Using CGM to predict progression to type 1 diabetes in children with positive islet autoantibodies

Children who have been identified as having multiple islet autoantibodies are at high risk of developing clinical type 1 diabetes. Among those who do develop the condition, there is a period of impaired fasting glucose or impaired glucose tolerance that may precede type 1 diabetes by several months or even years.

Since 2017, the ASK study in Colorado, USA, has screened 22 566 children (aged 1–17 years) for islet and coeliac autoantibodies. The aim of the present study was to assess whether continuous glucose monitoring (CGM) metrics could be used to predict progression to clinical type 1 diabetes in those who were at high risk.

Out of the screened population, 158 children were identified as being islet autoantibody-positive and thus eligible for CGM; of these, 91 wore a CGM device for a 7–10-day period and were followed for development of type 1 diabetes for a median of 6 months (range, 0.2–34 months). Of these, 16 (18%) progressed to diabetes, in a median of 4.5 months (range, 0.4–29 months). Age, sex, ethnicity and BMI were comparable between those who progressed to diabetes and those who did not.

Baseline HbA_1c was higher in progressors than non-progressors (38 vs 33 mmol/mol [5.6% vs 5.2%]), and progressors had higher glycaemic variability and mean sensor glucose. Progressors spent 21% of time with glucose levels >140 mg/dL (7.8 mmol/L), compared with 3% for non-progressors. A cut-off value of 10% of time above 7.8 mmol/L had 88% sensitivity and 91% specificity for prediction of type 1 diabetes. The risk of progressing to type 1 diabetes in 1 year was 80% in those with >10% of time above 7.8 mmol/L, while it was only 5% in those with ≤10% of time above this level.

Thus, the authors recommend this cut-off as a marker of high risk of progressing to type 1 diabetes in children with positive islet autoantibodies. Various other CGM metrics were also accurate predictors; in particular, >15% time above 7.8 mmol/L had lower sensitivity but higher specificity, and the authors propose this metric as a predictor of developing diabetes within the next 6 months.

This study also raises some interesting questions about the utility of islet autoantibody screening. Current ADA guidelines recommend screening in first-degree relatives of people with type 1 diabetes. Notably, early identification reduces the risk of diabetic ketoacidosis (DKA) at diagnosis, which in turn is associated with better long-term glycaemic control, in the order of 3–15 mmol/mol over 15 years (Duka et al, 2017). To date, the rate of DKA at diagnosis in the ASK study has been 6%, compared with 62% in Colorado as a whole. Autoantibody screening has been shown to be cost-effective if it decreases the relative risk of DKA by 20% and thus lowers HbA_1c by 1 mmol/mol.

Click here to read the study in full.