Intraoperative corticosteroid administration for resectable gastric cancer: a multicenter, randomized, open-label, phase II/III study.

Background: Excessive surgical stress induces inflammatory cytokine release, negatively impacting prognosis in patients with malignancies. This study aimed to determine whether the anti-inflammatory effect of a corticosteroid (CS) would improve prognosis when administered intraoperatively to patients with resectable gastric cancer.

Methods: In this multicenter, randomized, open-label, phase II/III study, patients with cStage II-III gastric cancer were randomized to CS administration or non-administration (control) groups. Patients in the CS group received 5 mg/kg of methylprednisolone just before the skin incision during surgery. The primary endpoints were the highest postoperative serum level of C-reactive protein (CRP_max) in the phase II portion, and recurrence-free survival (RFS) in the phase III portion.

Results: Between December 2016 and February 2019, 410 patients were enrolled. In the phase II portion, CRP_max was significantly lower in the CS group than in the control group (mean, 10.7 vs 14.3 mg/dL, respectively; P = 0.009). In the phase III portion, 3-year RFS rates in the CS (n = 202) and control (n = 204) groups were 67.2% and 63.0%, respectively, indicating no significant difference (hazard ratio, 0.807 [95% confidence interval, 0.590-1.105]; log-rank P = 0.182). Subgroup analysis showed that both histological type and clinical stage had significant interactions with RFS, suggesting a potential survival benefit of CS administration in patients with differentiated histological-type or cStage III gastric cancer.

Conclusions: Intraoperative CS administration mitigated postoperative CRP elevation but did not result in significantly improved survival in patients with cStage II-III gastric cancer. The study is registered with UMIN-CTR, number UMIN000024465.