This study aims to explore the impact of interleukin (IL)-10 single nucleotide polymorphisms (SNPs) and its interaction with environment on the risk of systemic lupus erythematosus (SLE). Chi-square testing method was used to investigate whether the distributions for genotype of four SNPs were differed from Hardy-Weinberg equilibrium (HWE). Logistic regression was used to test the association between IL-10 SNPs and SLE risk. The best interaction combinations between IL-10 SNPs and environmental factors were assessed by generalized multifactor dimensionality reduction (GMDR). Both rs1800896-G and rs1800871-T alleles were associated with increased risk of SLE, the odds ratios (ORs) (95% confidence interval (CI)) for the two SNPs were 1.68 (1.25–2.09) and 1.47 (1.12–1.94), respectively. Then, we used the GMDR method to analyze the high-order interactions of four SNPs within IL-10 gene and environmental factors on SLE risk. We found a significant interaction combination (two-locus model with P = 0.001) between rs1800896 and smoking, after adjusting for gender, age, body mass index (BMI), and alcohol drinking. We also used two-variable stratified analysis by logistic regression to analyze the synergistic effect between two variables (rs1800896 and smoking), which had significant significance in GMDR model. We found that current smokers with rs1800896-AG or GG genotype have the highest SLE risk, compared with never smokers with the rs1800896-AA genotype, OR (95% CI) = 2.24 (1.52–3.58). The rs1800896-G and rs1800871-T alleles and interaction between rs1800896 and current smoking were all associated with increased risk of SLE.

本研究旨在探讨白介素（IL）-10单核苷酸多态性（SNP）及其与环境的相互作用对系统性红斑狼疮（SLE）风险的影响。采用卡方检验方法研究四种SNP的基因型分布是否与Hardy-Weinberg平衡（HWE）不同。Logistic回归用于检验IL-10 SNP与SLE风险之间的关联。IL-10 SNP与环境因素之间的最佳相互作用组合是通过广义多因素降维（GMDR）进行评估的。rs1800896-G和rs1800871-T等位基因均与SLE风险增加相关，两个SNP的优势比（OR）（95％置信区间（CI））分别为1.68（1.25–2.09）和1.47（1.12–1.94） ， 分别。然后，我们使用GMDR方法分析了IL-10基因中的四个SNP和环境因素对SLE风险的高阶相互作用。在调整了性别，年龄，体重指数（BMI）和饮酒之后，我们发现rs1800896与吸烟之间存在显着的相互作用组合（P = 0.001的两基因座模型）。我们还通过逻辑回归使用二变量分层分析来分析两个变量（rs1800896和吸烟）之间的协同效应，这在GMDR模型中具有重要意义。我们发现，与从未吸烟者和rs1800896-AA基因型为OR（95％CI）= 2.24（1.52-3.58）的吸烟者相比，目前具有rs1800896-AG或GG基因型的吸烟者的SLE风险最高。rs1800896-G和rs1800871-T等位基因以及rs1800896与当前吸烟之间的相互作用均与SLE风险增加相关。