[128] Diagnostic Uncertainty of Fibroepithelial Lesions on Core Biopsy Results in Increased Likelihood of Excision and Final Diagnosis of Phyllodes Tumor

Stephanie JT Chen, Judy C Pang, Lili Zhao, Julie M Jorns. University of Michigan, Ann Arbor, MI

Background: Fibroepithelial lesion (FEL) is a broad term encompassing fibroadenoma (FA) and phyllodes tumor (PT). FA and PT are treated differently but can have overlapping histologic features, making diagnosis challenging on core biopsy. The designation of FEL may be used when there is diagnostic uncertainty. Study aims were to examine the frequency and significance of core biopsy diagnosis (CBD) of FELs and investigate which features may better predict PT vs. FA.
Design: An electronic database query (1/07-1/12) identified patients with CBD of FEL (N=603). Cases were separated into 4 groups: 1) FA (N=573); 2) FEL, favor FA (N=19); 3) FEL, favor PT (N=10) and 4) PT (N=1). In excised cases, biopsy and excision diagnoses were compared and clinicopathologic features assessed. Core biopsies from cases with follow-up excision were blindly reviewed by 2 breast pathologists for overall favored diagnosis and histologic features (based on a scoring system for architecture, cellularity, stromal atypia and stromal mitoses).
Results: Of cases with known follow-up 45/573 (7.9%) FA and 29/29 (100%) group 2-4 cases were later excised. Biopsy-excision correlation is shown in Table 1. Of note, 1 group 3 patient had excision elsewhere with unknown results.

Table 1. Biopsy-Excision Correlation
1) FA44 (97.8%)1 (2.2%)
2) FEL, favor FA13 (72.2%)5 (27.8%)
3) FEL, favor PT4 (45.4%)5 (55.6%)
4) PT01 (100%)

Among excised patients time from biopsy to excision was significantly decreased if CBD was not FA (groups 2-4) vs. FA, with mean times to excision of 4.0 and 13.9 months, respectively (p=0.002). Final diagnosis of PT was associated with a 1.8 fold increase in tumor size (p=0.003). Age at diagnosis and laterality were not associated with biopsy or final diagnosis. Blinded review of core biopsies resulted in accurate prediction of 6/12 (50%) excision-confirmed PT cases. Histologic features analysis is currently pending statistical analysis.
Conclusions: FELs are histologically heterogeneous and thus are subject to sampling error, resulting in possible diagnostic uncertainty on core biopsy. Intermediate CBD categories of FEL (groups 2-3) are useful in identifying patients that have increased risk of PT and require excision; however, these diagnoses should be used prudently as they may result in unnecessary surgical intervention.
Category: Breast

Monday, March 4, 2013 9:30 AM

Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 16, Monday Morning


Close Window