Background: With increasing interest in organ-preserving strategies for potentially curable esophageal cancer, real-world data is needed to understand the impact of pathological tumor response after neoadjuvant chemoradiotherapy (CRT) on patient outcome. The objective of this study is to assess the association between pathological tumor response following CROSS neoadjuvant CRT and long-term overall survival (OS) in a nationwide cohort. Material and methods: All patients diagnosed in the Netherlands with potentially curable esophageal cancer between 2009 and 2017, and treated with neoadjuvant CRT followed by esophagectomy were included. Through record linkage with the nationwide Dutch Pathology Registry (PALGA), pathological data were obtained. The primary outcome was pathological tumor response based on ypTNM, classified into pathological complete response (ypT0N0) and incomplete responders (ypT0N+, ypT+N0, and ypT+N+). Multivariable logistic and Cox regression models were used to identify predictors of pathological complete response (pCR) and survival. Results: A total of 4946 patients were included. Overall, 24% achieved pCR, with 19% in adenocarcinoma and 42% in squamous cell carcinoma. Patients with pCR had a better estimated 5-year OS compared to incomplete responders (62% vs. 38%, p< .001). Of the patients with incomplete response, ypT+N+ patients (32% of total population) had the lowest estimated 5-year OS rate, followed by ypT0N+ and ypT+ N0 (22%, 47%, and 49%, respectively, p< .001). Adenocarcinoma, well to moderate differentiation, cT3-4, cN+, signet ring cell differentiation and lymph node yield (≥15) were associated with lower likelihood of pCR. Conclusion: In this population-based study, pathological tumor response based on the ypTNM-stage was associated with different prognostic subgroups. A quarter of patients achieved ypT0N0 with favorable long-term survival, while one-third had an ypT+N+ response with very poor survival. The association between pathological tumor response and long-term survival could help in more accurate assessments of individual prognosis and treatment decisions.