Information The online version contains supplementary material available at 10.1007/s11239-022-02723-4.


Introduction
Antithrombotic treatment in atrial fibrillation (AF) patients with an indication for oral anticoagulation (OAC) undergoing percutaneous coronary intervention (PCI) poses a significant therapeutic dilemma. On the one hand, this subgroup of patients has an increased ischemic risk, including both stroke and stent thrombosis (ST) [1], while, on the other hand, the "traditional" approach of triple antithrombotic therapy (TAT) consisting of dual antiplatelet therapy (DAPT) with aspirin and a P2Y 12 inhibitor plus OAC is associated with high bleeding rates [2]. Recent randomized controlled trials (RCTs) compared dual antithrombotic therapy (DAT), with a non-vitamin K oral anticoagulant (NOAC) plus single antiplatelet therapy (SAPT), to TAT, with vitamin K antagonist (VKA) plus DAPT, and make a compelling case regarding safety of DAT with an antiplatelet agent, mainly clopidogrel, when compared with TAT [3][4][5][6]. However, there are concerns in case of aspirin omission, that patients on solely clopidogrel treatment might be exposed in increased risk of ST [1,7]. Among clopidogrel-treated patients, high platelet reactivity (PR) is commonly encountered, with incidence varying from 16 to 50% [8,9]. Τhere is paucity of data concerning the incidence of high on-treatment platelet reactivity (HTPR) in AF patients treated with NOACs undergoing PCI [10]. In spite of HTPR risk, clopidogrel is the preferred agent in this clinical context, with the novel P2Y 12 inhibitors, ticagrelor and prasugrel, underrepresented in the aforementioned trials [3][4][5][6]11]. In this scenario, platelet function testing (PFT) might serve as a clinical tool to guide tailoring of antithrombotic regimen, i.e. intensification of treatment via escalation to a more potent agent or aspirin addition in case of initial DAT selection. Furthermore, PFT has been suggested as a useful tool to identify patients with an increased risk of bleeding [12,13].
In this background, this study aims to investigate PR in AF patients on OAC undergoing PCI, in the context of a real-life contemporary registry, and its potential impact on antithrombotic agent choices and patient outcomes.

Methods
Greek AntiPlatElet Atrial Fibrillation (GRAPE-AF) registry is a nationwide, prospective, single-arm, observational, noninterventional, cohort study, enrolling patients with non-valvular AF on OAC undergoing PCI with stent implantation in 18 PCI centers in Greece. The study, with rationale and design previously described in detail, aims to provide "real world" data regarding antithrombotic therapy [14]. Overall, 653 patients have been recruited (exclusion criteria in supplementary Table 1). A written informed consent was obtained from all participants. The study complies with the declaration of Helsinki and was approved by the Ethic Committee of each participating hospital. GRAPE-AF study is registered at ClinicalTrials.gov: NCT03362788.
Between March 2018 and October 2019, in the context of GRAPE-AF registry, we assessed PR in anticoagulated AF patients undergoing PCI in 6 hospitals capable of performing PFT using the VerifyNow assay (Accriva Diagnostics, San Diego, CA). Eligible patients, treated with a P2Y 12 inhibitor, namely clopidogrel or ticagrelor (per treating physicians' choice), as part of DAT with OAC or TAT with OAC plus aspirin, were submitted in PFT pre-discharge, after index PCI. Results were available to the attending physician before patient's discharge and used upon his discretion.
For PR assessment, peripheral venous blood samples were drawn in 2 ml Greiner Bio-One Vacuette® partial fill blood collection tubes containing 3.2% sodium citrate, 2-4 h after P2Y 12 inhibitor maintenance dose and at least 12 h post -loading dose (600 mg of clopidogrel or 180 mg of ticagrelor) in case of P2Y 12 inhibitor naive patients. PR was measured via the point of care P2Y 12 VerifyNow platelet function assay by an investigator blinded to the antithrombotic regimen. Results were reported in P2Y 12 reaction units (PRU) and percent inhibition, which was calculated as: ([BASE -PRU]/ BASE) × 100. A PRU > 208 and a PRU < 85 were considered as HTPR and low on-treatment platelet reactivity (LTPR), respectively [8].
Follow-up visits were performed through telephone or personal contact in 1, 6 and 12 months after enrollment. Collected data included major adverse cardiovascular and cerebrovascular events (MACCE), bleeding complications and side effects of antithrombotic therapy, as well as information regarding patients' adhesion or changes to the prescribed treatment regimen.
The prespecified primary endpoints used for sample size calculation of this sub-study was PR level and HTPR rate between patients treated with clopidogrel vs. ticagrelor. Secondary endpoints included bleeding (per the Bleeding Academic Research Consortium [BARC] definition) and major adverse cardiovascular events (MACE), i.e. cardiovascular death, myocardial infarction, stroke, stent thrombosis or unplanned coronary revascularization rates, during the first year post index PCI.
Continuous data are presented as means ± standard deviation (SD) and compared using the Student's t-test whereas categorical variables are presented as frequencies and group percentages and compared using the chi-square or Fisher exact test. Kaplan-Meier curves for bleeding-free survival were obtained for each P2Y 12 inhibitor subgroup and compared using the log-rank test. If multiple bleeds were recorded for a patient, categorization was based on the first bleeding event at the highest level of severity experienced. All tests were two-tailed, and statistical significance was considered for P values < 0.05.
Approximately 600 patients during 1 year were initially planned to participate in GRAPE-AF registry. Out of them, 70% and 30% were predicted to receive clopidogrel and ticagrelor respectively, as part of their antithrombotic therapy. Taking into account that PR assessment wouldn't be available in all participating centres, we planned to assess PR in 120 patients. Out of them 36 and 84 were expected to receive ticagrelor and clopidogrel respectively. Assuming a PR of 35 ± 30 PRU and 100 ± 70 PRU for ticagrelor and clopidogrel respectively, group sample sizes of 36 and 84 would achieve > 95% power to detect a difference of at least 65.0 PRU between groups, with estimated SD of 35.0 and 70.0 respectively and with a significance level (alpha) of 0.05, assuming that the actual distribution is not normal [14]. When assuming an HTPR rate of 3% and 25% in ticagrelor and clopidogrel group respectively, the same group sample sizes were predicted to achieve 88% power to detect an absolute difference between the group proportions of 22% with a two-sided Fisher's exact test, at a 0.05 significance level.

Results
Overall, 653 patients have been enrolled in the GRAPE-AF registry. Out of these, a total of 101 patients with PFT were included in the present sub-study. Table 1 presents characteristics of study population per P2Y 12 inhibitor group at PFT with 66 and 35 patients treated with clopidogrel and ticagrelor, respectively. All patients received periprocedurally aspirin and till discharge. There was a tendency for more common use of ticagrelor in acute coronary syndrome than in stable coronary artery disease patients and ticagrelor use in the form of DAT (without aspirin at discharge). At PFT, VKA were administered to 13 patients (12.9%), while 88 (87.1%) received NOAC; 55 patients received rivaroxaban, 23 dabigatran and 10 apixaban (Suppl. Figure S1).
In patients suffering from any bleeding, PR at PFT was 119 ± 83 PRU compared to 124 ± 83 PRU in those who did not (P = 0.748). Mean PR in patients with BARC ≥ 2 bleeding events was 105 ± 96 PRU compared to 126 ± 80 PRU in patients with minor or no events (P = 0.370).
Regarding thrombotic complications, there were 4 patients who suffered an ischemic event under clopidogrel, 2 patients under ticagrelor and 1 patient with HTPR at discharge, under Rivaroxaban 15 mg monotherapy at 7th month.

Discussion
During a transitional period for antithrombotic therapy from TAT to DAT, while incorporating NOAC instead of VKA, we have described the use of PFT in AF patients undergoing PCI, in the context of a real-life contemporary registry. In our study, PR and subsequent incidence of HTPR was greatly reduced in ticagrelor-treated patients compared to clopidogrel-treated patients. The observed HTPR incidence in clopidogrel-receiving anticoagulated patients is consistent with studies examining PR in patients on clopidogrel without anticoagulation [8,9]. Among 103 patients on clopidogrel and rivaroxaban, HTPR and LTPR rates of 11.7% and 24.3% were described [10]. In contrast, the respective rates in our clopidogrel plus rivaroxaban-receiving patients were 32.4% and 14.7%; differences in populations and methodology used may have been implicated. Overall, observed PR levels were consistent with lack of effect of OACs in ADP-induced platelet reactivity measured by VerifyNow. Similar results have been reported in different clinical settings [15][16][17]. However, using other tests more sensitive to OAC effect might have unveiled differences. Moreover, PR decrease with ticagrelor was rather homogeneous, independently of anticoagulant administered. Of importance, our study provides for the first time, to our knowledge, data on PR under full-dose anticoagulation with rivaroxaban or dabigatran plus ticagrelor, in sizeable groups of patients. Data on dual pathway inhibition with the use of a either clopidogrel or ticagrelor in adjunct to a vascular dose of rivaroxaban (2.5 mg bid) have been recently reported in stable atherosclerotic heart disease patients; compared with DAPT, similar effects on platelet-mediated global thrombogenicity but reduced thrombin generation was demonstrated [18].

3
The potential clinical usefulness of PFT to guide individualized antithrombotic treatment in anticoagulated patients undergoing PCI is unclear [9,13,19]. In a randomized controlled trial involving Chinese patients and receiving warfarin and clopidogrel, PFT using vasodilator-stimulated phosphoprotein testing to up-titrate clopidogrel dosage in those with high platelet reaction index was applied. PFT-guided therapy resulted in decrease of MACCE, but no difference in survival or major bleeding events [20]. In our study, due to its observational nature, physicians' behavior in response to patient's PR result was variable; almost half of the patients remained on clopidogrel despite of HTPR demonstration. In addition, LTPR on ticagrelor practically had no impact on P2Y 12 inhibitor selection. This likely reflects the well described very low PR achieved in the great majority of non-anticoagulated ticagrelor-treated patients and the possible lack of predictive value for subsequent bleeding events [21][22][23]. Although in one third of HTPR patients aspirin was added post PFT, overall rates of TAT or DAT were not affected by HTPR or LTPR status. Clinical equipoise among physicians regarding use of this information in the absence of definitive clinical trial data demonstrating the efficacy of this approach is hypothesized.
In our population PFT failed to effectively differentiate subsequent bleeding events, although there was a trend of lower PR in patients with BARC ≥ 2 bleeding events compared to those with minor or no events, as well as, a trend of higher BARC ≥ 2 bleeding rates in LTPR patients. Differences in age, index event and aspirin administration at discharge between groups, as well as selection bias, have probably affected bleeding outcomes.
The pivotal trials comparing TAT versus DAT did not include comprehensive PFT or genotyping in their design; thus, a recommendation for routine testing in this setting cannot be established and PFT clinical usefulness remains unproved. Nevertheless, a low PR under clopidogrel in an anticoagulated patient who is -by default-in high bleeding risk, might dictate omitting aspirin and DAT adoption. In contrary, detecting HTPR under clopidogrel, could point towards the adoption of a more intense antiplatelet regimen, via continuation of aspirin -in the form of TAT-or switching to ticagrelor. Considering the frequent occurrence of HTPR on clopidogrel, significant concerns have been raised that the suggested by the guidelines default strategy of DAT with NOAC and clopidogrel, leaves up to one-third of patients without adequate platelet inhibition [24]. Although randomized controlled trials definitively showed the superiority of DAT over TAT regarding bleeding events, they were underpowered for ischemic events. A higher risk of mainly stent-related ischemic events associated with DAT over TAT was described in a recent meta-analysis, although not confirmed in another [25,26]. Escalation of antiplatelet therapy, even PFT-guided, might be considered in selected high-risk patients to avoid falling into the concept of sacrificing efficacy for safety [27].
PFT to identify LTPR and accordingly de-escalate antithrombotic therapy by choosing DAT to avoid bleeding may be more realistic, since bleeding side-effects are common in such patients. Nevertheless, the number of patients treated with TAT seems to be continuously decreasing as more clinicians adopt DAT at discharge following PCI. In similar clinical scenario, genotyping has been proposed as an independent factor in tailoring of antithrombotic therapy [10].

Limitations
Nonrandomized, registry data, with lack of statistical adjustments for differences in compared group characteristics are presented. Moreover, due to the nature of the study, results are depicted in an observational, descriptive manner. PFT sub-study was led to premature discontinuation due to completion of the main study and slower recruitment in substudy, pertinent to unavailability of PFT in participant centers. The small sample size renders the study underpowered for clinical events and hinders propensity score matching. Nevertheless, significant differences in prespecified primary endpoints were achieved. The results of this study are insufficient to demonstrate a clinical benefit of PFT in this group of patients, though are hypothesis generating. Apixaban was underrepresented in our study, given the fact that patients' enrollment was initiated prior to publication of the Antithrombotic Therapy after Acute Coronary Syndrome or PCI in Atrial Fibrillation (AUGUSTUS) trial [5]. Moreover, edoxaban is currently not available, and prasugrel use is limited in this context, due to lack of adequate data. Finally, patients of GRAPE-AF were recruited during a transitional period regarding trends in antithrombotic therapy and important scientific data accruement. Current patterns of antithrombotic treatment in AF patients post-PCI may therefore differ following the clearly demonstrated superiority of DAT with clopidogrel over TAT regarding safety and the proposal of DAT as the default strategy after one week or post discharge with class I recommendation [28].

Conclusion
In AF patients on OAC undergoing PCI, PFT applied in a real-life contemporary registry demonstrated the more potent platelet inhibition provided by ticagrelor and no impact of selected OAC on ADP-induced platelet reactivity measured by VerifyNow. PFT might represent a tool for tailoring of antithrombotic therapy although further and larger studies are required to elucidate its role. With the foregoing limitations, this study could be informative for the practicing 1 3 clinician balancing between the ischemic and bleeding hazard in such patients.
Funding Open access funding provided by HEAL-Link Greece. GRAPE AF registry was financially supported by the Hellenic Cardiac Society, which was not involved in study design; in the collection, analysis, and interpretation of data; in writing the report; and in the decision to submit the article for publication.

Declarations
Conflict of interest Dr. Alexopoulos has received lecturing honoraria/ advisory board fees from Astrazeneca, Bayer, Boehringer Ingelheim, Pfizer, Medtronic, Biotronik, Chiesi Hellas. Other authors have no disclosures.

Ethical approval
The study complies with the declaration of Helsinki and was approved by the Ethic Committee of each participating hospital.

Informed Consent A written informed consent was obtained from all participants.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.