P1.227. A Composite Body Composition Score Independently Predicts Long-Term Survival After Esophagectomy for Cancer

Abstract

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – long term outcomes Background Body composition (BC) measures are established determinants of esophageal cancer outcomes. Beyond known survival predictors such as cancer stage and performance status, a composite BC score independently predicts post-resection survival in lung cancer. We adapted this score for esophageal cancer patients to evaluate its association with long-term survival following esophagectomy. Methods This is a single-institution retrospective analysis of patients with esophageal cancer who underwent resection from 2000 to 2014. Visceral fat area, vertebral trabecular bone density, and normal-density skeletal muscle area were measured at the L3 vertebral level on preoperative computed tomography scans using a semi-automated technique. Measurements were normalized for age, sex, and height and divided into quartiles; the lowest quartile was designated as abnormal. We adapted an established BC scoring system originally developed for lung cancer patients; one point was assigned to each component that fell in the lowest quartile, and points were summed to generate a composite BC score (range 0-3). Scores were grouped as 0+1 and 2+3 for analysis. The relationship between BC score and overall survival was assessed using Cox proportional hazards modeling. Results Among 156 patients, 127 were male (81%), and the mean age was 63 ± 10 years. Induction therapy was administered to 109 patients (70%). Median follow-up was 5.4 years, and 109 patients (70%) had died at the time of last follow-up. BC scores were 0 in 42.3%, 1 in 41.7%, and 2 or 3 in 15.4% of patients. The distribution of individual BC components changed across BC scores; as BC score increased, the proportion attributed to visceral fat rose while that of normal-density skeletal muscle declined (Figure 1, p=0.0104). Patients with a higher BC score had significantly worse survival (Figure 2, p=0.026); 5-year survival was 57% for scores 0-1 versus 48% for scores 2-3. The BC score independently predicted long-term survival; patients with scores 0-1 had a 51% lower risk of death than those with scores 2-3 (HR 0.49, 95% CI 0.295-0.819, p=0006). Conclusion Patient body composition influences post-esophagectomy outcomes. A composite BC score derived from visceral fat, bone density, and skeletal muscle independently predicts long-term survival following esophagectomy for cancer. Incorporating this score into clinical practice may identify high-risk patients, guide personalized prehabilitation programs, support shared decision-making, and improve prognostic counseling for patients undergoing esophageal cancer treatment.