Purpose: Cancer stroma plays an important role in the progression of cancer. Although alterations in miRNA expression have been explored in various kinds of cancers, the expression of miRNAs in cancer stroma has not been explored in detail.Experimental Design: Using a laser microdissection technique, we collected RNA samples specific for epithelium or stroma from 13 colorectal cancer tissues and four normal tissues, and miRNA microarray and gene expression microarray were carried out. The expression status of miRNAs was confirmed by reverse transcriptase PCR. Furthermore, we investigated whether miRNA expression status in stromal tissue could influence the clinicopathologic factors.Results: Oncogenic miRNAs, including two miRNA clusters, miR-17-92a and miR-106b-25 cluster, were upregulated in cancer stromal tissues compared with normal stroma. Gene expression profiles from cDNA microarray analyses of the same stromal tissue samples revealed that putative targets of these miRNA clusters, predicted by Target Scan, such as TGFBR2, SMAD2, and BMP family genes, were significantly downregulated in cancer stromal tissue. Downregulated putative targets were also found to be involved in cytokine interaction and cellular adhesion. Importantly, expression of miR-25 and miR-92a in stromal tissues was associated with a variety of clinicopathologic factors.Conclusions: Oncogenic miRNAs were highly expressed in cancer stroma. Although further validation is required, the finding that stromal miRNA expression levels were associated with clinicopathologic factors suggests the possibility that miRNAs in cancer stroma are crucially involved in cancer progression.

An increasing number of studies have focused on circulating microRNAs (cmiRNA) in cancer patients’ blood for their potential as minimally-invasive biomarkers. Studies have reported the utility of assessing specific miRNAs in blood as diagnostic/prognostic biomarkers; however, the methodologies are not validated or standardized across laboratories. Unfortunately, there is often minimum limited overlap in techniques between results reported even in similar type studies on the same cancer. This hampers interpretation and reliability of cmiRNA as potential cancer biomarkers. Blood collection and processing, cmiRNA extractions, quality and quantity control of assays, defined patient population assessment, reproducibility, and reference standards all affect the cmiRNA assay results. To date, there is no reported definitive method to assess cmiRNAs. Therefore, appropriate and reliable methodologies are highly necessary in order for cmiRNAs to be used in regulated clinical diagnostic laboratories. In this review, we summarize the developments made over the past decade towards cmiRNA detection and discuss the pros and cons of the assays.