Background: Recurrent high-risk non-muscle-invasive bladder cancer (HR-NMIBC) poses significant therapeutic challenges due to frequent recurrences. Bacillus Calmette-Guerin (BCG) therapy limitations including global shortages, resistance and adverse effect. Patients often decline cystectomy due to significant impact on quality of life and complications, underscoring the need for effective bladder-preserving strategies. Methods: In a retrospective single-center analysis, 43 recurrent HR-NMIBC patients received intravenous tislelizumab every 3 weeks for up to 17 cycles alongside intravesical chemotherapy. Intravesical instillation agents included gemcitabine, epirubicin or mitomycin C administered per protocol. The primary endpoint was 3-month complete response rate, defined as absence of tumor confirmed by biopsy, cystoscopy, cytology and imaging. Secondary endpoints encompassed duration of response, survival metrics, and adverse events graded per CTCAE v5.0. Results: Among 43 recurrent HR-NMIBC patients with median age 67 years, 88.4% achieved 3-month complete response. Responders maintained a 24-month sustained response rate of 71.1% during median 28.4-month follow-up. The entire cohort demonstrated 36-month progression-free survival, overall survival, and cancer-specific survival rates of 82.8%, 85.0%, and 95.0% respectively. Subgroups with variant histology, BCG-unresponsive/intolerant disease, or multiple prior recurrences showed reduced responses. Treatment-related adverse events occurred in 90.7% of patients, with urinary frequency and hematuria being most common, which might stem from intravesical chemotherapy exposure. Grade 3 events affected 18.6% of patients, while immune-related adverse events led to discontinuation in 4.7%. No Conclusion: Tislelizumab plus intravesical chemotherapy achieves high response rates and durable survival benefits with manageable toxicity in recurrent HR-NMIBC, representing a promising non-BCG-dependent bladder-preserving strategy for surgically ineligible patients. (c) 2025 Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.