Background: Three randomised trials (GEST, JACCRO PC-01, and GEMSAP) were conducted to evaluate the efficacy of gemcitabine plus S-1 (GS) vs gemcitabine alone in patients with advanced pancreatic cancer (PC). In this pooled analysis, the efficacy and safety of GS vs gemcitabine were evaluated.
Methods: Additional follow-up was conducted and survival data were updated in each study. A total of 770 patients (gemcitabine 389; GS 381) were included in the pooled analysis. The efficacy and safety data were analysed according to disease extent: locally advanced PC (LAPC) or metastatic PC (MPC).
Results: There were 738 (95.8%) overall survival events. In patients with LAPC (n=193), the median survival was 11.83 months for gemcitabine and 16.41 months for GS (hazard ratio (HR)=0.708; 95% confidence intervals (CI), 0.527-0.951; P=0.0220). In patients with MPC (n=577), the median survival was 8.02 months for gemcitabine and 9.43 months for GS (HR=0.872; 95% CI, 0.738-1.032; P=0.1102). The rate of grade 3/4 toxicity (rash and thrombocytopenia in LAPC; rash, diarrhoea, vomiting, and neutropaenia in MPC) was significantly higher for GS than for gemcitabine.
Conclusions: Gemcitabine plus S-1 is a viable treatment alternative to gemcitabine, which is one of the standard treatments in patients with LAPC.