Intratumoral Heterogeneity of HER2 in Invasive Breast Carcinoma: How Much Testing Is Enough?
Gillian C Bethune, Martin C Chang. Mount Sinai Hospital, Toronto, ON, Canada; University of Toronto, Toronto, ON, Canada
Background: The prognostic and predictive value of HER2 (human epidermal growth factor receptor 2/ERBB2) warrants routine testing in invasive breast carcinoma. Current testing guidelines recommend the reporting of intratumoral heterogeneity; however, the clinical significance of heterogeneity remains unclear. There is some evidence that HER2-heterogeneous tumours fall between "positive" and "negative" cases with respect to prognosis and treatment response. The aim of this study is to evaluate the value of testing multiple blocks within a single tumour, with an emphasis on detecting intratumoural heterogeneity.
Design: As approved by the institutional research board, we searched (2006-2012) for breast excisions in which a single focus of invasive breast carcinoma was tested for HER2 on more than one block. We identified 151 consecutive cases (144 with 2 blocks; 7 with 3 blocks); an audit of HER2 results was performed on each case. Cases with IHC having regional heterogeneity (clusters of HER2 positive cells) were tested by FISH. 'Genetic heterogeneity' was defined according to CAP 2009 consensus (at least 5% but fewer than 50% of nuclei with HER2/CEP17 ratio of >2.2).
Results: Multiple blocks were tested in tumors ranging from 0.5 to 10.8 cm in size (mean 2.3 cm). Reasons for additional testing included: large tumour size, equivocal result on a single block, discordance with a prior biopsy, phenotypic or genetic heterogeneity, and clinician request. Testing of multiple blocks yielded no additional HER2 information in 146/151 (97%) of cases. Of these, 122/146 (84%) were HER2-negative by IHC, 20/146 (14%) were HER2-positive by IHC, and 4/146 (3%) were HER2-equivocal by IHC and FISH. Although regional and/or genetic heterogeneity were seen in 2/146 (1%) and 4/146 (3%) of these cases, the heterogeneous pattern could be seen on both blocks tested. Only 5/151 (3%) cases differed between blocks of the same tumor--all were tested by both IHC and FISH. These cases usually had high grade (4/5), heterogeneous IHC staining (4/5), and genetic heterogeneity if evaluated by FISH (4/4). By contrast, only a minority (2/5) showed morphologic heterogeneity. The tumors in these discordant cases were not significantly larger (0.9 to 4.6 cm, mean 3.0 cm).
Conclusions: In the vast majority of cases, unifocal invasive breast carcinomas do not differ in HER2 status when multiple blocks are tested. Cases with all of the following: equivocal HER2 status, high grade, and heterogeneity in IHC/FISH, are more likely to yield a different result on testing of a second block.
Monday, March 4, 2013 9:15 AM
Proffered Papers: Section B, Monday Morning