A blood test cleared in the United States last year as a diagnostic aid for Alzheimer's disease is not yet approved for clinical use in Australia, even though Australian researchers have been central to validating the science behind it.
What pTau217 actually measures
Phosphorylated tau 217 is a fragment of the tau protein that begins leaking from dying brain cells decades before symptoms appear. The Lumipulse G pTau217/β-Amyloid 1-42 Plasma Ratio, cleared by the US Food and Drug Administration on 16 May 2025, measures two such fragments in a standard venous blood draw and combines them into a single ratio [1]. The test runs on Fujirebio's fully automated Lumipulse G1200 platform, already installed in many hospital and reference laboratories worldwide, and returns a result in roughly 35 minutes [1].
The key detail is the ratio. pTau217 on its own flags neuronal injury; β-amyloid 1-42 in plasma reflects the sticky plaques that define Alzheimer's pathology. When the two are combined, the test's positive percent agreement with amyloid PET or lumbar puncture reached 91.7% and its negative percent agreement 97.3% in the FDA's pivotal study of 499 patients [11]. The US agency cleared the test only for adults aged 55 and over who are already presenting with signs and symptoms of cognitive decline, and explicitly as an adjunct to other clinical information, not a stand-alone diagnostic [1].
The biology matters because it explains why a blood draw can replace a brain scan. Phosphorylated tau 217 becomes detectable in blood as tau pathology begins in the medial temporal lobe, often 10 to 20 years before clinical symptoms emerge [11]. By the time a patient walks into a memory clinic worried about their keys, the disease has usually been running for years.
The Australian evidence base
Australian researchers have been central to the case for pTau217. The Florey Institute of Neuroscience and Mental Health showed in October 2024 that the Lumipulse assay runs on standard clinical chemistry analysers using samples from the Australian Imaging, Biomarkers and Lifestyle (AIBL) cohort, the country's flagship longitudinal Alzheimer's study [4]. The finding, published in the Lancet group's eBioMedicine, matters because it means the test could be run in routine pathology labs rather than only specialist research facilities, with obvious implications for regional and remote Australia, where PET scanners are scarce and lumbar puncture is rarely performed [4].
The next step came from CSIRO. In October 2025, a CSIRO-led study reported that combining plasma pTau217 with the Aβ42/Aβ40 ratio reached approximately 93% accuracy for identifying brain amyloid, with 88% sensitivity and 95% specificity in the broader study cohort and 99% sensitivity in an intention-to-test analysis [3]. The biomarker panel matched the accuracy of PET scans and CSF testing, currently the two gold-standard confirmatory methods, both of which are expensive, invasive, and largely unavailable outside major metropolitan hospitals [3].
A separate 1,767-patient multicentre validation published in Nature Medicine in April 2025, drawn from five independent cohorts across Sweden, Spain and Italy, reported AUCs of 0.93 to 0.96 for plasma p-tau217 against amyloid PET, with 89 to 91% accuracy in secondary care and 85% in primary care, rising to 92 to 94% using a two-cutoff approach [2]. Performance was largely unaffected by chronic kidney disease, sex, APOE genotype or cognitive stage, but was lower in patients aged 80 and over [2].
There is a caveat worth flagging. Independent real-world evaluations published after FDA clearance suggest the cleared cut-points may need recalibration in routine clinical populations, where comorbidity burden, age and pre-test probability differ from the pivotal trial [10]. In older patients and those with chronic kidney disease, accuracy has been reported as low as 83% in some cohorts [10]. Professor Christopher Rowe, director of the Australian Dementia Network, has noted that local accuracy above 90% against amyloid and tau PET has been demonstrated in several Australian papers, but that the test's real-world performance will depend on who is being tested [7].
Why timing matters: the anti-amyloid era
The blood test has arrived at a moment when Australia has, for the first time in a quarter-century, drugs that aim to slow the underlying disease. In March and September 2025, the Therapeutic Goods Administration registered two anti-amyloid therapies, donanemab (marketed as Kisunla) and lecanemab (marketed as Leqembi), for adults with mild cognitive impairment or mild Alzheimer's dementia and confirmed amyloid pathology [5][6][9]. Both are the first new disease-modifying treatments approved in Australia in more than 25 years [5].
Lecanemab was approved in 51 countries before reaching Australia, on its third submission to the TGA [6]. Both drugs require confirmation of amyloid pathology by PET scan or CSF testing before a patient can start treatment [9], and PBS listing has not been recommended for either, leaving Australian patients to pay out of pocket or through private insurance; published reports put per-patient treatment costs in the tens of thousands of dollars [5][6]. Both also carry a risk of amyloid-related imaging abnormalities, brain swelling or microbleeds, with a fatal outcome rate of approximately 0.6% reported in long-term UK trial data [6]. The lecanemab trials reported a mean slowing of clinical decline of approximately six months on the CDR-SB at 18 months [6], which is precisely the population a cheap, scalable screening test is designed to find.
This is where pTau217 earns its place. A negative blood result, with 97.3% negative agreement in the FDA pivotal study, could safely rule out amyloid pathology and spare patients a PET scan or lumbar puncture [11]. A positive result would still need confirmatory testing before anti-amyloid therapy, but it would triage the waiting list and shift the bottleneck from diagnosis to treatment access.
When Australians might actually get it
Here the picture is less encouraging. pTau217 is currently labelled for research use only in Australia. Clinical use requires Therapeutic Goods Administration approval of the in-vitro diagnostic, ethics-committee sign-off, or a hospital exemption, and there is no Medicare rebate [7]. Professor Rowe has described the lack of MBS funding as the most significant impediment at present [7].
Estimated patient cost for a domestically performed pTau217 test is A$200 to A$350; sending samples to US laboratories for ALZpath testing would cost considerably more [7]. The Florey and ADNeT consortium has secured Department of Health and Aged Care funding to trial pTau217 blood testing in Australian primary care, with the goal of reducing diagnostic delay [7]. ADNeT clinicians suggest patient access in Australia is feasible within 12 to 24 months if TGA in-vitro diagnostic approval and a Medicare rebate are secured, though both remain pending [7].
Use in asymptomatic people is not yet endorsed. The FDA cleared the test only for adults already showing signs of cognitive decline [1], and Alzheimer's Association commentary emphasises its role as an adjunct, not a stand-alone diagnostic. Some peer-reviewed studies have reported that pTau181 and pTau217 predict asymptomatic amyloid accumulation as accurately as amyloid PET, raising the prospect of future pre-symptomatic use, particularly for stratifying risk before anti-amyloid therapy, but this remains a research question rather than a clinical recommendation.
The stakes are demographic. An estimated 433,300 Australians were living with dementia in 2024, with Alzheimer's disease accounting for around 70% of cases [8]. The Australian Institute of Health and Welfare projects that number will more than double to roughly 812,500 by 2054 [8]. Dementia is already the second leading cause of death in Australia and the leading cause of disease burden among people aged 65 and over, with an estimated 1.6 million Australians involved in dementia care in 2024 [8]. A blood test that costs less than a PET scan and runs on equipment already in the country's pathology labs is the kind of infrastructure change the next two decades will require.
The science is now reasonably settled. Will Australia's regulators and funders be able to move as fast as the disease itself?