Adjuvant immune checkpoint inhibitors for urothelial carcinoma: systematic review and Meta-analysis

Purpose: To compare disease-free survival (DFS), overall survival (OS), and adverse events (AEs) among muscle-invasive urothelial carcinoma (MIUC) patients receiving adjuvant immune checkpoint inhibitors (ICIs) versus placebo/observation following radical surgery. Methods: This was a systematic review/meta-analysis of all published phase 3 randomized controlled trials. MEDLINE, EMBASE, and Cochrane were searched from inception until April 4, 2024. Pooled hazard ratios (HR) and relative risks (RR), plus confidence intervals (CI), were generated using frequentist random-effects modeling. Results: Three trials were identified: IMvigor010, CheckMate 274, and AMBASSADOR. In the overall cohort, adjuvant ICIs significantly improved DFS by 23% (HR = 0.77, 95% CI = 0.65–0.90). No DFS benefit was observed in patients with upper tract disease (HR = 1.19, 95% CI = 0.86–1.64). The highest magnitude of DFS benefit was observed among patients who had received prior neoadjuvant chemotherapy (HR = 0.69) and pathologic node-positive disease (HR = 0.75). A similar DFS benefit was observed irrespective of tumor PD-L1 status. Pooled OS demonstrated a 13% non-significant benefit (HR = 0.87, 95% CI = 0.75–1.01). Grade ≥ 3 immune-mediated AEs occurred in 8.6% and 2.1% of ICI and placebo/observation patients, respectively (RR = 4.35, 95% CI = 1.02–18.5). AEs leading to treatment discontinuation occurred in 14.3% and 0.9% of patients, respectively. Conclusion: Adjuvant ICIs confer a DFS benefit following radical surgery for MIUC, particularly among node-positive patients and those who received prior neoadjuvant chemotherapy. The lack of benefit for upper tract disease suggests that alternate adjuvant approaches, including chemotherapy, should be considered for these patients. Tumor PD-L1 status is not a predictive biomarker, highlighting the need for biomarkers in this setting.