The human epidermal growth factor receptor 2 (HER2) is overexpressed or gene amplified in about 15 % of breast cancers and is associated with aggressive tumors with early distant metastasis. Twenty years ago the recombinant monoclonal antibody trastuzumab targeting HER2 was approved and thus the prognosis improved dramatically. Since then, numerous agents targeting HER2 have been discovered. Accurate HER2 diagnostics is crucial for selecting patients for HER2-targeted therapy. HER2 status is determined by immunohistochemistry (IHC) of the HER2 protein and in situ hybridization (ISH) of gene copies in routine pathology. Unfortunately, HER2 scoring is prone to interobserver variability, high costs and time-consuming methods. The main aim of this project is to improve HER2 diagnostics.
The project cohort consists of HER2 positive breast cancer patients in Stockholm (Stockholm HER2 cohort) treated with trastuzumab with clinicopathological data and outcome data. From archived FFPE tumor tissue, new HER2 testing includes IHC and ISH. In addition, we investigate HER2 gene expression through targeted gene expression assays as an inexpensive and efficient alternative method for determining HER2 status with potential use in routine pathology. Based on DNA sequencing, mutations involved in resistance to anti-HER2 therapy are investigated. Furthermore, machine learning methods for ISH scoring are applied to improve pathological assessment of HER2 scoring.
Within this project HER2 status is investigated on a protein, RNA and DNA level. The overall aim of the project is to improve diagnostics and knowledge of therapy resistance in HER2 positive breast cancer. If we improve the diagnostics we can customize HER2-targeted therapy, which in turn may reduce the risk of side effects and lower health care expenses. In the end, accurate treatment will be given to the right patient.