Continuous glucose monitoring using the FreeStyle Libre Pro iQ in presymptomatic type 1 diabetes.

AIMS/HYPOTHESIS: The aim of the study was to evaluate the ability of the FreeStyle Libre Pro iQ continuous glucose monitoring (CGM) system to discriminate between presymptomatic stages of type 1 diabetes and stratify progression to stage 3 type 1 diabetes. The study also aimed to compare the performance of the FreeStyle Libre Pro iQ with that of the Dexcom G6.

METHODS

From December 2023 to December 2025, 50 children and adolescents without islet autoantibodies (control participants) and 89 with early-stage (stage 1 or 2) type 1 diabetes underwent CGM with the FreeStyle Libre Pro iQ. Participants with early-stage type 1 diabetes were followed up for the development of stage 3 clinical type 1 diabetes.

RESULTS

CGM data were unavailable or incomplete for 13 participants (9%), leaving 126 (91%) for analysis (73 female, median age 10.0 years [IQR 7.9-12.8]). The CGM parameters used-namely the SD, CV, time with glucose levels above 7.8 mmol/l (140 mg/dl; TA140) and 8.9 mmol/l (160 mg/dl; TA160) and time in tight range-differed significantly across participants with no early-stage, stage 1 or stage 2 type 1 diabetes, and between participants with a single dysglycaemia abnormality and those with multiple dysglycaemia abnormalities. One or more of these parameters exceeded the 97.5th percentile of control values in 10 (22%) of the 45 individuals with stage 1 type 1 diabetes and in 13 (52%) of the 25 individuals with stage 2 type 1 diabetes (p=0.017). Two or more of the parameters exceeded the 97.5th percentile threshold in six (13%) of the individuals with stage 1 type 1 diabetes and 11 (44%) of the individuals with stage 2 type 1 diabetes (p=0.0077). The 97.5th percentile threshold for TA140 (i.e. >15%) was exceeded by none of the 45 individuals with stage 1 type 1 diabetes, 10 (40%) of the 25 individuals with stage 2 type 1 diabetes and six of the eight individuals who progressed to clinical diabetes within 1 year, identifying these individuals as rapid progressors. FreeStyle Libre Pro iQ values differed markedly from those previously observed using the Dexcom G6 in a similar cohort. Using the 97.5th percentile of values of control individuals to define CGM abnormalities enabled harmonisation across systems and supported the consistent identification of high-risk individuals. With this approach, the sensitivity for identifying those who progressed to clinical diabetes within 1 year ranged from 75% to 100% for the FreeStyle Libre Pro iQ and from 50% to 100% for the Dexcom G6.

CONCLUSIONS/INTERPRETATION: These findings support CGM as a promising tool for staging and risk stratification in early-stage type 1 diabetes, while highlighting the need for system-specific interpretation before use in therapeutic decision-making.