New conditional recommendations support the use of blood-based biomarkers to aid in the diagnostic workup of Alzheimer’s disease among patients with cognitive impairment evaluated in specialized memory care settings.
The guidance identifies minimum diagnostic performance thresholds for blood-based biomarkers (BBMs) to be used either as a triaging tool or a confirmatory test, based on sensitivity and specificity.
The guideline was based on a systematic review of 49 observational studies evaluating 31 BBM tests. The mean sample size was 560 participants, with study sizes ranging from 70 to 2,244. Reported participant age ranged from 62.6 to 85.9 years.
In 32 studies that reported apolipoprotein E ε4 genotype, the proportion of ε4 carriers ranged from 27.1% to 56.2%. Male representation in studies ranged from 33.8% to 60.0%.
The panel used the Grading of Recommendations, Assessment, Development, and Evaluation framework and an evidence-to-decision process to assess BBM test performance. Blood tests measuring plasma phosphorylated tau (p-tau181, p-tau217, p-tau231, %p-tau217) and the Aβ42/Aβ40 ratio were evaluated against reference standards: cerebrospinal fluid biomarkers, amyloid positron emission tomography imaging, or neuropathology.
To qualify as a triaging test, a BBM had to demonstrate 90.0% or more sensitivity and 75.0% or more specificity. For confirmatory testing, both sensitivity and specificity needed to meet or exceed 90.0%. Most performance assessments were based on a single cutoff, typically derived from Youden’s index.
Although some tests met or exceeded the minimum thresholds for triaging or confirmatory use, diagnostic accuracy varied widely across the 31 BBMs evaluated. Reported sensitivity ranged from 49.31% to 91.41%, and specificity from 61.54% to 96.72%. Most tests achieved these thresholds using a single cut-point, and more clinically relevant two-cutoff strategies were not evaluated. Many commercially available BBMs did not meet the recommended performance levels. The panel issued two conditional recommendations based on low-certainty evidence: (1) BBMs meeting triage-level performance may help rule out Alzheimer’s disease (AD) pathology; (2) BBMs meeting confirmatory-level thresholds may substitute for cerebrospinal fluid or amyloid positron emission tomography in confirming AD pathology.
“A BBM test should not be obtained before a comprehensive clinical evaluation by a healthcare professional, and test results should always be interpreted within the clinical context,” said Sebastian Palmqvist, MD, PhD, of Lund University and Skåne University Hospital in Malmö, Sweden, and colleagues.
“The panel urges clinicians to consider the pretest probability of AD pathology for each patient when deciding whether or not to use a BBM test,” and not as a standalone diagnostic tool.
The certainty of evidence was rated as low to very low due to risks of bias, including lack of prespecified cutoffs and unblinded test interpretation. Only studies focused exclusively on cognitively impaired populations were included; studies that combined impaired and unimpaired participants were excluded.
Most tests were evaluated using single-threshold performance, and multi-threshold or biomarker combination approaches were not analyzed due to limited data.
BBMs may assist in diagnosing Alzheimer’s disease in patients with mild cognitive impairment or dementia, provided they meet predefined accuracy thresholds.
The panel advised that these tests be ordered only after a comprehensive clinical evaluation by trained specialists and interpreted in context. As new evidence emerges, recommendations will be updated through a linked living guideline platform.
Full disclosures can be found in the published study.
Source: Alzheimer’s & Dementia
