PURPOSE: To prospectively evaluate prognostic capabilities of non-muscle invasive bladder cancer (NMIBC) biomarkers for predicting disease recurrence or progression after radical TURB (transurethral resection of bladder tumor). METHODS: Evaluated biomarkers included blood: plasminogen activator inhibitor 1 (PAI-1), soluble urokinase plasminogen activator receptor (suPAR), interleukin 8 (IL-8) and urine: IL-8, vascular endothelial growth factor (VEGF) and apolipoprotein E (APOE). Blood and urine samples acquired before TURB for NMIBC from 223 subjects were analysed. The primary outcome was tumor recurrence or progression. RESULTS: After 3 months follow-up with cystoscopy or TURB- 92 patients were tumor free (Group 1). In 131 subjects (Group 2) a recurrence of NMIBC (n = 120) or progression to muscle invasive bladder cancer (MIBC) (n = 11) has been observed. No major clinical differences between these two groups were spotted. The group 2 has presented with significantly higher concentrations of blood IL-8 and suPAR as well as urine VEGF and APOE. The serum IL-8 and urinary VEGF showed the highest prognostic abilities with AUROC of 0.611 (95% CI: 0.534-0.687, p = 0.0044) and 0.632 (95% CI: 0.557-0.707, p = 0.0006), respectively. Multivariable machine learning models which included all investigated biomarkers and European Organisation for Research and Treatment of Cancer (EORTC) risk scores have allowed to discriminate the two patient entities with AUROC of 0.84 (95% CI: 0.73-0.95, p <
0.0001). CONCLUSIONS: The assessed biomarkers alone have shown unsatisfactory prognostic capabilities to be used for prognostication of outcomes after TURB. More complex multivariable prediction models may improve their prognostic performance. TRIAL REGISTRATION: The study was retrospectively registered at clinicaltrails.gov with National Clinical Trial number (NCT): NCT06235853.