Genetic Association and Clinical Relevance of TNFSF13B/BAFF and PADI4 Polymorphisms in ANCA-Associated Vasculitis: A Case–Control Study with Genetic Model Analysis in Guangxi Population

Objective: TNFSF13B, which encodes B-cell-activating factor (BAFF) and peptidylarginine deiminase 4 (PADI4), plays crucial roles in the pathogenesis of ANCA-associated vasculitis (AAV). This study investigated the associations of single-nucleotide polymorphisms (SNPs) in TNFSF13B/BAFF and PADI4 genes with AAV susceptibility, clinical phenotypes, and disease activity in a Guangxi Chinese population. Methods: A case–control study included 324 AAV patients and 324 healthy controls. After propensity score matching (201 pairs), genomic DNA was genotyped for TNFSF13B/BAFF rs3759467 (formerly rs386492354) and rs1041569, and PADI4 rs11203366 and rs874881 using multiplex PCR and high-throughput sequencing. Genetic associations were analyzed via logistic regression, subgroup, haplotype, and clinical correlation analyses. For each of the four SNPs separately, machine learning models (logistic regression, SVM, Random Forest, XGBoost) were built and evaluated via 5-fold cross-validation. No formal adjustment for multiple comparisons was applied due to the exploratory nature of this study. Results: For TNFSF13B/BAFF, the rs3759467 C allele was protective (dominant model OR = 0.60, p = 0.011; log-additive OR = 0.71, p = 0.020; CA haplotype OR = 0.71, p = 0.019), while the rs1041569 T allele was a risk factor (dominant model OR = 1.70, p = 0.016). Subgroup analysis revealed stronger protective effects of rs3759467 in females, Han ethnicity, and MPA patients, and stronger risk effects of rs1041569 in Han ethnicity and MPA patients. Haplotype CA was protective (OR = 0.71, p = 0.019), and TT was risk-associated (OR = 1.55, p = 0.017). Both TNFSF13B/BAFF SNPs were associated with rash and hemoptysis incidence (p < 0.05). rs1041569 was also associated with RBC (red blood cell) count and HB (hemoglobin) levels (p < 0.05). For PADI4, rs11203366 and rs874881 showed no association with AAV susceptibility (all p > 0.05). However, their genotypes were associated with disease activity (BVAS, Birmingham Vasculitis Activity Score), RBC count, and HB levels (p < 0.05). Although machine learning was applied to explore predictive patterns, its performance was suboptimal (AUC < 0.6), indicating limited clinical applicability. Accordingly, the primary findings rely on the genetic model analysis, and the machine learning results should not be overinterpreted as clinically actionable. SHAP analysis indicated that risk-associated genotypes contributed most to model predictions. Conclusions:TNFSF13B/BAFF gene polymorphisms rs3759467 and rs1041569 were associated with AAV susceptibility in this Guangxi cohort, influencing clinical manifestations like rash, hemoptysis, and anemia severity. PADI4 polymorphisms rs11203366 and rs874881 are not associated with susceptibility but may correlate with disease activity and hematological parameters. These findings highlight the ethnic and clinical subtype specificity of genetic influences in AAV. Due to the lack of external validation, these findings are exploratory and require replication.