ObjectiveTo compare the differential therapeutic effects of Bacillus Calmette-Guérin (BCG) instillation and radical cystectomy (RC) for high-risk non-muscle–invasive urothelial cancer (NMIBC) classified as high-grade T1 in initial and repeat transurethral resection of bladder tumors (TURBT) and to construct a prediction model.MethodsWe retrospectively analyzed the clinical data of patients with malignant bladder tumors treated at the First Affiliated Hospital of Soochow University from January 2016 to December 2017 and compared the differences in 1-year, 2-year, 3-year, 5-year, and comprehensive overall survival (OS) and progression-free survival (PFS) between BCG instillation treatment and RC treatment. Survival curves were drawn to show differences in OS and PFS between the two groups. Concurrently, univariate and multivariate COX analyses were performed to identify risk factors affecting OS and PFS, and a nomogram was created.ResultsIn total, 146 patients were included in the study, of whom 97 and 49 were in the BCG and RC groups, respectively. No statistical differences were observed in the 1- and 2-year OS and PFS between the two groups, whereas significant statistical differences were found in the 3-year, 5-year, and comprehensive OS and PFS. Survival curves also confirmed the statistical differences in OS and PFS between the BCG and RC groups. Multivariate COX analysis revealed that the treatment method, concomitant satellite lesions, and albumin-to-alkaline phosphatase ratio (AAPR) were independent risk factors affecting OS and PFS. The nomogram that was further plotted showed good predictive ability for OS and PFS.ConclusionFor patients who exhibit high-level T1 pathology after both initial and repeat TURBT, especially those with low AAPR, and concomitant satellite lesions, choosing RC as a treatment method offers a better prognosis.