Back to 78th Annual Meeting
Back to Program Outline

MicroRNA Profiling Can Distinguish Normal From Neoplastic Urothelium
Benjamin Keenan, M.A.1, Derek Beaton, M.A.2, Van Le, B.A.1, Yasmin Akbari, B.A.1, Joon Hee Lee, B.A.1, Tanmay Parekh, B.A.1, Travis Antes, PhD3, Louis Liou, M.D./PhD1.
1Boston University, Boston, MA, USA, 2University of Massachusetts, Dartmouth, MA, USA, 3System Biosciences, Mountain View, CA, USA.

BACKGROUND:MicroRNA plays an important role in the control of gene expression, including many oncogenes and tumor suppressor genes. Early diagnosis of urothelial carcinoma (UC) is important for successful treatment, but currently accurate tests capable of distinguishing between low grade and high grade UC are limited. We aim to identify microRNA markers that distinguish low grade and high grade UC from each other as well as from healthy bladder urothelium, in order to develop a fast and accurate diagnostic and prognostic test for UC.
METHODS:IRB approval was obtained at our institution. Urothelium tissue samples taken from healthy patients as well as patients with high or low grade UC were flash frozen and stored at -80°C. RNA was extracted from these samples using the mirVana miRNA Isolation Kit (Ambion, California). Extracted total RNA was then quantified and reverse transcribed to cDNA using Quantimir (Systems Bioscience, California). Real-Time PCR was performed in triplicate 384-well plates using primers from the oncomir kit (Systems Bioscience, California). We identified genes of interest by performing a signal-to-noise-ratio one-sample t-test against the endogenous control U6 and a model normal patient. The delta CT method is used to compute and normalize values in our data using the endogenous RNA U6 as an internal control. A one sample t-test is used to compare normalized/transformed data from each cancer group to a mean that is calculated by a priori averaging of all our normal samples (post normalization/transformation).
RESULTS: A number of microRNAs showed statistically significant (p<0.05) changes in expression between cancer tissue and normal tissue (~350 for high grade tumors and ~150 for low grade). Using hierarchical clustering and analysis of fold changes, we narrowed these candidates down to 16 microRNA of interest (9 distinguishing between high grade and normal urothelium and 7 distinguishing between low grade and normal). Interestingly, most of the low grade-associated miRs were up-regulated while those associated with high grade are down-regulated compared to their expression in normal urothelium. (See Table I)
CONCLUSIONS: MicroRNA expression quantification shows promise as a potential diagnostic tool to identify and diagnose early stage UC. We examine the specific microRNA expression differences between high grade and low grade UC in order to better diagnose UC and give more accurate prognosis. We have shown these differences to be clear when comparing both high and low grades with normal urothelium, but further validation in large cohorts and in a multi-institutional setting will need to be done.

High Grade Distinguishing miRs
miRp-valueFold Change vs. Normal
'hsa-miR-196b'2.34E-260.09308861
'hsa-miR-143'1.33E-360.02948114
'hsa-miR-23b'3.19E-290.06032224
'hsa-miR-30a-3p'2.57E-270.06435693
'hsa-miR-302a*'2.42E-260.11070085
'hsa-miR-329'9.30E-210.11815555
'hsa-miR-339'4.36E-190.18845472
'hsa-miR-373*'9.43E-200.20813928
'hsa-miR-496'1.78E-150.16520724

Low Grade Distinguishing miRs
miRp-valueFold Change vs. Normal
'hsa-miR-142-5p'0.0438374811.8381426
'hsa-miR-29b'0.038471862.44437238
'hsa-miR-325'0.017733424.32463512
'hsa-miR-143'3.97E-080.21138269
'hsa-miR-425-5p'0.033676642.84153053
'hsa-miR-365'0.0065591219.7338189
'hsa-miR-429'0.0027701210.2700075

Back to 78th Annual Meeting
Back to Program Outline