Sequential paclitaxel followed by tegafur and uracil (UFT) or S-1 versus UFT or S-1 monotherapy as adjuvant chemotherapy for T4a/b gastric cancer (SAMIT): a phase 3 factorial randomised controlled trial

Background The prognosis for locally advanced gastric cancer is poor despite advances in adjuvant chemotherapy. We did the Stomach cancer Adjuvant Multi-Institutional group Trial (SAMIT) to assess the superiority of sequential treatment (paclitaxel then tegafur and uracil [UFT] or paclitaxel then S-1) compared with monotherapy (UFT or S-1) and also the non-inferiority of UFT compared with S-1. Methods We did this randomised phase 3 trial with a two-by-two factorial design at 230 hospitals in Japan. We enrolled patients aged 20-80 years with T4a or T4b gastric cancer, who had had D2 dissection and a ECOG performance score of 0-1. Patients were randomly assigned to one of four treatment groups with minimisation for tumour size, lymph node metastasis, and study site. Patients received UFT only (267 mg/m(2) per day), S-1 only (80 mg/m(2) per day) for 14 days, with a 7-day rest period or three courses of intermittent weekly paclitaxel (80 mg/m(2)) followed by either UFT, or S-1. Treatment lasted 48 weeks in monotherapy groups and 49 weeks in the sequential treatment groups. The primary endpoint was disease-free survival assessed by intention to treat. We assessed whether UFT was non-inferior to S-1 with a non-inferiority margin of 1 33. This trial was registered at UMIN Clinical Trials Registry, number C000000082. Findings We randomly assigned 1495 patients between Aug 3, 2004, and Sept 29, 2009. 374 patients were assigned to receive UFT alone, 374 to receive S-1 alone, 374 to received paclitaxel then UFT, and 373 to receive paclitaxel then S-1. We included 1433 patients in the primary analysis after at least 3 years of follow-up (359, 364, 355, and 355 in each group respectively). Protocol treatment was completed by 215 (60%) patients in the UFT group, 224 (62%) in the S-1 group, 242 (68%) in the paclitaxel then UFT group, and 250.(70%) in the paclitaxel then S-1 group. 3-year disease-free survival for monotherapy was 54 0% (95% CI 50.2-57.6) and that of sequential treatment was 57.2% (53.4-60.8; hazard ratio [HR] 0.92, 95% CI 0.80-1.07, p=0.273). 3-year disease-free survival for the UFT group was 53.0% (95% CI 49.2-56.6) and that of the S-1 group was 58 2% (54.4-61.8; HR 0.81, 95% CI 0.70-0.93, p=0.0048; p(non-inferiority)=0.151). The most common grade 3-4 haematological adverse event was neutropenia (41 [11%] of 359 patients in the UFT group, 48 [13%] of 363 in the S-1 group, 46 [13%] of 355 in the paclitaxel then UFT group, and 83 [23%] of 356 in the paclitaxel then S-1 group). The most common grade 3-4 non-haematological adverse event was anorexia (21 [6%], 24 [7%], seven [2%], and 18 [5%], respectively). Interpretation Sequential treatment did not improve disease-free survival, and UFT was not non-inferior to S-1 (and S-1 was superior to UFT), therefore S-1 monotherapy should remain the standard treatment for locally advanced gastric cancer in Japan.