Abstract P5-02-01: Comparison of ESR1, PGR, HER2 and KI67 expression by central IHC and MammaTyper® RT-qPCR kit in the FinHer trial

Abstract

Abstract Background: Subtyping of breast cancer has become an integral part of standard evaluation of breast cancer patients1. However, assaying of ER, PgR and Her-2/neu by immunohistochemistry (IHC) carries an up to 20% risk of erroneous results2,3. Moreover, reliable assessment of Ki-67 by IHC in grade 2 breast cancer is challenging due to high intra- and interobserver variations4,5. Interobserver concordance for ESR1, PGR, HER2, and KI67 determination on mRNA level using MammaTyper® reagents were 96.8%, 97.2%, 100%, 97.6%, respectively, based on predefined cutoffs (Laible et al., abstract submitted). Here we tested the clinical concordance and prognostic value of MammaTyper® determinations in the FinHer trial patient population. Methods: RNA was extracted from formalin-fixed paraffin-embedded (FFPE) breast tumor tissue, and candidate gene expression was analyzed using the RNXtract® IVD and MammaTyper® IVD kits (BioNTech Diagnostics GmbH, Mainz) from 791 patients who participated in the FinHer trial. RNA levels of ESR1, PGR, HER2, and KI67 mRNA expression were measured using RT-qPCR,normalized according to the 40-DDCT method and compared with IHC or CISH results. Associations with distant disease-free survival (DDFS) and overall survival (OS) were assessed using the log-rank test. Results: ESR1, PGR, HER2, and KI67 mRNA levels exhibited strong correlations with the respective clinical assays (each p-value <0.0001). The concordance rate of the mRNA assay and the clinical asssay was 92% for ESR1, 92% for HER2, 83% for PGR, and 68% for KI67 RNA. Using predefined cut-off values for ESR1, PGR, and KI67, the mRNA levels were prognostic for DDFS (p=0.002, p=0.005, and p=0.0005) and OS (p<0.0001, p=0.0001, and p=0.0024, respectively), whereas HER2 mRNA expression was not (p=0.17 and 0.11, respectively). Unexpectedly, the HER2 mRNA expression distribution of the ER-negative cancers was bimodal with little overlap between the HER2-low and HER2-high subsets, while in ER-positive cancers HER2 mRNA distribution was almost unimodal and in between the two subpopulations of ER-negative cancers. When different cut-offs were used for ER-positive and ER-negative cancers, tumor HER2 mRNA was also significantly associated with DDFS (p=0.031) and OS (p=0.018). Interestingly, 17% of ESR1 mRNA-negative and HER2 mRNA-negative patients exhibited high mRNA expression of PGR, and such patients had high 5-yr DDFS and OS (>95%). Conclusions: Determination of ESR1, PGR, HER2, and KI67 without macrodissection of routine whole tissue FFPE specimens results in highly concordant results when compared to clinical assays. A significant minority of HER2 negative breast cancers expressed PGR mRNA despite low ESR1 mRNA levels and had superior outcome. HER2 mRNA levels differed substantially between ER-positive and ER negative tumors. This may explain why HER2 determination using a single cut-off for HER2 mRNA with the Oncotype DX assay frequently results in a false negative finding6. The MammaTyper-defined ESR1, PGR, HER2, and KI67 expression showed strong correlations with the corresponding clinical assays and survival. 1) Goldhirsch et al., Annals of Oncology 2013 2) Hammond et al., JCO 2010 3) Wolff et al., JCO 2007 4) Varga et al., PloS 2012 5) Polley et al., JNCI 2013. Citation Format: Ralph M Wirtz, Pirkko-Liisa Kellokumpu-Lehtinen, Jorma Isola, Vesa Kataja, Petri Bono, Taina Turpeenniemi-Hujanen, Sirkuu Jyrkkiö, Harri Sitho, Sebastian Eidt, Ugur Sahin, Heikki Joensuu. Comparison of ESR1, PGR, HER2 and KI67 expression by central IHC and MammaTyper® RT-qPCR kit in the FinHer trial [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P5-02-01.