Abstract

Abstract Introduction: Recent studies have reported that serum microRNAs (miRNAs) are potentially useful biomarkers for cancer. However, the detection system using serum miRNAs is not established in colorectal cancer (CRC). The aims of this study are 1) to identify specific serum miRNAs and establish the discriminant model for CRC detection and 2) to validate the miRNAs and the discriminant model using a large cohort. Methods: First, we identified serum miRNAs related with the presence of CRC and constructed the discriminant model for CRC detection using the publicly-released serum miRNA database including 50 patients with CRC and 90 healthy individuals (GSE59856 and GSE73002). Comparing serum miRNA levels between patients with CRC and healthy individuals, we picked up miRNAs (p<0.05 by t-test and fold change > 2). Using these miRNAs, Fisher’s linear discriminant analysis was performed and the diagnostic models were constructed using less than 7 miRNAs. Subsequently, 1117 patients of National Cancer Center Hospital with CRC were enrolled as a validation cohort. Patients with the following criteria were excluded; (i) patients simultaneously or previously diagnosed as the other cancers, (ii) patients who were treated for CRC previously and (iii) patients with familial adenomatous polyposis or Lynch syndrome. Control blood samples were obtained from patients without any history of cancer who were admitted or referred to National Center for Geriatrics and Gerontology or Yokohama Minoru Clinic between 2010 and 2015. A total of 1013 CRC patients and 4384 non-cancerous patients were analyzed. Total RNA was extracted from 300 micro L of serum and comprehensive miRNA expression analysis was performed using a 3D-Gene microarray. ROC analysis was performed to evaluate the previous discriminant models. The sensitivity in each pathological stage and location of CRC was also investigated.This research is partially supported by the “Development of Diagnostic Technology for Detection of miRNA in Body Fluids” grant from the Japan Agency for Medical Research and Development (AMED). Results:First we picked up 30 miRNAs for CRC diagnosis. Fisher’s analysis revealed 107611 candidates of numerical formulas, that showed > 80% of sensitivity and specificity. We could narrow these candidates to 43 formulas with validation analysis. By ROC analysis, the AUC, sensitivity and specificity of the discriminant formula with best performance was 0.904, 79.6% and 86.5%, respectively. The sensitivities of each pathological stage were as follows; pStage 0: 79.0%, pStage I: 90.7%, pStage II: 85.1%, pStage III: 73.7%, pStage IV: 62.2%. The sensitivity of CRC in the right side of the colon was 79.4%, whereas that in the left side was 79.7%. Conclusions:We identified novel serum miRNAs for CRC detection. Our discriminant using these miRNAs can diagnose CRC including early stage. Furthermore, the sensitivity was high irrespective of the tumor location. Citation Format: Hiroyuki Takamaru, Yutaka Saito, Taku Sakamoto, Seiichiro Abe, Masayoshi Yamada, Takeshi Nakajima, Kazuki Sudo, Ken Kato, Junpei Kawauchi, Satoko Takizawa, Hiromi Sakamoto, Motohiro Kojima, Atsushi Ochiai, Shumpei Niida, Hideshi Ishii, Tomoko Takamaru, Juntaro Matsuzaki, Takahisa Matsuda, Takahiro Ochiya. Novel serum microRNAs that enable liquid biopsy for colorectal cancer: validation study of large cohort [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5678. doi:10.1158/1538-7445.AM2017-5678

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call