Correlation between HER2 determined by fluorescence in situ hybridization and reverse transcription-polymerase chain reaction of the oncotype DX test.

The human epidermal growth factor receptor 2 (HER2) gene is amplified and its protein product overexpressed in about 20% of invasive breast cancers. Despite more than a decade of efforts to standardize HER2 testing, controversy persists regarding the most optimal testing method. Recently, Oncotype DX reports have begun including HER2 results in addition to the previously reported Recurrence Score. We compared HER2 results obtained by fluorescence in situ hybridization (FISH) in our laboratories with HER2 results obtained by reverse transcription-polymerase chain reaction (RT-PCR) as documented in the Oncotype DX report. We then sought to identify potentially significant characteristics in the discrepant cases. We identified breast cancer patients with estrogen receptor-positive, lymph node-negative tumors who had Oncotype DX testing performed between September 2008 and March 2012. Patient and tumor characteristics including HER2 FISH and Oncotype DX test results were recorded. Image analysis was performed on cases with discrepancy between the HER2 FISH and Oncotype DX HER2 results to determine the relative proportion of invasive tumor. Eight of 194 (4.1%) cases showed discrepancy between HER2 FISH and Oncotype DX RT-PCR results. Although the overall percent agreement (96%) and percent negative agreement (100%) were high, percent positive agreement was only 50%. Three of 8 (38%) discrepant cases showed heterogeneous amplification by FISH. Seven of 8 (88%) discrepant cases had <50% invasive tumor in the Oncotype DX tissue block. Percent positive agreement between HER2 FISH and Oncotype DX RT-PCR is low. Multiple factors may contribute to this discrepancy including a suboptimal microdissection and possibly heterogeneous amplification of HER2 gene in some cases.