Cilostazol combined with P2Y12 receptor inhibitors: A substitute antiplatelet regimen for aspirin‐intolerant patients undergoing percutaneous coronary stent implantation

Abstract Background Cilostazol combined with P2Y12 receptor inhibitor has been used as a substitute regimen for aspirin‐intolerant patients undergoing percutaneous coronary stent implantation on a small scale. Its exact impact on platelet functions and clinical benefits of aspirin‐intolerant patients is unknown. Hypothesis Cilostazol combined with P2Y12 receptor inhibitors could be used as a substitute antiplatelet regimen for aspirin‐intolerant patients undergoing percutaneous coronary stent implantation. Methods In this multicenter prospective cohort trial, patients undergoing elective percutaneous coronary stent implantation were assigned to the cilostazol group (cilostazol plus P2Y12 receptor inhibitors), based on aspirin intolerance criteria, or the aspirin group (aspirin plus P2Y12 receptor inhibitors). Platelet PAC‐1, CD62p, and vasodilator‐stimulated phosphoprotein phosphorylation (VASP‐P) were detected by flow cytometry. The primary endpoints were major adverse cardiovascular and cerebrovascular events (MACCE) including all‐cause death, acute myocardial infarction, emerging arrhythmia, nonfatal stroke, and heart failure. The secondary endpoints were the Bleeding Academic Research Consortium (BARC) bleeding events. Results One hundred and fifty‐four aspirin‐intolerant percutaneous coronary stent implantation patients and 154 matched aspirin‐tolerant patients from a total of 2059 percutaneous coronary stent implantation patients were enrolled. The relative activation level of PAC‐1, CD62p, and platelet reaction index reflected by the VASP‐P test were similar in the two groups (p > .05). After 12 months of follow‐up, the incidence of all‐cause death was 1.9% in the cilostazol group and 1.3% in the aspirin group (risk ratio [RR], 1.500; 95% confidence interval [CI], 0.254–8.852; p = 1.000); the incidence of acute myocardial infarction was 0.6% in the cilostazol group and 1.3% in the aspirin group (RR, 0.500; 95% CI, 0.046–5.457; p = 1.000). No significant difference was seen in other MACCE events, or in any types of BARC bleeding events. Conclusions Cilostazol combined with P2Y12 inhibitors was not inferior to aspirin‐based standard therapy and could be used as a reasonable substitute antiplatelet regimen for aspirin‐intolerant patients undergoing percutaneous coronary stent implantation, but again with limitations, which required a larger sample and longer follow‐up to confirm its efficacy.


| INTRODUCTION
Dual antiplatelet therapy consisting of aspirin and P2Y 12 receptor inhibitors has become a standard treatment to prevent thrombotic complications for patients undergoing percutaneous coronary stent implantation. 1,2 However, aspirin-intolerant patients undergoing stent implantation with previous peptic ulcer, erosive gastritis, gastrointestinal bleeding, gout, and aspirin-related mucocutaneous or respiratory hypersensitivity were facing a higher risk of aspirinrelated adverse reactions, which might outweigh the antiplatelet benefit of standard dual antiplatelet therapy (DAPT). [3][4][5] Several measures have been attempted to deal with the dilemma, but neither adding proton pump inhibitors nor aspirin desensitization could address aspirin intolerance situation. [6][7][8] Cilostazol acts as an antiplatelet drug by selectively inhibiting the activity of phosphodiesterase III, further inhibiting the decomposition of cAMP. 9 Since cilostazol does not affect the cyclooxygenase pathway, it exerts an antiplatelet effect without increasing the risk of gastrointestinal bleeding and other adverse reactions relating to aspirin. 10 Nonaspirinbased DAPT, such as cilostazol combined with P2Y 12 receptor inhibitors, has been used for aspirin-intolerant patients on a small scale. 11 However, the efficacy and safety, especially the antiplatelet effects, have not been truly verified by large-scale clinical trials.

| Grouping and treatment
Patients were divided into the cilostazol group and the aspirin group according to the assessment of whether patients could tolerate aspirin at admission. Since there has never been a formal definition of "aspirin intolerance" before, we redefined the clinical criterion based on previous cross-sectional investigations and related cohort trials. 11,13 "Aspirin intolerance" was defined as (1) aspirin-related gastrointestinal discomforts like acid reflux, vomiting, abdominal pain; (2) history of peptic ulcer, atrophic gastritis, erosive gastritis, antral gastritis, and reflux esophagitis; (3) gastrointestinal bleeding history or comorbid high bleeding risk, positive fecal occult blood during hospitalization (at least consecutive two times); (4) history of subtotal gastrectomy or gastric polypectomy within 3 years; (5) hyperuricemia/gout; (6) aspirin-exacerbated respiratory disease including severe rhinitis, bronchospasm, and aspirin-induced respiratory diseases 14 ; (7) aspirin-related skin and mucosal hypersensitivity reactions including rash, urticaria, edema, and allergic shock. Patients who met any aspirin intolerance criteria above were assigned to the cilostazol group (cilostazol plus clopidogrel or ticagrelor; the maintaining dose was 50 mg bid, 75 mg qd, 90 mg bid for cilostazol, clopidogrel, and ticagrelor, respectively). Once a patient was enrolled into the cilostazol group, one aspirin-tolerant patient would be matched into the aspirin group by propensity score matching in a 1:1 ratio to ensure comparability between groups. The maintaining dose was 100 mg qd, 75 mg qd, 90 mg bid for aspirin, clopidogrel, and ticagrelor, respectively, in the aspirin group. CD62P reflected the expression of platelet P-selectin, while PAC-1 reflected the level of activated platelet glycoprotein GPIIb/IIIa fibrinogen receptor. [15][16][17][18] Meanwhile, platelet vasodilator-stimulated phosphoprotein phosphorylation (VASP-P) was measured to detect the platelet VASP phosphorylation level, which reflected the platelet reaction index (PRI) indirectly. 19,20 VASP-P measurement followed the protocol provided by the Biocytex kit.

| Follow-up and endpoints
The primary endpoint (efficacy endpoint) was the first occurrence of major adverse cardiovascular events (MACCE), including all-cause death, acute myocardial infarction, emerging arrhythmia, nonfatal stroke, heart failure. The secondary endpoint (safety endpoint) was Bleeding Academic Research Consortium (BARC) bleeding events following BARC standards and other skin/mucosal adverse reactions (rash, urticaria, conjunctivitis, angioedema), headache, and gastrointestinal discomfort. 21 All outcomes were adjudicated according to standard definitions by an independent committee blinded to treatment assignment. All enrolled patients were followed 12 months after discharge. Endpoints were acquired through telephone interviews and outpatient clinics。

| Statistical analysis
Baseline characteristics were summarized for the two groups by treatment allocation. The rate of MACCE was estimated to be 13% in the cilostazol group in 1 year of follow-up. The study would have 80% power to detect the difference with a two-sided α level of .05.
The sample size was estimated to be 132 in each group. Assuming a dropout rate of 10%, a total of 145 patients in each group were required. Propensity score matching was used to account for the differences in baseline characteristics between the cilostazol group and the aspirin group. 22

| Baseline characteristics
From January 2018 to January 2020, a total of 2059 patients underwent elective percutaneous coronary stent implantation. Among them, 172 patients were given cilostazol combined with P2Y 12 receptor inhibitors due to a clear reason for aspirin intolerance. Of the 172 patients, 160 met our inclusion criteria and entered the cohort.
In the initial multivariable analysis, variables consisting of age, chronic kidney disease, smoking history, low-density lipoprotein cholesterol, and P2Y 12 receptor inhibitor were found with significant differences between the two groups. After 1:1 propensity score matching, a total of 154 patients in the cilostazol group and matched 154 patients in the aspirin group successfully entered the final comparison of followup ( Figure 1 and Table S1), and all baseline characteristics were wellbalanced between the two groups (Table 1). Meanwhile, 96 patients' venous blood (1:1 paired from the two groups) were collected for antiplatelet effect detection by flow cytometry after informed consent.

| Flow cytometry test
Forty-eight patients in the cilostazol group and 48 matched patients in the aspirin group participated in this study part. The relative activation level of PAC-1 in the cilostazol group was 10.95 ± 7.85%, the relative activation level of PAC-1 in the aspirin group was 9.09 ± 8.09%. There was no significant difference between the two groups (p = .171). The relative activation level of platelet CD62p in ZHAO ET AL. | 191 the cilostazol group was 17.87 ± 8.97%, which was 14.51 ± 8.87% in the aspirin group (p = .055) ( Figure 2). The PRI measured by the VASP-P method in the cilostazol group was 52.60 ± 22.58%, which was similar to that of the aspirin group (49.51 ± 23.76%, p = .507) (Table S2).

| DISCUSSION
Aspirin intolerance has become an increasing heightened risk with the emergence of rising coronary heart disease prevalence, especially in Southeast Asian populations. 24,25 In western countries, the prevalence of aspirin intolerance was estimated to range from 0.6% to 1.5%. However, a multicenter observational study in Japan involving 947 patients found that up to 30% of patients taking low-dose aspirin clinically showed aspirin intolerance, while another cohort study in China revealed 9.9% of aspirin-intolerant patients with CAD in one single center, mainly manifested as severe gastrointestinal reactions. 13,[25][26][27] The concept of "Aspirin intolerance" has not been universally acknowledged before. It is often mislabeled as "aspirin hypersensitivity" or "aspirin resistance". 28, 29 Dai defined "aspirin intolerance" as "any conditions that prevent patients from long-term use of low-dose aspirin", 13 while earlier researchers conflated it with aspirin-exacerbated respiratory disease. 28    gastritis, antral gastritis, and reflux esophagitis (33%), followed by high gastrointestinal bleeding events (27%). Remarkably, some patients were included in the cilostazol group with more than one intolerance factor, among which hyperuricemia was the most common comorbid factor (Table 3 and Figure S2). The mechanism of aspirin intolerance has not been well explained yet. Polymorphisms of enzymes like UDP-glycolaldehyde transferase, cytochrome P450, and heterologous/medium-chain fatty acid CoA ligase were thought to be the underlying mechanism of aspirin intolerance. 32,33 Cilostazol is a specific and strong inhibitor of PDE3 in platelets and smooth muscle cells, where it diminishes intracellular calcium.
In platelets, cilostazol inhibits both primary and secondary platelet aggregation induced by ADP, arachidonic acid, collagen, and  19,20,37 Besides, cilostazol was also found to have the ability to phosphorylate VASP through promoting PKA activation and further to inhibit platelet aggregation. 38

| CONCLUSION
In patients with aspirin intolerance undergoing percutaneous coronary stent implantation, cilostazol combined with P2Y 12 receptor inhibitors was not inferior to aspirin combined with P2Y 12 receptor inhibitors on MACCE and BARC bleeding events. Similar inhibition of PAC-1, CD62p, and VASP-P was also found in the two groups.
Cilostazol combined with P2Y 12 inhibitors could be a reasonable substitute antiplatelet regimen for aspirin-intolerant patients undergoing percutaneous coronary stent implantation, but again with limitations, which required a larger sample and longer follow-up to confirm its efficacy.

ACKNOWLEDGMENTS
This study was supported by a grant from the National Natural

CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.