Immunohistochemistry detection of putative miR-200c and miR-203 targets in breast cancer patients

Part of this thesis will be published in the following: Gomes, B.C., Santos, B. 2015. Methods for studying microRNAs expression and their targets in formalin-fixed, paraffin-embedded (FFPE) breast cancer tissues. In Methods in Molecular Biology: Cancer Drug Resistance (Rueff, J. & Rodrigues, A.S. eds), Springer Protocols.

The morphological variability and clinical evolution of breast cancer have prompted researchers to find a strategy to classify the disease and to possibly define supportive prognostic and predictive indicators. Recently, some studies have focused on the putative utility of miRNA as a novel class of cancer markers. Since potential targets of miRNAs are often provided only by bioinformatic tools there is a gap in this area of study. Therefore, parallel to the quantification of the expression of miR-203 and miR-200c in tumor tissue from patients from Central Lisbon Hospital with breast cancer, the aim of this dissertation was to analyze the expression of their putative targets – ATM, BMI1, SIX1 and SOX2 – by immunohistochemistry. 45 samples were analyzed corresponding to 43 patients whose mean age at diagnosis was 62 years. The most common type was invasive carcinoma NOS (71,1%) followed by invasive lobular carcinoma (8,9%). 86,4% of samples were ER positive, 79,1% PR positive, 13,6% HER2 positive and 45,5% high ki67. miR-200c was downregulated in 12,8% of the samples and upregulated in 23,1%. Comparatively, 20,5% of tumors presented miR-203 downregulation and 30,8% upregulation. Anti-ATM and anti-BMI1 antibodies didn’t perform properly thus, they were not assessed. Regarding to SIX1 and SOX2, only 13.3% and 8.9% of tumors were positive, respectively. Furthermore, a statistically significant association between the expression of both proteins and various clinicopathological parameters was not found, except for the number of pregnancies that seems to be associated with SIX1 positivity (p = 0.034). Regarding the relationship between levels of miRNAs and expression of their putative targets, there was no statistically association. In the future a bigger sample size should be used to increase the robustness of results and patient’s follow-up would allow evaluating the association between SIX1 and SOX2 with therapeutic outcome.