Diagnostic microRNA markers to screen for sporadic human colon cancer in stool: I. Proof of principle.
To present proof-of-principle application for employing micro(mi)RNAs as diagnostic markers for colon cancer, we carried out global microarray expression studies on stool samples obtained from fifteen individuals (three controls, and three each with TNM stage 0-1, stage 2, stage 3, and stage 4 colon cancer), using Affymetrix GeneChip miRNA 3.0 Array, to select for a panel of miRNA genes for subsequent focused semi-quantitative polymerase chain reaction (PCR) analysis studies. Microarray results showed 202 preferentially expressed miRNA genes that were either increased (141 miRNAs), or reduced (61 miRNAs) in expression. We then conducted a stem-loop reverse transcriptase (RT)-TaqMan® minor groove binding (MGB) probes, followed by a modified qPCR expression study on 20 selected miRNAs. Twelve of the miRNAs exhibited increased and 8 decreased expression in stool from 60 individuals (20 controls, 20 with tumor-lymph node-metastatic (TNM) stage 0-1, 10 with stage 2, five with stage 3, and 5 with stage 4 colon cancer) to quantitatively monitor miRNA changes at various TNM stages of colon cancer progression. We also used laser-capture microdissection (LCM) of colon mucosal epithelial tissue samples (three control samples, and three samples from each of the four stages of colon cancer, for a total of 15 samples) to find concordance or lack thereof with stool findings. The reference housekeeping pseudogene-free ribosomal gene (18S rRNA), which shows little variation in expression, was employed as a normalization standard for relative PCR quantification. Results of the PCR analyses confirmed that twelve miRNAs (miR-7, miR-17, miR-20a, miR-21, miR-92a, miR-96, miR-106a, miR-134, miR-183, miR-196a, miR-199a-3p and miR214) had an increased expression in the stool of patients with colon cancer, and that later TNM carcinoma stages exhibited a more pronounced expression than did adenomas. On the other hand, eight miRNAs (miR-9, miR-29b, miR-127-5p, miR-138, miR-143, miR-146a, miR-222 and miR-938) had decreased expression in the stool of patients with colon cancer, which was also more pronounced from early to later TNM stages. Results from colon mucosal tissues were similar to those from stool samples, although with more apparent changes in expression. Cytological studies on purified stool colonocytes that employed Giemsa staining showed 80% sensitivity for detecting tumor cells in stool smears. The performance characteristics of the test confirmed that stool is a medium well-suited for colon cancer screening, and that the quantitative changes in the expression of few mature miRNA molecules in stool associated with colon cancer progression provided for more sensitive and specific non-invasive diagnostic markers than tests currently available on the market. Thus, a larger prospective and properly randomized validation study of control individuals and patients exhibiting various stages of colon cancer progression (TNM stages 0-IV) is now needed in order to standardize test conditions, and provide a means for determining the true sensitivity and specificity of a miRNA screening approach in stool for the non-invasive detection of colon cancer, particularly at an early stage (0-I). Eventually, we will develop a chip to enhance molecular screening for colon cancer, as has been accomplished for the detection of genetically-modified organisms (GMOs) in foods.