Colorectal Cancer (CRC) is a heterogeneous disease. Microarray molecular profiling has been used to identified tumor subtypes with different clinical behavior (1,2). Our main objective was to find out whether microRNAs (miR) were able to classify CRC tumors in the same way as mRNA and to identify miR targets in the colon tumors associated to the poor outcome High-stroma/CMS4 subtype.
mRNA and miR expression were analyzed in 88 colorectal tumors (24 Dukes A, 26 B, 19 C, 19 D) using gene expression and miR microarrays. Hierarchical clustering was used to classified tumors based on miR expression. Association or miR subtypes to mRNA subtypes and to CMS was carried out. TALASSO software (http://talasso.cnb.csic.es) was used to find miR-mRNA interactions. Candidate gene-miR interactions were scored and biologically validated in 293T cell line.
Expression profiling of miRs revealed three molecular subtypes clearly associated with those obtained by mRNA expression profiling (p = 0.000 for HC subtypes and p = 0.001 for CMS subtypes). miR subtype 1 associates with the Low-stroma/CMS2 Subtypes, miR subtype 2 correlates with the High-stroma/CMS4 subtypes and miR subtype 3 is associated with Mucinous-MSI/CMS1 subtypes. 788 genes showed significant interaction with 176 miRs (p
1. There is a clear correlation between mRNA classification and 3 novel miR subtypes. 2. miR subtype 1 is associated with Low-Stroma/CMS1 subtype, miR subtype 2 with High-stroma/CMS4 subtypes and miR subtype 3 with Mucinous-MSI/CMS1 subtypes. 3. High-stroma gene (SCL6A6) shows specific interaction with miR-30b.
Clinical trial identification
Legal entity responsible for the study
Instituto de Investigación Sanitaria San Carlos, Hosptal Clínico San Carlos
Fundación Mutua Madrileña, Fundación Rodríguez Pascual, Fundación Merck Salud.
All authors have declared no conflicts of interest.