Identification of type 1 diabetes risk phenotypes using an outcome-guided clustering analysis
Authors' conclusion
Demographic, metabolic, immunological and genetic markers may be used to identify clusters of distinctive characteristics and different risks of progression to type 1 diabetes among autoantibody-positive individuals with a family history of type 1 diabetes. The results also revealed the heterogeneity in the population and complex interactions between variables.
Study context
- Population
- Initially non-diabetic islet autoantibody-positive relatives of individuals with T1D, TrialNet Pathway to Prevention studyvalidation n=706
- Participants
- 1,123 relatives
- Exposure
- Group A vs Group B vs Group C
- Outcome
- Time to development of T1D
Methodological detailsoutcome-guided clustering, independent validation dataset, A decision rule for assigning individuals to clusters was developed
- Outcome-guided clustering
- Independent validation dataset
- A decision rule for assigning individuals to clusters was developed
Main finding
As reported in the abstract
Diabetes-free survival probability · 2 years
| Group | Value | Median glucose mean AUC | Median IA-2A titre |
|---|---|---|---|
| Group A | 0.42 (95% CI 0.34–0.51) | 9.48 mmol/l (IQR 9.16-10.02) | |
| Group B | 0.73 (95% CI 0.67–0.80) | 287 DK units/ml (IQR 250-319) | |
| Group C | 0.84-0.99 |
Value
Group A0.42 (95% CI 0.34–0.51)
Group B0.73 (95% CI 0.67–0.80)
Group C0.84-0.99
Median glucose mean AUC
Group A9.48 mmol/l (IQR 9.16-10.02)
Group B— not stated in the abstract
Group C— not stated in the abstract
Median IA-2A titre
Group A— not stated in the abstract
Group B287 DK units/ml (IQR 250-319)
Group C— not stated in the abstract
- Six clusters with varying T1D risks, categorised into three groups
- 6
Observational data: associations, not effects.
Further findings
- Validation dataset confirmed the clusters can identify individuals with similar characteristics
Original publication
Identification of type 1 diabetes risk phenotypes using an outcome-guided clustering analysis. Diabetologia. Published 6 August 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →