MiRNAs as oncologic biomarkers for breast cancer
McDermott, Ailbhe
McDermott, Ailbhe
Loading...
Repository DOI
Publication Date
2014-08-18
Type
Thesis
Downloads
Citation
Abstract
Understanding of the molecular complexity underlying breast cancer has greatly advanced in recent years. Distinct phenotypic subtypes have been identified each of which carry prognostic and predictive implications. Despite these advances, delays in diagnosis resulting in haematogenous dissemination and development of distant metastases remains the leading cause of death for breast cancer patients. This prevalent heterogeneous disease is in critical need of circulating biomarkers to expedite accurate diagnosis and provide more specific individualised treatment modalities in an effort to alleviate the global social and economic burden of this cancer. The knowledge that miRNA expression is altered in breast cancer, coupled with their presence in circulation and promising biomarker characteristics has placed them at the forefront of translational research with applicability to diagnostics, prognostics and prediction. This study was undertaken to evaluate the potential of miRNAs as circulating diagnostic and predictive biomarkers in breast cancer, and to advance understanding of their downstream targets and cellular location. Accurate miRNA quantification requires rigorous normalisation of raw data using stably expressed endogenous controls. At the outset of this study there was a dearth of systematically tested control miRNAs for use in the circulation. Using both miRNA microarrays and RQ-PCR, miR-16 and miR-425 were thus identified as the most stably expressed endogenous controls for RQ-PCR in blood specimens for breast cancer studies. To identify circulating miRNA expression patterns associated with Luminal A (ER+PR+HER2/neu-) tumours, expression profiling of 667 miRNAs on 20 blood samples (10 Luminal A; 10 controls) was performed. Artificial neural networking analysis of the dataset identified 76 differentially expressed miRNAs. Validation in a larger cohort revealed a profile (miR-29a, miR-181a and miR-652) with sensitivity and specificity of 77% and 74%, respectively, to identify Luminal A tumours, providing insight into the molecular mechanisms underpinning this common subtype. In the next phase of the study, miRNAs belonging to the miR-15 family were shown to be aberrantly expressed in blood and tissue of women with breast cancer. Combination of miR-15a, miR-195 and miR-497 provided a sensitivity and specificity profile of 82% and 76.5%, respectively, for breast cancer detection. In an effort to elucidate the functional role of the miR-15 family in carcinogenesis in silico analysis was performed to identify miRNA-mRNA duplex interactions, which were further examined by comparison of the expression of putative target genes (CCND1, FGF2 and BCL2) with miRNA expression in breast tumours. To gain insight to the cellular location of miR-195 and miR-497, miRNA in situ hybridisation was performed. In this manner miR-195, was identified as being localised to stromal and tumour epithelial cells. Both miR-195 and miR-497 expression in vitro were affected by treatment with a demethylation agent, indicating a role for epigenetic modification, possibly explaining their altered expression in the breast cancer state. Finally, the application of miRNAs as predictive biomarkers for guiding and monitoring response to chemotherapy was assessed by tracking their expression in serial blood samples taken from patients undergoing neoadjuvant treatment for breast cancer. The data presented herein reveals that miRNAs are exciting molecules with the potential to revolutionise breast cancer diagnosis and treatment in pursuit of individualised targeted cancer treatment for patients.
Publisher
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivs 3.0 Ireland