Sequential gemcitabine and platinum versus first-line combination of gemcitabine and platinum for advanced pancreatic cancer treatment: a retrospective study

The purpose of this study was to investigate the impact of combinational versus sequential gemcitabine and platinum on prognosis of advanced pancreatic cancer. ;Two hundred and three patients with advanced pancreatic cancer were selected. They were divided into GemP (first-line gemcitabine and platinum), Gem-then-P (sequential gemcitabine and platinum), Gem/other (first-line gemcitabine-based therapy without subsequent platinum), and Gem (first-line gemcitabine-based therapy without subsequent systemic therapy) groups. The Kaplan-Meier method and log-rank test were used for survival analyses. Cox regression model and propensity score matching were used for prognostic analyses. ;The median survival was 12.5 months [95 % confidence interval (CI), 11.2-13.7] in the GemP group (N = 65), 8.3 months (95 % CI 5.0-11.7) in the Gem-then-P group (N = 35), 11.6 months (95 % CI 4.6-18.5) in the Gem/other group (N = 26), and 4.7 months (95 % CI 3.3-6.0) in the Gem group (N = 77) (P < 0.001). Considering the GemP and Gem-then-P groups, performance status, serum creatinine, and response to first-line treatment were independent prognostic factors for overall survival in the multivariate analysis. No specific factors were identified for predicting the choice between GemP and Gem-then-P. ;First-line gemcitabine and platinum-based combinations were not superior to sequential gemcitabine and platinum for overall survival. The best sequence of chemotherapy for advanced pancreatic cancer should be explored in future clinical trials