Identification of a six-miRNA panel in serum benefiting pancreatic cancer diagnosis

Abstract

Pancreatic cancer (PC) has posed a great health threat to a growing number of people all over the world. Detection of serum miRNAs, being sensitive, noninvasive, and easy to obtain, has a great potential of being a novel screening method for PC patients. In this study, we investigated miRNA expression levels in serum by qRT-PCR. The study was divided into four phases: the screening, training, testing, and external validation stage. We firstly chose candidate miRNAs using Exiqon panels in the screening phase. Then, a total of 129 PC serum samples and 107 normal controls (NCs) were further analyzed in the following training and testing phases to identify differently expressed miRNAs. A cohort of 30 PC serum samples vs 30 NCs was used to confirm the diagnostic value of the identified miRNAs in the external validation phase. Moreover, miRNA expressions in additional 44 PC tumor tissue samples and the matched adjacent normal tissue samples as well as 32 pairs of serum-derived exosomes samples were also further explored. As a result, we identified six significantly upregulated miRNAs in the serum of PC: let-7b-5p, miR-192-5p, miR-19a-3p, miR-19b-3p, miR-223-3p, and miR-25-3p. A six-miRNA panel in serum was then established. The area under the receiver operating characteristic curves (AUC) for the panel was 0.910 for the combined training and testing phases, which showed higher diagnostic value than the individual miRNA. Prognostic value prediction using Cox's proportional hazards model and Kaplan-Meier curves showed that increased serum miR-19a-3p was closely related to worse overall survival (OS). In addition, significant upregulation of miR-192-5p, miR-19a-3p, and miR-19b-3p was observed in both PC tissue and serum-derived exosomes samples. In conclusion, we identified a six-miRNA (let-7b-5p, miR-192-5p, miR-19a-3p, miR-19b-3p, miR-223-3p, and miR-25-3p) panel in the serum for PC early and noninvasive diagnosis.

Keywords: biomarker; pancreatic cancer; qRT-PCR; serum miRNA.

Figure 1

The flow chart of experiment design. PC: pancreatic cancer; NC: normal control

Figure 2

Expression levels of six miRNAs in the serum of 129 PC patients and 107 NCs (in the training and testing phases). N: normal control; T: tumor. Horizontal line: mean with 95% CI

Figure 3

ROC curve analyses of the six‐miRNA signature to discriminate PC patients from NCs. (A) The combined two cohorts of training and testing phases (129 PC vs 107 NCs); (B) external validation phase (30 PC vs 30 NCs). AUC: areas under the curve; ROC curve: receiver operating characteristic curve

Figure 4

Expression levels of six miRNAs in 44 PC tumor tissues and the matched adjacent normal tissues. T: tumor; N: normal control

Figure 5

Expression levels of six miRNAs in 32 PC serum‐derived exosomes and 32 NCs. T: tumor; N: normal control

Figure 6

Heat‐maps of pathway investigation using KEGG (A) and GO (B) analyses. KEGG: Kyoto Encyclopedia of Genes and Genomes; GO: Gene Ontology