Cluster Analysis of Younger-Onset Type 2 Diabetes in an Asian Cohort Reveals Distinct Subgroups With Differential Pathophysiology and Outcomes
Diabetes, 2025
Tan C., Kee K., Zheng H., Wong K., Chan W., Song Y., Ang K., Subramaniam T., Sum C., Lim S.
| Disease area | Application area | Sample type | Products |
|---|---|---|---|
Metabolic Diseases | Patient Stratification | Plasma | Olink Target 96 |
Abstract
Younger-onset type 2 diabetes (T2D, onset <40 years) represents a growing global health challenge, characterized by heterogenous pathophysiology and accelerated complications. Current ‘one-size-fits-all” treatment approaches may be inadequate for this population. To address this heterogeneity, we performed clinical variable-based clustering using BMI, onset age, HbA1c and HOMA2 indices in 717 participants across discovery and validation cohorts. Three distinct subgroups were identified: mild obesity-related diabetes (MOD), severe insulin-deficient diabetes (SIDD) and severe insulin-resistant diabetes with insulin insufficiency (SIRD-II). Over median follow-up of 2.8 years, SIRD-II demonstrated 11-fold increased risk of progressive chronic kidney disease, while both SIDD and SIRD-II showed threefold increased risk for progressive albuminuria compared with MOD. SIRD-II also demonstrated 3.5-fold and 2.3-fold higher 10-year cardiovascular risk compared with SIDD and MOD respectively. Metabolomic analysis revealed distinct signatures: SIDD exhibited lower levels of lipids, amino acids, and inflammatory markers, while SIRD-II demonstrated elevated glucose, lipids, and branched-chain amino acids, suggesting glucolipotoxicity. Proteomics analysis validated previously reported biomarkers (IGFBP1, RTN4R, PLXNB2) and identified additional molecules (CDHR2, ERBB4, DPP6) that may shed light on disease mechanisms. In conclusion, younger-onset T2D exhibits distinct subgroups with differential pathobiology, molecular signatures, and clinical outcomes, suggesting the need for personalised precision diabetes care.
ARTICLE HIGHLIGHTS
To understand the heterogeneity of younger-onset type 2 diabetes, clinical data-driven clustering was performed, which identified three distinct subgroups that were replicated in an independent cohort. Compared with the mild obesity-related diabetes (MOD) subgroup, both severe insulin-deficient diabetes (SIDD) and severe insulin-resistant diabetes with insulin insufficiency (SIRD-II) subgroups had higher risk of developing diabetes-related complications. Differential molecular signatures confirmed the biological distinctiveness of younger-onset T2D subgroups and highlight potential mechanisms, such as glucolipotoxicity stress that may drive complications in the SIRD-II subgroup. Proteomic analyses validated previously reported biomarkers and identified novel candidates, providing a foundation for future mechanistic studies.