Dose-Volume Predictors for Radiation Esophagitis in Breast Cancer Patients Undergoing Hypofractionated Regional Nodal Radiotherapy

Abstract

Radiation esophagitis (RE) is often overlooked in breast cancer radiotherapy. This study aimed to assess the incidence and dose-volume predictors of RE in breast cancer patients undergoing hypofractionated regional nodal irradiation (RNI). Eligible patients were included who received intensity-modulated radiotherapy (RT) at the chest wall, the supraclavicular/infraclavicular fossa, level II axilla, and/or the internal mammary chain after mastectomy. The prescribed dose was 43.5 Gy in 15 fractions. The dose constraint for the esophagus was maximum dose <48 Gy. RE was evaluated weekly during RT and at 1 and 2 weeks, followed by 3 and 6 months after RT, and was graded according to the Common Toxicity Criteria for Adverse Events v3.0. The esophagus was contoured from the lower border level of the cricoid cartilage to the lower margin of the aortic arch. Esophageal total volume, mean dose (Dmean), maximum dose (Dmax), and the relative and absolute volumes receiving at least 5-45 Gy by 5 Gy increments (RV5-RV45 and AV5-AV45) were evaluated. Univariable and multivariable logistics regression analyses were performed to determine risk factors for RE, and receiver operating characteristic curves were obtained to identify the thresholds of esophageal dosimetric parameters. In total, 298 patients were included between May 8, 2020 and January 5, 2022 (minimum post-RT follow-up: 6 months). A total of 153 (51.3%) patients had left-sided breast cancer and 145 (48.7%) patients received internal mammary nodal irradiation (IMNI). Grade 2 and 3 RE incidence was 40.9% (122/298) and 0.3% (1/298), respectively. No grade 4 or 5 RE was observed. All RE cases resolved within 1 month after RT, and the median duration of RE was 3 weeks (range, 1-5). Based on univariable analyses, tumor laterality (p < .001), IMNI (p = .056) and esophageal Dmean, Dmax, RV10-RV40, and AV10-AV40 were risk factors of ≥grade 2 RE. Esophageal RV10-RV40 and AV35-AV40 were significantly associated with the risk of ≥grade 2 RE after adjusting for tumor laterality and IMNI. Based on multivariable analyses, RV25 and AV35 were optimum dose-volume predictors for ≥grade 2 RE at thresholds 20% for RV25 (35.9% vs. 60.9%, p = .04) and 0.27 mL for AV35 (31.0% vs. 54.6%, p = .04). RE is common in breast cancer patients undergoing hypofractionated RNI. With the same esophageal contouring standard, maintaining the upper esophageal V25 at <20% and V35 at <0.27 mL may decrease the risk of RE and improve the quality of life of patients.