The Diagnosis of Phyllodes Tumour amongst Participants of Mammographic Screening
Gelareh Farshid. BreastScreen SA, Wayville, South Australia, Australia
Background: Recent studies have sought discriminatory features to distinguish phyllodes tumours (PT) from fibroadenomas on needle biopsies. This distinction is imprecise but impacts patient care. Excision is advised for PTs but is not mandatory for fibroadenomas, particularly in the setting of screening where there is an imperative to limit surgery for benign processes, such as fibroadenomas. We wished to review the diagnosis of PT made in the setting of population based breast cancer screening.
Design: We searched our prospective database for cases with a diagnosis of PT during 01/01/1989 to 22/08/2012. We tabulated demographic, imaging, biopsy and outcome data and followed the women through their subsequent course.
Results: During this 23 year period we performed a total of 1,185,686 mammographic screening examinations. Of these 14298 women (1.2%) underwent biopsy. PT was diagnosed in 30 women, reflecting an overall rate of 2.53 per 100,000 women screened (0.0025%). Of these 13 were benign, 8 borderline and 9 malignant. Follow up of at least 12 mths is available in 20 cases. One woman died of metastatic PT 43 months later, 3 died of inter-current disease, 1 developed breast cancer and the remaining 15 are alive and well. PT was diagnosed during screening in 22 women (73.3%) and after symptoms arose in the ensuing 8-69 mths following a negative screen in the remaining 8 women. A circumscribed mass, mean size 34.7mm (std 17.8mm), was the dominant imaging finding. The imaging grade was 3 in 10 cases, 4 in 7 cases and 5 in 5 cases. Enlargement was documented in 21 cases (70%). A fibroepithelial lesion was specified on cytology in 5 of 23 cases. A biphasic malignancy, mucinous lesion, fat necrosis, hyperplasia and a papillary lesion were other cytologic considerations. Core biopsy favoured PT in 8 of 15 cases, fibroadenoma in 4, sarcoma in 2 and a fibroepithelial lesion in 1 case.
Conclusions: Our findings document the extreme rarity of PT in the setting of population based screening. This contrasts with the high prevalence of fibroadenomas. This finding is plausible, since their large size at diagnosis favours presentation as a palpable mass and prior studies have established a mean age of 40 at diagnosis, younger than the target age group for screening (50-69). While the limited precision of percutaneous needle biopsies in subclassifying fibroepithelial lesions remains an enduring challenge, awareness of the extremely low prior probability of PT, and the even more remote odds of diagnosing malignant PTs among screen detected lesions may assist in avoiding unnecessary surgery in equivocal cases.
Monday, March 4, 2013 1:00 PM
Poster Session II # 43, Monday Afternoon