BACKGROUND: Tislelizumab, an anti-programmed cell death protein 1 monoclonal antibody, has demonstrated efficacy and safety in solid tumors. This analysis evaluates its long-term safety. PATIENTS AND METHODS: A retrospective analysis was performed on data from 8 randomized phase III clinical trials involving patients with advanced gastrointestinal (GI) or lung cancers who received tislelizumab. Endpoints included immune-mediated adverse events (imAEs) by tumor type, race, and treatment duration, including early (within 1 year) and delayed (>1 year) imAEs. RESULTS: In all, 2636 patients (1415 with GI cancer, 1221 with lung cancer) were included in the analysis, with a median follow-up of 15.4 months. Most imAEs were low-grade. In GI cancers, 32.5% of patients experienced any-grade imAEs (7.3% grade ≥ 3). In lung cancers, 39.6% reported any-grade imAEs (8.2% grade ≥ 3). The most frequently reported imAEs were skin toxicity, hypothyroidism, pneumonitis, and hyperthyroidism. The incidence of imAEs was slightly higher in Asian patients compared with non-Asian patients (34.3% vs 26.9% in GI cancer and 35.5% vs 29.7% in lung cancer, respectively), but the incidence of grade ≥ 3 imAEs remained similar (7.5% vs 6.9% in GI cancer and 6.1% vs 7.2% in lung cancer, respectively). Most imAEs occurred within 6 months of treatment initiation. Delayed imAEs were infrequent and predominantly occurred while patients were still receiving tislelizumab therapy. CONCLUSIONS: Tislelizumab's safety profile, as monotherapy or combined with chemotherapy, remains consistent with previous reports across tumor types. Delayed imAEs were relatively infrequent, with no new signals identified in this long-term safety analysis.