Do blood-based biomarkers diagnose AD or just inform on an individual’s risk?

medwireNews: Blood-based biomarkers are a useful tool for stratifying an individual’s risk for Alzheimer’s disease (AD) in the early stages of cognitive decline, but do they actually “diagnose” AD in the truest sense? This was the quandary posed by a debate at the 18th World Congress on Controversies in Neurology in London, UK, between Robert Perneczy (Ludwig-Maximilians-University Munich, Germany) and Arfan Ikram (Erasmus University Medical Center, Rotterdam, the Netherlands). Blood biomarkers are “an important tool” In his argument in support of using serum markers such as phosphorylated (p)-tau in diagnosing AD, Perneczy commented that there has been a shift toward a biological, rather than a clinical, diagnosis of AD, which is why “blood biomarkers are emerging as an important tool to diagnose AD in the earliest stages,” with the core plasma and cerebrospinal fluid (CSF) markers being amyloid-beta proteinopathy (Aβ42) and p-tau (p-tau 217, 181 and 231). He referred to research suggesting that biomarkers “influence clinical decisions,” namely a finding by Gil Rabinovici and team that a positive amyloid positron emission tomography (PET) result doubled the use of anti-AD drugs among thousands of Medicare beneficiaries with mild cognitive impairment (MCI), from about 40% before PET to 80% afterwards. A negative amyloid PET in the same group had less of an effect, decreasing the use of anti-AD medications from 27% to 24%. Research has also shown that plasma p-tau 181 correlates significantly with CSF p-tau 181 and tau and Aβ PET, increasing in line with the progression from cognitively healthy and preclinical AD to MCI and AD, Perneczy commented. It can also be used to differentiate between subgroups of dementia patients with a similar accuracy to that of CSF biomarkers, and positivity for p-tau is associated with an increased risk for clinical progression, he added. Such findings have been replicated in more recent studies using real world data, Perneczy pointed out, with plasma Aβ42/40 ratio and p-tau 217 found to correlate with Aβ PET in a study by Matthew Meyer et al that has just published. However, Perneczy warned that plasma biomarkers can be influenced by factors such as BMI and creatinine levels and he said that it is something “we need to be aware of as we start using these blood-based biomarkers in the clinic.” The use of blood-based biomarkers is important because in addition to the recent approvals of the new anti-amyloids we now have a lot of new compounds currently being assessed in clinical trials, he highlighted, quoting that there are currently 36 drugs in phase 3 and 87 in phase 2. He said that an appropriate use for blood-based biomarkers in primary care, as described by Oskar Hansson et al, is in the creation of algorithms, alongside digital cognitive tools, to stratify patients with cognitive symptoms in to those with a low probability of AD, who “don’t need to worry,” a medium probability, who should be considered for advanced testing at a memory clinic, and a high probability, who would make good candidates for disease-modifying treatment. He then sees blood biomarkers being used mainly in assessments in secondary care, alongside detailed cognitive tests and brain magnetic resonance imaging (MRI) to exclude patents with a low likelihood of AD pathology before brain PET or CSF testing is undertaken by specialists so that they can be referred for continued monitoring in primary care. Diagnosis or risk profiles? Arguing that serum markers are not useful for diagnosing AD, Arfan Ikram gave an epidemiologist’s viewpoint. He said that, by definition, for blood biomarkers to diagnose AD theyould need to “define the presence of disease,” which he said current biomarkers do not; “they reflect the underlying disease process” and risk for the disease. He explained that for early diagnosis, a biomarker would need to be an essential cause for the disease, in that it must always be present for the disease to occur, but people without tau, for instance, can still develop dementia. Similarly, a biomarker for diagnosis could be a sufficient cause, so by itself it is sufficient for the disease to be present, except people can have tau and not develop AD, Ikram pointed out. Ikram referenced the National Institute on Aging and Alzheimer’s Association collaboration on a framework toward a biological definition of AD, which states that if there is biomarker evidence of both beta amyloid (Aβ) and pathologic (p)-tau then the term AD can be applied. However, he highlighted that this biologically based definition of AD does not depend on either of these biomarkers being causal. The presenter stressed that using such a biomarker definition of AD diagnosis is “dangerous,” as it could lead to the unnecessary treatment of people who would never develop clinical AD and precludes further clinical research toward new risk factors for AD. Rather, Ikram believes that we should use these biomarkers to determine “risk profiles” for individuals. He explained: “Do not ‘etiologically’ subtype disease but diagnose disease or syndrome on the presence of putative risk factors or biomarkers, simply say, for instance, that this is dementia in presence of amyloid and vascular pathology but absence of tau.” medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2024 Springer Healthcare Ltd, part of the Springer Nature Group This article was originally published on Journey to a New Era for Alzheimer’s Disease, an independent education program developed by Springer Healthcare IME. This program is supported by an educational grant from Lilly. This content is intended for healthcare professionals outside of the UK. CONy 2024; London, UK: March 21–23

Keynote webinar | Spotlight on sleep in brain health

Springer Medicine

Do blood-based biomarkers diagnose AD or just inform on an individual’s risk?

Related topics

Keynote webinar | Spotlight on sleep in brain health

Springer Medicine

Please log in to get access to this content

Related topics