Clinical biomarkers in adjuvant chemotherapy for gastric cancer after D2 dissection by a pooled analysis of individual patient data from large randomized controlled trials.

Tsuburaya, Akira; Guan, Jia; Yoshida, Kazuhiro; Kobayashi, Michiya; Yoshino, Shigefumi; Tanabe, Kazuaki; Yoshikawa, Takaki; Oshima, Takashi; Miyashita, Yumi; Sakamoto, Junichi; Tanaka, Shiro

BACKGROUND Adjuvant therapy for gastric cancer is a standard among the world with no regimen selection criteria. Also, prognostic factors except for tumor staging have not been established. We aimed to identify prognostic and predictive markers for gastric cancer adjuvant therapy from large randomized controlled trials with standard lymph node dissection. METHODS Three studies: ACTS-GC, CLASSIC, and SAMIT were found and selected for a pooled analysis, following PRISMA guideline. The integrity of individual participant data (IPD) was verified in the eligible 3527 patients registered, and fixed-effect model was used. The primary endpoint was relapse-free survival (RFS) and the secondary endpoint was overall survival (OS). RESULTS Age was a significant prognostic factor in addition to tumor stages both in "surgery alone" and "adjuvant" groups. Adjuvant therapy was effective for every TN stage; however, it tended to be more effective in T1-2 than in T3-4. Also, it was more effective in low- or middle-BMI than in high-BMI group with Hazard ratio [HR]s: 0.58, 0.58, and 1.05, respectively. Capecitabine plus oxaliplatin (CAPOX) was more effective than S-1 for T1-2, N2-3, and differentiated type with HRs between 0.59 and 0.70, but with no difference among TNM stages. Combining histology to TN; the HRs in differentiated T1-2 N1-3 groups were between 0.29 and 0.45. For T3-4 N0-1 group, S-1 was likely to be effective, not significant. CONCLUSIONS Age is a significant prognostic factor both in surgery alone and adjuvant group. CAPOX is more effective for differentiated T1-2 tumors with lymph node metastasis.