Abstract LB-172: A randomized phase II study of S-1 versus capecitabine as first-line chemotherapy in the elderly metastatic gastric cancer patients with/without poor performance status: clinical and pharmacogenetic results.

Introduction: This study investigated the efficacy and safety of S-1 vs. capecitabine in elderly patients (pts) with metastatic gastric cancer (MGC) with/without poor performance status (PS), and examined the association between CYP2A6 polymorphisms and treatment outcomes. Methods: Eligibility criteria included MGC, measurable lesion(s), no prior chemotherapy for MGC, and age of 70-85 yrs with ECOG PS 0-2 or age ≥ 65 and 2 bid or capecitabine 1250 mg/m 2 bid on days 1-14 every 3 weeks. The primary endpoint was overall response rate (ORR) with time to progression (TTP), overall survival (OS), quality of life (QOL), safety, and the association between CYP2A6 genotypes ( *1 , *4 , *7 , *9 , *10 ) and treatment outcomes as secondary endpoints. Results: From May 2007 to July 2010, 107 pts were randomized to receive either S-1 (n=53) or capecitabine (n=54). Baseline characteristics were well balanced between the two arms (S-1:capecitabine): median age, 72 yrs (range, 64-81):71 yrs (65-78); male, 74%:81%; and PS 0/1, 85%:83%. The median number of cycles was 4 (range, 1-26) in S-1 arm and 5 (1-32) in capecitabine arm. The incidence of G3/4 neutropenia was 3.8% in each arm, and the most common G3/4 non-hematological toxicities included anorexia (21.2% with S-1 vs. 7.5% with capecitabine), fatigue (13.5% vs. 15.1%), and nausea (7.7% vs. 1.9%). S-1 arm had higher incidences of vomiting (40.4% vs. 17.0%; P =0.010) and tearing (51.9% vs. 28.3%; P =0.017), and lower incidence of HFS (25.0% vs. 58.5%; P =0.001) than capecitabine arm. Of the 49 pts assessable for response in each arm, the ORR was 28.6% (95% CI, 15.9-41.3%) in S-1 arm and 26.5% (95% CI, 14.1-38.9%) in capecitabine arm. The median TTP was 3.2 months (95% CI, 1.6-4.8 months) in S-1 arm and 3.4 months (95% CI, 2.5-4.3 months) in capecitabine arm ( P =0.825), and median OS was 8.5 months (95% CI, 6.1-10.9 months) and 10.3 months (95% CI, 7.1-13.5 months), respectively ( P =0.691). CYP2A6 polymorphisms were associated with the efficacy of S-1. Pts having two variant alleles ( V / V ) had inferior efficacy to those having no or one variant allele ( W / W or W / V ): median TTP 2.3 vs. 4.1 months ( P =0.062), and median OS 6.4 vs. 11.5 months ( P =0.034). However, the efficacy of capecitabine did not differ according to CYP2A6 genotypes: median TTP was 3.6 months in V / V pts vs. 3.3 in W / W or W / V ( P =0.257), and median OS was 11.6 vs. 10.2 months ( P =0.756), respectively. In pts with V / V genotype, capecitabine arm had better TTP (median, 3.6 vs. 2.3 months; P =0.062) and OS (median, 11.6 vs. 6.5 months; P =0.090) than S-1 arm. Conclusions: Both S-1 and capecitabine were active and tolerable for elderly MGC pts with/without poor PS with different toxicity profiles. Genotyping for CYP2A6 might provide useful information for treatment selection. Citation Format: Sook Ryun Park, Sun-Young Kong, Mi-Jung Kim, Min Kyeong Kim, Byung-Ho Nam, Mihee Choi, Young-Iee Park. A randomized phase II study of S-1 versus capecitabine as first-line chemotherapy in the elderly metastatic gastric cancer patients with/without poor performance status: clinical and pharmacogenetic results. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr LB-172. doi:10.1158/1538-7445.AM2013-LB-172