Impact of first-line cisplatin versus non-cisplatin based chemotherapy on progression-free survival in patients with advanced urothelial carcinoma previously treated with perioperative cisplatin based chemotherapy.

335 Background: Perioperative cisplatin based chemotherapy (PCBC) is a standard of care in the management of muscle invasive urothelial carcinoma (UC). Cisplatin based (C) therapy also represents the historical first line treatment of metastatic disease. There is however no data to guide the optimal choice of first line chemotherapy regimen – C re-treatment vs other second-line or non cisplatin regimens (NC) –in UC patients who relapse after receiving PCBC. This multicenter retrospective study compares C vs NC first line treatment on progression-free survival (PFS) for patients (pts) with advanced UC after PCBC and cystectomy. Methods: Data were collected for patients who received various first-line chemotherapies for advanced UC following previous PCBC therapy. Cox proportional hazards models were used to investigate the prognostic ability of type of peri-operative / first-line chemotherapy, visceral metastasis, ECOG status, time from prior chemotherapy (TFPC), anemia, leukocytosis and albumin on PFS. Results: Data were available for 145 pts from 12 centers. The mean age was 62 years, 113 (77.9%) were men and ECOG-PS was 0 or >0 in 74 (51.0%) and 61 (42.0%) patients. Ninety-one (62.8%) pts received C first line, the median number of cycles was 4 (range 1-17) and the median TFPC was 6.2 months (range 1-154). Median overall survival was 86 weeks (95% CI 70-106) and median PFS was 24 (95% CI 18-27) weeks. Time from perioperative chemotherapy (TFPC) (>52 weeks vs ≤52 weeks; HR 0.63 p=0.027) and ECOG-PS at first line (1+ vs 0; HR 1.73 p=0.010), were prognostic of PFS. No significant effect was noted for C vs NC first line (p=0.70); however, among patients with TFPC >52 weeks, patients with NC had worse PFS (median 4.6 months, 95% CI 1.8-12.2) than those who received C (median 8.1, 95% CI 3.2-16.3). Conclusions: There is no evidence to suggest overall superiority of C vs NC based first line chemotherapy or a second-line regimen in patients with advanced UC who received prior PCBC. However, those with TFPC >52 weeks should probably receive C first line chemotherapy given better PFS with C.