[2007] Proposal for a New Classification of “Equivocal” Fibroepithelial Lesions (FELs)

Erika Resetkova, Nour Sneige. University of Texas MD Anderson Cancer Center, Houston, TX

Background: Management of “equivocal” FELs with unusual histomorphology or abnormal clinical and imaging findings diagnosed by core biopsy (CB) is still controversial. These lesions are difficult to classify even upon excision and infrequently subjected to second opinion.
Design: Final excisional specimens of equivocal FELs diagnosed on CB were blindly re-reviewed by two additional experienced breast pathologists, in order to evaluate inter-observer variability and to assess importance of each individual histologic parameter for final classification. The institutional review board of The University of Texas M.D. Anderson Cancer Center approved this study. The reviewers based their diagnoses on systematic evaluation of numerous histologic criteria (predominant architectural pattern, margins, satellite nodules, stromal overgrowth, stromal cellularity and stromal atypia, mitotic activity, stromal periductal condensation and hyalinization, stromal heterogeneity, presence of heterologous components and/or fat, and degree of epithelial atypia). The inter-observer variability results were analyzed using percentage agreement and the Kappa test.
Results: Sixty five surgical cases of FELs were re-reviewed. We report a very high inter-observer variability between the original pathology diagnosis and two experienced breast pathologists. 38% of cases were upgraded either from FA to benign PT, or from benign PT to PT of uncertain malignant potential. Based on re-review, different clinical management would be recommended in 20% of cases. The most significant histologic parameters consistently yielding to an upgrade by expert pathologists were focal infiltration at the tumor periphery (p=0.01), presence of satellite nodules (p=0.05), and stromal heterogeneity (p=0.05).
Conclusions: Due to the difficulty to reach consensus on diagnosis of equivocal FELs and a high inter-observer variability, we propose a new classification and recommend an algorithm for their clinical management.
Category: Quality Assurance

Tuesday, March 5, 2013 9:30 AM

Poster Session III # 280, Tuesday Morning


Close Window