Expression of cell-free miRNAs from plasma of Brazilian breast cancer patients

in this study, we aimed to determine the cell-free miRNA (cf-miRNA) profile between five cohorts of BC patients according to main molecular subtypes. These samples were classified according to their age at diagnosis in two groups: Young (diagnosed before 40 years old) and Elderly (diagnosed after 40 years old). We isolated total miRNA from the plasma of these patients and then assessed their miRNA levels using a high-throughput and high-resolution technique based on digital barcode hybridization. We collected plasma samples of breast cancer patients diagnosed at Hospital das Clinicas, Medicine School (Universidade de São Paulo). Using 200 uL of this plasma, we isolated cell-free miRNA (cf-miRNA) following the instructions of the miRNeasy Serum/plasma kit (cat. 217184, Qiagen). Then, we quantified the cf-miRNA by spectrophotometry and used 40 or 120 µL of each sample were concentrated up to 4 µL using the Eppendorf® 5301 concentrator for 20 or 35 minutes at 45 °C following the manufacturer’s recommendations. After concentration, 0.5 µL of each sample was loaded into the NanoDrop™ 8000 Spectrophotometer (ThermoFisher) to verify the miRNA mass. Then, ~25 ng of cf-miRNA were hybridized for 16.5 hours with molecular barcoding for 827 experimentally validated human miRNAs (from miRBase v21) using the nCounter Master Mix (NanoString® Technologies, cat. NAA-AKIT-012). Subsequently, miRNAv3 NanoString® cartridges (NanoString® Technologies, cat. CSO-MIR3-12) were loaded with the mix per sample, sealed, and transferred to an nCounter® Digital Analyzer device (NanoString® Technologies) for data collection. miRNA expression data were analyzed in the nSolver™ Data Analysis v.4.0 software (NanoString® Tech.) with the default protocol by the manufacturer. For cf-miRNA, we normalize the expression of human miRNAs by the geometric mean of all negative (negative normalization) and positive (positive normalization) probes. Then, we used the top 15 more stable regions (up to 15% of the coefficient of variation) to normalize the content set data. Data were exported in comma-separated value (*.csv) format for further analysis.