| dc.description.abstract | HER2 is over-expressed in 20-25 % of breast cancers. Due to the increase in proliferation and survival signalling resulting from HER2 over-expression, this sub-type is associated with more aggressive tumour progression and poor prognosis. HER2 targeted-therapy has significantly improved patient prognosis, however, despite all progress only a subgroup derives optimal benefit, whereas others have refractory disease or develop resistance. It is therefore necessary to identify new targets to improve patient outcome. A group of non-coding RNAs, miRNAs, are often aberrantly expressed in tumours and miRNA expression profiles have been seen to evolve over the course of treatment implicating them in therapeutic resistance. Using the lapatinib sensitive, BT-474, and resistant, BT-474/L, HER2 expressing cells, 44 miRNAs were found to be upregulated. The region with the highest number of differentially expressed miRNAs is the 14q32 region. RTqPCR confirmed upregulation of 14q32 miRNAs in the BT-474/L compared to BT-474 as well as 2 other pairs (HCC1954, HCC1954/L, and SKBR-3, SKBR- 3/L) linking increased miRNA expression with acquired resistance to lapatinib. As the 14q32 region is regulated by DNA imprinting, we explored epigenetic changes between the sensitive and resistant lines. Global methylation reversal cased upregulation of all three miRNAs in the sensitive cells. This suggests loss of methylation is a key in controlling 14q32 miRNA expression. Since miRNAs are not suitable therapeutic targets, differential gene expression analysis combined with in silico analysis was used to identify targets of the miRNAs. Silencing of five target genes seemed to decrease sensitivity of the cells to HER2-targeted treatment. Analysis of the NeoALTTO clinical trial data suggests high expression of four of these genes, SOCS2, BASP1, NEDD4L, and SH3BGRL could be linked to better prognosis. These results suggest upregulation of 14q32 miRNAs caused by loss of epigenetic regulation leads to decreased expression of SOCS2, BASP1, NEDD4L, and SH3BGRL. This loss could contribute to HER2-targeted therapy resistance. Therefore expression levels of 14q32 miRNAs could be used as a prognostic biomarker to identify patients likely to relapse. | en_US |
