Comparison of Sequential Intravesical Gemcitabine and Docetaxel vs Bacillus Calmette-Guérin for the Treatment of Patients With High-Risk Non–Muscle-Invasive Bladder Cancer

IMPORTANCE: Due to the ongoing bacillus Calmette-Guérin (BCG) shortage, sequential intravesical gemcitabine and docetaxel has been increasingly used as first-line therapy for high-risk non–muscle-invasive bladder cancer (NMIBC). However, data directly comparing these 2 therapies are lacking. OBJECTIVE: To compare the outcomes of patients with high-risk NMIBC treated with gemcitabine and docetaxel vs BCG. DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study was conducted from January 1, 2011, to December 31, 2021. The median (IQR) duration of follow-up was 23 (12-33) months for patients receiving gemcitabine and docetaxel and 49 (27-79) months for patients receiving BCG. All patients were treated at the University of Iowa tertiary care center. A total of 312 patients with high-risk treatment-naive NMIBC were included; 174 patients were treated with BCG therapy and 138 were treated with gemcitabine and docetaxel therapy. EXPOSURES: After undergoing complete transurethral resection of bladder tumor, patients received either sequential intravesical gemcitabine, 1 g, and docetaxel, 37.5 mg, or 1 vial of BCG. Induction treatments were administered once per week for 6 weeks. Maintenance regimens were initiated if the patient was disease free at the first follow-up visit. MAIN OUTCOMES AND MEASURES: The primary outcome was high-grade recurrence-free survival (RFS). Survival probabilities were estimated using the Kaplan-Meier method. Cox regression models were used to evaluate the association of covariates with outcomes. Adverse events were reported using the Common Terminology Criteria for Adverse Events, version 5. RESULTS: Among 312 patients, the median (IQR) age was 73 (66-79) years; 255 patients (81.7%) were male and 292 (93.6%) were White. Baseline clinicopathological characteristics such as sex, smoking status, and pretreatment tumor pathology were similar between treatment groups. High-grade RFS estimates were 76% (95% CI, 69%-82%) at 6 months, 71% (95% CI, 64%-78%) at 12 months, and 69% (95% CI, 62%-76%) at 24 months in the BCG group and 92% (95% CI, 86%-95%) at 6 months, 85% (95% CI, 78%-91%) at 12 months, and 81% (95% CI, 72%-87%) at 24 months in the gemcitabine and docetaxel group. Multivariable Cox regression analyses controlled for age, sex, treatment year, and presence of carcinoma in situ revealed that treatment with gemcitabine and docetaxel was associated with better high-grade RFS (hazard ratio, 0.57; 95% CI, 0.33-0.97; P = .04) and RFS (hazard ratio, 0.56; 95% CI, 0.34-0.92; P = .02) than treatment with BCG. Induction therapy for BCG was associated with greater treatment discontinuation than induction therapy for gemcitabine and docetaxel (9.2% vs 2.9%; P = .02). CONCLUSIONS AND RELEVANCE: In this cohort study, gemcitabine and docetaxel therapy was associated with less high-grade disease recurrence and treatment discontinuation than BCG therapy. These findings suggest that, while awaiting results from an ongoing randomized clinical trial during the current BCG shortage, use of gemcitabine and docetaxel can be considered for recommendation in updated practice guidelines.