MG53 is a myokine modulating insulin signaling in several tissues; its relationship to glucose tolerance or risk of developing type 2 diabetes mellitus (T2DM) is unknown. This observational, prospective study aimed at evaluating the relationship between MG53 and glucose tolerance, testing whether its circulating levels may be associated with disease progression in a cohort at high risk of T2DM.

Five hundred and fifteen subjects who underwent a deep characterization of their glucose tolerance in the years 2003–2005 participated in this study. MG53 levels were measured at baseline. Glucose tolerance status was available over a follow-up of 15 ± 2 years for 283 of them; their vital status as of December 2020 was also retrieved.

MG53 levels were significantly lower in subjects with normal glucose tolerance than in subjects with impaired glucose regulation (IGR) or T2DM. Individuals in the highest MG53 levels quartile had more frequently 1h-post load glucose ≥ 155 mg/dL (54% vs 39%; p = 0.015), worse proportional control of β-cell function (p < 0.05–0.01), as determined by mathematical modeling, and worse Disposition Index (DI) (0.0155 ± 0.0081 vs 0.0277 ± 0.0030; p < 0.0001). At follow-up, baseline MG53 levels were higher in progressors than in non-progressors (120.1 ± 76.7 vs 72.7 ± 63.2 pg/ml; p = 0.001; ROC curve area for incident diabetes of 0.704). In a multivariable regression with classic risk factors for T2DM and DI, MG53 remained independently associated with progression with T2DM.

MG53 may be a novel biomarker of glucose dysregulation associated with β-cell dysfunction, likely improving our ability to identify, among high-risk subjects, those more likely to develop T2DM.