Abstract P6-05-28: Nicastrin RNA in situ hybridization (RNAScope®) reveals estrogen receptor negative breast cancer patients at risk of metastatic relapse and could serve as companion diagnostic for anti-nicastrin monoclonal antibody therapy

Abstract

Abstract Introduction of targeted therapies into oncology requires co development of a drug and its companion diagnostic to achieve clinical efficacy and minimise side effects. Given the aggressive nature of estrogen receptor alpha negative (ERa-ve) breast cancer (BrCa), novel therapeutics are needed. Gamma-secretase enzyme is a relevant target in cancer, and within it, Nicastrin (NCSTN) is amenable to therapeutic intervention using monoclonal antibodies (mAb). NCSTN gene is amplified in a some breast tumours, which correlates with high NCSTN mRNA and adversely impacts overall survival (TCGA, Nature 2012). Immunohistochemistry revealed NCSTN protein overexpression in 47.5% of BrCa (n = 1000), conferring worse prognosis in the ERα-ve cohort. NCSTN genetic depletion in triple negative BrCa cells attenuated tumor growth in vitro and in vivo, invasive capacity, mesenchymal features and cancer stem-cell propagation (Filipovic et al; Lombardo et al). We have developed and characterized anti-NCSTN mAbs in terms of their therapeutic properties, functional effects in vitro and in vivo, and binding epitopes. Here, we have used a novel, bright-field chromogenic in situ hybridization technique (RNAscope), to quantify single-cell NCSTN mRNA levels in ERα-ve BrCa patients (n = 311), and propose this test as a companion diagnostic for targeted anti-NCSTN therapy. Methods:Paraffin BrCa tissue microarrays sections (Nottingham Tenovus ERα-ve cohort; n = 311) were analyzed by RNAScope using a NCSTN-specific probe (GenBank accession number NM_015331.2, and probe region: nt 158-1306). Visualized as brown punctate dots, RNA staining in individual tumor cells was scored semi-quantitatively on a 1-4 scale: 1 = 0-5 dots/cell, 2 = 6-10 dots/cell, 3 = 11-15 dots/cell, 4 = >16 dots/cell. H-score (100-400) capturing the percentage of tumor cells at each staining level was calculated and correlated with clinical and tumour characteristics. Results: NCSTN high H-score >120 correlated with high proliferative index (Ki67) (p = 0.002), high nuclear pleomorphism (p = 0.03) and mitotic score (p = 0.02) and membranous expression of the NCSTN protein (p = 0.05). It further correlated with cytoplsmic BRCA1 positivity (p = 0.025), which is associated with BrCa cell invasion. Importantly, in Kaplan-Meier analyses NCSTN Score>120 conferred a higher probability of metastatic relapse at 10-year follow up (p = 0.019), and retained significance in the Cox regression multivariate analysis, along with tumor grade, stage and size (p = 0.04). Conclusion: RNAScope successfully detected NCSTN mRNA in paraffin embedded BrCa tissues. High NCSTN RNAScope levels may be a novel independent prognostic factor in this patient cohort and one that correlates with established markers of aggressive disease. We will also present data on anti-NCSTN mAb efficacy in cell systems based on NCSTN RNAScope score. We propose that detection of high NCSTN levels by RNAScope in primary BrCa removed at surgery or in subsequent biopsies of disease relapse can be further developed to serve as a clinical companion diagnostic test to assess patient eligibility for treatment using anti-NCSTN therapy. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-05-28.