PrecivityAD2 is headed for a Japanese regulatory bid after observational studies in symptomatic adults found strong agreement with amyloid PET. A blood route to identifying brain amyloid could streamline confirmatory testing and treatment discussions.
C2N Diagnostics and LSI Medience plan to pursue Japanese regulatory approval for PrecivityAD2, a blood-based assay designed to classify whether a symptomatic patient is likely to have brain amyloid pathology. A report describing the planned Japanese application says the companies intend to seek authorization; the available evidence does not establish that regulators in Japan have approved the test.
The proposal matters because blood collection is operationally simpler than amyloid positron emission tomography (PET) or cerebrospinal-fluid testing. If its performance transfers reliably into Japanese clinical practice, PrecivityAD2 could help clinicians determine which patients need further biomarker assessment or are candidates for treatment discussions. Its evidence base, however, is specifically about classifying amyloid-PET status in symptomatic adults—not using a blood result alone to determine the cause of cognitive symptoms.
The clinical evidence behind the application
The principal evidence comes from a human observational study rather than a treatment trial. A peer-reviewed clinical-validation study of 583 people evaluated for suspected Alzheimer’s disease compared the test’s classification with amyloid PET, the study’s reference standard. Fifty-three per cent of participants were PET-positive.
PrecivityAD2 achieved 88% overall agreement with PET and an area under the receiver operating characteristic curve, or AUC, of 0.94. The AUC indicates strong separation between PET-positive and PET-negative participants across different score thresholds; it should not be read as meaning the assay was “94% accurate.” The 88% agreement figure reflects the proportion classified consistently with PET at the study’s selected decision threshold.
A separate validation using stored human plasma samples reported 91% accuracy at the prespecified cut point, with 90% sensitivity and 92% specificity. In simplified terms, under those study conditions, sensitivity of 90% would correspond to identifying about 90 of every 100 PET-positive people, while specificity of 92% would correspond to correctly classifying about 92 of every 100 PET-negative people.
The second cohort strengthens the evidence that the algorithm can reproduce its classification performance beyond the primary dataset. Both studies nevertheless remain diagnostic-accuracy research: they assessed concordance with amyloid status, not whether testing improved symptoms or long-term clinical outcomes.
Two biological signals, calculated from four analytes
PrecivityAD2 is sometimes described as a two-marker test because its algorithm combines the plasma amyloid-beta 42/40 ratio with percentage phosphorylated tau217, or %p-tau217. The underlying laboratory process measures four analytes: Aβ42, Aβ40, phosphorylated tau217 and non-phosphorylated tau217.
According to the assay’s peer-reviewed analytical description, the targets are isolated through immunoprecipitation and measured using liquid chromatography–tandem mass spectrometry. The resulting values are used to calculate the two biomarker ratios, which are then integrated into an Amyloid Probability Score 2, or APS2.
The combined score brings together related but distinct biological information. The Aβ42/40 ratio is associated with cerebral amyloid deposition, while %p-tau217 captures an Alzheimer’s-related phosphorylation signal. APS2 is therefore an algorithmic estimate of likely PET classification, not a direct image of plaques in the brain.
That distinction is clinically important. Amyloid is a core feature of Alzheimer’s disease biology, but a positive amyloid result does not by itself establish that Alzheimer’s pathology is the sole cause of an individual’s cognitive symptoms. Clinical history, cognitive assessment and other investigations remain part of the diagnostic process.
How the result could alter an Alzheimer’s work-up
A validated blood test could function as an earlier step in a staged assessment. A result indicating low probability of amyloid might allow a clinician to redirect parts of the evaluation, while a high-probability result could support referral for confirmatory testing or specialist review where required. This could be particularly useful as biomarker confirmation becomes more relevant to treatment eligibility and monitoring decisions.
The reported sensitivity and specificity are more informative than a single headline accuracy figure, but they do not directly tell a patient the probability that their result is correct. That probability—expressed as positive or negative predictive value—also depends on how common amyloid positivity is in the population being tested.
The primary cohort, for example, consisted of symptomatic people already being evaluated for suspected Alzheimer’s disease, and just over half were PET-positive. A clinic applying the assay to a population with a substantially lower prevalence would generally see a different proportion of positive results confirmed by PET, even if sensitivity and specificity remained similar. The published results therefore fit a symptomatic diagnostic pathway more directly than broad screening of people without cognitive concerns.
Errors also remain possible at the reported performance levels. False-negative results could delay further amyloid assessment, while false-positive results could lead to additional testing and anxiety. The test’s value in practice will depend not only on analytical performance but also on where it is placed in the pathway and how clinicians interpret its probability score alongside the rest of the assessment.
Japan will test how well the evidence travels
The peer-reviewed studies provide clinically relevant human validation, but they do not report a Japan-specific clinical-validation cohort. This does not negate the existing findings; it means the published record does not yet show how closely the reported performance will transfer to the patients, referral patterns and laboratory workflows encountered in Japan.
Transportability matters for any diagnostic algorithm. The prevalence and stage of disease among referred patients affect predictive values, while sample handling, laboratory quality controls and the mix of comorbid conditions can influence real-world performance. Japanese regulatory review will determine whether the analytical and clinical evidence is adequate for the proposed use and whether additional local data are needed.
For now, the development is an approval plan rather than an authorization. Its significance lies in the possibility of moving a technically sophisticated, PET-linked blood assay into another national clinical system. The studies offer a substantial basis for that review: two human cohorts found strong classification performance against amyloid PET. The next meaningful measure will be whether that performance can be maintained in the patients and clinical pathways for which the test would be approved.