Comparison between neoadjuvant chemotherapy followed by surgery and definitive chemoradiotherapy for survival in patients with resectable esophageal squamous cell carcinoma (JCOG2305A).

417 Background: Neoadjuvant triplet chemotherapy followed by surgery (Neo-S) has become the standard treatment for resectable esophageal squamous cell carcinoma (ESCC) based on the results of JCOG1109. However, definitive chemoradiotherapy (CRT) remains an option for organ preservation with curative intent. This analysis aimed to explore subgroups of patients undergoing CRT, including salvage treatment, to identify those with survival outcomes potentially equivalent to Neo-S. Methods: Pooled data from two clinical trials of patients with resectable (T1N1-3M0 and T2-3NanyM0) ESCC. JCOG0909 was a single-arm trial that evaluated upfront CRT consisting of 5-FU (1000 mg/m 2 on day1-4, 29-32) and cisplatin (80 mg/m 2 on day1, 29) combined with radiotherapy at a dose of 50.4 Gy, including salvage treatment. If there was no progression at the time of salvage treatment, it was not regarded as an event for progression-free survival (PFS) in JCOG0909. JCOG1109 compared neoadjuvant treatment with doublet chemotherapy, triplet chemotherapy, and doublet chemotherapy with radiotherapy followed by surgery. The patients were enrolled in JCOG0909 between 2010 and 2014 and in JCOG1109 between 2012 and 2018. Subgroup analyses of clinical data, including tumor location, cT, cN, and clinical stage, were conducted on overall survival (OS) and PFS using Cox proportional hazards regression models for patients with JCOG0909 and JCOG1109. When a subgroup with a multivariable hazard ratio (HR) of more than 0.91 was detected, CRT including salvage treatment was considered not to be inferior to Neo-S in that subgroup. Results: Ninety-three patients in the CRT group from JCOG0909 and 196 patients in the Neo-S group from the triplet chemotherapy arm of JCOG1109 were included in this analysis. Although there were no significant differences, PFS and OS tend to be better in the Neo-S group than in the CRT group (PFS, HR 0.72, 95% confidence interval [CI] 0.50-1.03; OS, HR 0.89, 95% CI 0.59-1.36). For patients with Ut/Mt or cT1-2 disease, OS in the Neo-S group had a HR of 0.91 or more compared with those in the CRT group (Table). Conclusions: Subgroups in which OS after CRT including salvage treatment was not inferior to that of Neo-S were identified. Hazard ratios of Neo-S to CRT by subgroups in multivariable Cox regression. Progression-free survival Overall survival Subgroups Hazard ratio 95% confidence interval Hazard ratio 95% confidence interval Ut/Mt 0.85 0.56-1.28 1.13 0.70-1.83 Lt 0.45 0.21-0.97 0.44 0.19-1.00 cT1-2 0.88 0.46-1.68 1.45 0.65-3.24 cT3 0.66 0.42-1.02 0.72 0.44-1.16 cStage IB+II 0.79 0.45-1.36 1.06 0.56-2.00 cStage III 0.75 0.45-1.23 0.92 0.52-1.62 Overall 0.72 0.50-1.03 0.89 0.59-1.36 Ut upper thoracic esophagus, Mt middle thoracic esophagus, Lt lower thoracic esophagus.