Reduced Severe Toxicities in Elderly Esophageal Cancer Patients Treated with Intensity Modulated Radiation Therapy: A Population-Based Analysis

Abstract

Intensity-modulated radiation therapy (IMRT) is increasingly being utilized in esophageal cancer (EC). Given the dosimetric advantage of IMRT, the purpose of this study was to evaluate the overall toxicity rate of EC patients treated with IMRT or 3D conformal radiation therapy (CRT) in a population-based analysis. We analyzed 2,578 EC patients aged >65 years using the combination of two claims databases between 2002 and 2009: the Surveillance, Epidemiology, and End Results (SEER)-Medicare and the Texas Cancer Registry (TCR)-Medicare databases. Nine adverse events after radiation therapy were identified: heart failure (HF), myocardial infarction (MI), acute respiratory distress syndrome (ARDS), pneumonias, atrial fibrillation (AF), pleural effusion (PE), pericardial effusion (PCE), unscheduled hospital admission (UHA), and emergency department (ED) visits. Chi-square analysis was utilized to investigate the associations between treatment and patients’ clinical parameters. Kaplan-Meier curves were performed to predict time to onset of each of the adverse effects. Three groups of adverse effects were categorized through the combination of the nine adverse effects: cardiac-relevant toxicities group (HF, MI, AF and PCE), pulmonary-relevant toxicities group (ARDS, pneumonias and PE), and hospitalized-relevant toxicities group (ED visits and UHA). Cox regression models were applied to compare the three groups of adverse events, respectively. Of 2,578 EC patients identified, IMRT and CRT groups were well balanced on baseline clinical information including age, gender, race, stage, tumor grade and pretreatment diseases. IMRT was associated with reduced UHA rate and lower overall adverse events when compared with patients treated with CRT (p < 0.05). Kaplan-Meier analysis showed no significant difference in time to the onset of any adverse effect when treated with IMRT vs. CRT. For time to cardiac-relevant toxicities analysis using the Cox regression modeling method, age, stage and Charlson comorbidity index (CCI) were independent prognostic factors (p < 0.05). For time to pulmonary-relevant toxicities analysis, CCI and patients undergoing surgery after radiotherapy were two independent prognostic factors (p < 0.05). Age, race, and CCI were found as independent indicators in time to hospitalized-relevant toxicities analysis (p < 0.05). Compared with EC patients treated with CRT, patients treated with IMRT had reduced unscheduled hospital admission rate and lower overall adverse events. Future study will expand the sample size of IMRT and include younger patients to further evaluate the comparison of survival, toxicity-relevant factors between IMRT and CRT treated patients.