Distinct epigenetic ageing patterns are associated with heterogeneity in kidney function decline in type 2 diabetes.

Biological ageing is a heterogeneous process that shapes susceptibility to chronic disease. However, whether ageing patterns diverge within clinically defined type 2 diabetes (T2D) subgroups remains unclear. We investigated whether epigenetic age acceleration (EAA) differs across T2D phenotypes and whether these patterns relate to subsequent renal vulnerability. We included 607 multi-ethnic Asians with recent-onset T2D previously classified into three clinically distinct subgroups: mild obesity-related diabetes (MOD), mild age-related diabetes with insulin insufficiency (MARD-II), and severe insulin-resistant diabetes with relative insulin insufficiency (SIRD-RII). EAA was quantified using multiple epigenetic clocks. Kidney function was assessed longitudinally using eGFR slope over a median follow-up of 7.3 years, serving as a maker of cumulative systemic ageing burden. SIRD-RII, despite being chronologically younger, showed accelerated ageing across multiple second-generation clocks, whereas MARD-II, despite being chronologically older, showed comparatively lower levels of EAA. Among all epigenic clocks examined, GrimAge2logA1c was both highest in SIRD-RII and the only clock associated with kidney function decline (eGFR slope, β [SE] = -0.113 [0.040], p = 4.77 × 10-3). Epigenome-wide analyses identified subgroup-specific, glucose-responsive loci associated with accelerated ageing. Hypomethylation at TXNIP (cg19693031) characterised the SIRD-RII subgroup and was associated with both faster ageing pace and kidney function decline. Exploratory longitudinal within-person changes in glucose-linked epigenetic ageing were correlated with changes in TXNIP methylation within SIRD-RII (r = -0.901, p = 1.93 × 10-6). These findings suggest that T2D subgroups represent distinct epigenetic ageing phenotypes that may be associated with renal vulnerability. Integrating epigenetic ageing metrics with clinical stratification may enhance identification of individuals at risk for accelerated systemic ageing, informing geroscience-guided intervention strategies early in disease development.