Computed tomography-measured body composition parameters do not influence survival in non-metastatic colorectal cancer.

BACKGROUND: There is an increasing interest in studying the impact of altered body composition parameters and colorectal cancer (CRC) treatment outcomes. The aim of this study is to explore the impact of computed tomography (CT)-measured visceral obesity, sarcopenia and myosteatosis on survival of non-metastatic CRC. METHODS: Consecutive patients with stage I-III CRC who underwent curative-intent treatment between January 2010 and December 2015 were included. By measuring the visceral fat area, and the skeletal muscle index and radiodensity in the pre-operative staging CT, patients were classified as visceral obese, sarcopenic or myosteatotic. The associations between CT-based body composition parameters and survival were assessed using log-rank tests and a Cox regression analysis. RESULTS: Of 359 patients, 263 (73.3%) were visceral obese, 85 (23.7%) sarcopenic and 80 (22.3%) myosteatotic. Overall (OS), cancer-specific (CSS) and disease-free survivals (DFS) at 5 years were 78.8%, 84.7% and 75%, respectively. Myosteatosis and the combination of myosteatosis and visceral obesity were associated with a reduced DFS (hazard ratio 1.67; 95% confidence interval 1.06-2.61 and hazard ratio 1.85; 95% confidence interval 1.15-2.96, respectively). However, after performing a multivariate analysis including other relevant clinicopathological factors, none of the body composition parameters were associated with any long-term outcome measures, even after stratifying by cancer stage. CONCLUSIONS: CT-measured body composition parameters do not independently influence survival in non-metastatic CRC. There is a need for larger cohort studies with standardized patient selection and methodology to confirm these findings.