Proof-of-concept Application of Continuous Glucose Monitoring Data Analytics to Identify Diabetes Glucotypes

Abstract

Background

In this proof-of-concept study, we evaluated if monogenic diabetes resulting from mutations of the HNF-1α gene (HNF1A-MODY) has a distinctive continuous glucose monitoring (CGM) glucotype, in comparison to type 1 diabetes (T1D).

Methods

Using CGM data from 5 subjects with HNF1A-MODY and 115 subjects with T1D, we calculated multiple glucose metrics, including measures of within- and between-day variability (such as coefficient variation for each hour [CV_{b_1h}]).

Results

The MODY and T1D cohorts had minimum CV_{b_1h} of 11.3 ± 4.4 and 18.0 ± 4.9, respectively (P = .02) and maximum CV_{b_1h} of 33.9 ± 5.0 and 50.3 ± 10, respectively (P < .001). All subjects with HNF1A-MODY had a minimum %CV_{b_1h} ≤ 17.3% and maximum %CV_{b_1h} ≤ 37.1%. In contrast, only 12 of 115 subjects with T1D had both a minimum and maximum %CV_{b_1h} below these thresholds (P < .001).

Conclusion

HNF1A- MODY is characterized by a low hourly, between-day glucose variability. CGM-derived glucose metrics may have potential applicability for screening for atypical diabetes phenotypes in the T1D population.