Establishment and validation a prognostic nomogram for early-onset colon cancer patients in stage I-II

Abstract

Abstract Purpose The study aims to establish and validate a nomogram model to predict the survival of early-onset colon cancer patients in stage I-II. Methods The eligible patients were extracted from the Surveillance, Epidemiology, and End Results (SEER) database between 2012 to 2015.Patients were be divided into training group and validation group in a 7:3 ratio randomly. The significant prognostic factors were identified through univariate and multivariate analysis and they were be constructed nomogram model. The predictive performance of the nomogram was evaluated by the concordance index (C-index), calibration plots, decision curve analysis (DCA). Results Our study totally enrolled 3528 early-onset CC patients in stage I-II, of 2469 patients in the training cohort and 1059 in the validation cohort. Race, Marital status, and T stage were be confirmed the independent prognostic factors after univariate and multivariate analysis. The C-index of nomogram were 0.724 and 0.692 in the training and validation cohorts, respectively. Likewise, the calibration plots showed the good consistency in the probability of 3-, 5-years OS between the actual observation and the nomogram prediction in the training. The DCA showed the good clinical practical efficacy of the nomogram model. Moreover, the patients be divided into two different risk score groups in two cohorts based the nomogram. The low-risk group had a better survival than the high-risk group. Conclusion We developed and validated a meaningful prognostic nomogram model for stage I-II colon cancer patients of early-onset, which clinicians could apply it to make better decisions for individual patient.