Blood test for Alzheimer’s Disease coming to fruition

Biomarkers are essential for the diagnosis of neurodegenerative diseases such as Alzheimer’s Disease (AD) in vivo, as the brain can only be accessed after death for (histo)pathological confirmations of the presence of a disease. A correct, etiological diagnosis of dementia is particularly essential now disease-modifying therapies to treat the underlying AD pathophysiology become available. A blood test for AD could be a non-invasive, simple to perform, relatively inexpensive and scalable tool to assist diagnosis, as well as to assist prognosis and monitoring of disease progression across the (pre-)clinical phases of AD. Furthermore, it could expedite clinical trials that investigate efficacy of potential disease-modifying drugs, by helping in patient selection, and in monitoring of therapeutic efficacy. Technological advancements enabled us to measure AD-related proteins in the blood with high enough sensitivity to obtain meaningful results in clinical samples. In this thesis, we developed and validated various novel blood-based biomarkers for different contexts of use in AD employing high-sensitivity assays, namely amyloid beta (Abeta), phosphorylated tau (P-tau), total tau (T-tau), glial fibrillary acidic protein (GFAP), neurofilament light (NfL) and brain-derived neurotrophic factor (BDNF) for diagnosis, prognosis and/or monitoring. We addressed analytical, pre-analytical and clinical aspects, with the ultimate goal to reach implementation of AD blood-based biomarkers into clinical and research practice. The key findings of this thesis are: 1. Plasma Abeta42 and Abeta40 can be measured robustly by several assays, one of which we developed and thoroughly characterized ourselves in collaboration with diagnostic companies. 2. Plasma P-tau isoforms, such as P-tau181, P-tau217 and P-tau231, can be measured robustly by several assays with comparably good performance. 3. Plasma Abeta42, Abeta40 and T-tau levels are sensitive to variations in sample handling from collection until biobank storage, mainly to delayed processing when kept at room temperature, while plasma biomarkers P-tau181, GFAP and NfL levels appear robust to pre-analytical variations. 4. Plasma Abeta42/40, GFAP and P-tau have diagnostic value in identifying ongoing amyloid pathology across the AD continuum, including the pre-dementia stages of AD. Plasma T-tau and plasma NfL levels are not associated with ongoing amyloid pathology. 5. There are cross-sectional relations of plasma GFAP and plasma NfL with measures of disease severity, such as syndrome diagnosis, neuropsychological test performance and atrophy ratings on MRI, while this is less apparent for plasma Abeta42/40. 6. GFAP, P-tau181 and NfL have value in differential diagnosis, with higher GFAP and P-tau181 levels observed in individuals with AD-dementia as compared to individuals with frontotemporal dementia (FTD) or dementia with Lewy bodies (DLB), and with higher NfL levels observed in individuals with FTD as compared to individuals with AD-dementia or DLB. 7. Plasma Abeta42/40, serum GFAP and serum NfL, but not plasma T-tau, have prognostic value, with associations with rates of cognitive decline and risk of incident dementia in cognitively unimpaired individuals with subjective cognitive decline. 8. Plasma BDNF level is associated with risk of dementia in cognitively unimpaired individuals with ongoing amyloid pathology, similar to what we observed for the BDNF Val66met genotype, which was a risk factor for dementia. 9. Serum NfL levels increase steeper over time in cognitively unimpaired individuals with subjective cognitive decline who are progressing to dementia as compared to those that remain non-demented during follow-up. 10. Appreciating the complexity of AD and the overlap of its pathology with other forms of dementia, a panel of blood-based biomarkers could be beneficial over single markers. Depending on the context of use, e.g., identification of ongoing amyloid pathology, differential diagnosis of AD versus other types of dementia, prognosis of dementia risk among cognitively unimpaired, or longitudinal monitoring of disease progression, different combinations of blood-based biomarkers might be needed.