This interview was conducted by Gustav Ceder for Biobanking.com. It has been edited for length and clarity.
Henrik Zetterberg is Professor of Neurochemistry at University College London (UCL) and at the University of Gothenburg, and a Senior Consultant in Clinical Chemistry at Sahlgrenska University Hospital. His research focuses on fluid biomarkers for neurodegenerative diseases, especially Alzheimer’s disease.
For years, total tau in blood proved too noisy to be clinically useful: much of it comes from the body’s periphery rather than the brain, muddying Alzheimer’s disease (AD) readouts. New assays that selectively measure brain derived tau—and, specifically, brain-derived phosphotau217 (ptau217), aim to fix that. For biobanks, the implications are immediate: large cohorts with stored samples and imaging/cognition data are uniquely positioned to validate cut points, compare against PET (an imaging technique that uses radiotracers to visualize biological processes in the body), and assess performance in real-world populations.
In this Q&A, Professor Henrik Zetterberg explains what brain derived tau means, why peripheral tau confounds older tests, how nextgen assays are built, and what biobanks can do now, including microsampling at scale.
Q: In plain terms, what does it mean that some tau in blood comes specifically from the brain—and how do newer technologies separate brain from peripheral tau?
Henrik Zetterberg: “This is knowledge known for some time, now back in the limelight with brain derived tau (BDtau). Tau is made by many cell types in the body. Neurons are the main producers because they use tau to stabilize axons, which is fundamental to neuronal biology.” “There is also tau in the enteric nervous system and in peripheral nerves, and some tau in blood cells such as platelets as well as in melanocytes. When we originally developed tau biomarkers, we used cerebrospinal fluid (CSF).”
“CSF is on the brain side of the blood–brain barrier, and we obtained good results with those biomarkers. When we moved to blood, the first assays performed poorly: total tau in blood only vaguely associated with CSF tau levels.”
“Closer study suggested that approximately 80% of tau in the bloodstream does not come from the brain; it originates from the periphery.”
“We then developed phosphotau assays, which are somewhat more CNS related even in blood. However, phosphorylation of peripheral nerve tau—irrelevant to Alzheimer’s disease—can confound measurements. So we set out to target differences between central (brain) and peripheral tau.”
“Brain derived tau lacks a specific amino acid sequence, creating a neoepitope junction. We generated antibodies to that junction, yielding an assay that selectively measures brain derived tau and not peripheral tau.”
“This assay associates strongly with Alzheimer’s related brain pathophysiology.”
Q: Is this available on multiplex platforms?
Henrik Zetterberg: “There are a couple of assays, but the only one that currently exists in a multiplex setting with other markers is the NULISA (Alamar Biosciences) version present in their CNS 120 panel.”
Q: What clinical blind spots can brain‑derived p‑tau help address compared with standard blood p‑tau markers?
Henrik Zetterberg: “We have sometimes observed unexpectedly high phospho-tau concentrations in people we do not think have Alzheimer’s disease, which is problematic if phospho-tau markers are used broadly, for example in primary care. Peripheral neuropathies such as Charcot MarieTooth disease and ALS (amyotrophic lateral sclerosis) can show increased blood phospho-tau.”
“That increase likely reflects phosphorylation of peripheral ‘big tau.’ While that may be useful for developing biomarkers for those diseases, it obscures the relationship with Alzheimer’s disease and makes the data noisier.”
“We do not yet know the extent in conditions such as diabetic peripheral neuropathy, which is common among people with cognitive problems. This is where brain derived ptau217 should excel, giving stronger confidence that a biomarker abnormality reflects Alzheimer’s related brain changes.”
Q: From a biobanking perspective, what is the value of measuring brain derived ptau217 in these clinical cohorts?
Henrik Zetterberg: “My hope is that brain derived ptau217 could reliably reflect the number of neurons affected by amyloid and Alzheimer’s pathophysiology. With strong validation data from cohorts and biobank samples, we could implement it clinically, using an absolute concentration to help indicate how advanced Alzheimer’s disease is in terms of brain tissue engagement.”
“This new NULISA assay will be very valuable when examining cohorts with biobanked samples where participants have also undergone amyloid PET and tau PET imaging and good longitudinal cognitive assessment.”
“That will allow us to validate—this is still somewhat a hypothesis—and derive cut points relative to amyloid PET and tau PET, which are now the gold standards to stage Alzheimer’s disease.”
“Combined with cognitive assessment and a complete medical workup, such an accurate test could be sufficient for clinicians to prescribe advanced treatments, making it easier for healthcare systems to use these drugs.”
“Then perhaps amyloid PET will not be necessary to prescribe these treatments. A surrogate for amyloid PET could be CSF sampling, but both are resource intensive and costly.” “I see this biomarker as a potential alternative—existing phosphotau markers already work somewhat in this way, but eliminating false positives would be a major advance.”
“Then I think we would have enough confidence to make decisions on the basis of blood tests only.”
Q: As biomarker discovery moves toward multi-omic integration and large-scale data sharing, what role do you see for biobanks in enabling interoperable, ethically governed data ecosystems that accelerate Alzheimer’s validation globally?
The biobanks have been essential for this development, and they will become even more important. Key will be that samples and data can be shared because complex panels need a lot of replication and validation work before getting deployed in clinical practice.
Q: What should biobank managers do now?
Henrik Zetterberg: “From my point of view: if you run a cohort and biobank samples focused on Alzheimer’s onset and progression, you should obtain ptau217 measurements—specifically brain derived ptau217—fairly quickly, because it looks like the most reliable biomarker of Alzheimer’s related pathophysiology.”
“And if it is a community or population based cohort, you would also want to use additional markers to examine other disease processes. Panel based testing would be valuable.”
“The important thing is to spread the message that these assays exist now—you do not have to wait to use them. The field has developed rapidly, and we have learned a great deal from this new class of refined biomarkers where several confounders have been mitigated.”
About the Author
Biobanking.com delivers valuable news and information on biobanking, biorepositories, cryopreservation, and biosample/biospecimen management, while connects buyers with top vendors globally.
