This study applies K-means clustering to urinary inflammatory and oxidative stress biomarkers in IC/BPS patients, aiming to provide a new perspective on disease classification and its clinical relevance.

We retrospectively analyzed urine samples from 127 IC/BPS patients and 30 controls. Urinary levels of 10 inflammatory cytokines (Eotaxin, IL-2, IL6, IL-8, CXCL10, MCP-1, MIP-1β, RANTES, TNFα, and NGF) and 3 oxidative stress markers (8-hydroxy-2-deoxyguanosin [8-OHdG], 8-isoprostane, and total antioxidant capacity [TAC]) were quantified. K-means clustering was performed to identify biomarker-based patient subgroups, and associations with clinical characteristics and treatment outcomes within each cluster were examined.

IC/BPS patients exhibited significantly elevated urinary levels of Eotaxin, MCP-1, NGF, 8-OHdG, 8-isoprostane, and TAC compared to controls (all p < 0.05). K-means clustering identified four distinct subgroups (Figure 1). A summary of each cluster’s key characteristics is presented in Figure 2. Cluster 4 patients, characterized by elevated oxidative stress markers (8-OHdG, 8-isoprostane, TAC) and inflammatory cytokines (Eotaxin, IL-6, CXCL10, MCP-1, and RANTES), exhibited lower MBC and lower VAS pain scores. In contrast, Cluster 2 exhibited the lowest levels of most biomarkers among all clusters. Cluster 1 showed the intermediate biomarker levels, higher than Cluster 2 but lower than Cluster4, including Eotaxin, IL-2, CXCL10, RANTES, and 8-OHdG. Correlation analysis revealed distinct cluster-specific associations between biomarker levels and clinical parameters, including VAS pain score, MBC, the grade of glomerulation, and treatment outcomes (GRA). In Cluster 4, MIP-1β negatively correlated with GRA (r=-0.597, p=0.015), suggesting its potential role in predicting treatment response.

Traditional IC/BPS classification is based on clinical symptoms and/ or cystoscopic findings. However, these methods may not capture the molecular complexity of the disease. Our study demonstrated that applying K-means clustering to urine biomarkers provides deeper insights into IC/BPS patient subtypes, revealing significant variations in inflammatory and oxidative stress marker levels among clusters.

Notably, Cluster 4 patients, characterized by elevated oxidative stress markers (8-OHdG, 8-isoprostane, TAC) and inflammatory cytokines (Eotaxin, IL-6, CXCL10, MCP-1, and RANTES), exhibited lower MBC and lower VAS pain scores. Furthermore, 8-isoprostane levels showed a moderate to strong positive correlation with ICPI (r = +0.573) and VAS (r = +0.468), which was significantly stronger than the weak correlations observed in previous studies of the entire ESSIC type 2 IC/BPS cohort.7, 8 These findings suggest that Cluster 4 patients experience greater oxidative stress and more severe bladder inflammation, which manifest as reduced bladder capacity and differences in pain perception. 8-isoprostane may serve as a potential biomarker for disease severity in this subgroup. Moreover, MIP-1β levels demonstrated a strong negative correlation with GRA (r = -0.597), suggesting its potential as a predictor of poor treatment response in Cluster 4 patients.

The findings of this study underscore the heterogeneity of IC/BPS and highlight the potential of biomarker-based clustering as an objective approach to disease classification. By analyzing urine biomarkers with K-means clustering, we identified distinct subgroups with unique clinical characteristics and treatment responses. This biomarker-based approach offers a more objective and reproducible method for disease stratification compared to traditional classifications based solely on clinical presentation.

Applying K-means clustering to urine inflammatory and oxidative stress biomarkers provides a new perspective on disease classification, identifying IC/BPS subtypes with distinct clinical and biochemical characteristics. This approach may refine disease phenotyping and guide personalized treatment strategies in the future.