Discovery of a new phospho-HER2+/FISH- molecular subtype of human breast cancer by functional pathway mapping

Abstract

11009 Background: Effective treatment of breast cancer through the targeting of the HER2 protein in human breast cancer represents a major advance in oncology, yet the identification of responders has relied on non-quantitative methods (IHC) that measure the erbB2 protein or indirect genomic analysis (FISH) of the erbB2 gene that cannot predict protein pathway activation status. We utilized a new quantitative protein microarray assay to measure total and phosphorylated HER2 in the context of broad-scale EGFR signal pathway mapping in order to generate a new molecular characterization scheme for human breast cancer. Methods: Pure tumor epithelium from 149 frozen pre-treatment human breast cancer tissue specimens (from the I-SPY TRIAL: CALGB 150007/150012, ACRIN 6657) were procured via Laser Capture Microdissection and protein pathway mapping was performed using Reverse Phase Protein Microarrays (RPMA) whereby the activation of 40 key signaling proteins was quantitatively measured at once. Results: While phospho-HER2 and total HER2 as measured by RPMA had excellent concordance with FISH (95%) and IHC (94%), of the 63 cases where both clinical FISH and IHC status of c-erbB2 were known, we discovered that 5/45 (11%) of the FISH-/IHC- cases had phosphorylated HER2 levels as high or higher than the FISH+/IHC positive patients. These results were confirmed and validated by independent analysis with quantitative Western Blot using a separate biopsy specimen from the same patients. Conclusions: A new molecular phenotype of human breast cancer has been identified whereby total levels of HER2 are low yet levels of the phosphorlyated receptor are very high. This molecular phenotype is not detectable by FISH analysis nor by measurement of the total HER2 protein itself. Given the central importance of phosphorylation on effective signal transduction of the EGFR family, we are planning to determine the clinical significance of this finding by retrospective analysis of banked material with outcome, and a prospective clinical trial in I-SPY 2. Support: ACRIN U01 CA079778 ; CALGB CA31964, CA33601; NCI SPORE CA58207. [Table: see text]