Identification of microRNAs for the early diagnosis of colorectal cancer (CRC)

Introduction: microRNAs (miRNAs) are small non-coding RNAs involved in the development of various cancers. Quantitative real-time PCR (qPCR) is the assay commonly used to investigate miRNA expression and qPCR-low-density arrays are the most used technique for both identification and validation of modulated miRNAs. One crucial pre-processing step for miRNA analysis is data normalization, aimed at reducing nonbiological sources of variation [1]. This process would allow to identify a small set of miRNAs to be used for data normalization in subsequent validation studies [2]. Materials and methods: we analyzed the expression levels of 381 human miRNAs on TaqMan Array MicroRNA Card A v.2 (Applied Biosystems), on a cohort of 60 plasma samples (38 precancerous lesions/cancer and 22 without lesion) from individuals enrolled in the CRC screening program of the Milan Local Health Authority that underwent colonoscopy at our Institute (INT) after a positive fecal occult blood test (FIT+). Starting from these data, we developed and applied a data-driven normalization method able to identify a small set of reference miRNAs to use for data normalization in the subsequent validation studies [2,3]. Briefly, by considering the miRNAs expressed in all the samples, the relative expression of each miRNA was first computed according to their mean expression value [4] and the best subset of miRNAs that resemble this value was selected [2,3]. Results and discussion: we identified 4 housekeeping miRNAs suitable for data normalization and 7 miRNAs significantly different in subjects with precancerous/cancerous lesions versus subjects without lesions. We also identified 4 miRNAs related to presence of initial adenoma, one linked to advanced adenoma and 8 to presence of cancerous lesion. We are now constructing of a Custom TaqMan Array Cards including the identified miRNAs and some other miRNAs of interest, i.e those haemolysis-related and miR-378 [5], with the aim of validating these biomarkers on a prospective cohort of 120 FIT+ subjects that underwent colonoscopy at INT. Conclusion: the adopted strategy allowed the identification of reference miRNAs to be used for data normalization and the identification of modulated miRNAs that will be validated in larger prospective series. Acknowledgements: This work was supported by grants from Associazione Italiana per la Ricerca sul Cancro (AIRC) (Grants No.10529 and No. 12162 to MA Pierotti).