Final Overall Survival Analysis of the SOGUG Phase 2 MAJA Study: Maintenance Vinflunine Versus Best Supportive Care After First-Line Chemotherapy in Advanced Urothelial Carcinoma

Current first-line treatment for advanced urothelial carcinoma has a limited duration of response. The MAJA study aimed to compare vinflunine (VFL) plus best supportive care (BSC) maintenance therapy versus BSC alone in 88 patients after first-line treatment with cisplatin/gemcitabine. Results demonstrated a progression-free survival benefit and a positive OS trend (with limited power due to the small sample size) with VFL in maintenance therapy with no unexpected long-term adverse effects. Introduction: The MAJA study compared vinflunine (VFL) plus best supportive care (BSC) maintenance therapy versus BSC alone in advanced urothelial carcinoma responsive to first-line chemotherapy. The primary end point of progression-free survival was achieved. We present the final overall survival (OS) and long-term follow-up safety analyses. Patients and Methods: Patients were enrolled, and a subsequent post hoc analysis was performed on the basis of radiologic response or stabilization to first-line cisplatin/gemcitabine (CG) chemotherapy (4-6 cycles), according to Response Evaluation Criteria in Solid Tumors (RECIST). VFL + BSC versus BSC alone were randomly assigned until disease progression. Results: At final analysis, 58 patients (66.7%) had died while 29 (33.3%) had survived; the BSC arm had higher mortality (VFL thorn BSC, n = 26, 59.1% vs. BSC, n = 32, 74.4%). Median follow-up of surviving patients was 38.8 months (interquartile range, 23.8-61.6). Median OS was 16.7 months (95% confidence interval, 0-34.5) in VFL and 13.2 months (95% confidence interval, 6-20.4) in the BSC groups (hazard ratio, 0.736; 95% confidence interval, 0.44-1.24, P=.182). Post hoc group division did not affect median OS in either study arm. Conclusion: Final analysis supported a benefit of VFL in maintenance therapy in patients with disease control after first-line treatment with CG, with no unexpected long-term adverse effects. The study was insufficiently powered to show a significant OS advantage. (C) 2020 Elsevier Inc. All rights reserved.