Multimodal neuroimaging related to cerebrospinal fluid biomarkers and cognitive function in Alzheimer's disease

Abstract

Background Alzheimer's disease (AD) is characterized by amyloid-β (Aβ) and tau protein accumulation, reflected in cerebrospinal fluid (CSF) analysis. However, the interplay among CSF biomarkers, neuroimaging, and cognition remains elusive. Objective To explore associations among neuroimaging features, CSF biomarkers, and cognitive performance in AD. Methods Sixty patients with clinically diagnosed AD showing Aβ pathology in CSF underwent neuroimaging assessment of gray matter volume using T1-weighted MRI, cerebral blood flow (CBF) using single-photon emission computed tomography, and white matter hyperintensities (WMHs) using T2-weighted or fluid-attenuated inversion recovery images. Partial least square (PLS) regression identified imaging findings related to CSF biomarkers and Mini-Mental State Examination (MMSE) scores. Structural equation modeling (SEM) explored associations between factors with variable importance in projection (VIP) scores above 1.5 in PLS regression. Results Lateral temporal and occipital gray matter volumes positively correlated with MMSE scores (VIP = 1.95, 1.53), whereas WMHs in parietal and frontal periventricular regions were negatively associated with CSF Aβ 42 (VIP = 1.54, 1.58). Lateral temporal CBF was also associated with MMSE scores (VIP = 2.22). SEM analysis showed that reduced CSF Aβ 42 was linked to increased WMHs (p = 0.028), which correlated with each region (p < 0.005) and explained the reduced MMSE score (p = 0.013). Lateral temporal CBF correlated with temporo-occipital gray matter volume (p < 0.001) and influenced the MMSE score (p < 0.001). Conclusions This study suggests that amyloid pathology via WMHs and neurodegeneration of the lateral temporal lobe independently contribute to cognitive impairment in patients with AD.