Login / Signup

Plasma miRNA-based signatures in CRC screening programs.

Susanna ZanuttoChiara Maura CiniselliAntonino BelfioreMara LecchiEnzo MasciGabriele DelconteMassimo PrimignaniGiulia TosettiMarco Dal FanteLinda FazziniAldo AiroldiMarcello VangeliFrancesca TurpiniGiovanni Giuseppe Rubis PassoniPaolo ViaggiMonica ArenaRoberta Ilaria Olimpia MottaAnna Maria CantùCristiano CrostaGiuseppe De RobertoFrancesca IannuzziAndrea CassinottiValentina Dall'OlioLaura TizzoniGabriella SozziEmanuele MeroniLuigi BisantiMarco Alessandro PierottiPaolo VerderioGariboldi Manuela
Published in: International journal of cancer (2019)
Colorectal cancer (CRC) screening programs help diagnose cancer precursors and early cancers and help reduce CRC mortality. However, currently recommended tests, the fecal immunochemical test (FIT) and colonoscopy, have low uptake. There is therefore a pressing need for screening strategies that are minimally invasive and consequently more acceptable to patients, most likely blood based, to increase early CRC identification. MicroRNAs (miRNAs) released from cancer cells are detectable in plasma in a remarkably stable form, making them ideal cancer biomarkers. Using plasma samples from FIT-positive (FIT+) subjects in an Italian CRC screening program, we aimed to identify plasma circulating miRNAs that detect early CRC. miRNAs were initially investigated by quantitative real-time PCR in plasma from 60 FIT+ subjects undergoing colonoscopy at Fondazione IRCCS Istituto Nazionale dei Tumori, then tested on an internal validation cohort (IVC, 201 cases) and finally in a large multicenter prospective series (external validation cohort [EVC], 1121 cases). For each endoscopic lesion (low-grade adenoma [LgA], high-grade adenoma [HgA], cancer lesion [CL]), specific signatures were identified in the IVC and confirmed on the EVC. A two-miRNA-based signature for CL and six-miRNA signatures for LgA and HgA were selected. In a multivariate analysis including sex and age at blood collection, the areas under the receiver operating characteristic curve (95% confidence interval) of the signatures were 0.644 (0.607-0.682), 0.670 (0.626-0.714) and 0.682 (0.580-0.785) for LgA, HgA and CL, respectively. A miRNA-based test could be introduced into the FIT+ workflow of CRC screening programs so as to schedule colonoscopies only for subjects likely to benefit most.
Keyphrases