Plasma MicroRNA Signature Validation for Early Detection of Colorectal Cancer. Issue 1 (January 2019)

• Record Type: Journal Article
• Authors: Herreros-Villanueva, Marta; Duran-Sanchon, Saray; Martín, Ana Carmen; Pérez-Palacios, Rosa; Vila-Navarro, Elena; Marcuello, María; Diaz-Centeno, Mireia; Cubiella, Joaquín; Diez, Maria Soledad; Bujanda, Luis; Lanas, Angel; Jover, Rodrigo; Hernández, Vicent; Quintero, Enrique; José Lozano, Juan; García-Cougil, Marta; Martínez-Arranz, Ibon; Castells, Antoni; Gironella, Meritxell; Arroyo, Rocio
• Is Part Of: Clinical and translational gastroenterology. Volume 10:Issue 1(2019)
• Publication Date: 2019-01
• Subjects: Stomach -- Diseases -- Periodicals; Intestines -- Diseases -- Periodicals; Gastroenterology; Gastrointestinal Diseases; Liver Diseases; Intestines -- Diseases; Stomach -- Diseases; Periodical; Periodicals; Fulltext; Internet Resources; Periodicals; Electronic journals; 616.33
• DOI: 10.14309/ctg.0000000000000003
• Languages: English
• ISSNs: 2155-384X

Abstract

Abstract : OBJECTIVES: Specific microRNA (miRNA) signatures in biological fluids can facilitate earlier detection of the tumors being then minimally invasive diagnostic biomarkers. Circulating miRNAs have also emerged as promising diagnostic biomarkers for colorectal cancer (CRC) screening. In this study, we investigated the performance of a specific signature of miRNA in plasma samples to design a robust predictive model that can distinguish healthy individuals from those with CRC or advanced adenomas (AA) diseases. METHODS: Case control study of 297 patients from 8 Spanish centers including 100 healthy individuals, 101 diagnosed with AA, and 96 CRC cases. Quantitative real-time reverse transcription was used to quantify a signature of miRNA (miRNA19a, miRNA19b, miRNA15b, miRNA29a, miRNA335, and miRNA18a) in plasma samples. Binary classifiers (Support Vector Machine [SVM] linear, SVM radial, and SVM polynomial) were built for the best predictive model. RESULTS: Area under receiving operating characteristic curve of 0.92 (95% confidence interval 0.871–0.962) was obtained retrieving a model with a sensitivity of 0.85 and specificity of 0.90, positive predictive value of 0.94, and negative predictive value of 0.76 when advanced neoplasms (CRC and AA) were compared with healthy individuals. CONCLUSIONS: We identified and validated a signature of 6 miRNAs (miRNA19a, miRNA19b, miRNA15b, miRNA29a, miRNA335, and miRNA18a) as predictors that can differentiate significantly patients with CRC and AA from those who are healthy. However, large-scale validation studies in asymptomatic screening participants should be conducted.