A nomogram for predicting the malignant diagnosis of BI-RADS US category 4A lesions in women with dense breast tissue

Background: Biopsy has been the recommended for BI-RADS category 4A, 4B and 4C. However, the malignancy rate of 4A lesions is very low (2-10%). Therefore, the majority of biopsies of category 4A lesions are benign it would take unnecessary costs and make patients anxiety. This study aimed to establish a nomogram, incorporating ultrasound and mammographic as well as patients' information, to predict the individual likelihood of malignancy of BI-RADS US 4A lesions in diagnostic setting of women with dense breast. To identify patients at lower risk of malignancy and thus avoid unnecessary biopsy. Methods: Nomogram was based on an analysis of 418 BI-RADS US 4A patients with dense breast tissues treated at Sun Yat-sen Memorial Hospital. They also got physical examination and mammography. All patients were confirmed pathology by breast biopsy or surgical excision. Multivariate logistic regression was used to identify statistically significant variables (P < 0.05), and included in the nomogram. The predictive accuracy and discriminative ability were determined by concordance index (C-index) and calibration curve. Model was subjected to external validation using 97 patients from the second affiliated Hospital of Guanghzou medical university. Results: On multivariable analysis, independent risk factors were history of breast cancer, Ultrasound features (Margin, Shape, Direction, Lymph nodes), Calcifications of mammography, which were in nomogram. The calibration curves for probability of malignancy showed optimal agreement between nomogram prediction and actual observation. The C-index of nomogram for predicting malignancy in training cohort was 0.81 (95%CI: 0.73-0.89), validation cohort was 0.77 (95%CI: 0.58-0.95). In addition, it showed good performance in stratifying different risk groups of patients both in training and validation cohorts. Conclusions: We developed a well discriminated and calibrated nomogram that can provide individual prediction of malignancy of BI-RADS US 4A with dense breast based on history of breast cancer, Ultrasound features (Margin, Shape, Direction, Lymph nodes), Calcifications of mammography. Our nomogram may help to identify patients at lower risk of malignancy and avoid unnecessary biopsy. Legal entity responsible for the study: Sun Yat-sen Memorial Hospital, Sun Yat-sen University. Funding: Guangdong, Science and Technique Department, 2017B030314026 and the National Natural Science Foundation of China (81272893). Disclosure: All authors have declared no conflicts of interest.