ABSTRACT
BACKGROUND: Phyllodes tumors represent less than 1% of all breast neoplasms and can mimic fibroadenoma on core needle biopsy (CNB). The treatment of fibroepithelial (FE) neoplasms identified on CNB is controversial. We sought to identify factors that were associated with phyllodes tumors after CNB suggested FE neoplasm. MATERIALS AND METHODS: A retrospective database was queried for all patients diagnosed with FE neoplasm on CNB at Ben Taub General Hospital over a 10-y period. One hundred twenty-three patients were identified and demographic, clinical, and outcome data were analyzed. RESULTS: Of the 123 patients, 46 (37%) were found to have fibroadenomatous features and 59 (48%) were found to have FE features. All went on to have surgical excision. Forty (38%) contained phyllodes tumors, and 65 (62%) found no phyllodes tumor on final pathology. There were significant differences in the median size of the masses (4 cm versus 2.4 cm P < 0.002) and density of the masses (P < 0.001) between the group that contained phyllodes tumors and the group that did not on preoperative imaging. Further evaluation did not show any significant differences on preoperative imaging between benign and borderline/malignant phyllodes tumors. Hispanic ethnicity correlated with a higher chance of phyllodes tumor after CNB (P < 0.001). CONCLUSIONS: Women commonly present to clinics for evaluation of palpable breast masses. Based on the results of CNB, clinical decisions can be made to help direct treatment. If CNB identifies phyllodes tumor, surgical excision remains the standard of care; however, patients with suspicious FE neoplasms represent a treatment dilemma as many will prove to be benign. Preoperative size and the density of the mass on imaging and ethnicity were associated with phyllodes tumors on final pathology.
