Nomogram predicting short- and long-term outcomes in colon cancer based on CT body composition.

Low skeletal muscle index (SMI) and hypoalbuminemia were independent risk factors for postoperative complications in colon cancer patients undergoing radical resection.

• The original cohort comprised 475 patients (272 males, 203 females; mean age 64.8 ± 11.9 years).
• Postoperative complications occurred in 85 patients (17.8%) in the original cohort.
• Multivariate analysis identified low SMI as an independent risk variable for postoperative complications (P = 0.025).
• Hypoalbuminemia was also identified as an independent risk variable for postoperative complications (P = 0.048).
• Analysis was performed using logistic regression models.

Low IMAT (myosteatosis) was an independent prognostic variable associated with worse recurrence-free survival in colon cancer patients.

• Low IMAT was associated with worse RFS with HR 2.919 (95% CI: 1.423–5.985, P = 0.003).
• This finding was identified via Cox proportional hazards multivariate analysis.
• The median follow-up was 51 months (IQR, 37.5–62.25).
• IMAT refers to intermuscular adipose tissue measured on CT imaging.

High skeletal muscle index (SMI) was an independent prognostic variable associated with better recurrence-free survival in colon cancer patients.

• High SMI was associated with improved RFS with HR 0.450 (95% CI: 0.247–0.821, P = 0.009).
• This finding was identified via Cox proportional hazards multivariate analysis.
• The median follow-up was 51 months (IQR, 37.5–62.25).
• Both SMI and IMAT were derived from CT-based body composition analysis.

The nomogram developed for predicting recurrence-free survival demonstrated strong predictive performance in both the original and validation cohorts.

• In the original cohort (n = 475), AUCs for 1-, 3-, and 5-year RFS were 0.885, 0.867, and 0.868, respectively.
• In the validation cohort (n = 209, from Weihai Central Hospital Affiliated to Qingdao University), AUCs for 1-, 3-, and 5-year RFS were 0.784, 0.817, and 0.897, respectively.
• The nomogram was based on independent predictors of RFS identified by Cox proportional hazards regression.
• The nomogram was described as providing 'strong predictive performance for RFS.'

The study design was a retrospective two-cohort analysis of colon cancer patients who underwent radical surgical resection at two institutions.

• The original cohort included 475 patients from the Affiliated Hospital of Qingdao University.
• The validation cohort included 209 patients from Weihai Central Hospital Affiliated to Qingdao University.
• CT-based body composition parameters (SMI and IMAT) were the primary exposure variables.
• Cox proportional hazards and logistic regression models were used to analyze associations between body composition and outcomes.