Aims/hypothesis: We aimed to analyse the association between plasma circulating microRNAs (miRNAs) and the immunometabolic profile in children with type 1 diabetes and to identify a composite signature of miRNAs/immunometabolic factors able to predict type 1 diabetes progression.
Methods: Plasma samples were obtained from children at diagnosis of type 1 diabetes (n = 88) and at 12 (n = 32) and 24 (n = 30) months after disease onset and from healthy control children with similar sex and age distribution (n = 47). We quantified 60 robustly expressed plasma circulating miRNAs by quantitative RT-PCR and nine plasma immunometabolic factors with a recognised role at the interface of metabolic and immune alterations in type 1 diabetes. Based on fasting C-peptide loss over time, children with type 1 diabetes were stratified into the following groups: those who had lost >90% of C-peptide compared with diagnosis level; those who had lost <10% of C-peptide; those showing an intermediate C-peptide loss. To evaluate the modulation of plasma circulating miRNAs during the course of type 1 diabetes, logistic regression models were implemented and the correlation between miRNAs and immunometabolic factors was also assessed. Results were then validated in an independent cohort of children with recent-onset type 1 diabetes (n = 18). The prognostic value of the identified plasma signature was tested by a neural network-based model.
Results: Plasma circulating miR-23~27~24 clusters (miR-23a-3p, miR-23b-3p, miR-24-3p, miR-27a-3p and miR-27b-3p) were upmodulated upon type 1 diabetes progression, showed positive correlation with osteoprotegerin (OPG) and were negatively correlated with soluble CD40 ligand, resistin, myeloperoxidase and soluble TNF receptor in children with type 1 diabetes but not in healthy children. The combination of plasma circulating miR-23a-3p, miR-23b-3p, miR-24-3p, miR-27b-3p and OPG, quantified at disease onset, showed a significant capability to predict the decline in insulin secretion 12 months after disease diagnosis in two independent cohorts of children with type 1 diabetes.
Conclusions/interpretations: We have pinpointed a novel miR-23a-3p/miR-23b-3p/miR-24-3p/miR-27b-3p/OPG plasma signature that may be developed into a novel blood-based method to better stratify patients with type 1 diabetes and predict C-peptide loss.
Keywords: Biomarkers; Cardiovascular risk; Diabetic complications; Immunometabolism; Insulin secretion; Osteoprotegerin; microRNA.