T1-weighted/T2-weighted ratio reflects microstructural changes in Alzheimer's disease

The T1-weighted/T2-weighted ratio (T1w/T2w-ratio) has been considered as a relatively simple neuroimaging approach to map myelin content in the aging and diseased brain. However, we hypothesize that any process affecting the tissue microstructural integrity also contributes to observed changes in the T1w/T2w-ratio. Here, we aim to evaluate the ability of T1w/T2w-ratio to detect microstructural changes, specifically within the paradigm of Alzheimer's disease (AD).

Postmortem in-situ T1-weighted and T2-weighted MRI scans of 27 control and 51 AD brain donors were processed into standardized T1w/T2w-ratio images and parcellated into cortical regions according to the brainnetome atlas. Using an identical pipeline, antemortem MRI scans of 9 AD cases were processed. Immunohistochemical staining and digital quantification were performed for amyloid-beta (4G8), phosphorylated tau (AT8), neuro-axonal damage (NfL), myelin (PLP), microglia (IBA1), and iron (Meguro) in 9 cortical regions. Data was analysed using linear mixed models.

Whole cortex T1w/T2w-ratio was 15% lower in AD compared to controls and was 13% lower in the hippocampus of typical AD compared to atypical AD clinical phenotypes. Additionally, a strong correlation was found between antemortem and postmortem T1w/T2w-ratios (r = 0.85). Whole-brain associations between T1w/T2w-ratio and histological changes were found in both control and AD patients: T1w/T2w-ratio associated with myelin density (Controls: β = 0.229; AD: β = 0.312) and microglia density (Controls: β=-0.090; AD: β=-0.496). In AD but not in controls, T1w/T2w-ratio associated with amyloid-beta (β = 0.233), phosphorylated tau (β=-0.220), and neuro-axonal damage (β = 0.170). Iron load showed contrasting associations in controls and AD (Controls: β=-0.359; AD: β = 0.154).

The findings in this study suggest that the T1w/T2w-ratio is mainly associated with myelin density in healthy controls, but is a broader indicator of microstructural integrity in Alzheimer's disease, where it additionally largely associates with accumulation of amyloid-beta, phosphorylated tau and neuro-axonal damage. Therefore, the T1w/T2w-ratio is not a myelin-specific marker in AD, but could be a valuable neuroimaging tool for tracking longitudinal microstructural changes when interpreted alongside established AD pathology biomarkers.