Pretreatment MRI-Based Radiomics for Predicting Recurrence and Disease-Free Survival in Young Women With Breast...

This study investigated the associations of radiomics based on baseline MRI with recurrence and disease-free survival (DFS) after neoadjuvant chemotherapy (NAC) in young women with breast cancer.

In total, 181 women aged 40 years or younger with breast cancer who underwent MRI before NAC were allocated into the training (n=126) and test cohorts (n=55). Three radiomics signatures were built using the intratumoral, peritumoral, and combined regions of MR images. Univariate and multivariate logistic regression were performed to select independent risk factors to construct a clinical model. A nomogram model was developed by integrating the clinical model and the combined radiomics signature. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC). Multivariate Cox regression and Kaplan-Meier analyses were used to determine the associations of various models with DFS.

Among the radiomics signatures, the combined signature best predicted recurrence, with AUCs of 0.899 and 0.849 in the training and test cohorts, respectively. The nomogram model displayed the best performance in predicting recurrence in the training (AUC=0.925) and test cohorts (AUC=0.880). The nomogram model most accurately predicted DFS in the training (C-index=0.872) and test cohorts (C-index=0.846).

The nomogram model based on pretreatment breast MRI could effectively predict breast cancer recurrence in young women undergoing NAC and serve as a potential biomarker for the risk stratification of DFS.