Prognostic model for early-onset colorectal cancer with liver metastasis after primary tumor resection and chemotherapy

This study aims to develop a nomogram for predicting the overall survival (OS) of early-onset colorectal cancer with liver metastasis (EOCRC-LM) patients who underwent primary tumor resection combined with chemotherapy. Using the SEER database (2010-2015), we identified EOCRC-LM underwent primary tumor resection and chemotherapy. Eligible cases were randomly split into training/validation cohorts. Least Absolute Shrinkage and Selection Operator(LASSO) regression and multivariate Cox analysis identified independent OS predictors, which were integrated into a nomogram. Model performance was assessed via calibration curves, time-dependent Receiver Operating Characteristic, and Decision Curve Analysis (DCA), demonstrating clinical utility. A cohort of 1049 early-onset colorectal cancer patients with liver metastases underwent tumor resection and chemotherapy was analyzed using SEER data (2010-2015), randomly divided into training (n = 734) and validation (n = 315) cohorts. LASSO and multivariate Cox regression identified eight independent prognostic factors: marital status, tumor location, T/N stage, CEA level, lymph node count, and bone/lung metastases. The nomogram showed strong predictive accuracy, with training cohort Area Under the Curve(AUCs) of 0.70 (0.66, 0.73) (2-year), 0.72 (0.68, 0.76) (3-year), and 0.77 (0.72, 0.80) (5-year), and validation cohort AUCs of 0.68 (0.62, 0.74) , 0.71 (0.65, 0.77) , and 0.80 (0.75, 0.86) , respectively. Calibration curves confirmed close alignment between predicted and observed survival, while DCA validated clinical utility for personalized risk assessment. The clinical prediction model developed from SEER data for OS in EOCRC-LM patients underwent combined primary tumor resection and chemotherapy demonstrated robust performance in prognostic evaluation. This nomogram-based tool provides clinicians with reliable quantitative guidance and a theoretical foundation for implementing personalized therapeutic strategies, thereby addressing the critical need for precision medicine in this clinically heterogeneous population.