Development and external validation of a nomogram to predict survival and guide surgical strategy for colorectal cancer with synchronous liver and lung metastases: a SEER-based analysis

Abstract

Surgical management of colorectal cancer with synchronous liver and lung metastases (CRC-SLLM) remains challenging because of marked clinical heterogeneity, and reliable tools for individualized risk assessment are lacking. We aimed to develop and externally validate a nomogram for overall survival (OS) prediction and to explore its potential value in guiding selection for primary tumor resection (PTR). Patients diagnosed with CRC-SLLM between 2010 and 2015 were identified from the Surveillance, Epidemiology, and End Results (SEER) database and randomly assigned to a training cohort (n = 808) and an internal validation cohort (n = 345) in a 7:3 ratio, whereas external validation was performed using 65 patients from the Fourth Hospital of Hebei Medical University, China. After feature selection with least absolute shrinkage and selection operator (LASSO) regression, a multivariable Cox-based nomogram was constructed, and its performance was assessed using time-dependent receiver operating characteristic curves, calibration plots, decision curve analysis (DCA), and risk stratification. The final nomogram incorporated nine predictors: age, primary tumor site, histological grade, tumor size, N stage, bone metastasis, carcinoembryonic antigen status, metastasectomy, and chemotherapy. It showed stable discrimination across the training, internal validation, and external validation cohorts, with 12-month areas under the curve (AUCs) of 0.791, 0.714, and 0.778 and 24-month AUCs of 0.742, 0.740, and 0.730, respectively; calibration was good in all cohorts, and DCA supported potential clinical utility. Nomogram-based risk stratification separated patients into prognostically distinct low- and high-risk groups, and an exploratory analysis indicated that the survival benefit of PTR was largely confined to low-risk patients. This externally validated nomogram provides individualized survival prediction and risk stratification for patients with CRC-SLLM and may help identify those more likely to benefit from PTR, although this finding remains exploratory and warrants confirmation in prospective multicenter studies.