Gemcitabine plus carboplatin versus gemcitabine plus oxaliplatin in cisplatin unfit patients with advanced urothelial carcinoma: A randomized phase II study (COACH, KCSG GU10-16).
4534 Background: We investigated the activity and safety of first-line gemcitabine-oxaliplatin (GemOx) compared with gemcitabine-carboplatin (GCb) in cisplatin-ineligible patients with advanced UCCC. Methods: Treatment naïve, cisplatin-ineligible patients with advanced UCCC were randomly assigned to GemOx [gemcitabine 1,000 mg/m2, oxliplatin 100 mg/m2 on day 1 (D1) every 2 weeks] or GCb (gemcitabine 1,000 mg/m2 on D1 and D8, carboplatin AUCC 4.5 on D1 every 3 weeks) stratified by ECOG performance status (PS) and visceral metastases. The primary endpoint was objective response rate (ORR), and secondary endpoints were progression-free survival (PFS) and overall survival (OS). Results: Between January 2011 and Mar 2017, 80 patients were enrolled; 39 patients and 40 patients were allocated to GCb and GemOx arm, respectively. Median age was 72 years (range 46-84) in GCb arm and 72.5 (55-85) in GemOx arm. ECOG PS was 2 in 41% (GCb arm) and 43% (GemOx arm), and median GFR was 45 ml/min (interquartile range 36-56) in GCb and 47 ml/min (37-56) in GemOx arm. ORR were 48.7% and 55.0% in GCb and GemOx, respectively. With a median follow-up duration of 37.8 months, median PFS and OS in GCb and GemOx arm were 5.5 months (95% CI, 4.8-6.2) vs. 4.4 months (95% CI, 2.7-6.1) and 9.1 months (95% CI, 5.2-13.0) vs. 11.0 months (95% CI, 6.9-15.0), respectively. Grade 3 leukopenia, neutropenia and fatigue were significantly more common in GCb arm (26% vs. 3%, p=0.003; 33% vs. 10%, p=0.014; 15% vs. 3%, p=0.057) while any grade neuropathy was more common in GemOx (8% vs. 60%, p<0.001). Conclusions: GemOx has showed comparable efficacy with GCb and favorable hematologic toxicity profile. GemOx may be used as a new option for UCCC patients who are not suitable for platinum-containing chemotherapy. Clinical trial information: NCT01487915.