Martin Kristensson
Visiopharm ,Denmark
Title: Optimizing HER2 assessment in breast cancer - application of automated image analysis
Biography
Biography: Martin Kristensson
Abstract
Human epidermal growth factor receptor 2 (HER2) is a receptor for circulating growth factor, stimulating uncontrolled cell proliferation. Trastuzumab reacts with HER2 and arrests the cells in the G1 phase. Immune system-directed cell killing is probably also involved in the anti-cancer effect. In breast cancer, overexpression of HER2 occurs in 15%, and analysis of HER2 expression is therefore pivotal for treatment decision. In a cohort of 462 patients, stained with Roche’s ready-to-use HER2 test. The HER2-CONNECT algorithm evaluates the IHC staining reaction of HER2, based on cell membrane connectivity. The connectivity translates into the classic diagnostic score for HER2 expression (0, 1+, 2+ or 3+) in agreement with the ASCO/CAP guidelines. The automated reading was compared to manual reading, and for the borderline (2+) to manual reading of the HER2 gene expression on fluorescence in situ hybridization (FISH). Manual reading demonstrated a sensitivity of 85.0% and a specificity of 86.0% with 14.1% inconclusive samples. Using HER2-CONNECT, sensitivity increased to 97.6% and specificity to 95.5%. Less than 4.5% of the cases were deemed inconclusive. Total agreement when comparing HER2-CONNECT with manual IHC supplemented by FISH in borderline cases was 92.8%. Using the HER2-CONNECT digital image analysis algorithm based on IHC detection of HER2 protein expression on tumour cell membranes, less than 4.5% of the cases were inconclusive whether overexpressed or not. Compared to manual reading, the need for supplementary FISH testing is reduced by 67.7%. In the routine diagnostic setting, this would have significant impact on cost reduction and turn-around-time.
