Abstract 3967: miRNA reference samples for interlaboratory study

Abstract

Abstract The biomarkers of interest in this study are microRNA (miRNA), small non-coding RNAs involved in the regulation of gene expression. Currently, over 1000 human miRNAs have been identified. Several laboratories within the Early Detection Research Network are currently profiling miRNA as potential biomarkers. The reference sample paradigm presented and evaluated here is designed to provide measurement assurance in the earliest phases of a cancer biomarker development pipeline. Three commercially available samples of human total RNA (brain, liver, and placenta) were selected for detectable differences in miRNA expression based upon previously reported miRNA microarray experiments. For each round of this study, replicate samples of each tissue were aliquoted, blinded, and distributed to participating laboratories. The samples were then profiled using the standard workflows of each laboratory. This was repeated over multiple rounds of measurement. Raw data were processed in pair-wise comparisons between tissue types. Log2 transformed ratios were calculated for each detectable miRNA. For each of these miRNA ratios, the results among laboratories were compared to the consensus value to provide a measure of concordance and help identify laboratories that might need to make adjustments to their process. Results demonstrated repeatability within labs and reproducibility among labs with respect to direction and magnitude of differential miRNA expression between tissues for a set of miRNAs of interest (MOI). For the first round, 79% of reported results for all log2 ratio combinations and MOI were in agreement with the consensus values. Subsequent rounds demonstrated increased concordance, with 91 and 100% agreement, respectively. The data from the nine “neat” tissue total RNA samples were also used for in silico modeling to predict the values for two complex mixtures based on three components. The two mixture designs provide a pair of reference samples with designed-in reciprocal ratios of 2:1 for brain and placenta, as well as a shared 1:1 liver component. For two brain-selective and two placenta-selective MOI, 91% of reported values were in close agreement with the target values. Overall, measurement results were harmonious for a set of miRNAs of interest measured by all participants, both within rounds and between rounds. The laboratories were also able to demonstrate the ability to resolve the 2-fold differences created with complex mixtures. These results suggest that endogenous miRNAs present within total RNA may be suitable as QA/QC process controls. Citation Format: P. Scott Pine, Lynn R. Sorbara, Sudhir Srivastava, Marc Salit. miRNA reference samples for interlaboratory study. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3967. doi:10.1158/1538-7445.AM2015-3967