Ask a neurologist how far along a patient's Alzheimer's disease has progressed and, until recently, the honest answer involved a PET scanner, a radioactive tracer, and a bill most people cannot pay. Blood tests have started to change that. But a blood test that says "yes, this is Alzheimer's" is not the same as one that says "and you are at this particular point in it." A new analysis of the Swedish BioFINDER-2 cohort tries to close that gap by watching several forms of the tau protein rise and plateau at different moments along the disease.
The work, published in Nature Communications, comes from a group led by Lyduine Collij and Oskar Hansson at Lund University, with collaborators at Washington University in St. Louis. They used mass spectrometry, which measures proteins by their exact molecular weight, to quantify tau species in both blood plasma and cerebrospinal fluid. The question was not just whether these markers are abnormal. It was when each one becomes abnormal.
Not all tau moves at the same time
Tau can be chemically modified in many ways, and the team focused on two kinds of change. One is phosphorylation, where small chemical groups get attached at specific spots on the protein. The other involves a stretch of tau called the microtubule-binding region, or MTBR. Rather than measuring raw amounts, the researchers looked at phosphorylation occupancy, essentially what fraction of tau molecules carry a given tag at a given site.
To place each person on a timeline, the group estimated disease duration from amyloid and tau PET scans using an algorithm called SILA, then fit smooth curves through the biomarker data. A clear order emerged. Phosphorylation at site 217, written %p-tau217, moved first. It started climbing just before amyloid PET scans turned positive, making it an early sentinel. Other phosphorylated forms shifted later, covered a narrower range, and hit their ceiling sooner, meaning they stopped being informative once the disease advanced.
The later stages had their own markers. Changes in %p-tau205 and in MTBR-tau243 lined up with the onset of tau PET positivity. And MTBR-tau243, especially a plasma version the authors label eMTBR-tau243, tracked the actual burden of tau tangles in the cortex during the later phase of illness. The authors suggest that non-phosphorylated mid-region tau may work as a genuine late-stage readout, something the field has lacked.
Why staging is the hard part
Detecting Alzheimer's is now the easier problem. Staging it is where the clinical value hides. A drug trial wants to enroll people at the right point, since a therapy aimed at early amyloid changes will look useless if it is tested on patients whose brains are already full of tangles. A clinic wants to counsel families about what comes next. Both need a way to say where someone is, not just what they have.
The appeal of this study is that a panel of markers, each peaking at a different time, could act like hands on a clock. Read %p-tau217, %p-tau205, and eMTBR-tau243 together and you get a rough position along the disease cascade rather than a single yes-or-no. The researchers argue that these three, measured in plasma or spinal fluid, report on distinct biological events happening in sequence.
The gaps that remain
Some restraint is warranted here. This is one cohort, BioFINDER-2, drawn largely from a memory-clinic and research population in Sweden, so the same clock may tick differently in other groups. The disease-duration estimates lean on PET imaging and a modeling algorithm, not on watching individuals change over decades, so the timeline is reconstructed rather than directly observed. The most intriguing measurements still rely on mass spectrometry, which is not something a corner lab runs on demand. Turning eMTBR-tau243 into a routine assay is its own engineering project.
What the paper delivers is a map of which tau signals matter when. According to PubMed, the underlying data come from participants with paired plasma and CSF measurements and PET-based staging. If the pattern holds up in more diverse cohorts, the payoff is concrete. A tube of blood might one day tell a doctor not only that Alzheimer's is present, but how far it has already traveled.
Comments