AbstractIntroduction: There is great long term concern over the adverse cardiotoxic effects of radiation therapy (RT) treatment particularly in left-sided breast cancer (BC) patients (Chapman et al 2008). There is currently an urgent need to find a more specific, sensitive and clinically valid biomarker which can be used for the earlier detection of radiation induced heart disease (RIHD). MicroRNAs (miRNAs) have appeared as ideal candidate biomarkers for early stage RIHD diagnosis due to their non-invasive, fluid-, tissue-, organ- and disease specific nature. To date there is no study which focuses upon the utility of previously recognised cardiovascular disease linked miRNAs as biomarkers for RIHD and specifically in the context of left-sided breast irradiation. The aim of this work was to thus evaluate whether miRNAs which have been previously linked to cardiac damage are differentially expressed in BC patients treated for left-sided RT, and if they could provide clues of cardiotoxic change earlier and with higher specificity and sensitivity than traditional heart imaging and circulation protein based biomarkers.
Method: This study recruited samples from a larger research study which focused on the utility of Cardiac Magnetic Resonance (CMR) imaging in the detection of RIHD at an earlier stage in low-radiation dose treated left sided BC women at Oxford University Hospital NHS Foundation Trust. A total of 15 low-dose radiation treated left-sided BC women from the larger study were recruited to this current study. Total miRNA was isolated from 15 BC patients receiving left-side radiotherapy at timepoints: A (baseline pre-radiotherapy), B (24-72h post radiotherapy), C (3 months post radiotherapy) and D (6 months post-radiotherapy). Mean dose to the whole heart ranged from 0.6–2.8 Gray of radiation. Real time PCR was performed using TaqMan microRNA assays for microRNAs linked to cardiotoxicity (miR-1, -27a, -133a, -133b, -29b, -30a, -505, -144, -208a, -451 and -125a) using sno U6 as internal reference. The ΔΔCT calculation method was used. Students t-test was performed to analyse the data when time point A was compared to D alone (p<0.05). A one-way ANOVA was also conducted to explore individual relationships between A, B, C and D.
Results:None of the 15 BC patients were noted to have suffered myocardial damage from CMR imaging, BNP or troponin results. Significant miRNA expression changes were observed when utilising students t-test comparing time point A and D (n=8-15) for: miR-1 (increased nearly 3-fold, p=0.0378), miR-27a (decreased nearly 4-fold, P=0.0464), miR-451 (decreased more than 2-fold, P= 0.0321), and miR-125a (increased more than 3-fold, P=0.0098).miR-144 was also found to be upregulated from A (n=10) to C (n=7) (P≤0.01) and decreased back to baseline levels from C (n=7) to D (n=12) (P≤0.0001). Other statistically significant relationships include a more than 3-fold increase in the levels of miR-1 from time point B to D (p=0.002) and C to D (p=0.003). miR-125 levels also increased by more than 2-fold between time points B to D (p=0.031). miR-144a levels increased by more than 465-fold from time point A to C (p<0.01) and then decreased by more than 1600 fold from time point C to D (p<0.0001). miR-133b levels were found to significantly decrease by more than 40-fold between A (n=10) to B (n=8) (p=0.015). The expression levels of miR-133a, miR-29b, miR-30a and miR-505 were not altered 6 months post-radiotherapy. The expression of miR-208a was undetectable at all time points in this study (n=7-13).
Conclusion:This study has shown that miR-1, miR-27a, miR-125 and miR-451 were altered from the pre-radiation to the 6 month post-radiation stage in left sided RT BC patients, whereas imaging and circulation cardiac injury protein biomarkers did not change at all. In addition it was also observed that miR-144 and miR-133b shows a differential regulation pattern in response to RT. As none of the patients within this study developed clinically diagnosed cardiotoxicity as a result of radiation, the link between RIHD as a cause of these expressional changes cannot be established. However this study does highlight that miRNAs warrant further investigation in larger patient cohorts for potential as early stage biomarkers of RIHD as changes in expression patterns are seen.
|Date of Award||2017|
|Supervisor||Helen Maddock (Supervisor), Hardip Sandhu (Supervisor), Mayel Gharanei (Supervisor) & Tim James (Supervisor)|