Background: The incoming disease-modifying therapies against Alzheimer’s disease (AD)
require reliable diagnostic markers to correctly enroll patients all over the world. CSF AD biomarkers,
namely amyloid-β 42 (Aβ42), total tau (t-tau), and tau phosphorylated at threonine 181 (p-tau181),
showed good diagnostic accuracy in detecting AD pathology, but their real usefulness in daily clinical
practice is still a matter of debate. Therefore, further validation in complex clinical settings, that is patients
with different types of dementia, is needed to uphold their future worldwide adoption.
Methods: We measured CSF AD biomarkers’ concentrations in a sample of 526 patients with a clinical
diagnosis of dementia (277 with AD and 249 with Other Type of Dementia, OTD). Brain FDG-PET was
also considered in a subsample of 54 patients with a mismatch between the clinical diagnosis and the
Results: A p-tau181/Aβ42 ratio higher than 0.13 showed the best diagnostic performance in differentiating
AD from OTD (86% accuracy index, 74% sensitivity, 81% specificity). In cases with a mismatch
between clinical diagnosis and CSF findings, brain FDG-PET partially agreed with the p-tau181/Aβ42
ratio, thus determining an increase in CSF accuracy.
Conclusion: The p-tau181/Aβ42 ratio alone might reliably detect AD pathology in heterogeneous samples
of patients suffering from different types of dementia. It might constitute a simple, cost-effective
and reproducible in vivo proxy of AD suitable to be adopted worldwide not only in daily clinical practice
but also in future experimental trials, to avoid the enrolment of misdiagnosed AD patients.