Abstract P4-08-34: MammaTyper© – An in vitro quantitative local gene expression test as a predictor of OncotypeDX©- and EndoPredict©-results

Abstract

Abstract Background: Risk stratification in early-stage breast cancer patients still remains a clinical challenge. It is well known that expression of ER/ PR, HER2 and Ki-67 are valuable prognostic and predictive markers. According to the commonly used St. Gallen classification (2013), breast cancer treatment decisions can be based on four different subtypes (Luminal A-like, Luminal B-like [HER2+/-], HER2 + [non-luminal] and Basal-like). Immunohistochemistry is predominantly used to determine the receptor status, although semi-quantitative analysis shows significant inter-observer variability. Several quantitative gene expression tests – such as Oncotype DX© and EndoPredict© – are available as diagnostic tools but their use is often limited due to financial considerations. MammaTyper© has been shown to be a precise and reproducible biomarker determination tool to investigate the expression of ER/ PR, HER2 and Ki-67 and may prove to be a cost efficient quantitative diagnostic tool. Methods: This study tested ESR-1, PGR, MKI67 and ERBB2 mRNA-levels by RT-qPCR in 100 FFPT-samples using MammaTyper© kit. The test identified breast cancer subgroups according to St. Gallen classification. RT-qPCR results were correlated to IHC-test results via χ2-Test and to risk groups of Oncotype DX©- (n=55) and EndoPredict©-testing (n=45). Oncotype©-Recurrence Score levels were separated into low (RS <18), intermediate (RS 18-30) and high risk (RS ≥ 31). EndoPredict© EPScore© and EPclin Risk Score© cut off levels defining high and low risk levels were 7.0, respectively 3.3. Linear regression model was performed investigating the prediction of EPscore© and Recurrence Score© via MammaTyper© mRNA-results. Results: IHC-testing resulted in 43 Luminal A-like samples (43%) and 57 Luminal B-like [HER2 negative] (57%) tumors. MammaTyper ©-based classification showed significant correlation to IHC-based classification (χ2=12,68; p=0,005). EndoPredict©-data showed low-risk levels in 36 patients (80%) and high-risk status in 9 cases (20%). Correlation to mRNA-based St. Gallen-subtypes approved significance (χ2=17,32, p<0,001). In linear regression model only MKI67-mRNA showed independent correlation to EPScore© (p<0,001; R2 0,290) A total of 34 patients (61,8%) showed low risk, 13 (26,3%) intermediate risk (RS 18-30) and 8 (14,5%) high-risk recurrence score levels (RS≥31) in OncotypeDX©. MammaTyper© results correlated to OncotypeDX© risk-levels (χ2=27,98; p<0,001). In linear regression model mRNA-levels of ESR-1, PGR and MKI67 were significant predictors of OncotypeDX© RS (p-values: p=0,003, p<0,001, p=0,033, R2= 0,602) Conclusion: This study shows mRNA-testing is a valid marker to identify breast cancer subtypes. Additionally, MammaTyper© results are significantly correlated to Oncotype DX©- and EndoPredict©-results, indicating it might be an efficient as well as cost effective diagnostic instrument. However a large prospective multicenter study should be performed to prove its diagnostic validity in predicting the most appropriate and effective treatment in early-stage breast cancer. Citation Format: Eckhoff K, Pokorny J, Baumann K, Perner S, Pursche T, Rody A. MammaTyper© – An in vitro quantitative local gene expression test as a predictor of OncotypeDX©- and EndoPredict©-results [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-08-34.