e15578 Background: Although dCRT is the standard treatment for pts who have locally advanced unresectable EC or refuse surgery, the prognosis of these pts remains dismal. There are urgent needs to develop the novel treatment strategy based on prognostic stratification after dCRT. Methods: A total of 382 pts with locoregional EC without distant metastasis except for supraclavicular lymph node who received dCRT at Asan Medical Center in South Korea from 2006 to 2015 were included. Overall survival (OS) and progression-free survival (PFS) were analyzed using Kaplan-Meier method. Risk factors were analyzed using Cox regression. Risk scores were calculated by multiplying coefficients in Cox proportional hazard model. Results: Baseline characteristics were as follows: median age = 66 yrs (range: 40-85); male = 359 pts (94.0%); squamous cell carcinoma = 375 (98.2%); cTNM stage (AJCC 8th) I = 40 (10.5%), II = 122 (31.9%), III = 128 (33.5%), IV = 92 (24.1%). During median follow-up of 52.9 mo, median PFS was 13.5 mo (95% CI, 10.9-16.1), and median OS was 26.7 mo (95% CI, 19.8-33.7). In the univariate analyses, sex (only for PFS), weight loss (≥ 5 kg) during dCRT, cT stage, cN stage, cTNM stage, clinical response after dCRT, reason for dCRT were significant prognostic factors for PFS and OS. In the multivariate analyses, clinical response after dCRT, cTNM stage, and weight loss were independent prognostic factors for PFS and OS (Table). Risk-scoring model using these factors stratified pts into four groups: for median PFS (p < 0.0001), group 1 = 58.2 mo (95% CI, 43.5-73.0), group 2 = 17.0 mo (95% CI, 11.9-22.1), group 3 = 9.0 mo (95% CI, 7.0-11.1), and group 4 = 3.9 mo (95% CI, 3.7-4.2); for median OS (p < 0.0001), group 1 = 106.2 mo (95% CI, 44.9-167.6), group 2 = 38.0 mo (95% CI, 24.4-51.5), group 3 = 13.0 mo (95% CI, 8.5-17.6), and group 4 = 8.0 mo (95% CI, 7.4-8.6). Conclusions: In dCRT-treated locoregional EC pts, survival outcome significantly varied according to baseline clinical stage, treatment response, and dynamic change in body weight. Different treatment and surveillance strategies based on the risk score might be needed in these pts.[Table: see text]