Prognostic value of CD103+ tumor-infiltrating lymphocytes in bladder cancer: A systematic review and meta-analysis

CD103+ tissue-resident memory T cells are emerging mediators of antitumor immunity whose prognostic significance in bladder cancer, across muscle-invasive and metastatic urothelial carcinoma, remains undefined at a meta-analytic level. We performed a systematic review and meta-analysis of cohorts with histologically confirmed muscle-invasive bladder cancer or metastatic urothelial carcinoma, comparing high vs. low CD103+ tumor-infiltrating lymphocyte density assessed by immunohistochemistry or ITGAE mRNA expression, with overall survival endpoint. PubMed, Embase, Scopus, and Cochrane CENTRAL were searched from inception through March 2026. Pooled hazard ratios (HR) with 95% confidence intervals (CIs) were estimated using a restricted maximum likelihood random-effects model. Subgroup analyses were performed by treatment context, disease stage, and assessment method. The protocol was registered at OSF (u36xv). Three studies comprising 6 cohorts were included. High CD103⁺ infiltration was significantly associated with prolonged overall survival (HR: 0.50, 95% CI: 0.30-0.83). A significant treatment-context interaction (P = 0.005) favored patients undergoing surgery with adjuvant chemotherapy (HR: 0.20) over those receiving immunotherapy (HR: 0.69). Stage-stratified analysis showed a significant interaction (P = 0.003), with pronounced benefit in muscle-invasive bladder cancer (HR: 0.27) but a non-significant trend in metastatic urothelial carcinoma (HR: 0.79). Immunohistochemistry-based cohorts showed a numerically larger effect than ITGAE mRNA cohorts. High CD103⁺ tumor-infiltrating lymphocyte infiltration may be associated with improved overall survival in bladder cancer, particularly in muscle-invasive bladder cancer patients undergoing surgery with adjuvant chemotherapy. Prospective validation in larger, stage-homogeneous cohorts is warranted to establish CD103 as a prognostic biomarker. These findings are hypothesis-generating and CD103 assessment is not currently ready for clinical use.