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

Abstract

This study investigated the efficacy and safety of S-1 versus capecitabine in elderly patients with metastatic gastric cancer (MGC), and examined the association between cytochrome P450 2A6 (CYP2A6) polymorphisms and treatment outcomes. MGC patients 70-85 years old with Eastern Cooperative Oncology Group (ECOG) performance status 0-2 or 65-70 years old with ECOG PS 2 were randomized to receive S-1 40 mg/m, twice daily, or capecitabine 1250 mg/m, twice daily, on days 1-14 every 3 weeks. From May 2007 up to July 2010, 107 patients were enrolled. G3/4 neutropenia developed in 3.8% of each arm, and the most common G3/4 nonhematological toxicities were anorexia and fatigue. Vomiting and tearing were more frequent with S-1 and hand-foot syndrome with capecitabine. The primary endpoint, the overall response rate, was 26.4% (14/53, 95% confidence interval: 14.5-38.3%) in S-1 and 24.1% (13/54, 95% confidence interval: 12.7-35.5%) in capecitabine, both of which exceeded the null hypothesis response rate of 10%. The median time to progression (TTP; 3.2 vs. 3.4 months, P=0.813) and overall survival (OS; 8.5 vs. 10.3 months, P=0.691) were similar in both arms. CYP2A6 polymorphisms were associated with S-1 efficacy. In the S-1 arm only, patients with CYP2A6 variant/variant alleles had worse TTP and OS than those with wild/wild or wild/variant alleles, and in multivariate analysis, the CYP2A6 genotype was predictive for TTP and OS. Both S-1 and capecitabine were active and tolerable for elderly MGC patients. The CYP2A6 genotyping might guide treatment selection.