Exploring molecular pathways in low-grade oestrogen receptor positive breast cancer: a biomarker driven approach. (25th February 2016)

• Record Type: Journal Article
• Authors: Mukherjee, Abhik; Craze, Madeleine; Joseph, Chitra; Nolan, Chris; Green, Andrew; Rakha, Emad; Ellis, Ian Ogilvie
• Is Part Of: Lancet. Volume 387(2016)Supplement 1
• Page Start: S76
• Publication Date: 2016-02-25
• Subjects: Medicine -- Periodicals; Medicine -- Periodicals; Medicine; Medicine; Electronic journals; Periodicals; 610.5
• DOI: 10.1016/S0140-6736(16)00463-3
• Languages: English
• ISSNs: 0140-6736

Abstract

Abstract: Background: The identification of the drivers of neoplasia and progression in low-grade oestrogen receptor (ER) positive luminal breast cancers compared with high-grade luminal breast cancers is essential. A surrogate biomarker profile is needed to distinguish between low-grade and high-grade cancers to help clinicians administer or avoid additional therapy. This study aimed to investigate molecular determinants of low-grade luminal breast cancers from a biomarker database. Methods: 214 biomarkers from a well-annotated series of 1845 primary breast cancers were analysed for correlations with grade of breast cancer. Proteins with significant associations with grade 1 ER-positive cases (n=388) versus high-grade ER-positive cases (n=1183) were interrogated for pathway enrichment analysis (corrected for false discovery rate), with the STRING 10 platform (open access web-resource) incorporating Gene Ontology (GO) and KEGG. Findings: Univariate analysis (p range=0·041–0·0001) identified 28 markers with significant association for grade 1 ER-positive tumours (eg, androgen receptor [AR], FHIT, FOXA1) and 54 with a significant negative association (eg, PTEN, GATA3, KAPNA1, p53). Pathways unifying the positively associated proteins in low-grade breast cancers revealed enrichments in intracellular steroid hormone receptor signalling (p=0·003), mammary gland development (p=0·008), AR binding (p=0·017) (GO). For negatively associated proteins, significant enrichments included cell proliferation (p=0·0002), radiation response (p=0·0003), T cell differentiation (p=0·004), and double-strand break repair (p=0·0004) (GO), and the HIF1α (p=0·00003), p53 (p=0·0016), ErbB (p=0·002), and JAK-STAT (p=0·01) (KEGG) pathways. Interpretation: These preliminary findings are the first, to our knowledge, to unify biomarkers for grade-related analysis in breast cancer, with use of protein based data. Within the constraints of selection bias, data mining from immunohistochemistry of multiple biomarkers in relation to the features and behaviour of breast cancer hold promise. Funding: Pathological Society of Great Britain and Ireland, National Institute for Health Research.