Background Muscle depletion is characterized by reduced muscle mass (myopenia), and increased infiltration by intermuscular and intramuscular fat (myosteatosis). This study examined the role of particular body composition profiles as prognostic markers for patients with colorectal cancer undergoing curative resection. Methods Patients with colorectal cancer undergoing elective surgical resection between 2006 and 2011 were included. Lumbar skeletal muscle index (LSMI), visceral adipose tissue (VAT) surface area and mean muscle attenuation (MA) were calculated by analysis of CT images. Reduced LSMI (myopenia), increased VAT (visceral obesity) and low MA (myosteatosis) were identified using predefined sex‐specific skeletal muscle index values. Univariable and multivariable Cox regression models were used to determine the role of different body composition profiles on outcomes. Results Some 805 patients were identified, with a median follow‐up of 47 (i.q.r. 24·9–65·6) months. Multivariable analysis identified myopenia as an independent prognostic factor for disease‐free survival (hazard ratio (HR) 1·53, 95 per cent c.i. 1·06 to 2·39; P = 0·041) and overall survival (HR 1·70, 1·25 to 2·31; P < 0·001). The presence of myosteatosis was associated with prolonged primary hospital stay (P = 0·034), and myopenic obesity was related to higher 30‐day morbidity (P = 0·019) and mortality (P < 0·001) rates. Conclusion Myopenia may have an independent prognostic effect on cancer survival for patients with colorectal cancer. Muscle depletion may represent a modifiable risk factor in patients with colorectal cancer and needs to be targeted as a relevant endpoint of health recommendations. Myopenia affects survival