Measuring Diagnostic Accuracy and Precision of Core Needle Biopsies of Fibroepithelial Breast Lesions to Inform Surgical Management.

AimsFibroepithelial lesions (FELs) of the breast are a heterogeneous group of neoplasms that present diagnostic challenges, particularly on core needle biopsy. This study aimed to evaluate the diagnostic accuracy of breast core biopsy in detecting phyllodes tumors (PTs) using traditional and evidence-based methods, and to identify additional clinical factors that may aid surgical decision-making.Methods and ResultsConsecutive core biopsy specimens from 2017 to 2019 were reviewed, with surgical pathology or clinical follow-up as reference standards. Discordant lesions underwent re-review. Diagnostic accuracy was assessed using Bayesian methods, calculating likelihood ratios (LRs) and integrating them with Fagan nomograms to estimate post-test probabilities. Logistic regression identified factors that improved diagnostic precision. Among 425 patients with FELs on biopsy (369 fibroadenoma [FA], 9 PT, 47 indeterminate), 91 underwent excision and 337 had follow-up (mean 36 months). The overall upgrade rate to PT was 19.7%, but only 7% of FA patients upgraded, reflecting a low false-negative rate. Of nine biopsy-diagnosed PTs, seven were confirmed (false-positive rate 22%). Diagnostic LRs demonstrated strong performance: FA (LR = 0.06) effectively ruled out PT, indeterminate FELs (LR = 6) warranted excision, and PT (LR = 78) strongly supported surgical management. Older age and larger tumor size were significantly associated with PT, while BIRADS, needle gauge, clinical presentation, and prior breast cancer were not predictive.ConclusionsCore biopsy effectively stratifies FELs and guides management. While indeterminate lesions remain challenging, integrating clinical factors such as age and size enhances decision-making, reduces overtreatment of FA, and ensures timely diagnosis of PT.