RESUMO
Immunomodulation combined with antigen therapy holds great promise to arrest autoimmune type 1 diabetes, but clinical translation is hampered by a lack of prognostic biomarkers. Low-dose anti-CD3 plus Lactococcus lactis bacteria secreting proinsulin and IL-10 reversed new-onset disease in nonobese diabetic (NOD) mice, yet some mice were resistant to the therapy. Using miRNA profiling, six miRNAs (i.e., miR-34a-5p, miR-125a-3p, miR-193b-3p, miR-328, miR-365-3p, and miR-671-3p) were identified as differentially expressed in plasma of responder versus nonresponder mice before study entry. After validation and stratification in an independent cohort, plasma miR-193b-3p and miR-365-3p, combined with age and glycemic status at study entry, had the best power to predict, with high sensitivity and specificity, poor response to the therapy. These miRNAs were highly abundant in pancreas-infiltrating neutrophils and basophils with a proinflammatory and activated phenotype. Here, a set of miRNAs and disease-associated parameters are presented as a predictive signature for the L. lactis-based immunotherapy outcome in new-onset type 1 diabetes, hence allowing targeted recruitment of trial participants and accelerated trial execution. ARTICLE HIGHLIGHTS: Low-dose anti-CD3 combined with oral gavage of genetically modified Lactococcus lactis bacteria secreting human proinsulin and IL-10 holds great promise to arrest autoimmune type 1 diabetes, but the absence of biomarkers predicting therapeutic success hampers clinical translation. A set of cell-free circulation miRNAs together with age and glycemia at baseline predicts a poor response after L. lactis-based immunotherapy in nonobese mice with new-onset diabetes. Pancreas-infiltrating neutrophils and basophils are identified as potential cellular sources of discovered miRNAs. The prognostic signature could guide targeted recruitment of patients with newly diagnosed type 1 diabetes in clinical trials with the L. lactis-based immunotherapy.