Development and validation of nomograms for prediction of overall survival and cancer-specific survival of patients of colorectal cancer.

Abstract

Nomograms are intuitive tools for individualized cancer prognosis. We sought to develop a clinical nomogram for prediction of overall survival and cancer-specific survival for patients with colorectal cancer. Patients with colorectal cancer diagnosed between 1988 and 2006 and those who underwent surgery were retrieved from the Surveillance, Epidemiology, and End Results database and randomly divided into the training (n=119 797) and validation (n=119 797) cohorts. Log-rank and multivariate Cox regression analyses were used in our analysis. To find out death from other cancer causes and non-cancer causes, a competing-risks model was used, based on which we integrated these significant prognostic factors into nomograms and subjected the nomograms to bootstrap internal validation and to external validation. The 1-, 3-, 5- and 10-year probabilities of overall survival in patients of colorectal cancer after surgery intervention were 83.04, 65.54, 54.79 and 38.62%, respectively. The 1-, 3-, 5- and 10-year cancer-specific survival was 87.36, 73.44, 66.22 and 59.11%, respectively. Nine independent prognostic factors for overall survival and nine independent prognostic factors for cancer specific survival were included to build the nomograms. Internal and external validation CI indexes of overall survival were 0.722 and 0.721, and those of cancer-specific survival were 0.765 and 0.766, which was satisfactory. Nomograms for prediction of overall survival and cancer-specific survival of patients with colorectal cancer. Performance of the model was excellent. This practical prognostic model may help clinicians in decision-making and design of clinical studies.