Nomograms Predicting Long-Term Survival in Patients With De Novo Metastatic Colon Cancer: A Population-Based Analysis

Abstract

AIM: Our study aims to evaluate the overall survival (OS) and cancer-specific survival (CSS) of patients with colon cancer who present with distant metastasis, and to construct a prognostic nomogram for forecasting long-term survival outcomes.

METHODS: This population-based cohort analysis involved patients identified with de novo metastatic colon cancer between 2010 and 2015, utilizing data from the Surveillance Epidemiology and End Results (SEER) database.

RESULTS: The analysis comprised 6857 individuals diagnosed with de novo metastatic colon cancer and divided evenly into training and validation sets. Results from multivariate Cox regression analysis revealed that both OS and CSS were independently influenced by histological grade, patient age, T and N stage, presence of distant metastasis, perineural invasion, levels of carcinoembryonic antigen (CEA), receipt of chemotherapy, and surgery. Additionally, race emerged as a predictive factor for CSS but not for OS. The investigation successfully crafted a predictive nomogram capable of estimating personalized long-term survival probabilities, with a concordance index (C-index) of approximately 0.72 in both training and validation cohorts. By incorporating various clinicopathological characteristics, this nomogram effectively stratifies patients into distinct risk groups, each with a unique prognostic outlook.

CONCLUSIONS: This investigation sheds light on prognostic factors that impact the survival of patients with newly diagnosed metastatic colon cancer. Nomograms also enable accurate prediction of individual long-term survival for patients with de novo metastatic colon cancer.