Phase III randomized controlled trial to confirm efficacy of surgery plus postoperative dose-modified S-1 adjuvant chemotherapy against surgery alone for vulnerable elderly patients with pathological stage II/III gastric cancer: JCOG1507 (BIRDIE trial).

Abstract

347 Background: In Japan, the standard treatment for resectable gastric cancer is primary gastrectomy followed by adjuvant chemotherapy based on S-1. However, the survival benefit of adjuvant chemotherapy in elderly patients, particularly those considered vulnerable, remains uncertain. Methods: JCOG1507 (BIRDIE trial) was a phase III randomized controlled trial to confirm the superiority of dose-modified S-1 adjuvant chemotherapy to surgery alone for vulnerable elderly patients with pathological stage II/III gastric cancer. The trial was conducted at 45 institutions of the Stomach Cancer Study Group of the Japan Clinical Oncology Group. Vulnerability to S-1 was defined as: (1) weight loss <15% and creatinine clearance (Ccr) ≥30 mL/min after total gastrectomy, or (2) weight loss <15% and 30 ≤ Ccr <80 mL/min after the other gastrectomy. Patients in the control arm were followed without S-1, while those in the experimental arm received one year of dose-modified S-1 monotherapy at a reduced dose according to the Ccr and type of gastrectomy. The primary endpoint was overall survival (OS). Secondary endpoints included relapse-free survival (RFS), time-to-treatment-failure, treatment continuity, dose intensity, and safety. Assuming a 3-year OS of 55% in the control arm and a 10% improvement with the experimental arm, the sample size was set at 370 (one-sided α=0.05, power 75%). This trial was registered with UMIN (000025742). Results: Patient accrual began in January 2017, but the planned sample size was reduced to 170 due to slow accrual (α=0.1, power 70%), and enrollment was closed to 167 patients in January 2024. At the second planned interim analysis, conducted after completion of protocol treatment for all enrolled patients on June 21, 2025, the 3-year OS was 63.8% (95% CI, 51.0–74.1) in the control arm and 63.4% (95% CI, 50.6–73.7) in the experimental arm (stratified HR, 1.272; 95% CI, 0.768–2.107). As the probability of demonstrating superiority at the final analysis was considered low, the trial was terminated early due to futility. For RFS, the 3-year RFS was 52.5% (95% CI, 40.6–63.1) in the control arm and 57.9% (95% CI, 45.5–68.4) in the experimental arm (HR, 0.888; 95% CI, 0.567–1.389). Conclusions: Dose-modified S-1 adjuvant therapy did not improve OS in vulnerable elderly patients with pStage II/III gastric cancer. Surgery alone remains the standard treatment for this population. Clinical trial information: 000025742 .