The global burden of ischemic stroke (IS) keeps rising [1], with forecasts suggesting that IS-related deaths will increase to approximately 4.9 million worldwide by 2030 [2, 3]. IS is a complex disorder that can be classified into five etiological subtypes according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification: large-artery atherosclerosis (LAA), small-vessel occlusion (SVO), cardioembolism (CE), undetermined, and other determined etiology [4]. The clinical manifestations, targeted therapeutic strategies, and prognostic outcomes across the different etiological subtypes are distinct [4, 5], underscoring the significance of subtype-specific research in IS. The primary cause of LAA stroke is atherosclerotic plaque formation and subsequent rupture of large cerebral arteries [6]. LAA typically manifests with severe focal neurological deficits (e.g., hemiplegia, aphasia), accompanied by a high susceptibility to early neurological deterioration [4].

The P2Y purinoceptor 12 (P2Y12 receptor) gene (herein referred to as the P2Y12 gene) localizes to chromosome 3, long arm bands q21 to q25 (3q21-q25) and encodes a 342-residue, G-protein-coupled receptor on the surface of platelets, microglia, and vascular smooth muscle cells (VSMCs) [7]. The adenosine diphosphate (ADP)-mediated P2Y12 receptor signaling pathway can promote platelet activation and aggregation, pro-inflammatory cytokine release, formation of platelet-leukocyte aggregates, and VSMC migration and proliferation [8]. These processes contribute to the progression of atherosclerosis and thrombosis. Consistent with the essential role of the P2Y12 receptor in atherosclerosis and pathological thrombosis, one of its single-nucleotide polymorphisms (SNPs), i-T744C (rs2046934, a T$\to$C transition at 744 nucleotides (nt) downstream of intron 5’s start site), was observed to significantly influence ADP-induced platelet aggregation [9, 10, 11]. i-T744C was also associated with increased susceptibility to atherosclerosis and related complications, such as coronary artery disease (CAD) and myocardial infarction [12, 13]. Prior studies linking P2Y12 polymorphisms to IS susceptibility yielded inconsistent results [14, 15]. However, these studies did not stratify IS into subtypes, thus potentially overlooking a possible subtype-specific effect of P2Y12 polymorphisms. Investigation of the association of P2Y12 polymorphisms with specific IS subtypes can overcome the potential confounding effect of subtype heterogeneity, thereby improving the accuracy of genetic risk stratification. As far as we are aware, no such studies have yet been conducted.

Nomograms are a clinically valuable instrument for integrating diverse data types and can be used for multi-disease risk assessment with increased precision. The LAA subtype is a multi-factorial disease attributable to an interplay of multiple factors [16]. Integration of these variables into a predictive model may enhance the diagnostic accuracy of LAA. These considerations motivated us to develop a nomogram that integrates well-established genetic and clinical variables, enabling quantification of the weighted contribution of each input variable and making it intuitive for clinical practice.

Therefore, the aim of this study was to elucidate the relationship between two common P2Y12 polymorphisms [i-T744C and C34T (rs6785930, a C$\to$T transition at 34 nt downstream of exon 2’s start site)] and the susceptibility to three major IS subtypes (LAA, SVO, and CE), as defined by TOAST criteria. We also developed a quantitative genetic-clinical nomogram, with the aim of improving the ability to predict LAA risk.

This study provides the first evidence of a significant association between the P2Y12 i-T744C polymorphism and an elevated risk of LAA stroke. Furthermore, we developed a nomogram model that incorporates traditional risk factors along with the i-T744C polymorphism for predicting LAA stroke. Patients with high scores in this predictive nomogram have a high probability of developing LAA stroke, and may therefore derive clinical benefit from early prediction and preventive care.

Studies have established that the P2Y12 receptor, encoded by the P2Y12 gene, plays an essential role in atherosclerosis and atherothrombosis. Activation of P2Y12 on platelets triggers the phosphoinositide 3-kinase (PI3K)/protein kinase B (PKB/AKT) pathway via coupled G protein $\beta$$\gamma$ (G$\beta$$\gamma$) subunits, resulting in platelet activation and thrombosis [11, 19]. Moreover, activated P2Y12 facilitates the release of a variety of pro-inflammatory factors [8], thereby contributing to inflammatory cascades linked to atherosclerosis. Recent research has demonstrated that P2Y12 receptors participate in regulating VSMC migration into the intima and plaque, which contributes to the development of atherosclerosis and its complications [20].

Other investigations have shown that P2Y12 genetic variants were associated with susceptibility to atherosclerosis-related disorders and also influenced platelet reactivity. Fontana et al. [10] reported that four variants in the P2Y12 gene—i-T744C, i-C139T (a C$\to$T transition at 139 nt downstream of intron 5’s start site), i-ins801A (a single-nucleotide A insertion at 801 nt downstream of intron 5’s start site), and G52T (a G$\to$T transition at 52 nt downstream of exon 2’s start site)—were in linkage disequilibrium. An individual who carries all mutant alleles in the four polymorphisms was classified as an H2 haplotype carrier. Otherwise, they were classified as having the H1 haplotype. Most studies have used the i-T744C or G52T SNPs to represent the haplotypes H1 and H2 [13, 14, 15]. The P2Y12 H2 haplotype was reported to be positively correlated with ADP-induced platelet aggregation and was accompanied by a significant reduction in the cyclic adenosine monophosphate (cAMP) concentration [21]. A case–control study reported that H2 allele carriers displayed significant susceptibility to peripheral arterial disease compared with H1 allele carriers [22]. Several reports in distinct populations have revealed that patients with mutant genotypes of the i-T744C polymorphism have a higher propensity for CAD or acute coronary syndrome [12, 23]. Notably, there is still controversy regarding the association between P2Y12 polymorphisms and IS. Lu et al. [14] found that the H2 allele may confer an elevated risk of cerebral infarction. In support of this, a meta-analysis indicated the P2Y12 i-T744C polymorphism may be predictive of IS risk [24]. In contrast, a prospective cohort study conducted in white males yielded negative results [15]. Previous study has suggested that predisposition to certain IS subtypes is genetically determined, with distinct mechanisms underlying their development [17]. Consequently, the lack of specific IS subclassification and instead the investigation of overall IS association with P2Y12 polymorphisms may lead to potential bias in studies of genotype-phenotype associations. Unlike other IS subtypes, the pathophysiological foundation of LAA stroke is the development of platelet-rich thrombus at the site of the ruptured atherosclerotic lesion [25]. Our analysis indicated the i-T744C polymorphism is an independent risk factor for the LAA subtype, but not for the SVO or CE subtypes. This finding supports the hypothesis that the P2Y12 i-T744C polymorphism exerts subtype-specific effects on LAA stroke, but not on other IS subtypes. The theoretical basis for this observation is grounded in the role of the i-T744C SNP in regulating platelet aggregation and mediating the development of atherosclerosis. However, future studies should also perform functional experiments to elucidate the molecular mechanisms underlying the P2Y12 i-T744C-LAA stroke genotype-phenotype association.

Consistent with previous findings [15], the P2Y12 C34T polymorphism did not correlate with any IS subtypes in the current study. The P2Y12 C34T polymorphism has also been investigated for its association with clopidogrel resistance (CR). The results showed the T allele of C34T polymorphism was associated with increased susceptibility to CR and concomitant adverse cardiac or cerebrovascular outcomes in Chinese cohorts [26]. However, conflicting results have been reported in other ethnic groups [27], and further studies should focus on correlations between the C34T polymorphism and adverse clinical outcomes in clopidogrel-treated stroke patients.

The usefulness of nomograms for the diagnosis and prognostic evaluation of IS and its subtypes has previously been reported through the integration of relevant risk factors [28, 29, 30]. For example, Chen et al. [28] devised a nomogram with good discrimination for the prediction of early IS. Their nomogram comprised 6 clinical parameters: gender, diabetes, family history, coronary heart disease, smoking, and age. Similar to a report in the literature [16], we identified age, hypertension, smoking, and HDL-C as four clinical correlates of LAA stroke. Additionally, we also found that the 744C polymorphism may be a potential predictor for the development of LAA stroke. IDI and NRI-common metrics quantify the incremental predictive value of one model over another in clinical research [31]. These suggested the i-T744C polymorphism might offer important additional clinical information for LAA stroke risk prediction. Based on the above findings, we constructed a novel nomogram for early LAA stroke prediction that includes the i-T744C polymorphism and clinical variables. This may serve as a reliable tool for personalized LAA stroke monitoring, as evidenced by its satisfactory calibration, discriminative capacity (AUC of 0.8 in the training cohort and 0.723 in the validation cohort), and favorable net clinical benefit. Our nomogram provides an immediate predicted risk of LAA stroke in clinical practice based on patient-specific variables, helping clinicians to rapidly identify high-risk individuals (predicted probability $>$30%). Once identified, these individuals may warrant closer clinical follow-up and timely intervention. Importantly, the visual nomogram offers a straightforward quantitative reference for shared decision-making by the clinician and patient, thereby improving adherence to preventive interventions in the susceptible population. We anticipate that this predictive model will be continuously upgraded and refined for future clinical applications by incorporating new variables and expanding the sample size to enhance its accuracy and practicality.

Our study had several limitations. First, the sample sizes of the stroke subtype and control groups were relatively small. Further studies with larger and ethnically diverse cohorts are needed to confirm our results. Second, complex gene-environment interactions drive the development of different stroke subtypes, whereas our predictive nomogram focuses on limited variables and is exclusively applicable to LAA stroke. Therefore, future studies should incorporate more genetic and environmental variations, as well as systematic analysis of gene-environment interaction effects within stroke subtypes. Finally, this analysis was limited to genotype-phenotype associations and requires further in-depth exploration of the potential mechanisms underlying the observed associations.

Our findings revealed that the P2Y12 i-T744C polymorphism was associated with LAA stroke, with the C allele being a significant predisposing factor. We constructed a nomogram model that combines the i-T744C polymorphism and clinical variables. This model displayed a favorable capacity to discriminate individual LAA risk, and could thus help to identify high-risk individuals and enable personalized prevention. Our findings require further validation in large-scale, multi-ethnic studies.

The datasets used and analyzed in the current article can be available from the corresponding author upon reasonable request.

CW—carried out the experiment and wrote the manuscript, YC, XW—collected the samples, JC—analyzed and interpreted the data, ZZ—designed the study and revised the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

This study was conducted in accordance with the Declaration of Helsinki. All procedures were approved by the Ethics Committee of Quanzhou First Hospital (Approval Number: [2024] K112). Written informed consent was signed by all participants before enrollment in the study.

We are grateful to the patients for their participation in our study.

This work was funded by the Startup Fund for scientific research, Fujian Medical University (2023QH1312).

The authors declare no conflict of interest.