Abstract
BACKGROUND: In the JCOG1108/WJOG7312G trial, a combination (FLTAX) of 5-fluorouracil (FU) /leucovorin (FL) and paclitaxel (PTX) did not show superiority in overall survival (OS) to FL in untreated patients with severe peritoneal metastasis of gastric cancer (GC-SPM), some of whom received second-line chemotherapy with PTX after FL. This post hoc study aimed to investigate the clinical implications of using both FU and PTX either sequentially or in combination for patients with GC-SPM. METHODS: A total of 94 patients were enrolled and categorized into the following three subgroups: patients treated with (1) FL followed by PTX (FL/PTX, N = 25), (2) FL followed by best supportive care (BSC) (FL/BSC, N = 21), and (3) FLTAX (N = 48). OS was compared between the subgroups. By comparing baseline factors between the FL/PTX and FL/BSC subgroups, factors preventing the sequential use of PTX (SUP) were explored using logistic regression model. The efficacy of FL and FLTAX was compared according to the presence of risk factors preventing SUP. RESULTS: The FL/PTX subgroup showed better and equivalent OS compared to the FL/BSC (median 7.8 vs. 2.0 months, p < 0.01) and FLTAX (median 7.8 vs. 8.0, p = 0.49) subgroups, respectively. Glasgow Prognostic Score 2 and initially unresectable disease were identified as risk factors preventing SUP. Absence of both risks predicted SUP with a sensitivity of 13% and a specificity of 100%, whereas absence of any risks predicted SUP with a sensitivity of 67% and a specificity of 62%. FLTAX showed better OS than FL in patients with one or two of these risks but worse OS in those with none. CONCLUSIONS: Although sequential use of FU and PTX showed equivalent survival to FLTAX in patients with GC-SPM, FLTAX might be preferable given the difficulty in selecting patients likely to receive sequential use at the initiation of first-line chemotherapy.