JASPAC 01: Randomized phase III trial of adjuvant chemotherapy with gemcitabine versus S-1 for patients with resected pancreatic cancer.

Abstract

4008 Background: Adjuvant chemotherapy with gemcitabine (G) has been standard treatment for resected pancreatic cancer (PC). In the GEST study, S-1 (S) had shown non-inferiority to G in overall survival (OS) for unresectable PC. The aim of this phase III study is to investigate non-inferiority of S to G on OS as adjuvant chemotherapy for resected PC. Methods: Patients (pts) after macroscopically curative resection of PC with an ECOG PS of 0-1 and adequate organ functions were randomly assigned to G (1000 mg/m2, iv, d1, 8 and 15, q4w, for 6 courses) or S (80/100/120 mg/day based on BSA, po, d1-28, q6w, for 4 courses) with balancing by surgical margins (R), nodal status (N) and institution. Primary endpoint was OS. With 180 pts per arm, the study had 80% power to prove non-inferiority with a margin of hazard ratio (HR) 1.25 on the basis of expected HR 0.87, with 0.05 two-sided alpha. Secondary endpoints were relapse-free survival (RFS), safety, and quality of life (EQ-5D). One interim analysis was planned after 180 deaths. Results: From 4/2007 to 6/2010, 385 pts were enrolled from 33 hospitals in Japan. 378 pts (G/S: 191/187) were included in the full analysis set. Pts characteristics (G/S) were well balanced (PS0: 67%/70%, R0: 86%/88%, N0: 38%/36%). Based on the interim analysis with 205 OS events, IDMC recommended to publish the results. OS at 2-years were 53% for G and 70% for S. HR for S to G was 0.56 (95% CI, 0.42-0.74, p<0.0001 for non-inferiority, p<0.0001 for superiority). On subgroup analysis, HRs for R0/R1, N0/N1 pts were 0.57 (95% CI, 0.42-0.78)/0.53 (0.27-1.05), 0.48 (0.28-0.83)/0.58 (0.41-0.80), respectively. RFS at 2-years were 29% for G and 49% for S. HR of relapse for S to G was 0.56 (95% CI, 0.43-0.71, log-rank p<0.0001). Incidences of grade 3/4 toxicities in G/S were leukopenia 39%/9%, hemoglobin decrease 17%/13%, thrombocytopenia 9%/4%, elevated AST 5%/1%, fatigue 5%/5%, and anorexia 6%/8%. Relative dose intensity of G/S was 84%/97%. EQ-5D QOL score in S was significantly better than that in G (p<0.0001). Conclusions: S-1 adjuvant chemotherapy is shown non-inferior, and furthermore, even superior to GEM. S-1 is considered as the new standard treatment for resected PC pts. Clinical trial information: UMIN000000655.