Abstract
OBJECTIVE: (18)F-FDG PET is widely used to accurately stage numerous types of cancers. Although (18)F-FDG PET/CT features of tumors aid in predicting patient prognosis, there is increasing interest in mining additional quantitative body composition data that could improve the prognostic power of (18)F-FDG PET/CT, without additional examination costs or radiation exposure. The aim of this study was to determine the association between overall survival and body composition metrics derived from routine clinical (18)F-FDG PET/CT examinations. METHODS: Patients who received baseline (18)F-FDG PET/CT imaging during workup for newly diagnosed esophageal adenocarcinoma (EAC) were included. From these studies, psoas cross-sectional area (CSA), muscle attenuation (MA), SUV(mean), and SUV(max) were obtained. Correlation with overall survival was assessed using a Cox Proportional Hazards model, controlling for age, body mass index, (18)F-FDG dose, glucose level, diabetes status, in-hospital status, and tumor stage. RESULTS: Among the 59 patients studied, psoas MA and SUV(max) were found to be significant predictors of survival (HR 0.94, 95% CI 0.88-0.99, p = 0.04, and HR 0.37, 95% CI 0.14-0.97, p = 0.04, respectively) and remained independent predictors. Psoas CSA and SUV(mean) did not significantly influence survival outcomes. CONCLUSIONS: Characterization of psoas muscles as a surrogate marker for sarcopenia on baseline (18)F-FDG PET/CT imaging is relatively easily obtained and may offer additional prognostic value in patients with EAC.