Smoking and outcomes following personalized antiplatelet therapy in chronic coronary syndrome patients: A substudy from the randomized PATH‐PCI trial

Abstract Background This is a sub‐analysis of the Personalized Antithrombotic Therapy for Coronary Heart Disease after PCI (PATH‐PCI) trial in China to explore the relationship between smoking and outcomes following personalized antiplatelet therapy (PAT) in chronic coronary syndrome (CCS) patients undergoing percutaneous coronary intervention (PCI). Methods As a single‐center, prospective, randomized controlled and open‐label trial, the PATH‐PCI trial randomized CCS patients undergoing PCI into standard group or personalized group guided by a novel platelet function test (PFT), from December 2016 to February 2018. All patients were divided into smokers and nonsmokers according to their smoking status. Subsequently, we underwent a 180‐day follow‐up evaluation. The primary endpoint was the net adverse clinical events (NACE). Results Regardless of smoking status, in the incidence of NACE, there was a reduction with PAT but that the reductions are not statistically significant. In the incidence of bleeding events, we found no statistically significant difference between two groups (smokers: 2.0% vs. 1.4%, HR = 1.455, 95% confidence interval [CI]: 0.595−3.559, p = .412; nonsmokers: 2.2% vs. 1.8%, HR = 1.228, 95% CI: 0.530−2.842, p = .632). In smokers, PAT reduced major adverse cardiac and cerebrovascular events (MACCE) by 48.7% (3.0% vs. 5.9%, HR = 0.513, 95% CI: 0.290−0.908, p = .022), compared with standard antiplatelet therapy (SAT). PAT also reduced the major adverse cardiovascular events (MACE) but there was no statistically difference in the reductions (p > .05). In nonsmokers, PAT reduced MACCE and MACE by 51.5% (3.3% vs. 6.7%, HR = 0.485, 95% CI: 0.277−0.849, p = .011) and 63.5% (1.8% vs. 4.9%, HR = 0.365, 95% CI: 0.178−0.752, p = .006), respectively. When testing p‐values for interaction, we found there was no significant interaction of smoking status with treatment effects of PAT (p int‐NACE = .184, p int‐bleeding = .660). Conclusion Regardless of smoking, PAT reduced the MACE and MACCE, with no significant difference in bleeding. This suggests that PAT was an recommendable regimen to CCS patients after PCI, taking into consideration both ischemic and bleeding risk.


| INTRODUCTION
Percutaneous coronary intervention (PCI) effectively improves symptoms and quality of life in patients with coronary artery disease (CAD) and may reduce cardiovascular events in a subset with higher risk CAD. 1,25][6] The results of a meta-analysis of data from 20 743 patients showed that the guided antiplatelet strategy reduced the risk of major adverse cardiovascular events (MACE) more than standard antiplatelet therapy (SAT), including reducing the incidence of cardiac death, recurrent infarction, stroke, and in-stent thrombosis. 7oking is a well-recognized and preventable complex cardiovascular risk factor.In 1996, a study by Waters et al. suggested that smoking increased the risk of CAD and morbidity and mortality in CAD patients. 83][14][15] Recently, Ramtowski et al evluated the effect of smoking and smoking cessation on platelet reactivity and aggregation with Verify Now and several other measures of platelet function.They finally found an increase in platelet reactivity and the risk of thrombotic complications in CAD patients following PCI after smoking cessation. 16However, no one has yet to look specifically at smokers and genetic predisposition for clopidogrel high in-treatment platelet reactivity.We investigated this issue through sub-analysis of the Personalized Antithrombotic Therapy for Coronary Heart Disease after PCI (PATH-PCI) study in China, to examine the effect of smoking on the safety and efficacy of personalized antiplatelet therapy (PAT) in chronic coronary syndrome (CCS) patients after PCI.All patients aged ≥ 18 who were confirmed to have CAD were enrolled.CCS was described as a records of coronary revascularization or acute coronary syndrome (ACS)＞3 months in the past or a medical diagnosis of CAD with angiographically documented coronary artery stenosis of at least 70% diameter narrowing or a left main stenosis increased than 50% amenable to PCI as American Heart Association category.The trial's detailed methodology, inclusion and exclusion criteria, and main results were published earlier. 17In this exploratory sub-analysis, we probed the effect of smoking situation on the safety and efficacy of PAT in study groups.

| Randomization and study groups
We randomized all patients before they underwent PCI in this trial into two study groups: either (I) personalized group guided by PFT (n = 1146) or (II) standard group (n = 1139).The personalized group received PFT to evaluate the maximum aggregation rate (MAR) and took ticagrelor (90 mg twice daily) if the platelet maximum aggregation rate (MAR) > 55%, or 75 mg/day clopidogrel if the MAR ≤ 55%, as described previously. 17Without testing MAR, the standard group patients took the SAT for 6 months after PCI.The SAT included a clopidogrel (75-mg/day) maintenance dosage plus an aspirin (100 mg/day) The other therapies, including angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, b-blockers, statins, and stents, were chosen by experienced physicians.We divided all patients into two subgroups for analysis according to their smoking status: smokers (S-PAT, n = 597; S-SAT, n = 573) and nonsmokers (NS-PAT, n = 549; NS-SAT, n = 566).For smoking, we adopted the World Health Organization (WHO) definition.Smokers were defined as those who had been smoking until admission and those who had quit within 6 months before admission (Current Smoking status).Nonsmokers included patients who stopped smoking at least 6 months earlier than inclusion in the trial.

| Study endpoints
The primary endpoint was the net adverse clinical events (NACE), which were combination of ischemic events (cardiac death, MI, stroke, stent thrombosis, urgent revascularization) or bleeding events (Bleeding Academic Research Consortium [BARC] class ≥ 2 bleeding events, as BARC criteria defined 20 ).The secondary endpoint was the individual components of the primary endpoint.All-cause death was defined as an unequivocal, noncardiac-cause death, as previously described. 17Cardiac death, all-cause death re-infarction, stent thrombosis and target vessel revascularization (TVR) were regarded as MACE. 14Major adverse cardiac and cerebrovascular events (MACCE), described as MACE, plus stroke.We followed up with patients at baseline, 1 and 6 months after recruitment through message, telephone contact or office visits.

| Statistical analyses
We used SPSS version 22.0 (SPSS Inc.) when analyzing data.
Continuous variables were performed as mean ± standard deviation (X ± S), and we compared continuous variables using t-tests with normal distribution or Mann−Whitney U-tests with nonnormal distribution.Categorical variables were expressed as percentages and compared by the chi-square test or Fisher exact test.We used Kaplan-Meier survival analysis and log-rank tests to analyze and assess the relationship between treatment and outcomes.With the adjustment of the confounders, such as sex, age, diabetes, hypertension and LDL-C, multivariate Cox regression for interaction testing was performed to evaluate the difference of outcomes.All tests were two-sided, and differences were statistically significant if p < .05.

| Study population
We enrolled 2285 CCS patients in the PATH-PCI study in China.We then randomly assigned 1146 patients to personalized group and 1139 patients to standard group.Patients were then subdivided into two subgroups based on smoking status (Figure 1).Table 1 showed the baseline characteristics of patients and the baseline demographics were essentially balanced.One thousand one hundrend seventy patients (51.2%) were smokers and 1904 patients (83.3%) were male.Smokers were significantly older, and more often had a diabetes and hypertension presentation.For nonsmokers, there was no difference, except in a relatively high percentage of the male study population.The angiographic of the study groups were displayed in Table S1.They were equally distributed between smokers and nonsmokers, except for stent expansion pressure, which was higher in personalized group than standard group.Stent use was more frequent in the NS-PAT cohort, but the difference was not statistically significant due to a Type 2 error (N too small).

| Clinical outcomes
Focusing on the efficacy, for the incidence of NACE, we found a reduction with PAT but that the reductions are not statistically significant (smokers, 4.7% vs. 6.8%,p = .142;nonsmokers, 5.5% vs. 8.1%, p = .085)(Figure 2 and Table 2).With regard to safety, for major bleeding, there was no difference between the groups in smokers (2.0% vs. 1.4%, p = .412)or nonsmokers (2.2% vs. 1.8%, p = .632)(Figure 2 and Table 2).Interaction tests shown that there was no significant interaction of smoking status with treatment effects of PAT (p int-NACE = 0.184, p int-bleeding = 0.660).

| Smokers
As shown in Figure S1 and Table 2, among smokers, the rates of MACCE were significantly different between S-SAT and S-PAT (3.0% vs. 5.9%, HR = 0.513, 95% CI, 0.290−0.908,p = .022).The difference in the incidence of other secondary endpoints between the groups was not statistically significant (Figure S1, S2 and Table 2; all p > .05).
T A B L E 1 Baseline characteristics of the study cohort in smokers and nonsmokers.It was worth noting that the incidence of stroke was decreased in personalized group compared to standard group (0.3% vs. 1.6%,HR = 0.217, 95% CI, 0.047−1.004,p = .051).

| Nonsmokers
As shown in Figure S1 and Table 2, among nonsmokers, for MACCE, the personalized group was lower than standard group (3.3% vs. 6.7%,HR = 0.485, 95% CI, 0.277−0.849,p = .011).In addition, we found a lower incidence of MACE in personalized group, compared with standard group (1.8% vs. 4.9%, HR = 0.365, 95% CI, 0.178−0.752,p = .006).However, we did not find significant difference for other secondary endpoints (Figure S1, S2 and Table 2; all p > .05).Notably, it showed a trend for a lower rate of TVR in personalized group (0.2% vs. 1.4%,HR = 0.128, 95% CI, 0.016−1.020,p = .052).These results were similar to the previously published in the primary analysis of this clinical trial. 17d the effectiveness of PAT was furthermore emphasized via our analyses.Cox proportional hazards model results suggested PAT was a protective factor for MACCE in CCS patients after PCI, regardless of whether they smoked (Table 3).

| DISCUSSION
In this study, we set out to estimate the effect of smoking on the efficacy and safety of PAT in Chinese CCS patients after PCI.While there was a significant reduction in secondary endpoints of MACE and MACCE, and no significant difference in bleeding, the primary endpoint of NACE was not significant meaning that the nonsignificant increase in bleeding offset the significant reduction in MACE.Additionally, PAT may be an independent protective factor against postoperative MACCE.PAT was an recommendable regimen to CCS patients after PCI, taking into consideration both ischemic and bleeding risk.
Smoking has been identified as a significant contributing factor to cardiovascular morbidity and mortality.It affects vascular function and hemodynamics and is linked to platelet activation. 21Compared with nonsmokers, smokers seemed to have a higher short-term mortality rate after PCI. 22Based on the evidence of an observation study, the incidence of MACCE in patients after PCI was increased 1.060 times by smoking. 23Smoking increased the production of clopidogrel active metabolites by inducing CYP1A2 activity. 24This effect was more notable in high-risk patients, especially those with ACS, a circumstance defined as the "smoking paradox." 25 The PARADOX study suggested that nonsmokers who received clopidogrel had a weaker antiplatelet effect than smokers, nonsmokers saw little or no benefit from clopidogrel treatment, especially in ACS patients. 26Sia et al. suggested the seemingly benefits of smoking disappeared upon adjustment. 27The GRAVITAS trial proposed that the more effective antiplatelet therapy may eliminate this paradox. 28,29However, East Asian Paradox proposed that more potent P2Y12 receptor inhibitors may significantly increase the risk of severe bleeding, but do not reduce the incidence of ischemic events after PCI, as thrombogenicity may be related to race. 30,31Recent evidence suggests the opportunity for a personalized long-term antithrombotic monotherapy with aspirin or P2Y12 receptor inhibitor according to the patient's characteristics. 32These findings emphasized that clinicians should carefully consider the benefits and risks of antiplatelet drugs in different smoking situation.
Ticagrelor or prasugrel can be used when clopidogrel is ineffective or increases the risk of ischemia.And the use of ticagrelor is gradually increasing. 33In the multicenter, randomized, double-blind study, named ONSET/OFFSET, compared with clopidogrel, ticagrelor was identified as a faster and more powerful P2Y12 receptor inhibitor for patients with CCS. 34The PLATO trial proposed that ticagrelor reduced the ischemic events without increasing major bleeding compared with clopidogrel in patients with ACS, regardless of smoking status. 35Later studies were again confirmed similar results in CCS patients. 36In the current post hoc analysis, if MAR ≥ 55%, patients received ticagrelor 90 mg twice daily plus aspirin 100 mg once daily.
Regardless of smoking status, we found there was a reduction in the incidence of NACE for patients with PAT, but that the reductions are not statistically significant.And we found no statistically significant difference between two groups in the incidence of bleeding events.Among smokers, the rate of MACCE were significantly lower in PAT group than SAT group without a significant increase in bleeding events.In addition, among nonsmokers, the occurrences of MACCE and MACE were reduced with no increase in bleeding in PAT group.However, a post-hoc analysis from the TROPICAL-ACS trial suggested that DAPT stepdown therapy was equally safe and effective regardless of whether the patient smoked or not. 37When testing p-values for interaction, we also did not find a significant interaction of smoking status with treatment effect of PAT and SAT.Interestingly, our cox proportional hazards model results suggested PAT was a protective factor for MACCE in CCS patients after PCI.
Therefore, this post hoc analysis identified that although smoking influenced the vascular function and hemodynamics, it could not translate into a significant effect on the worse outcomes.These findings underscore the message about the effectiveness and safety of PAT, and revealed the harmfulness of smoking to the public and the need to quit smoking.Based on the real world, our study provided essential insights into the efficacy and safety of PAT guided by PFT in CCS patients after PCI.

| LIMITATIONS
Some potential limitations need to be considered.First, the PATH-PCI study is an open-label, single-center study in a single ethnic group that was predominantly male, the results may not be (and probably are not) generalizable to the other populations.It is necessary to conduct multicenter and large-sample clinical trials to further investigate the efficacy and safety of personalized therapy based on platelet function.Second, we performed a subgroup analysis according to smoking status, which is a post hoc analysis with all its inherent limitations.Third, the smoking status in this study was self-reported by patients, and there was no biochemical test involving cigarette components.Finally, we measured platelet function in patients with the PL-12 platelet function analyzer only in the initial state, did not retest during subsequent follow-ups, and did not study the pharmacokinetics of clopidogrel or ticagrelor.

| CONCLUSION
To sum up, smoking is a modifiable risk factor.Regardless of smoking, PAT reduced the MACE and MACCE, with no significant difference in bleeding.This suggests that PAT was an recommendable regimen to CCS patients after PCI, taking into consideration both ischemic and bleeding risk.

2 | METHODS 2 . 1 |
Study patients The Personalized Antithrombotic Therapy for Coronary Heart Disease after PCI (PATH-PCI) in CCS patients undergoing PCI (ChiCTR-INR-16010077) a randomized and open-label trial, was implemented at the First Affiliated Hospital of Xinjiang Medical University.The Institutional Ethics Committee of the First Affiliated Hospital of Xinjiang Medical University had approved the protocol of this study.All patients gave written informed consent before being included in the study.

Ying
Pan and Ting-Ting Wu drafted the manuscript, designed the study and provided methodological expertize.Tuo Yan, Chang-Jiang Deng, Yi Yang, Shun Wang, and Xian-Geng Hou drafted the tables and figures and performed the statistical analysis.Xiang Xie and Ying-Ying Zheng designed the study and made critical revision of the manuscript.All authors contributed to the acquisition of data and final approval of the version to be published.All authors agreed to be accountable for all aspects of the work.
Primary and secondary endpoints of following up for 180 days.Multivariable Cox regression analysis of MACCE according to smoking status.
Cumulative Kaplan−Meier estimates of the time to the first adjudicated occurrence of NACE and bleeding in smokers and nonsmokers.NACE, net adverse clinical events.T A B L E 2Abbreviations: MACCE, major adverse cardiovascular and cerebrovascular events: cardiac death, MI, ST, stroke, or urgent revascularization; MACE, major adverse cardiovascular events: cardiac death, MI, ST, or urgent revascularization; MI, myocardial infarction; NACE, net adverse clinical events; ST, stent thrombosis; TVR, target vessel revascularization.T A B L E 3