Background: Two randomized phase III trials of first-line chemotherapy for advanced gastric cancer (JCOG9205 and JCOG9912) conducted by the Japan Clinical Oncology Group used 5-fluorouracil continuous infusion (5-FUci) as the control arm. New active agents (e.g., S-1, irinotecan, and taxanes) were introduced as second-line chemotherapy in the late 1990s after JCOG9205. This combined analysis evaluated whether patients in the 5-FUci arm of JCOG9912 exhibited better survival after adjusting for baseline factors and also investigated the cause of survival prolongation.
Patients and methods: The subjects were patients assigned to the 5-FUci arms who met the eligibility criteria of both JCOG9205 and JCOG9912. Overall survival (OS), time to treatment failure (TTF), and survival after treatment failure in the first-line chemotherapy (OS-TTF) were compared after adjusting baseline characteristics using the Cox proportional hazard model. Second-line chemotherapy details were also reviewed.
Results: The combined analysis included 89 and 230 patients in JCOG9205 and JCOG9912, respectively. After adjusting baseline characteristics, TTF was similar between groups (HR 0.95; 95 % CI, 0.73-1.26). However, both OS (HR, 0.74; 95 % CI, 0.56-0.99) and OS-TTF (HR, 0.76; 95 % CI, 0.57-1.01) were longer in JCOG9912. More patients in JCOG9912 received second-line chemotherapy (83 vs. 52 %) with new drugs (77 vs. 10 %) than in JCOG9205. OS-TTF was substantially prolonged in patients who received second-line chemotherapy (HR, 0.66; 95 % CI, 0.46-0.95).
Conclusion: OS and OS-TTF were longer in JCOG9912 than JCOG9205. Second-line chemotherapy with new drugs is a potential reason for the observed prolongation of survival.