Association of Preoperative Hematological and Inflammatory Biomarkers with Pathological Tumor Stage in Bladder Cancer: A Retrospective Cohort Study

Abstract Background Systemic inflammatory and hematological parameters derived from routine laboratory tests have been investigated as potential indicators of tumor-related biological characteristics in bladder cancer. However, their association with pathological tumor stage remains incompletely characterized. This study evaluated the association of routinely available preoperative hematological and inflammatory biomarkers with pathological tumor stage in patients undergoing transurethral resection of bladder tumor (TUR-BT). Methods This retrospective single-center cohort study included 182 patients with primary bladder cancer who underwent TUR-BT between January 2018 and January 2026. For the ordinal stage analysis, patients with pTa, pT1, and pT2 disease were included (n = 179); patients with pT4 (n = 2) or pTis (n = 1) were not included in the three-level ordinal analysis. Preoperative hemoglobin, red cell distribution width (RDW), neutrophil-to-lymphocyte ratio (NLR), systemic immune-inflammation index (SII), systemic inflammation response index (SIRI), and De Ritis ratio were evaluated. Tumor-related characteristics, including tumor size, number, multifocality, location, and pathological grade, were also recorded. Continuous variables were compared using the Kruskal–Wallis test with Dunn–Bonferroni post-hoc analysis. Spearman correlation analysis and age-adjusted ordinal logistic regression were performed. Multicollinearity was assessed using variance inflation factors. Results Among the 179 patients included in the ordinal stage analysis, 74 had pTa, 67 had pT1, and 38 had pT2 disease. Age, hemoglobin, RDW, NLR, SII, SIRI, and De Ritis ratio differed significantly across pathological stages. Median hemoglobin decreased from 14.9 g/dL in pTa to 14.2 g/dL in pT1 and 13.2 g/dL in pT2 disease (p = 0.002), whereas NLR, SII, SIRI, RDW, and De Ritis ratio increased with advancing stage. Higher pathological stage was also associated with larger tumor size, higher pathological grade, and greater multifocality. In the age-adjusted ordinal logistic regression model, age (OR = 1.041, 95% CI: 1.020–1.063, p < 0.001) and SII (OR = 1.001, 95% CI: 1.000–1.001, p = 0.032) were statistically significant, whereas hemoglobin, RDW, NLR, SIRI, and De Ritis ratio were not. The multivariable analysis included 178 patients with complete data for all model variables. Multicollinearity was most pronounced among NLR, SII, and SIRI. Conclusions Several routinely available hematological and inflammatory parameters were associated with pathological tumor stage in unadjusted analyses. After adjustment for age and the other evaluated biomarkers, age and SII remained statistically associated with higher pathological stage, whereas hemoglobin was no longer significant. These findings should be considered exploratory associations rather than evidence of validated predictive performance. Prospective studies incorporating established clinical, cystoscopic, cytological, and radiological variables are required to determine whether these biomarkers provide incremental information beyond conventional preoperative assessment.