Risk prediction of esophageal squamous cell carcinoma recurrence in patients who underwent esophagectomy after receiving neoadjuvant treatment: a nationwide retrospective study in Japan

Background: Although neoadjuvant treatment (NAT) has become the standard of care for patients with locally advanced esophageal cancer, the risk of recurrence remains high. The risk must be predicted accurately, so that appropriate adjuvant therapy can be planned. We aimed to develop a model predicting recurrence of esophageal squamous cell carcinoma (ESCC) in patients who received NAT before esophagectomy. Methods: This nationwide study included 3874 patients from 85 institutions. Patients who underwent NAT and then surgery for ESCC were eligible. We developed a Cox proportional hazards model and created a nomogram to predict disease recurrence after NAT and curative esophagectomy. Results: Of the patients, 268 (6.9%), 1280 (33.0%), 2006 (51.8%), and 320 (8.3%) had clinical stage I, II, III, and IV tumors, respectively. The 5-year recurrence rate was 45.1% (95% confidence interval 43.4%–46.7%). Multivariable analysis revealed that body mass index, type of neoadjuvant treatment, primary tumor location, operative blood loss, pathological tumor stage, pathological therapeutic effect, and leakage were independently associated with disease recurrence. Using 13 commonly measured perioperative variables, we created a predictive nomogram, and the area under the curve was 0.783 (95% confidence interval 0.766–0.800). This nomogram was also adequately validated internally and had excellent calibration capacity (calibration slope, 0.992). Conclusions: The model developed in this study adequately predicted ESCC recurrence in patients who underwent NAT and then esophagectomy. Further research with this nomogram is needed to assess the effect of adjuvant therapy in patients at high risk for recurrence.