P-61 Impacts of salvage chemotherapy after nivolumab therapy (NIVO): A REVIVE substudy

The primary endpoints met in the REVIVE study have been previously reported, demonstrating the chemotherapeutic efficacy after the progression on NIVO therapy in patients with advanced gastric cancer (AGC). Therefore, the current study evaluated the survival from the NIVO therapy initiation in all registered participants of the REVIVE trial. The REVIVE trial was a prospective, multicenter, observational study evaluating the efficacy and safety of chemotherapy for NIVO-refractory or NIVO-intolerant patients with AGC (UMIN000032182). We primarily register the patients who underwent NIVO therapy as primary registration, and the patients formally were registered as formal registration. The previously reported main study analyzed data of formally registered patients who underwent chemotherapy with irinotecan, trifluridine/tipiracil hydrochloride, or oxaliplatin combination regimens. In this study, patients who discontinued NIVO therapy for any reason at data cutoff among primarily registered patients were selected. The survival of patients who received the best supportive care (cohort A) was compared to those included in the main study (cohort B). Of 395 primarily registered patients, 108 patients in cohort A and 199 patients in cohort B were included, respectively. Those receiving other chemotherapeutic regimens (N = 47) or continuing NIVO therapy (N = 38) were excluded. Median overall survival (OS) and time to treatment failure (TTF) were 9.3 (95% confidence interval [CI], 8.3–10.2) and 1.8 (95%CI, 1.6–2.2) months at 234 and 307 events, respectively, from the initiation of NIVO therapy in the whole population. The objective response rate (ORR) and disease control rate (DCR) were 9.1% and 43.0%, respectively. Patients in cohort B had significantly better prognosis in OS (median, 12.2 vs. 4.8 months; hazards ratio [HR], 0.43 [95%CI 0.34–0.57]; p < 0.01). However, the difference in short-term efficacies was not observed: ORR, 6.5% vs. 10.6%; DCR, 38.0% vs. 45.7%; and median TTF, 1.9 vs. 1.8 months (HR, 1.06 [95%CI 0.84–1.35]; p = 0.62) (cohort A vs. B). The post-progression survival (PPS) from the date of NIVO therapy discontinuation was significantly better in cohort B than in cohort A (median PPS, 8.1 vs. 1.9 months; HR, 0.22 [95%CI 0.17–0.30]; p < 0.01). The proportion of patients who received the best supportive care after NIVO therapy was similar, regardless of the effectiveness of NIVO therapy (transition rate: 25% in complete response or partial response [responders, N = 28] and 33% in stable or progressive disease [non-responders, N = 267]). The difference of OS from the initiation of NIVO therapy in responders was not observed between cohorts A and B; however, the OS of cohort B in non-responders was significantly longer than that of cohort A (median OS, 10.8 vs. 4.8 months; HR, 0.60 [95%CI 0.52–0.69]; p < 0.01). Salvage chemotherapy as much as possible after NIVO therapy could improve the AGC prognosis.