Original Article
Development and validation of a novel nomogram for predicting distant metastasis-free survival among breast cancer patients
Abstract
Background: Distant metastasis (DM) from breast cancer has a poor prognosis. Our objective was to develop and validate a nomogram to predict individual distant metastasis-free survival (DMFS) and risk stratification in non-metastatic breast cancer patients.
Methods: A nomogram was based on an analysis of 1,201 breast cancer patients treated at Sun Yat-sen Memorial Hospital from 2001 to 2014. Using univariate and multivariate analyses to identify the predictors, this model was externally validated in an independent cohort of 538 patients from the Guangdong General Hospital between 2004 and 2012. The predictive discrimination and calibration ability of this nomogram were assessed using concordance index (C-index), risk group stratification, and calibration curve.
Results: The 5-year DMFS in the training and validation cohorts were 95.74% and 91.02%, respectively. On multivariable analysis of training cohort, the prognostic factors in the nomogram comprised age, tumor size, lymph node status, molecular subtype, and lymphovascular invasion (LVI). The C-index of our model was 0.75 [95% confidence interval (CI): 0.67–0.83] for the training cohort and 0.71 (95% CI: 0.64–0.78) for the validation cohort. The calibration curves for 5-year DMFS showed good agreement between the model prediction and actual observation. Based on the risk stratification, Kaplan-Meier curves indicated that the low-risk group had significantly better prognosis than the high-risk group (P<0.001).
Conclusions: Our nomogram can provide an individual prediction of 5-year DMFS in non-metastatic breast cancer patients. This prognostic tool may help clinicians to make appropriate treatment regimens and optimal surveillance plans.
Methods: A nomogram was based on an analysis of 1,201 breast cancer patients treated at Sun Yat-sen Memorial Hospital from 2001 to 2014. Using univariate and multivariate analyses to identify the predictors, this model was externally validated in an independent cohort of 538 patients from the Guangdong General Hospital between 2004 and 2012. The predictive discrimination and calibration ability of this nomogram were assessed using concordance index (C-index), risk group stratification, and calibration curve.
Results: The 5-year DMFS in the training and validation cohorts were 95.74% and 91.02%, respectively. On multivariable analysis of training cohort, the prognostic factors in the nomogram comprised age, tumor size, lymph node status, molecular subtype, and lymphovascular invasion (LVI). The C-index of our model was 0.75 [95% confidence interval (CI): 0.67–0.83] for the training cohort and 0.71 (95% CI: 0.64–0.78) for the validation cohort. The calibration curves for 5-year DMFS showed good agreement between the model prediction and actual observation. Based on the risk stratification, Kaplan-Meier curves indicated that the low-risk group had significantly better prognosis than the high-risk group (P<0.001).
Conclusions: Our nomogram can provide an individual prediction of 5-year DMFS in non-metastatic breast cancer patients. This prognostic tool may help clinicians to make appropriate treatment regimens and optimal surveillance plans.
