Study on breast carcinoma Her2/neu and hormonal receptors status assessed by automated images analysis systems: ACIS III (Dako) and ScanScope (Aperio)

Abstract

Her-2/neu is overexpressed in 20-30% of breast cancer patients and is associated with a more aggressive disease. Identification of Her-2/c-erbB-2-neu overexpression is based on immunohistochemical [ihc] detection of protein and/or gene amplification in fluorescence in situ hybridization test (FISH). Also Estrogen receptors [ER] and Progesterone receptors [PR] are the prognostic and predictive biomarkers, recently analysed by ihc methods. Subjective, manual scoring of the ihc Her-2/neu expression and expression of the ER/PR reported as the percentage of immunopositive cells are the most common mode of interpretation among pathologists. Automated microscopy and computerised processing have provided increased accuracy in quantification and standardisation.

The aims of our study were: to evaluate the scoring reproducibility of Her-2 /neu ihc expression tested by two automated systems: ACIS (Dako) and ScanScope (Aperio); to estimate the ER/PR expression in ihc staining methods with different anti-ER/anti-PR antibodies (the monoclonal and the ER/PR pharmDx TM Kit) by the ACIS system. Her-2/neu ihc expression was measured in 114 primary invasive breast carcinomas by the manual and the automated scoring (ACIS and Aperio system). 106 slides stained ihc with two types of anti-ER/anti-PR antibodies entered the quantisation. The results of our investigations showed very high reproducibility of Her-2/neu scores in intra- and interobserver analysis by ACIS evaluation. The major concordance was present in strong 3+ ihc cases; very small discordance was shown by cases with low expression of Her-2/neu. The accuracy of scoring by the Aperio was little lower in comparison to ACIS but it might result from the smaller and variable series of samples analysed by Aperio. The concordance in scoring of two automated systems was 86.5% (p<0.0001; gamma=0.887); the discordance was referred only to the lower expression of Her-2/neu. The concordance in manual scoring performed by the single observer and the panel was 84.2% (p<0.0001, gamma = 0.99); the discordance comprised a few cases with strong expression (2+ vs 3+). Very high intra- and interobserver reproducibility of the ER/PR ihc measurements was present in the readers results (referred to the percentage of immunoreactive carcinomatous cell population in the breast carcinomas acc. to the ACIS algorithm). No differences were disclosed in the percentage of ER-immunoreactive and PR-immunoreactive carcinomatous cell populations when used 2 different type of antibodies, in the ACIS automated method.