Prognostic significance of sarcopenia diagnosed based on the anthropometric equation for progression-free survival and overall survival in patients with colorectal cancer.

Abstract

The purpose of this study was to investigate the prognostic significance of sarcopenia diagnosed based on anthropometric equations for progression-free survival (PFS) and overall survival (OS) in patients with colorectal cancer (CRC). A total of 1,441 CRC patients who underwent surgical treatment between January 2012 and December 2016 were enrolled in this study. Sarcopenia was diagnosed according to validated anthropometric equations. The Kaplan-Meier method with the log-rank test was used to estimate the survival curve. Cox proportional hazards regression models with forward selection were used to evaluate risk factors affecting the prognosis of CRC patients. R package "survival" was used to build the prognostic nomograms to predict 1-5 years of PFS and OS in CRC patients. The concordance index (C-index) and calibration curve were used to evaluate the prognostic accuracy of the prognostic nomogram. Two hundred and seventy-one patients (18.8%) were diagnosed with sarcopenia. Sarcopenia was significantly associated with advanced age, large tumor size, and high mortality. Compared with the non-sarcopenia patients, the PFS of sarcopenia patients was worse (5-year PFS, 48.34 vs. 58.80%, p = 0.003). Multivariate survival analysis showed that patients with sarcopenia had a higher risk (23.9%) of adverse PFS (HR, 1.239; 95%CI: 1.019-1.505, p = 0.031) than patients without sarcopenia. The OS of patients with sarcopenia was significantly worse than that of patients without sarcopenia (5-year OS: 50.92 vs. 61.62%, p = 0.001). In CRC patients, sarcopenia was independently associated with poor OS (HR: 1.273, 95%CI: 1.042-1.556, p < 0.001). Moreover, sarcopenia effectively differentiated the OS of CRC patients in the normal carcinoembryonic antigen (CEA) subgroup but not in the high CEA subgroup. Notably, sarcopenia can provide effective prognostic stratification in CRC patients at different pathological stages. Nomograms that integrated prognostic features were built to predict the risk of adverse outcomes in CRC patients. The C-index and calibration curves showed that these nomograms had good prediction accuracy. Internal validation confirmed that our nomogram has wide application potential. Sarcopenia diagnosed based on anthropometric equations is an independent risk factor for PFS and OS in CRC patients.