Abstract
Abstract
Objective
To evaluate the performance of radiologists’ radiologic-pathologic (rad-path) assessment in BI-RADS 4C and 5 breast lesions with benign core biopsy pathology.
Methods
Institutional pathology records were reviewed, analyzing surgical excision results for benign core biopsies after BI-RADS 4C and 5 assessments from 2019 to 2024, and the rad-path correlation issued by the radiologist prior to excision was recorded. Pathologic diagnoses and imaging data were retrieved from the electronic medical record.
Results
A total of 523 BI-RADS 4C and 5 core biopsies were identified during the study period. The malignancy rate following BI-RADS 4C assessment was 74.1% and following BI-RADS 5 assessment was 93.2%. Fifty-one benign image-guided biopsies underwent radiology-pathology correlation analysis. Radiologists categorized 19/51 benign pathology results as radiology-pathology discordant, with an upgrade rate to malignancy of 47.3%, and 32/51 as concordant, with an upgrade rate of 9.4%. All concordant upgrades occurred in high-risk lesions (atypical ductal hyperplasia, papilloma) for which excision had been recommended at the time of concordance assessment. Discordant benign pathology upgraded to malignancy included diagnoses such as dense stromal fibrosis, fibrocystic change, chronic inflammation, atypical ductal hyperplasia, hematoma, and papilloma.
Conclusion
Radiologist-performed rad-path correlation is highly reliable for benign BI-RADS 4C and 5 lesions. Radiology-pathology discordant benign biopsies carry a high risk of malignancy and warrant surgical excision. In contrast, radiology-pathology concordant benign results are highly predictive of final benign outcome, particularly when high-risk features are absent. These findings support the essential role of radiologist rad-path correlation in reducing unnecessary surgery and avoiding missed cancers.