Alzheimer Disease Blood-Based Biomarkers: Translation from Research into Clinical Use

Recent FDA approvals of new disease-modifying therapies for Alzheimer disease (AD) has increased the clinical demand for tests of amyloid pathology that can be scaled for wider use. Concurrent advances in blood-based biomarker (BBB) assays have made it possible to perform minimally invasive and highly accessible laboratory testing for core biomarkers of this disease. While there are currently no FDA-approved blood tests for AD, many of the major in vitro diagnostic (IVD) companies have products in development (including several with breakthrough device status), and thus new IVD products are anticipated in the not-too-distant future. Existing diagnostic approaches, such as cerebrospinal fluid (CSF) analysis and positron emission tomography (PET) imaging, are effective but are invasive and often more expensive (especially the latter) compared to blood tests. Additionally, these tests may not be widely accessible based on socioeconomic and geographic factors, as well as regulatory and reimbursement status. PET in particular requires administration of a radiolabeled tracer prior to imaging, necessitating proximity to a tracer production site or a well-validated supply chain, which often limits this technology to tertiary care centers in predominantly major urban centers. The future availability of plasma assays is poised to fundamentally change the diagnostic process for AD, as it may lower some of the barriers to testing faced by CSF and PET testing. Recent research advances have identified several promising blood biomarkers that offer similar diagnostic performance to CSF and PET imaging biomarkers. These blood biomarkers are associated with key pathological features of AD, such as amyloid-β (Aβ) plaques and tau tangles, can provide insights into disease progression and pathology burden. Revised research criteria for the diagnosis and staging of AD, recently published by an Alzheimer’s Association workgroup, introduced a more prominent role for BBBs. In particular, plasma phosphorylated tau was integrated into the biomarker staging system, which previously relied on PET and CSF biomarkers. Currently, BBBs are only recommended for use in a diagnostic capacity for individuals with symptoms suggestive of AD; however, other applications are under study. There is interest in the integration of BBBs at multiple points along the care pathway for individuals with AD, including for early and differential diagnosis, determining eligibility for specific treatments, monitoring response to therapy, and tracking disease progression. Biomarkers that could help determine the effectiveness of future classes of therapies that are currently in clinical trials and enable optimal prescribing of combinations for even greater efficacy are also highly desired. The hope is that integration of BBBs into the diagnostic paradigm provides practical and objective measures to personalize therapy and slows the progression of this fatal disease. Recent research presented at the Alzheimer’s Association International Conference (AAIC) 2024 further highlighted the potential of BBBs to revolutionize AD diagnosis and treatment. These tests, some of which demonstrated an accuracy of 90% in identifying AD in patients with cognitive symptoms seen in specialized memory care clinics, could replace current expensive and invasive methods, improving early detection, and reducing wait times for treatment. Additionally, the tests could enhance clinical trial recruitment by more easily identifying cognitively unimpaired individuals who might benefit from early intervention. While it is expected that such tests may change how AD is diagnosed and managed in the future, the transition from research to clinical use will not occur without challenges. As we gain greater clarity in which biomarker(s) will form the central diagnostic set, for instance, future efforts should focus on harmonization across IVD products. This will ensure comparability of results across labs and ease the synthesis of research findings. Furthermore, new tests must be validated across diverse populations and in various clinical scenarios. Interpretation of new types of data must be reliably integrated into a process that already includes established cognitive exams and neuroimaging protocols. Only a well-implemented combined approach will properly support clinical decision-making. By capturing the perspectives of a wide-ranging group of experts, this article aims to explain the latest advances in BBBs for AD while new capabilities are rapidly transitioning from research to clinical use. The contributors examine the current state of scientific understanding, key challenges that remain for practical implementation, and potential future directions in dementia care. Nick Ashton: Blood biomarkers should replicate what we observe in CSF. Therefore, the markers considered in blood should be the same: phosphorylated tau (p-tau), Aβ peptide 42/40 ratio (Aβ42/40), and total tau (t-tau) as AD-specific markers and neurofilament light (NfL) for neurodegeneration. It has long been known that the signal of t-tau does not translate well to blood due to peripheral expression. Efforts have been made to develop non-phosphorylated tau. Although the detection of Aβ42/40 in plasma has improved significantly, it is unlikely to be a primary diagnostic blood biomarker because of several confounding factors, such as poor robustness, pharmacodynamic effects, and pre-analytical instability. However, it is relevant for understanding novel disease-modifying therapies that shift Aβ production to a less pathogenic state, such as gamma-secretase modulators. This leaves p-tau as the principal biomarker in blood, particularly p-tau217, which has shown utility in identifying cerebral pathology at all stages of the disease continuum and outperforms other p-tau epitopes like p-tau181, p-tau231, and p-tau212. Therefore, plasma p-tau217 is anticipated to be an essential component of patient management and treatment decisions related to AD. NfL in blood shows limited disease specificity to AD but is a good indicator of axonal damage. NfL is useful in some differential diagnosis cases, such as atypical Parkinsonian disorders compared to Parkinson disease and dementia in Down syndrome. Mari DeMarco: In CSF, Aβ peptide 1-42 is the anchor biomarker for AD pathology (expressed as a ratio with either Aβ 1-40 or p-tau). In blood, tau proteoforms are poised to take over this key role. Major analytical challenges for the measurement of both Aβ and tau proteoforms relevant to AD pathology in plasma are their relatively lower concentration and the increased matrix complexity of plasma, as compared to CSF. In blood, there is also peripheral production of these biomarkers, i.e., non-disease–related production, which reduces mean differences in concentration between AD and non-AD pathology cohorts, compared to differences observed in CSF. For Aβ peptides, there is the additional challenge of degradation by endogenous proteases if the sample is not carefully collected, processed, and stored prior to analysis. Thus, based on pre-analytical, analytical, and diagnostic performance considerations, current research favors the translation of tau proteoforms into clinical care. While there are also composite-based biomarker algorithms incorporating demographics (e.g., age) and other biomarkers (e.g., apolipoprotein E [APOE] genotype/phenotype), p-tau is also the key component of these algorithms. Andy Saykin: The advent of valid BBBs is having a transformative impact on Alzheimer research and beginning to penetrate clinical care as well. Currently, PET and CSF are still the established assays with known validity. Amyloid (and tau) PET are not widely accessible and are typically costly, though recent federal payer decisions by the US Centers for Medicare and Medicaid (CMS) may lessen the impact for eligible older individuals. CSF assays are widely used in clinical settings, but many older adults are disinclined toward this procedure given the invasive nature and potential temporary discomfort. Reliable and valid BBBs could address access, cost, and invasiveness concerns, though few experienced clinicians would rely solely on BBBs for clinical decision-making at present. For research, plasma p-tau217 appears particularly promising and, strikingly, is a robust predictor of cerebral amyloid-β deposition. Plasma-based assays for Aβ40 and Aβ42 are available on multiple platforms, which vary in measurement properties, with the Aβ42/40 ratio often performing well. Mass spectrometry-based assays are probably the best approach among fluid biomarker assays, though immunoassays and related technologies are rapidly improving and can be accurate and cost-effective. Beyond plasma Aβ and tau, NfL and glial fibrillary acidic protein (GFAP) are generating considerable interest and helping to address broader questions of neurodegeneration (NfL) and components of inflammation (GFAP). The combination of Aβ, p-tau, NfL, and GFAP appears to be the present state of the art for research into AD and related dementias biomarkers, although there is also considerable interest in BBBs that assess other pathways in greater detail. Mari DeMarco: To assist in the translation of BBBs into clinical care in a manner that improves the well-being and care of persons with cognitive impairment we need to take a step back and look at the bigger picture, that is looking beyond only laboratory considerations and diagnostic metrics. As part of the implementation of AD CSF biomarkers in Canada, via the IMPACT-AD observation study (www.impactAD.org), we sought input about the impact of biomarker testing from a variety of sources including individuals living with neurodegenerative disorders. I would like to highlight insights gained from patients undergoing testing and their family members, as the lessons learned from this group can positively inform implementation of BBBs. For instance, in deciding whether to undergo biomarker testing, patients and their family members indicated that they placed substantial trust in their physician’s opinion of the value of testing and greatly relied on their expertise. As part of pre-test counseling, they valued having a confident understanding of the advantages and disadvantages of testing so that they approached the process with a realistic picture of what the outcomes could be. It is also important to note that individuals undergoing testing were ones that were seeking answers as to the reason for their declining brain health and, related to this, valued the outcomes of testing most for the clarity biomarker testing brought to their diagnostic journey. In the implementation of blood biomarkers for AD, we need to consider both sides of this patient–physician dyad and ensure resources and supports are in place for all. For physicians, this could mean clear guidelines around appropriate and inappropriate clinical scenarios for testing, and ample performance data, clearly presented, in cohorts relevant to the populations they serve. This in turn would better enable physicians to accurately advise their patients. For patients and their family members, we need to consider supports systems across the diagnostic journey. For instance, in a paradigm where an indeterminate or positive blood test outcome needs to be followed by a confirmatory amyloid PET or CSF test, it is important to ensure that realistic expectations have been set. This includes relaying possible test outcomes and their uncertainty, and ensuring care pathways downstream of biomarker testing are in place and their access defined (e.g., referrals and wait-times). Nick Ashton: Blood biomarker assays need to move from research platforms to fully automated instrumentation that can be utilized in random-access fashion for clinical use. This will ensure that blood biomarkers can then be measured at ad hoc routine clinics. Further, additional work is needed to fully understand the clinical impact of comorbidities on p-tau217. For example, there are multiple reports of increased p-tau217 with chronic kidney disease (CKD), which might impact on clinical interpretation. Lastly, it is urgent to understand if blood biomarkers like p-tau217 are similarly effective in underrepresented ethnic groups and whether reported cutoffs for AD are transferrable. Andy Saykin: The ideal BBB would be highly sensitive and specific for amyloid and/or p-tau accumulation, reliable, well validated against PET and neuropathology, and easily accessible through point of clinical care at a reasonable cost. I would like to highlight 2 challenges, among others. First, there is very little data on diverse individuals with nearly all major studies largely having been performed in people of European ancestry and typically with higher socioeconomic status. Emerging evidence suggests ethnoracial differences in performance across a range of AD and related disorders and other biomarkers. Whether this reflects differences in genetic architecture of neurodegenerative diseases as a function of ancestry, differential comorbidities and co-pathologies, or exposome and other social determinants of health-related issues is currently unknown. It is essential for a precision medicine of AD and related dementias to address this gap. The second issue I would highlight is that training for clinicians regarding appropriate use and limitations will be critical. The context of use, base rates, and known potential confounding factors all need to be The FDA has identified classes of biomarkers that are useful and validated for specific such as from diagnostic or early detection markers that current or biomarkers have other appropriate or potential and in this have with health in the blood tests without the appropriate use and limitations of these assays, and also not their patients about these and other health all may the of specific tests. I recently a very research in the a blood test that back positive for amyloid but their PET performed by Alzheimer’s and by a medicine was clearly also In this there were cognitive disease-modifying treatment in to biomarker either cognitive impairment or AD clinical trials in biomarker positive individuals are and if in or development of cognitive due to AD, may the appropriate use criteria for all classes of biomarkers. present this a research The challenge is the of the The utility of BBB tests will be largely on and by the clinical of new which is still Beyond this, from a payer what is the evidence to clinical benefit the results of BBB tests management for AD. BBBs must be validated through testing and compared to diagnostic for key challenge is in the US it to most only to about clinical As as BBBs are all they care about is what a reasonable does with the are more in a value positive and and Therefore, tests that can help to (e.g., for differential diagnosis and therapy will be the most to of BBBs for AD, we can from the to in look at the FDA-approved diagnostic paradigm in by and among Medicare and they are without an because the is and the diagnostic limits only to patients where the therapy could For AD, the current also have concerns, so BBBs could be as a of similarly treatment to only that could Andy Saykin: research is a range of biomarkers of all CSF, blood, While this is generating data for for precision medicine in the the needs a and of well-validated biomarkers. BBBs are to be can be used early in primary and are cost-effective. BBBs will need to be integrated with other cognitive and and are to have an role for and the of of considerations are that biomarkers are to be the optimal or for early detection is not the optimal for disease staging and of progression or Nick Ashton: it is currently recommended to use BBBs in combination with clinical There are at present no on how to use these in individuals there multiple of evidence to they will be useful in that on the of or would need with CSF or PET It must be that some which may be the of can be in CSF but not blood or Mari DeMarco: substantial benefit and major the implementation of BBBs is the potential for increased ease of to biomarker testing a or an of a radiolabeled tracer and imaging is relatively less invasive and requires less health In like where health resources are not across geographic ease of access also the potential for increased in health the published on diagnostic performance of plasma biomarkers and health in Canada, BBBs are best for use in specialized not primary care. a blood test could be useful as part of the diagnostic for people with cognitive with AD to and I help and not all other diagnostic This blood test could be used to determine if AD should remain as part of the differential and whether additional testing, such as AD CSF biomarkers across or amyloid PET only at a of in is that disease and pre-test can diagnostic a is the use of testing in appropriate clinical populations to ensure expected test to diagnostic testing for AD, blood tests are to be more available to patients. can be in a physician’s or even via if is The is lower but can expensive if beyond clinical The impact of is There is no no and of I that the future diagnostic will largely be defined by the clinical of new but if this does I there will be a for blood-based testing. In of I could BBBs used in the of therapy a payer point of people is probably as important as people given the and challenges of the current for AD. I could also a where a BBB could be used in of a PET can be for older people who are a little to is whether a BBB can that has been defined as an imaging could this a blood test, it would be better a patient in a with a would like to the limitations of CSF and PET and an that is based on other a of For example, it would be if a blood test could patients who are to benefit from treatment more accurately and, a where are most on that if were to a trial and only patients based on a would results as to a shift of clinical value in a biomarker or in a diagnostic that was for the of the is the of BBBs in the development and of disease-modifying therapies for Mari DeMarco: the in new amyloid there is a on early and accurate diagnosis of AD. In the have been by the and and in Canada, of these have with the clinical trial data for these and other the has learned key the of diagnosis to diagnostic and the need for early for the to and disease progression. the of CSF biomarkers into we have already observed the positive of increased diagnostic including increased diagnostic of physicians and downstream positive on patient care (e.g., of and the health (e.g., in need for other diagnostic the potential for BBBs to be in the diagnostic with their lower to access, we are that these positive can be to disease-modifying this the (e.g., to the people persons with biomarker evidence of AD at the early in the pathological as is Nick Ashton: on the a blood p-tau217, can accurately in or amyloid It is not clear how accurate a blood test for p-tau217 would be for trial that with tau (e.g., in the clinical trial for Lastly, it is anticipated that blood tests would be useful in monitoring disease in such trials or could be used as an outcome While promising on the group the data suggests this may not be possible with the current markers we Andy Saykin: BBBs can be very for cognitively patients for tau and other and helping to individuals who can benefit from PET and/or CSF present BBBs are not validated for use in from other established biomarkers of and However, as assays improving and additional data is collected, I it is they will for diagnostic in the future. For patients where the and of is imaging will remain a major diagnostic Mari DeMarco: new disease-modifying therapies on the on for new scenarios biomarkers for of treatment and monitoring for For the given the to on therapy and of it would be to have biomarkers that could provide objective and evidence for of treatment. For current clinical trial data for some a for while identified no to current therapies monitoring treatment. this with a blood test would be on patients and enable monitoring of for there are currently no markers to for the development and of The of new biomarkers in medicine also interest around and It is that of the core AD biomarkers in p-tau, only Aβ42 is there a and CSF tau efforts have long been in the but no are available and only data a t-tau have been For plasma AD biomarkers, analytical and biomarker (e.g., of analytical and diagnostic performance of various should As the data around specific markers and technologies for clinical implementation (e.g., plasma p-tau217 harmonization will a future Nick Ashton: have that blood is a useful for biomarkers. Thus, there is that it will be possible to biomarkers for other such as primary and as well as biomarkers like for (e.g., Parkinson disease and and for (e.g., and However, like p-tau, NfL, and Aβ, this will from identifying the biomarker in CSF with the of blood and the availability of (e.g., and the for blood to be the primary of biomarkers has been Lastly, regarding is still It requires that be and the blood be in a manner (e.g., and evidence suggests that blood that is may be a for a will be and validated the in which a combination of blood, and biomarkers will be used to help to factors at the disease to ensure Andy Saykin: most BBBs are to and this has In the future, combinations of genetic tests and may in blood that the of various brain and classes of brain on is toward central with in I important for BBB to look more and beyond the current of the available has been looking into this For as as AD, the are so there could be many other diagnostic The advances in testing is the of the tests, of to decision-making and patient health AD, Alzheimer blood-based CSF, cerebrospinal positron emission Aβ, amyloid-β p-tau, phosphorylated total NfL, neurofilament light imaging The that all on this have the criteria of eligibility for to the and of data, or analysis and of or the article for of the published and to be for all of the article thus ensuring that questions related to the accuracy or of part of the article are and who for has been from the Mari and all the from the support from multiple and is an of has on for and and from has from from and and has from and on for and has from is and from would like to the support and input from the and

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.