The clinical application value of body composition in predicting the prognosis of rectal cancer.

Abstract

Background: While computed tomography (CT)-based body composition has been studied for prognostic prediction in colorectal cancer, specific analyses for rectal cancer patients remain limited. This study aimed to investigate the relationship between CT-derived body composition indices and long-term postoperative outcomes in rectal cancer patients and to develop corresponding predictive models.

Methods: In this multicenter retrospective study, 696 patients who underwent radical surgery for rectal cancer between 2018 and 2021 were enrolled. Skeletal muscle index (SMI) and subcutaneous adipose tissue index (SATI) were calculated from preoperative CT scans at the third lumbar vertebra. Univariate and multivariate Cox regression analyses were performed to identify independent prognostic factors for recurrence-free survival (RFS) and overall survival (OS). Nomogram prediction models were constructed based on significant factors and validated internally using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).

Results: A total of 96 (13.8%) patients experienced recurrence and 89 (12.8%) died during follow-up. Multivariate analysis identified low SMI and high SATI as independent predictors of both poorer RFS (SMI: HR = 0.329, 95% CI 0.182-0.595; SATI: HR = 2.717, 95% CI 1.505-4.905) and OS (SMI: HR = 0.132, 95% CI 0.053-0.330; SATI: HR = 3.542, 95% CI 1.739-7.211), along with advanced T and N stages.Query The developed nomograms demonstrated good predictive accuracy. For RFS prediction, the area under the curve (AUC) values were 0.862, 0.846, and 0.824 for 3-, 4-, and 5-year predictions in the training set, and 0.825, 0.866, and 0.838 in the validation set. For OS prediction, the AUCs were 0.886, 0.898, and 0.875 (training set), and 0.876, 0.912, and 0.877 (validation set). Calibration curves and DCA indicated favorable model performance and clinical utility.

Conclusion: CT-derived body composition, specifically SMI and SATI, is associated with postoperative RFS and OS in rectal cancer patients. The established nomograms, integrating these indices with tumor stage, provide a valuable and individualized tool for prognostic assessment.