Aβ status assessment in a hypothetical scenario prior to treatment with disease‐modifying therapies: Evidence from 10‐year real‐world experience at university memory clinics

INTRODUCTION With the advent of disease‐modifying therapies, accurate assessment of biomarkers indicating the presence of disease‐associated amyloid beta (Aβ) pathology becomes crucial in patients with clinically suspected Alzheimer's disease (AD). We evaluated Aβ levels in cerebrospinal fluid (Aβ CSF) and Aβ levels in positron emission tomography (Aβ PET) biomarkers in a real‐world memory‐clinic setting to develop an efficient algorithm for clinical use. METHODS Patients were evaluated for AD‐related Aβ pathology from two independent cohorts (Ludwig Maximilian University [LMU], n = 402, and Medical University of Vienna [MUV], n = 144). Optimal thresholds of CSF biomarkers were deduced from receiver operating characteristic curves and validated against Aβ PET positivity. RESULTS In both cohorts, a CSF Aβ42/40 ratio ≥ 7.1% was associated with a low risk of a positive Aβ PET scan (negative predictive value: 94.3%). Implementing two cutoffs revealed 14% to 16% of patients with intermediate results (CSF Aβ42/40 ratio: 5.5%–7.1%), which had a strong benefit from Aβ PET imaging (44%–52% Aβ PET positivity). DISCUSSION A two‐cutoff approach for CSF Aβ42/40 including Aβ PET imaging at intermediate results provides an effective assessment of Aβ pathology in real‐world settings. Highlights We evaluated cerebrospinal fluid (CSF) and positron emission tomography (PET) amyloid beta (Aβ) biomarkers for Alzheimer's disease in real‐world cohorts. A CSF Aβ 42/40 ratio between 5.5% and 7.1% defines patients at borderline levels. Patients at borderline levels strongly benefit from additional Aβ PET imaging. Two‐cutoff CSF Aβ 42/40 and PET will allow effective treatment stratification.