Altered profile of serum microRNAs in pancreatic cancer-associated new-onset diabetes mellitus

Background: New-onset diabetes mellitus in pancreatic cancer has been recognized as a paraneoplastic phenomenon caused by the existence of the tumor. Circulating microRNAs (miRNAs) are emerging as non-invasive biomarkers for the detection of various cancers. In the present study, we hypothesized that a specific serum miRNA profile exists in pancreatic cancer-associated new-onset diabetes mellitus (PaC-DM). Methods: Initial screening of differentially expressed miRNAs in pooled serum samples from 25 PaC-DM patients, 25 non-cancer new-onset type 2 diabetes mellitus (T2DM) patients, and 25 healthy controls was performed by TaqMan low-density arrays (TLDA). A stem-loop quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was conducted to confirm the relative concentrations of candidate miRNAs in 80 PaC-DM, 85 non-cancer new-onset T2DM patients, and 80 healthy controls. Results: The TLDA identified 16 serum miRNAs that were significantly increased in PaC-DM samples. A combination of six serum miRNAs (miR-483-5p, miR-19a, miR-29a, miR-20a, miR-24, miR-25) was selected by qRT-PCR as a biomarker for PaC-DM. The area under the receiver operating characteristic curve (AUC) for the six-miRNA panel training and validation sets was 0.959 (95% confidence interval [CI] 0.890-1.028) and 0.902 (95% CI 0.844-0.955), respectively. The combination of these six miRNAs enabled the discrimination of PaC-DM from non-cancer new-onset T2DM with an AUC of 0.885 (95% CI 0.784-0.986) and 0.887 (95% CI 0.823-0.952) for the training and validation sets, respectively. Conclusion: The six-serum miRNA panel may have potential as a biomarker for the accurate diagnosis and discrimination of PaC-DM from healthy controls and non-cancer new-onset T2DM.