Impact of Total Protein Level and Dysglycaemia on the Ecacy of Clopidogrel in Patients Underwent Carotid Artery Stenosis with CYP2C19 Genetic Variants

BACKGROUND: CYP2C19polymorphisms are associated with the increased risk of major adverse cardiovascular/cerebrovascular events (MACCEs) in cerebral intervention. In this study, we wanted to investigate whether the CYP2C19 polymorphism and other nongenetic factors can inuence the incidence of MACCEs in cerebral artery stenosis disease patients. METHODS: A total of 164 patients underwent cerebral artery stenting and 138 patients underwent conservative treatment among 1406 who patients underwent CYP2C19 gene screening and were enrolled in this study. A Cox proportional hazards model and Kaplan–Meier analyses were used to assess the predictive value of CYP2C19 loss-of-function (LOF) allele (*2, *3) carrier status and other risk factors. RESULTS: The CYP2C19 *1/*1 genotype was observed to be the most predominant among the patients (41.96%). The patients who underwent conservative treatment and had glucose levels> 6.5 mmol/L were more likely to experience MACCEs (p = 0.022). CYP2C19 LOF allele variants (p = 0.032), total protein < 65 g/L (p = 0.017) and glucose > 6.5 mmol/L (p = 0.028) were associated with an increased risk of MACCEs in patients who underwent cerebral artery stents in the multivariable Cox analysis. CONCLUSION: CYP2C19 polymorphisms, total protein levels and glucose can impact the risk of MACCEs in patients who undergo cerebral artery stents. the tests stents. we that allele in the cerebral artery stent group. These could be driven by the treatment patients, the patient's is so poor that theCYP2C19 gene does not


Introduction
Cerebral artery stenosis is an important risk factor for ischaemic stroke 1 and can be divided into combined intracranial and extracranial arterial stenosis, intracranial arterial stenosis, and extracranial arterial stenosis 2 .The treatment of cerebral artery stenosis, including conservative treatment, carotid endarterectomy (CEA) and carotid artery stenting (CAS),is most commonly used in cerebral artery stenosis treatment.Meanwhile, dual antiplatelet therapy (DAPT) with aspirin and clopidogrel has been recommended for cerebral artery stenosis to reduce thromboembolic events 3 .
Clopidogrel, which is an orally administered prodrug, can be converted into an active metabolite by hepatic cytochrome P450 (CYP) 4 , and CYP2C19 is an important metabolizing enzyme.Among the 25 genetic variants of CYP2C19, the wild-type CYP2C19*1 allele is associated with functional CYP2C19-mediated metabolism, while CYP2C19*2 (c.681G > A; rs4244285) and *3(c.636G > A; rs4986893) alleles, which are the major CYP2C19 loss-of-function (LOF) variant alleles, can reduce the level of clopidogrel active metabolites in the blood and induce a decrease in clopidogrel function, resulting in inhibition of platelet aggregation [4][5][6][7] .
It has been reported that CYP2C19 polymorphism is involved in major adverse cardiovascular/cerebrovascular events (MACCEs) of cerebral intervention 8 .The CYP2C19 polymorphism in uencing the incidence of MACEs has been widely studied in coronary heart disease patients. It has been reported that patients with CYP2C19 LOF variant alleles who have undergone percutaneous coronary intervention (PCI) are more likely to experience MACCEs 9 , and 5-30% of PCI patients cannot respond to clopidogrel 10,11 . However, whether the CYP2C19 polymorphism can in uence the incidence of MACCEsin moderate-severe cerebral artery stenosis disease patients remains unknown.Additionally, many other factors can in uence the incidence of MACCEs after cerebral artery stenosis, including patient clinical characteristics such as dysglycaemia 12 and eGFR 13 .
In this study, to assess the risk predictors of the risk of MACCEs after cerebral artery stenosis, we analysed the potential predictive value of risk predictors such as CYP2C19 polymorphisms, liver function, kidney function and blood glucose level in the conservative treatment group and cerebral artery stent group.

Ethics statement
The Ethics Committee of First Hospital of Jilin University approved this project. All of the samples and data were collected after written informed consent was provided by the participants.The management and publicationof patient information in this research was strictly in accordance with the Declaration of Helsinki,including con dentiality and anonymity.

Study population
From October 2016 to December 2019, 1406 patients underwentCYP2C19 genotype screening at the genetic diagnosis centre of First Hospital of Jilin University. A total of 362 of them were diagnosed with cerebral artery stenosis in the neurosurgery department of our hospital. As shown in Fig. 1,302 patients were nally recruited in the study on the basis of the inclusion and exclusion criteria.
The inclusion criteria were as follows: 1) Patients were diagnosed with cerebral artery stenosis and underwent conservative treatment or cerebral artery stent; 2) patients received DAPT of clopidogrel (75 mg) and aspirin (100 mg) for at least 6 months.
According to the CYP2C19 loss-of-function allele (LOF) gene polymorphism, patients were divided into 2 groups according to their CYP2C19 genotype: a noncarrier group and a carrier group.Patients with wild-type CYP2C19 (no mutations) (*1/*1) were assigned to the noncarrier group.Patients with CYP2C19*2 or *3 were assigned to the carrier group.
MACCEs includedeath, stroke or stent thrombosis. All cerebral artery stenosis patients were diagnosed with moderate-severe cerebral artery stenosis disease. On account of status reasons, such as age factors, fragility of blood vessels and cardiac insu ciency, 143 patients could not tolerate the carotid intervention operation only orally with DAPT.
Aftercerebral artery stent operation, patients received 75 mg clopidogrel and 100 mg aspirin as dual antiplanet therapy for at least 6 months. Patients were followed up at 6 months. The CYP2C19 genotyping results and relevant clinical information were all recorded, including age, sex, liver function tests, kidneyfunction tests and other laboratory test values available in the database. All laboratory tests were performed at the First Hospital of Jilin University before treatment. Baseline variables are summarized with the use of descriptive statistics.

Genetic Analysis
We used a DNA array to screen the CYP2C19 genotypes. DNA was extracted from whole-blood samples with nucleic acid extracting reagent (BaiO Technology, Shanghai, China). The PCR program consisted of an initial step at 50 ℃ for 5 min, 94 ℃ for 5 min, 35 cycles at 94 ℃ for 25 s, 48 ℃ for 40 s and 72 ℃ for 30 s, and a nal extension at 72 ℃ for 5 min. We obtained images of the hybridization of the ampli cationproducts with the gene probes. The images and data were analysed by BaiO Array Doctor Version 2.0(BaiO Technology, Shanghai, China) software.

Statistical analysis
Data analysis was performed using SPSS 20.0 software (IBM, Armonk, NY, USA) and GraphPad Prism 5. Measurement data aredisplayed as the mean ± standard deviation, and categorical variables are expressed as the frequency and percentage. The distribution of categorical variables was compared among the study group using the chi-square test. Differences were considered statistically signi cant at p < 0.05.Kaplan-Meier analyses were used to generate survival plots of time to MACCEs during the 6-month follow-up period, and groups were compared by the log-rank test. Differences between the conservative treatment group and cerebral artery stent group in terms of the rate of MACCEs during the 6-month follow-up were assessed by a Cox proportional hazards regression model. Multivariable Cox regression analyses were adjusted by age, sex, AST,creatinine and BUN as a random effect. The interaction of CYP2C19 LOF allele carrierstatus with risk factors such as ALT, total protein and glucose was analysed by the above crude and multivariableCox models. Two-sided P < 0.05 was considered statistically signi cant.

Results
As shown in Fig. 1, the detailed patient selection process was described.From October 2016 to December 2019, 1406 patients underwent CYP2C19 genotype screening at the genetic diagnosis centre of First Hospital of Jilin University. A total of 362 of them were diagnosed with cerebral artery stenosis in the neurosurgery department of our hospital. 143 patients only orally took DAPT. Of these, 5 patients could not be followed up and were excluded. Therefore, 138 patients who took DAPT only orally for 6 months were nally recruited as the conservative treatment group. In addition, among the 219 patients who underwent cerebral intervention, 11 patients underwent carotid endarterectomy, 5 patients died in the hospital, 18 patients did not receive 75 mg clopidogrel for economic reasons and adverse effects, and 11 patients could not be followed up. Thus, a total of 55 patients were excluded from this study. Finally, 164 patients who underwent cerebral artery stenting and orally took DAPT for 6 months were recruited as the cerebral artery stent group.

Patient Baseline Characteristic
The baseline characteristics among individuals who orally took DAPT and underwent cerebral artery stenting by experiencing MACCEs are shown in Table 1.Patient demographics, medical history, CYP2C19 genotyping results and laboratory values were included in the analysis. In the conservative treatment group and cerebral artery stent group, there were no signi cant differences between the patients with and without MACCEs in age, sex, hypertension, diabetes mellitus, smoking or drinking. In the conservative treatment group, the glucose of the patients who experienced MACCEs was signi cantly higher than that of the non-MACCE patients (p = 0.007). There was no signi cant difference in the frequency of CYP2C19 LOF alleles (p = 0.536) in the conservative treatment group regardless of whether the patients experienced MACCEs. In addition, in the cerebral artery stent group, the frequency of CYP2C19 LOF alleles (p = 0.049) and the ALT (p = 0.019) of the patients who experienced MACCEs was signi cantly higher than that of the non-MACCE patients, and the total protein was signi cantly lower (p = 0.002). To identify the in uencing factors of the clinical endpoint after cerebral artery stenosis, we performed logrank tests.As shown in Table 2, in the conservative treatment group, we found that the patients whose glucose level was higher than 6.5 mmol/L were more likely to experience MACCEs (p = 0.022). In the cerebral artery stent group, we found that patients carrying CYP2C19 LOF alleles (*2 and *3) were more likely to experience MACCEsthan noncarriers (p = 0.020). We also found that the rates of MACCEs were signi cantly higher in patients whose ALT was higher than 35 U/L (p = 0.022),total proteinwas lower than 65 g/L (p = 0.020) or glucose was higher than 6.5 mmol/L (p = 0.019). Dysglycaemia may in uence the predictive value of CYP2C19 loss-of-function alleles in patients who underwent cerebral artery stents To further investigate how the in uencing factors impact the predictive value of CYP2C19 LOF alleles, we generated Kaplan-Meier curves to assess the association of CYP2C19 LOF alleles and the incidence of MACCEs strati ed by patient status. We found that the CYP2C19 LOF alleles were identi ed as a predictor of MACCEs at 6 months after carotid artery stenting only when the patients had glucose levels > 6.5 mmol/L (p = 0.0260) (Fig. 3).
CYP2C19 LOF allele variant, total protein level and dysglycaemiasynergically impact the risk of MACCEs in patients who underwent cerebral artery stents To further investigate the relationship between CYP2C19 LOF allele carriers and patient status with MACCE risk, the Cox proportional hazards regression model was used to assess the association.
As shown in Table 3, the univariable analysis revealed that glucose > 6.5 mmol/L (p = 0.033) was signi cantly associated with MACCEs at 6 months in the conservative treatment group.Similarly, onlyglucose > 6.5 mmol/L (p = 0.010) remained signi cant in the multivariable Cox analysis.Nevertheless, the univariable analysis revealed that CYP2C19 genotypes (p = 0.037), ALT > 35 U/L (p = 0.033), total protein < 65 g/L (p = 0.030) and glucose > 6.5 mmol/L (p = 0.028) were signi cantly associated with MACCEs at 6 months inthe cerebral artery stent group (Table 4). We found that CYP2C19 genotypes (p = 0.032),total protein < 65 g/L (p = 0.017) and (p = 0.028) remained signi cant in the multivariable Cox analysis (Table  4).Collectively, these results indicated that CYP2C19 LOF allele variants, total protein levels and dysglycaemiasynergically increase the risk of MACCEs.  The rate of MACCEs in the conservative treatment group was 14.49%, and that in the cerebral artery stent groupwas 9.14% (Table S1). A 403-patient study showed that MACCEs were observed in 8.19% of patients after treatment with CAS 17 , which is in line with our study. Considering the limited number of patients, the rate of MACCEs in this studymay not re ect the overall situation. This could also be driven by the patients who underwent conservative treatment, and carotid artery stents were almost all moderate-severe cerebral artery stenosis disease patients. Because conservative treatment patients whose own condition is not good enough to undergo surgery, they cannot tolerate carotid intervention operations, such as age factors, fragility of blood vessels and cardiac insu ciency, only orally. In this way, the rate of MACCEs in the conservative treatment group was higher than that in the cerebral artery stent group.
As a commonly prescribed antiplatelet, clopidogrel is usually used to prevent secondary ischaemia in patients treated by endovascular techniques.Composite MACCE outcomes, such as death, stroke, and stent thrombosis, limit the long-term success rate of interventions through the recurrence of symptoms. Although intervention has been associated with minimal complications, previous studies have reported that CYP2C19 loss-of-function polymorphisms may be a signi cant risk factor for in-stent restenosis 18 . Clopidogrel plus aspirin is used to reduce the risk of recurrent stroke. As one of the most prominent genetic polymorphisms, CYP2C19 polymorphisms can be used to explain a poor response to clopidogrel, and CYP2C19*2 is the strongest predictor of high residual platelet reactivity 19 .However, few studies have reported the signi cance of CYP2C19polymorphismsin affecting the incidence of MACCEs aftercarotid artery stenosis, especially carotid artery stents.
It has been reported that patients with CYP2C19 LOF variant alleles have a 3.58 times higher risk for death and stroke than patients with the CYP2C19 wild-type genotype 20 . A recent systematic review also demonstrated that CYP2C19 loss-of-function alleles were associated with clinical outcomes for ischaemic stroke 21 .In addition, a previous study reported that CYP2C19*2 was an independent risk factor for the primary outcomes of clopidogrel treatment in patients with acute ischaemic stroke 22 . In our study, the incidence of MACCEs in CYP2C19 LOF allele noncarriers (n = 343) was 7.14% (9/126), and in carriers, it was 14.69% (26/177). Patients with CYP2C19 LOF variant alleles had a 2.05 times higher risk of MACCEs than patients with the CYP2C19 wild-type genotype. Table 2 shows that there was no signi cant difference in the frequency of CYP2C19 LOF alleles in the conservative treatment group regardless of whether the patients experienced MACCEs. The frequency of CYP2C19 LOF alleles of the patients who experienced MACCEs was signi cantly higher than that of the non-MACCE patients. In the log-rank tests and multivariable Cox analysis, we found that CYP2C19 LOF allele variants were associated with an increased risk of MACCEs in patients who underwent cerebral artery stents. In this way, we infer that CYP2C19 LOF allele carriers are related to MACCE incidence in the cerebral artery stent group. These results could be driven by the fact that within conservative treatment patients, the patient's condition is so poor that theCYP2C19 gene does not work.
In addition, the baseline characteristics showed that in the conservative treatment group, the glucose of the patients who experienced MACCEs was signi cantly higher. In addition, in the cerebral artery stent group, the frequency of CYP2C19 LOF alleles and the ALT of the patients who experienced MACCEs was signi cantly higher, and the total protein was signi cantly lower. To identify the in uencing factors of the clinical endpoint after cerebral artery stenosis, we rst performed log-rank tests. We found that CYP2C19 LOF allele variants, high ALT levels, low total protein levels and high blood glucose levels were associated with an increased risk of MACCEs in patients who underwent cerebral artery stents. However, only a high blood glucose level is associated with an increased risk of MACCEs in patients who underwent conservative treatment.These ndings suggested that glycaemic control was key to MACCEprevention in all carotid artery stenosis patients.In addition, we generated Kaplan-Meier curves to assess the association of CYP2C19 LOF alleles and the incidence of MACCEs strati ed by patient status. We found that the CYP2C19 LOF allele carriers were identi ed as a predictor of the incidence of MACCEs at 6 months after carotid artery stenting only when the patients had high levels of glucose (glucose > 6.5 mmol/L).
A Danish cohortof nearly 60 000 patients with myocardial infarction demonstrated that the clinical e cacyof clopidogrel in diabetes mellitus (DM) was impaired 23 . It has also been reported that DM signi cantly increases the risk of stroke recurrence and poor outcome in the small-artery occlusion subtype 22 . In our study, the rate of MACCEs in CYP2C19 LOF allele carrierswas higher than that in noncarriers in patientswith high blood glucose levels who underwentcarotid artery stents, while in patients with normal blood glucose levels, it was not. This could be ascribed to high blood glucose levels decreasing the active metabolite of clopidogrel 24 , especially impairing the clopidogrel clinical e cacy of CYP2C19LOF allele carriers with genetically poor clopidogrel metabolism. The mechanism of this has been reported: the active metabolite of clopidogrel could increase plateletreactivity by upregulating platelet P-selectin 25 and proteinkinase C 26 via the glycation of platelet surface proteinsand osmotic effects of glucose 27 . These ndings remind usthat it is necessary to performCYP2C19genotype tests in patients who underwent cerebral artery stents with dysglycaemia.
Next, we used the Cox proportional hazards regression model to assess the association between CYP2C19 LOF allele carriers and patient status with MACCE risk.As shown in Table 3, the univariable analysis and multivariable Cox analysis revealed that glucose > 6.5 mmol/L was signi cantly associated with MACCEs at 6 months in the conservative treatment group. Nevertheless, the univariable analysis revealed that CYP2C19 genotypes, ALT > 35 U/L, total protein < 65 g/L and glucose > 6.5 mmol/L were signi cantly associated with MACCEs at 6 months in the cerebral artery stent group (Table 4). We found that CYP2C19 genotypes, total protein < 65 g/L and glucose > 6.5 mmol/L remained signi cant in the multivariable Cox analysis.
Collectively, these results indicated that CYP2C19 LOF allele variants, total protein levels anddysglycaemiasynergically impact the risk of MACCEs.On the one hand, malnutrition has an important prognostic value in patients who undergo cerebral artery stents 28 ,and total protein < 65 g/L may re ect themalnourished status of the patient. On the other hand, it could be due to poor kidney function, as the protein is lost in the urinary system, leading to a decrease in total protein. treatment resistance for clopidogrel. It has been reported that chronic kidney disease is associated with impaired drugabsorption and transport and platelet abnormalities, which are thought to decrease the response to clopidogrel 11,29 . In addition, CYP2C19 LOF wasreported to be associated with an increased risk of adverse cerebrovascularoutcomes in patients in the lowest quintile of eGFR.In this way, low total protein levels may re ect poor kidney function, and the response to clopidogrel is decreased.
In this way, we conclude thatCYP2C19 LOF allele variants, total protein levels and dysglycaemia synergically impact the risk of MACCEs in patients who underwent cerebral artery stents. In addition, it is necessary to perform CYP2C19 genotype tests in patients who underwent cerebral artery stents with dysglycaemiaand low total protein levels.
There are several limitations in our study. Considering that this is a single-centre study, further multicentre and large sample studies are needed to expand upon our ndings. Glucoselevels were assessed only at baseline, and a dynamic evaluationof glucose levels could be more informative. In addition, glucose levels can only re ect the current status, and glycated albumin and haemoglobin A1c can more accuratelyre ect the actual status of glycaemic control. Moreover, we did not assess the status of insulin resistance, which could lead to abnormal plateletP2Y12 receptor signalling in diabetic patients. Figure 1 Flow diagram of the study population recruitment process.

Figure 2
Distribution of the CYP2C19 genotype in patients.

Figure 3
Rates of MACCEs over 6 months of follow-up in patients underwent cerebral artery stent. CYP2C19 LOF alleles carriers including the genotypes *1/*2, *1/*3, *2/*2, *2/*3 and. *3/*3 The genotype of non-carriers is CYP2C19 *1/*1. The curves represent the percentage of patients surviving at the endpoints. The numbers below the survival curves are the numbers of patients in each group who survived at the endpoints and were still at risk over the follow-up period.