Biomarkers measured in cerebrospinal fluid (CSF), particularly amyloid beta 1-42 (Aβ42) and tau (total tau and phosphorylated tau), have been extensively studied to aid the diagnosis and prognostication of Alzheimer disease (AD). Altered CSF levels—classically low Aβ42 and elevated total or phosphorylated tau—are associated with the pathophysiology of AD and the tau/Aβ42 or tTau/Aβ42 ratios have been proposed to improve diagnostic accuracy versus single analytes. Multiple systematic reviews and trial programs have shown that CSF biomarker diagnostic accuracy is comparable to amyloid PET for detecting amyloid pathology, and CSF testing has been used in randomized controlled trials to determine eligibility for anti-amyloid therapies (for example, as an alternative to amyloid PET in pivotal lecanemab trials).
Compared with PET imaging, CSF biomarkers can provide equivalent information about the presence of amyloid pathology when interpreted with appropriate cutoffs and clinical context; therefore CSF testing has been adopted in some trial eligibility algorithms and in clinical guidance as an alternative to amyloid PET for confirming amyloid pathology prior to initiating amyloid-targeting therapies. Regulatory-cleared CSF assays (for example, Elecsys and Lumipulse CSF tests) report specified ratio cutoffs (examples include an Elecsys pTau181/Aβ42 cutoff of 0.023 and a tTau/Aβ42 cutoff of 0.28) and manufacturers explicitly advise that results must be interpreted in conjunction with the clinical evaluation and are not standalone diagnostic tests.
Blood- and urine-based biomarkers are an active area of research because of easier access to these matrices. Plasma assays for phosphorylated tau (pTau) and plasma Aβ42/40 ratios have shown promising analytic and early clinical validity and several blood tests have received Breakthrough Device designation; more recently, some plasma assays have obtained FDA clearance. Nevertheless, at the time of policy review the clinical utility of many blood and urine biomarkers as routine, standalone diagnostic tools remained limited: results require further validation, and pivotal studies demonstrating improved health outcomes from use in routine care are lacking.
Important limitations apply to many biomarker assays. There is substantial between-laboratory variability for CSF measurements and pre-analytical and analytical factors can affect results; standardization efforts (including Alzheimer’s Association and international consensus recommendations) aim to reduce this variability but differences persist. For blood and urine assays, heterogeneity of methods, lack of universal standardization, and limited external validation across platforms and populations restrict generalizability. These sources of variability and incomplete standardization are reasons why many tests remain investigational for certain uses and why CSF and PET confirmation are emphasized when required by therapy labels or trial protocols.