Cerebrospinal fluid tau and amyloid-β1-42 in patients with dementia

• Published in: Brain (London, England : 1878) Vol. 138; no. Pt 9; pp. 2716 - 2731
• Main Authors: Skillbäck, Tobias, Farahmand, Bahman Y, Rosén, Christoffer, Mattsson, Niklas, Nägga, Katarina, Kilander, Lena, Religa, Dorota, Wimo, Anders, Winblad, Bengt, Schott, Jonathan M, Blennow, Kaj, Eriksdotter, Maria, Zetterberg, Henrik
• Format: Journal Article
• Language: English
• Published: England 01.09.2015
• Subjects: Aged; Amyloid beta-Peptides - cerebrospinal fluid; Analysis of Variance; Clinical Medicine; Dementia - cerebrospinal fluid; Enzyme-Linked Immunosorbent Assay; Female; Humans; Klinisk medicin; Male; Medical and Health Sciences; Medicin och hälsovetenskap; Mental Status Schedule; Middle Aged; Neurologi; Neurology; Neurosciences; Neurovetenskaper; Peptide Fragments - cerebrospinal fluid; Retrospective Studies; Severity of Illness Index; Sweden; tau Proteins - cerebrospinal fluid; Sweden; dementia with Lewy bodies; frontotemporal dementia; Alzheimer’s disease; biomarkers; vascular dementia
• DOI: 10.1093/brain/awv181
• ISSN: 1460-2156

Progressive cognitive decline in combination with a cerebrospinal fluid biomarker pattern of low levels of amyloid-β1-42 and high levels of total tau and phosphorylated tau is typical of Alzheimer's disease. However, several neurodegenerative disorders may overlap with Alzheimer's disease both in regards to clinical symptoms and neuropathology. In a uniquely large cohort of dementia patients, we examined the associations of cerebrospinal fluid biomarkers for Alzheimer's disease molecular pathology with clinical dementia diagnoses and disease severity. We cross-referenced the Swedish Dementia Registry with the clinical laboratory database at the Sahlgrenska University Hospital. The final data set consisted of 5676 unique subjects with a clinical dementia diagnosis and a complete set of measurements for cerebrospinal fluid amyloid-β1-42, total tau and phosphorylated tau. In cluster analysis, disregarding clinical diagnosis, the optimal natural separation of this data set was into two clusters, with the majority of patients with early onset Alzheimer's disease (75%) and late onset Alzheimer's disease (73%) assigned to one cluster and the patients with vascular dementia (91%), frontotemporal dementia (94%), Parkinson's disease dementia (94%) and dementia with Lewy bodies (87%) to the other cluster. Frontotemporal dementia had the highest cerebrospinal fluid levels of amyloid-β1-42 and the lowest levels of total tau and phosphorylated tau. The highest levels of total tau and phosphorylated tau and the lowest levels of amyloid-β1-42 and amyloid-β1-42:phosphorylated tau ratios were found in Alzheimer's disease. Low amyloid-β1-42, high total tau and high phosphorylated tau correlated with low Mini-Mental State Examination scores in Alzheimer's disease. In Parkinson's disease dementia and vascular dementia low cerebrospinal fluid amyloid-β1-42 was associated with low Mini-Mental State Examination score. In the vascular dementia, frontotemporal dementia, dementia with Lewy bodies and Parkinson's disease dementia groups 53%, 34%, 67% and 53% of the subjects, respectively had abnormal amyloid-β1-42 levels, 41%, 41%, 28% and 28% had abnormal total tau levels, and 29%, 28%, 25% and 19% had abnormal phosphorylated tau levels. Cerebrospinal fluid biomarkers were strongly associated with specific clinical dementia diagnoses with Alzheimer's disease and frontotemporal dementia showing the greatest difference in biomarker levels. In addition, cerebrospinal fluid amyloid-β1-42, total tau, phosphorylated tau and the amyloid-β1-42:phosphorylated tau ratio all correlated with poor cognitive performance in Alzheimer's disease, as did cerebrospinal fluid amyloid-β1-42 in Parkinson's disease dementia and vascular dementia. The results support the use of cerebrospinal fluid biomarkers to differentiate between dementias in clinical practice, and to estimate disease severity.