Background: Colorectal cancer (CRC) is a major public health issue, and screening programs utilizing the fecal immunochemical test (FIT) and colonoscopy represent the primary strategy for early detection. Liquid biopsy performed on fecal samples obtained from FIT leftovers emerges as a promising approach for investigating non-invasive biomarkers, such as microRNAs (miRNAs). These molecules exhibit high stability and abundance in body fluids and are involved in a wide range of processes in colorectal carcinogenesis. Although FIT demonstrates good population acceptance, its accuracy is limited, yielding false-positive and false-negative results. Therefore, there is a need to develop more precise non-invasive tests to optimize the detection of CRC and its precursor lesions, aiming to enhance screening reliability and efficacy.
Objectives: To identify a fecal microRNA expression profile associated with colorectal cancer and its precursor lesions.
Methods: The study evaluated 155 individuals, including screening program participants referred for diagnostic colonoscopy (following a positive FIT) and patients with CRC treated at Hospital de Câncer de Barretos. All participants collected stool samples using FIT tubes. RNA was extracted, and miRNA expression profiles were quantified using the NanoString nCounter platform. Data processing included quality control and normalization performed in R using the Quantro and NanoStringNorm packages. Identification of differentially expressed miRNAs between groups was performed using the limma package. Selection of differentially expressed miRNAs was based on an adjusted p-value < 0.05, followed by a Fold Change (FC) > 1.5.
Results: The miRNA expression data underwent quality control, and technical replicates demonstrated high reproducibility. Various data normalization methods were tested. A method utilizing a set of eight miRNAs identified as housekeepers was selected to reduce technical variation among samples. Differential Expression (DE) analysis between Normal and Early Adenoma groups identified 89 upregulated miRNAs with significant adjusted p-values, nine of which exhibited an FC > 1.5. No significant miRNAs were identified in the comparisons between Normal and Advanced Adenoma groups, or between Advanced Adenoma and Early Adenoma groups. Comparison between Cancer and Early Adenoma groups showed 130 significant miRNAs; of these, seven were upregulated (FC > 1.5) and nine were downregulated (FC < -1.5). Finally, DE analysis between Normal and Cancer groups revealed 73 significant miRNAs, nine of which had an FC > 1.5.  
Conclusion: Detection of miRNAs in FIT leftovers is feasible and could be used to identify individuals with precursor lesions and CRC in the context of CRC screening.