Impact of IRS-1 rs956115 and CYP2C19 rs4244285 Genotypes on Clinical Outcome of Patients Undergoing PCI

Background insulin receptor substrate-1 (IRS-1) rs956115 is associated with vascular risk in patients with coronary artery disease (CAD) and concomitant diabetes. CYP2C19 rs4244285 modulates clopidogrel responsiveness and predicts outcome of CAD. We designed this study to explore the association between IRS-1 rs956115, CYP2C19 rs4244285, and platelet reactivity as well as 1-year outcome in patients with CAD undergoing percutaneous coronary intervention (PCI). Methods IRS-1 rs956115, CYP2C19 rs4244285 genotypes and platelet reactivity were assessed in 1611 post-PCI patients. Major adverse cardiovascular events (MACE) which were dened as a composite of cardiovascular death, myocardial infarction and ischemic stroke over 1-year were evaluated. One-way ANOVA was used to compare the platelet reactivity among different genotypes of rs956115 and rs4244285. Multivariable Cox proportional hazard model analysis was used to estimate the association between genotypes of rs956115 and rs4244285 and risk of MACE. Results At 1 month, patients with rs956115 CG genotype had signicantly lower level of residual ADP-induced platelet aggregation (PL ADP ) than those with CC genotype. PL ADP signicantly increased with the number of rs4244285 A alleles. Patients with rs956115 CG or GG genotype had a 2.09-fold higher risk of MACE than those with CC genotype (adjusted HR=2.09; 95%CI:1.04-4.19; P=0.0376), and those with rs4244285 GA genotype had a 2.19-fold higher risk than GG homozygotes (adjusted HR=2.19; 95%CI:1.13-4.24; P=0.0200). There was no signicant difference in risk between AA and GG homozygotes. No interaction between rs956115 and rs4244285 was observed. and genotypes associated with 2.09- and 2.19-fold cardiovascular risks respectively 1-year follow-up. The of to be independent of known clinical predictors, while that of rs4244285 by clopidogrel


Abstract
Background insulin receptor substrate-1 (IRS-1) rs956115 is associated with vascular risk in patients with coronary artery disease (CAD) and concomitant diabetes. CYP2C19 rs4244285 modulates clopidogrel responsiveness and predicts outcome of CAD. We designed this study to explore the association between IRS-1 rs956115, CYP2C19 rs4244285, and platelet reactivity as well as 1-year outcome in patients with CAD undergoing percutaneous coronary intervention (PCI).
Methods IRS-1 rs956115, CYP2C19 rs4244285 genotypes and platelet reactivity were assessed in 1611 post-PCI patients. Major adverse cardiovascular events (MACE) which were de ned as a composite of cardiovascular death, myocardial infarction and ischemic stroke over 1-year were evaluated. One-way ANOVA was used to compare the platelet reactivity among different genotypes of rs956115 and rs4244285. Multivariable Cox proportional hazard model analysis was used to estimate the association between genotypes of rs956115 and rs4244285 and risk of MACE.
Results At 1 month, patients with rs956115 CG genotype had signi cantly lower level of residual ADP-induced platelet aggregation (PL ADP ) than those with CC genotype. PL ADP signi cantly increased with the number of rs4244285 A alleles. Patients with rs956115 CG or GG genotype had a 2.09-fold higher risk of MACE than those with CC genotype (adjusted HR=2.09; 95%CI:1.04-4.19; P=0.0376), and those with rs4244285 GA genotype had a 2.19-fold higher risk than GG homozygotes (adjusted HR=2.19; 95%CI:1.13-4.24; P=0.0200). There was no signi cant difference in risk between AA and GG homozygotes. No interaction between rs956115 and rs4244285 was observed.
Conclusions In post-PCI patients, rs956115 GG/CG and rs4244285 GA genotypes were associated with 2.09-and 2.19-fold cardiovascular risks respectively at 1-year follow-up. The effect of rs956115 appeared to be independent of known clinical predictors, while that of rs4244285 GA could be mediated by lower clopidogrel response. Background Insulin receptor substrate-1 (IRS-1), a ligand of insulin receptor tyrosine kinase, plays a central role in insulin signal transduction system (1,2). Dysregulation of IRS-1 has been suggested as a common mechanism underlying insulin resistance which may lead to high platelet reactivity and low response to antiplatelet treatment in patients with type 2 diabetes mellitus (DM) (3,4).
CYP2C19 is one of the isoenzymes of hepatic cytochrome P450 (CYP450) system which plays a key role in the bioactivation of clopidogrel(5, 6). Carriers of CYP2C19 loss of function *2 (rs4244285) generate less amounts of active metabolite of clopidogrel than wild-type homozygotes, which subsequently resulting in a lower clopidogrel responsiveness and an increased risk of major adverse cardiac events in coronary artery disease (CAD) patients after percutaneous coronary intervention (PCI) (7)(8)(9).
This study examined the association between IRS-1 rs956115, CYP2C19 rs4244285 and platelet reactivity as well as major adverse cardiovascular events (MACE) in patients with CAD who had undergone PCI and were treated with aspirin and clopidogrel.

Methods
The data that support the ndings of this study are available from the corresponding author on reasonable request.

Study Design
This was a prospective single-center cohort study conducted in the First A liated Hospital of Nanjing Medical University, Nanjing, China. Complying with the Helsinki declarations and local regulations, the study was approved by the ethics committee of the First A liated Hospital of Nanjing Medical University.
Written informed consent was obtained from each patient.
The inclusion criteria were patients with CAD undergoing urgent or elective coronary stent implantation who were over 18 years old and planning to take dual antiplatelet treatment (DAPT) with clopidogrel 75 mg and aspirin 100 mg once daily for at least 1 year. Patients who met any of the following criteria were excluded: (1) allergic or intolerant to aspirin or clopidogrel; (2) at high risk of bleeding (e.g., platelet count <80×10 9 /L, known bleeding diathesis, active peptic ulcer, or with a history of cerebral hemorrhage within 1 year); and (3) planning to take drugs that could potentially interfere with the antiplatelet effects of aspirin (e.g., non-steroidal anti-in ammatory drugs) or clopidogrel (e.g., CYP3A inhibitors or CYP3A inducers). Baseline demographic and clinical characteristics, as well as medical and interventional treatments, were collected on a pre-speci ed case report form.

Sample Collection and Preparation
After receiving >5 days of aspirin and clopidogrel, blood was collected 2 hours post dosing (about 10am) from each patient into one 2-mL BD Vacutainer tube (Becton, Dickinson and Company, Franklin Lakes, NJ) containing 3.6 mg K2 EDTA and two 2-mL BD vacutainer tubes with 0.105 mol/L buffered sodium citrate (3.2%). Blood samples were transferred to the central laboratory within 1 hour after collection. Samples in EDTA tubes were frozen at −80°C for genotyping, whereas citrated samples were immediately processed for platelet aggregation studies. After centrifuging the citrated sample at 200g for 8 minutes at 22°C, platelet-rich plasma was carefully separated, and the remaining sample was centrifuged at 2465g for another 10 minutes to obtain plateletpoor plasma. The platelet count in platelet-rich plasma was standardized by addition of platelet-poor plasma to achieve a count of 250×10 9 /L. Platelet aggregation tests by light transmission aggregometry were performed within 3 hours of platelet-rich plasma preparation (10). At 1-month follow-up, patients received repeat blood collection for measurement of platelet reactivity as performed at baseline.

Platelet Reactivity Assay
Platelet aggregation testing was performed using a Chronolog Model 700 aggregometer (Chronolog Corporation, Havertown, PA). Immediately after preparation of platelet-rich plasma, 500 μL was transferred into each of the 2 test tubes, with 500 μL platelet-poor plasma as control. Platelet aggregation was induced using adenosine diphosphate (ADP) or arachidonic acid (AA) as agonists with nal concentrations of 5 umol/L and 1 mmol/L, respectively. The ADP and AA-induced platelet aggregations (PL ADP and PL AA , respectively) were recorded using the maximum platelet aggregation within 8 minutes.
Genotype Analysis IRS-1 (rs956115, C>G) and CYP2C19*2 (rs4244285, G>A) were genotyped using a custom-by-design improved multiplex ligation detection reaction technique (Genesky Biotechnologies Inc, Shanghai, China) based on the highly speci c double ligation and the multiplex uorescence polymerase chain reaction (11).
For quality control, repeated testing was performed randomly in 5% of samples with high DNA quality.

Clinical Follow-up
Patients were followed-up for 12 months by 2 investigators who were blinded to the results of platelet reactivity testing and genotyping. The patients were followed in the clinic or by telephone if they were unable to attend the clinic. The primary endpoint was de ned as the occurrence of MACE, a composite of cardiovascular death, myocardial infarction, and ischemic stroke within 12 months after PCI. The cardiovascular events in this study were de ned according to the American College of Cardiology 2001 (12).

Statistical Analysis
Continuous variables were described as mean ± standard deviation (SD) or median with interquartile range (IQR) when data did not follow a normal distribution, and differences between groups were analyzed by t test or nonparametric test. Categorized variables were expressed as numbers and percentages and were analyzed by χ 2 test or Fisher exact method. One-way ANOVA was used to compare the platelet reactivity among different genotypes of rs956115 and rs4244285. Multivariable Cox proportional hazard model analysis was used to estimate the association between genotypes of rs956115 and rs4244285 and risk of MACE reported as hazard ratio (HR) and 95% con dence intervals (CI). The model was adjusted for clinical covariables including age, previous myocardial infraction (MI), hypertension, diabetes mellitus, smoking status, previous PCI, left ventricular ejection fraction (LVEF), serum creatinine, low density lipoprotein, and diagnosis.
All analyses were performed using SAS, version 9.4 (SAS Institute, Cary, North Carolina) and gures were developed using R, version 3.2.0 (R Foundation for Statistical Computing, Vienna, Austria). A two-tailed P value of <0.05 was considered statistically signi cant.

Results
From Mar 2011 to September 2016, 2213 patients were consecutively screened, among whom 1614 patients who met the inclusion and the exclusion criteria were enrolled. Of the 1614 enrolled patients, 3 were not included in the nal analysis due to unsatisfactory blood sample quality. All the remaining patients completed the genotype assessment and 1-year clinical follow-up. Platelet aggregation testing were performed in 1175 patients at baseline and in 624 patients at 1-month follow-up ( Figure 1).

Patients' Characteristics
The baseline characteristics of patients included in this study are summarized in Table 1 Figure 2A). At 1-month follow-up, PL ADP was signi cantly different among the three genotypes (F=3.28, P=0.0381, Figure 2A). CG genotype was associated with a signi cantly lower PL ADP compared with CC genotype (P=0.0158, Figure 2A).
Regarding PL AA , there were no signi cant difference among the three genotypes of rs956115 either at baseline (F=2.73, P=0.0656, Figure S1A) or at 1-month follow-up (F=0.20, P=0.8180, Figure S1A). For rs4244285, PL ADP were signi cantly different among the three genotypes at baseline (F=53.27, P<0.001, Figure 2B) and 1-month follow-up (F=12.07, P<0.001, Figure 2B). By pairwise comparisons, the platelet reactivities corresponding to different genotypes of rs4244285 were all signi cantly different except the comparison between GA and AA at 1-month follow-up (P=0.4392, Figure 2B). As shown in Figure 2B, the platelet reactivity increased with the number of the A alleles of rs4244285. Regarding PL AA , there were no signi cant difference among the three genotypes of rs4244285 either at baseline (F=0.38, P=0.6870, Figure S1B) or at 1-month follow-up (F=0.78, P=0.4590, Figure S1B).   Figure 3B). No signi cant difference of MACE risk was found while using either dominant (P=0.0666; Table 2) or additive model (P=0.4936; Table   2).

Interaction Analysis
Among patients with GG genotype of rs4244285, those who had CG or GG genotype of rs956115 presented a 4.85-fold higher MACE risk than those who had CC genotype (adjusted HR=4.85, P=0.0081; Figure 4). By comparison, among patients with non-GG genotype of rs4244285, those who had CG or GG genotype of rs956115 presented a 1.40-fold higher risk than those who had CC genotype (adjusted HR=1.40, P=0.4764; Figure 4). The interaction between rs956115 and rs4244285 was none-statistically signi cant (P=0.1453; Figure 4).
between CG or GG and CC genotypes of rs956115 did not reach statistically signi cant in the diabetes subgroup ( Figure 5), the dominant model HR of MACE for patients with CG or GG genotype of rs956115 tended to be similar across subgroups. No signi cant interactions were observed in any of those subgroups except LVEF subgroup (Interaction P=0.0006) ( Figure 5).

Discussion
This study examined the impacts of IRS-1 rs956115 and CYP2C19 rs4244285 polymorphisms on clinical outcome of patients undergoing PCI and receiving DAPT treatment and found that G allele carriers of IRS-1 rs956115 had a 2.09-fold higher risk of MACE compared with non-carriers at 1-year follow-up. The rs4244285 GA genotype had a 2.19-fold higher risk than GG homozygotes. The effect of rs956115 was independent to known clinical covariables, while that of rs4244285 GA could be mediated by lower clopidogrel response.
Angiolillo et al. examined 7 single nucleotide polymorphisms (SNPs) of IRS-1 and found that rs956115 polymorphism was associated with a hyperreactive platelet phenotype and adverse cardiovascular outcomes in type-2 DM Caucasian patients concomitant with coronary artery disease (CAD) (13). However, uncertainty remains about the effects of IRS-1 rs956115 polymorphism on platelet function and cardiovascular outcome in non-selective CAD patients.
In this study, we found that rs956115 G allele was an independent prognostic factor of adverse cardiovascular outcomes in non-selective CAD patients, irrespective of CYP2C19*2 (rs4244285) polymorphism, diabetes mellitus and other known risk factors. Although rs956115 G allele didn't show a signi cant correlation with MACE in the subgroup of diabetes mellitus, our results showed the consistent tendency of almost all subgroups as shown in Figure 5.
Regarding the underlying mechanism, Angiolillo et al suggested that rs956115 polymorphism was associated with a hyperactive platelet phenotype in Caucasian type-2 DM patients (13). However, in a later study by Zhang et al, no association was observed between rs956115 polymorphism and platelet function pro le (14). Our results were in consistent with that of Zhang et al in a larger Chinese population, showing no signi cant difference in AA or ADPinduced platelet aggregation at baseline among different IRS-1 rs956115 genotypes. Moreover, ADP-induced platelet aggregation was even lower in rs956115 CG genotype compared with CC genotype at 1-month follow-up. Along with the results of Zhang's study, we suggest that the association between IRS-1 rs956115 polymorphisms and risk of MACE cannot be explained by impaired platelet reactivity to either clopidogrel or aspirin.
Theoretically, IRS-1 is one of the central nodes in insulin signaling network (15). It has been reported that IRS-1 is necessary for insulin-stimulated activation of phosphatidylinositol 3 kinase (PI3K)/AKT pathway and subsequent enhanced production of nitric oxide (NO) in endothelial cells (16), which plays a critical role in maintaining cardiovascular homeostasis (17). Previous studies have demonstrated that functional variants of IRS-1 directly impaired insulin regulated NO synthesis in cultured human endothelial cells (18,19). Considering the pivotal role of IRS-1 in PI3K/AKT signaling pathway of insulin, it may be reasonable to assume IRS-1 rs956115 polymorphism affects the same process or an unknown pathway and consequently impacts the clinical outcome of CAD patients.
Our results were consistent with previously reports and further con rmed that CYP2C19*2 (rs4244285) loss of function polymorphism is a strong predictor of impaired clopidogrel response and adverse clinical outcomes (7)(8)(9). This consistency, in turn, enhances the credibility of our results on rs956115. Meanwhile, from the interaction analysis, we did not nd a statistically signi cant interaction between IRS-1 rs956115 and CYP2C19 rs4244285 polymorphism, which proved IRS-1 rs956115 G allele to be an independent risk factor of MACE in CAD patients after PCI.

Conclusions
IRS-1 rs956115 G allele signi cantly increased the cardiovascular risk of post-PCI patients by 2.09-fold at 1-year follow-up, which was independent to CYP2C19 rs4244285 genotypes, pharmacological platelet response and known clinical covariables.

Declarations
Ethics approval and consent to participate Complying with the Helsinki declarations and local regulations, the study was approved by the ethics committee of the First A liated Hospital of Nanjing Medical University. Written informed consent was obtained from each patient.

Consent for publication
As stated above, informed consent on participation and publication was obtained from all participants.

Availability of data and materials
The datasets used during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.