HER2 Testing on Breast Carcinoma Metastases to Bone by Fluorescence In Situ Hybridization: A Study of 28 Cases with an Emphasis on Preanalytic Factors.
Brendan C Dickson, Martha Wood, Julita Mazurkiewicz, Martin C Chang. Mount Sinai Hosp, Toronto, ON, Canada; Univ of Toronto, ON, Canada
Background: Bone is a common site of metastasis for breast carcinoma. Treatment of metastatic disease with trastuzumab depends on HER2-amplification status, which can be assessed by fluorescence in situ hybridization (FISH). Recent studies have shown that bone marrow biopsies (BMBx) are an acceptable source of tumor tissue for FISH testing, when using highly standardized decalcification protocols (e.g. at least 85% success using EDTA, Zustin et al. 2009). Based on our experience as a reference center for HER2 testing, BMBx received from different centres vary in their handling. The purpose of this study is to examine differences in BMBx quality with respect to preanalytic factors.
Design: A five-year review of BMBx tested for HER2 by FISH was conducted as part of a QA assessment, with 28 cases identified, including 8 internal cases and 20 from 8 different referring centers. Each center was polled regarding the routine handling of bone marrow biopsies. For each case, the HER2 amplification ratio, slide quality, and preanalytic factors (pepsin time, fixation time, fixative, decalcification method) were recorded, if available.
Results: Of the 28 cases, 20 (71%) yielded readable FISH signals. The most common reasons for unreadable FISH were absent/faint signals (6/8), and hollow nuclei (2/8). Other cases were evaluable, with some being suboptimal due to autofluorescence (4/20), small sample (2), overlapping nuclei (3), and crush artifact (2). All of the unsatisfactory cases originated from 3 of the 9 labs and were decalcified in HCl (commercial solutions) for 2 hours. Of the readable cases, 18 were non-amplified, and 2 were HER2-equivocal. Of these, the slide quality did not correlate with pepsin time or fixation time. Decalcification solutions included neutral EDTA (1/20), formic acid (7/20), or HCl (12/20); however, 11/12 cases using HCl had a rapid process (30 min), or used dilute (0.1 N) solutions. Although readable, cases from 3 of 9 centers used a hematology protocol for BMBx including short fixation (60-90 min in formalin), or a variant fixative (B+).
Conclusions: BMBx are potential specimens for HER2 FISH testing, but may be susceptible to variation in preanalytic handling. Prolonged decalcification (e.g. 2 hrs. in HCl) results in higher failure rates. Documentation of the decalcification method should be considered for all BMBx submitted for HER2 FISH. Labs with universal protocols for BMBx optimized for hematologic evaluation should consider separate protocols for metastatic tumor.
Tuesday, March 1, 2011 9:30 AM
Poster Session III # 49, Tuesday Morning