Continuous Monitoring of Glycemic Variability to Predict Dys- and Hyperglycemia in Asymptomatic Type 1 Diabetes

Type 1 diabetes is a for now incurable disease caused by a major immune-mediated loss of insulin-producing pancreatic beta cells, can lead to potentially severe acute and chronic complications and requires a lifelong insulin treatment. Clinical onset of type 1 diabetes is preceded by an asymptomatic disease phase of highly variable duration, which is signaled by the presence of multiple (≥2) types of islet autoantibodies. Multiple autoantibody-positive individuals, with blood glucose levels still within normal limits (defined as stage 1 type 1 diabetes) have a 90% risk of developing symptomatic disease within the next 20 years. The development of dysglycemia (stage 2 type 1 diabetes) - i.e. disturbed blood glucose levels during an oral glucose tolerance test (OGTT) - dramatically raises the risk of impending clinical onset to 90% within 5 years. Identifying individuals at high risk of impending clinical onset of type 1 diabetes is important for early diagnosis, for reducing the incidence of inaugural ketoacidosis, and for enrolling participants of choice in immune intervention trials before and at clinical diagnosis, aiming to develop an effective cure or even better, prevention. The hyperglycemic clamp test is the gold standard for assessing beta cell function and has also been validated for estimating insulin action in parallel. Decreased clamp-derived measures of islet function and insulin action have been shown to outperform OGTT-derived variables for predicting progression to symptomatic disease. However, the repeated performance of both OGTTs and clamps are cumbersome and difficult to implement on a large scale in a seemingly healthy population. In contrast, continuous glucose monitoring (CGM) methods nowadays avoid the need of frequent calibration based on capillary blood measurements obtained by finger pricks. They also allow to detect more subtle glycemic fluctuations over longer observation periods and on a more frequent basis than achievable with OGTT, and to derive a wide variety of indices of glycemic variability. Preliminary findings suggest that repeated CGM metrics in stage 1 asymptomatic diabetes could represent a minimally invasive alternative to OGTT for early detection or prediction of stage 2 asymptomatic disease, provided that increased CGM-derived glycemic excursions can be shown to coincide with or precede OGTT-inferred dysglycemia, respectively, in a longitudinal study. In addition, it is anticipated that further increasing glycemic excursions may forecast impending clinical onset. Using the capacity of the nationwide network of the Belgian Diabetes Registry, we therefore propose a longitudinal study to measure variability of interstitial glucose levels with a userfriendly real-time CGM technology at regular intervals (every 6 months) during follow-up of multiple autoantibody-positive individuals (age 5-39 years; n>50) of type 1 diabetes patients, in comparison with single autoantibody-positive individuals (n>25). In parallel repeated OGTTs (age 5-39 years) and hyperglycemic clamp tests (age 12-39 years) will be performed for a period of at least 2-3 years. We anticipate that, especially when using up-to-date CGM technology, (i) the amplitude and time trends of various glycemic variability indices - alone or in combination - will differ between groups of individuals at moderate (single autoantibody-positive), high (stage 1), and very high (stage 2) risk of impending clinical onset (stage 3); (ii) (changes in) CGM-derived glycemic variability indices will be able to predict and/or diagnose stage 2 asymptomatic type 1 diabetes and clinical onset (stage 3) with a diagnostic efficiency equaling that of variables derived from (repeated) OGTT or clamp tests; (iii) correlating CGM metrics with hyperglycemic clamp- and OGTT-derived indices of beta cell function and insulin action will help to understand the relative contribution of both components to disease progression in general, and to glycemic variability in particular.