ESTRO 2025 - Abstract Book

S1821

Clinical – Upper GI

ESTRO 2025

3275

Digital Poster Cervical Esophagus and Stomach Radiation Dose on Anastomotic Leaks in Esophageal Cancer after Neoadjuvant Chemoradiation and McKeown Esophagectomy Yizhe Cheng 1,2 , Chen Yang 1,2 , Biqi Chen 1,2 , Baoqing Chen 1,2 , Qiaoqiao Li 1,2 1 Radiation Oncology, Sun Yat-sen University Cancer Center, Guangzhou, China. 2 Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou, China Purpose/Objective: Neoadjuvant chemoradiation (nCRT) followed by esophagectomy is the standard treatment for locally advanced esophageal cancer. However, emerging evidence suggests that nCRT may contribute to higher postoperative morbidity. This study aimed to assess the influence of radiation doses to the cervical esophagus and stomach on the occurrence of anastomotic leaks and overall prognosis in patients with esophageal squamous cell carcinoma (ESCC) undergoing trimodal therapy. Material/Methods: We conducted a retrospective analysis of detailed clinical and dose-volume data from T3-4 or N-positive thoracic ESCC patients treated with nCRT followed by McKeown esophagectomy with cervical anastomosis. Receiver operating characteristic analysis was used to determine optimal dose constraints, while logistic regression identified risk factors for anastomotic leaks. Results: The cohort included 218 ESCC patients, with a 5-year follow-up period. Anastomotic leaks, the most common complication, occurred in 21% of patients (46 cases). Patients with leaks, especially those with intrathoracic manifestations, experienced significantly longer hospital stays and increased need for ventilator support. However, no significant differences were observed in 30-day or 90-day mortality, or overall survival between patients with and without anastomotic leaks. A mean radiation dose limit of 39.4 Gy to the cervical esophagus was identified. The incidence of anastomotic leak was 34.3% in patients receiving doses above 39.4 Gy, compared to 17.4% in those receiving lower doses ( P = 0.027). Multivariate analysis confirmed that exceeding this threshold was a strong predictor of anastomotic leaks, while radiation doses to the stomach showed no significant effect. Robot-assisted minimally invasive esophagectomy, operating time longer than 4.9 hours, and diabetes were also identified as predictors of anastomotic leak. Conclusion: For ESCC patients undergoing nCRT and McKeown esophagectomy, limiting the mean radiation dose to the cervical esophagus to 39.4 Gy may reduce the risk of anastomotic leaks. Although the presence of anastomotic leaks did not impact overall survival, it was associated with prolonged postoperative recovery.

Keywords: anastomotic leak, radiation dose, prognosis