A Multi-institutional Analysis of Dosimetric Predictors of Toxicity Following Trimodality Therapy for Esophageal Cancer

Abstract

Abstract Purpose This multi-institutional review explored associations between radiation dose-volume histogram (DVH) parameters and cardiopulmonary toxicities with trimodality therapy for esophageal cancer. Methods and Materials We reviewed 465 consecutive patients with esophageal cancer treated with chemoradiation therapy followed by surgery at 2 tertiary-care institutions between 2007 and 2013. Using logistic regression, we assessed associations between lung and heart DVH parameters and cardiopulmonary toxicities and survival. Statistically significant variables were subsequently included in multivariable models, which incorporated age, smoking history, previous history of heart disease, and type of chemotherapy. Results The median age of the patients was 61 years (interquartile range, 54-68 years), and 86% were men. At baseline, 60% of the patients had known cardiac risk factors, 64% were current or former smokers, and 10% had other pulmonary comorbidities. Most patients had stage II to III (96%) adenocarcinoma (94%) of the distal esophagus. The radiation therapy (RT) modalities used were 3-dimensional conformal RT (38%), intensity modulated RT (41%), and proton therapy (20%). An increased heart dose was associated with increased risk of cardiac toxicity on univariable analysis (V20 Gy: odds ratio [OR], 1.20; 95% CI, 1.08-1.33; P = .001) (V30 Gy: OR, 1.24; 95% CI, 1.11-1.38; P Conclusions We observed an association between heart dose and cardiac toxicity for esophageal cancer. The risk of cardiac toxicity was 5%, 10%, and 15% when the heart V30 Gy dose was 14%, 20%, and 30%, respectively. For every 10% increase in V30 Gy, there was a corresponding 24% increase in the relative risk of cardiac toxicity.