TOPLINE

Type 2 diabetes (T2D) was associated with accelerated cortical thinning in seven frontal, parietal, and occipital brain regions over nearly three years in cognitively unimpaired older adults, independent of Alzheimer's disease (AD) pathology and cerebrovascular burden. 

METHODOLOGY

• Researchers conducted a longitudinal observational study of 1298 cognitively unimpaired adults (mean age 65 years, 305 with T2D, 869 female) from the Health and Aging Brain Study-Health Disparities cohort in the US who completed baseline and follow-up MRI scans.
• Researchers assessed the relationship between diabetes and cortical thickness changes across 34 brain regions.   
• The primary outcome was longitudinal change in cortical thickness across 34 bilateral cortical regions. 
• Researchers adjusted analyses for demographic factors and diabetes-related comorbidities and, for amyloid, tau, white matter hyperintensities, and APOE ε4 status. Mean follow-up duration was 2.95 years.

TAKEAWAY

• Adults with T2D showed faster cortical thinning in the superior frontal, postcentral, superior parietal, precuneus, cuneus, lingual, and pericalcarine regions than those without diabetes (false discovery rate [FDR]-corrected P < .05).
• These associations remained significant after accounting for socioeconomic factors, comorbidities, amyloid, tau, white matter hyperintensities, and APOE ε4 status.
• Higher baseline HbA1c was associated with faster cortical thinning across these regions, except the lingual gyrus (all FDR-corrected P < .032).
• T2D was associated with faster decline in processing speed, and cortical thinning in the identified regions accounted for approximately 13% of this effect.

IN PRACTICE

"Our study demonstrates that this regional pattern of cortical thinning partially mediated the accelerated decline in processing speed observed in diabetic individuals, highlighting these regions as promising neuroimaging targets pending replication in independent cohorts," the authors of the study wrote.

SOURCE

The study was led by Amaryllis A. Tsiknia, Mark and Mary Stevens Neuroimaging and Informatics Institute, University of Southern California, Los Angeles. It was published online on August 24, 2026, in Brain.

LIMITATIONS

The study was limited by the availability of up to four MRI scans per participant, the use of linear statistical models, and concurrent measurement of HbA1c and cortical-thickness changes. The analysis of processing speed and executive function was limited to the Digit 20
Symbol Substitution test and Trail-Making Test, which may have limited the characterization of T2D-related cognitive deficits.

DISCLOSURES

No specific funding information was provided in the study. The authors reported having no conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.