Association of Plasma Biomarkers With Longitudinal Atrophy and Microvascular Burden on MRI Across Neurodegenerative and Cerebrovascular Diseases.

Abstract

Background and objectives: Plasma biomarkers of Alzheimer disease (AD), neuroinflammation, and neurodegeneration are increasingly being used in clinical trials for diagnosis and monitoring of dementia. However, their association with longitudinal structural brain MRI changes, an important outcome measure across neurodegenerative and cerebrovascular diseases, is less known. We investigated how baseline plasma biomarkers reflect MRI markers of progression over time in patients with neurodegenerative and cerebrovascular diseases.

Methods: This longitudinal cohort study included patients from the Ontario Neurodegenerative Disease Research Initiative diagnosed with AD or mild cognitive impairment (AD/MCI), Parkinson disease (PD), frontotemporal dementia spectrum disorders (FTD), or cerebrovascular disease (CVD), followed annually for 2 years. Recruitment took place at specialized university-based dementia, movement disorders, and/or stroke clinics in the province of ON, Canada. MRI outcomes included markers of cerebral atrophy (ventricular CSF and regional gray matter volumes) and of small vessel disease pathology (white matter hyperintensity [WMH], perivascular spaces, and lacunar volumes). Hemorrhagic markers at baseline were also included. Plasma levels of glial fibrillary acidic protein (GFAP), neurofilament light chain (NfL), phosphorylated tau181 and tau217 (p-tau181, p-tau217), and β-amyloid (Aβ_42/40) were quantified from blood samples collected at baseline using Simoa and used as predictors in linear mixed models adjusted for time (months), age, sex, apolipoprotein E (APOE)-ε4 carrier status, kidney function, vascular risk factors, microtubule-associated protein tau (MAPT) diplotypes, waist-hip circumference ratio, and disease duration.

Results: We analyzed 1,240 MRIs from 473 patients (age: 69.2 ± 7.4 [range: 49-87]; 32.8% women). Elevated baseline levels of GFAP, NfL, p-tau181, and p-tau217, and to a lesser extent decreased levels of Aβ_42/40, were significantly associated with more cerebral atrophy and WMH burden at baseline (|B| = 0.02 to 1.69, p = 0.044 to <0.001) and with progression over time (|B| = 0.001 to 0.028, p = 0.049 to <0.001) in the pooled disease-agnostic group. Within disease-specific cohorts, GFAP and NfL were associated with cerebral atrophy and/or small vessel disease copathology in AD/MCI, PD, FTD, or CVD. P-tau181 and p-tau217 were associated with cerebral atrophy and/or small vessel disease copathology in AD/MCI, CVD, PD-MCI, or PD-dementia.

Discussion: Selected plasma biomarkers seem useful as prognosis and monitoring tools of longitudinal imaging changes within real-world populations of neurodegenerative and/or cerebrovascular diseases, and provide insight into overlap across diseases in shared pathologic burden.