Plasma Phospho-Tau217 Is a Promising Alzheimer’s Disease Biomarker

By Michael T. Lin, MD

Associate Professor of Neurology and Neuroscience, Weill Cornell Medical College

Dr. Lin reports no financial relationships relevant to this field of study.

SYNOPSIS: Validation and replication of findings in the use of the blood biomarker phospho-tau 217 for the diagnosis of Alzheimer’s disease holds out the possibility for a new era in the early diagnosis and treatment of this debilitating illness.

SOURCE: Palmqvist S, Janelidze S, Quiroz YT, et al. Discriminative accuracy of plasma phospho-tau217 for Alzheimer disease vs other neurodegenerative disorders. JAMA 2020; July 28. doi:10.1001/jama.2020.12134. [Online ahead of print].

The diagnosis of Alzheimer’s disease (AD) still is primarily clinical. Although approved biomarker tests are commercially available, they either are prohibitively expensive (amyloid PET imaging) or invasive (CSF Aβ42 and phospho-tau levels), and commonly are not obtained outside of research settings. A blood test potentially would be much more useful. Brain and cerebrospinal fluid (CSF) proteins do cross the blood-brain barrier and arachnoid granulations to enter the blood, but their usefulness depends on the degree of dilution and whether there are central nervous system (CNS) isoforms distinct from peripheral isoforms. Several recent publications and conference presentations show that plasma phospho-tau217 (Ptau217, tau phosphorylated at threonine 217) has impressive ability to discriminate AD from normal and non-AD neurodegenerative control subjects, as measured by area under receiver operating characteristic curve (AUC) of approximately 0.9.^1-3 Intuitively, the AUC is the probability that a random target subject will have a more abnormal test value than a random control subject.

Palmqvist and colleagues used an immunoassay to measure levels of plasma Ptau217 in 1,402 subjects from three separate cohorts, comparing performance with other biomarkers.¹ Cohort 1 consisted of 81 subjects (34 AD, 47 non-AD) who had subsequent autopsy verification of diagnosis. In this cohort, the AUC was 0.89 for distinguishing intermediate or high probability AD from non-AD, and 0.98 for distinguishing high probability AD from non-AD. These AUC values were significantly higher than those for other plasma biomarkers (Ptau181, neurofilament light chain [NfL]). Plasma Ptau217 levels also correlated significantly with brain tangle density in AD cases, but not in non-AD cases.

Cohort 2 consisted of 699 subjects from the Swedish BioFINDER-2 study. These subjects were classified by recognized clinical criteria into cognitively unimpaired, mild cognitive impairment (MCI), AD dementia, and other neurodegenerative diseases. All subjects had CSF biomarker measurements. In this cohort, the AUC was 0.96 for distinguishing AD dementia from other neurodegenerative diseases. This outperformed plasma Ptau181, plasma NfL, magnetic resonance imaging (MRI) volumetric measures, and even CSF Ptau181; only CSF Ptau217 (AUC 0.99) and brain tau positron emission tomography (PET) (AUC 0.98) performed better. Plasma Ptau217 also correlated well with amyloid pathology, with AUC of 0.87 for distinguishing positive from negative amyloid PET. By comparison, the CSF Aβ42:Aβ40 ratio had an AUC of 0.97, and CSF Ptau217 had an AUC of 0.93 for distinguishing amyloid PET status. Of note, plasma Ptau217 was not elevated in neurodegenerative disease with non-AD tau pathology, such as progressive supranuclear palsy or corticobasal syndrome.

Cohort 3 consisted of 622 subjects from the Colombian presenilin-1 E280A kindred. In this cohort, mean plasma Ptau217 levels were lowest in noncarriers, and then progressively higher in cognitively unimpaired carriers, and cognitively impaired carriers. Levels increased with age in carriers, and by age 24.9 years (about 20 years before median age of MCI onset), there was a significant difference from noncarriers. Plasma Ptau217 also correlated significantly with Mini-Mental State Exam (MMSE) scores and memory performance. Of note, this study was cross-sectional, not longitudinal.

Barthélemy and colleagues used a different, mass spectrometry approach to measure levels of plasma Ptau217, also comparing performance with other biomarkers.² They previously had reported that CSF Ptau217 was closely associated with amyloid plaques, and Ptau217 was better than tau phosphorylation at other sites (T181, S202, T205) at predicting amyloid plaques. In the current study, plasma Ptau217 correlated with CSF Ptau217 (Spearman rho = 0.78) better than plasma Ptau181 with CSF Ptau181 (Spearman rho = 0.68). Moreover, when normalized to total tau, plasma Ptau217/tau217 had a greater magnitude of change (+230% to 340%) from controls to amyloid-positive groups than plasma Ptau181/tau181 (+60% to 80%). Also, CSF and plasma Ptau217 were able to distinguish amyloid-positive, tau-PET-negative individuals from controls, suggesting that fluid Ptau biomarkers are changed before detectable tau aggregation, and that they reflect abnormalities in soluble tau occurring concomitantly with amyloid pathology. In a discovery cohort (n = 36), plasma Ptau217 had an AUC of 0.991 for distinguishing Aβ+ from Aβ- cases. In a validation cohort (n = 92), the AUC was 0.925, possibly lower because of a different plasma extraction method or lowering of plasma Ptau levels by concomitant CSF drainage. For both cohorts, Ptau217 outperformed Ptau181.

Thijssen and colleagues previously had shown that plasma Ptau181, measured by an immunoassay, was 3.5-fold higher in AD than in controls, and in a series of 404 cases could distinguish AD from clinically diagnosed frontotemporal dementia (FTD) (AUC 0.894) and autopsy-confirmed FTD (AUC 0.878).³ At the recent Alzheimer’s Association International Conference, they reported the results of plasma Ptau217 in a series of 210 cases (healthy controls, clinical AD spectrum, MCI, and FTD).⁴ Plasma Ptau217 was increased 5.7-fold in clinical AD relative to controls (AUC 0.92) and 5.0-fold relative to clinical FTD (AUC 0.95). Plasma Ptau217 was increased 3.3-fold in autopsy-confirmed AD relative to FTD (AUC 0.86). Again, Ptau217 was superior to Ptau181. Thus, multiple independent groups, using different biochemical methods, examining different populations, have shown diagnostic utility for Ptau217 as a plasma biomarker discriminating brain AD pathology from other pathology. Availability of such a test clearly would be a huge advance, making accurate diagnosis of AD much easier for both clinical and research purposes.

COMMENTARY

A number of issues remain to be worked out. The methodology should be standardized and replicable. With both immunoassays and mass spectrometry, a substantial proportion of samples were below the limit of detection. The discovery and validation cohorts in Barthélemy’s work may have given slightly different results because of a change in extraction method. All of the studies noted here used highly selected research populations, and they should be extended to real world clinical populations, more ethnically diverse subjects, and subjects with multiple pathologies. The study designs were cross-sectional, and inferences about the progression of Ptau217 levels over time should be validated by longitudinal studies. Plasma Ptau217 can detect brain AD pathology decades before symptoms develop. This raises several conundrums. Should testing be done in asymptomatic subjects, especially given that there currently is no effective disease-modifying intervention? Even in subjects who are symptomatic, should testing be done if there are explanatory comorbidities (depression, medications, sleep disorders, etc.), as is common? If multiple pathologies coexist, can the testing be used to judge the relative contribution of AD pathology? Can the testing be used to predict the rate of progression? It is worth reconsidering appropriate use criteria for AD biomarker testing, expanding to include use of blood biomarkers, which surely are coming soon.⁵

REFERENCES

1. Palmqvist S, Janelidze S, Quiroz YT, et al. Discriminative accuracy of plasma phospho-tau217 for Alzheimer disease vs other neurodegenerative disorders. JAMA 2020; July 28. doi:10.1001/jama.2020.12134. [Online ahead of print].
2. Barthélemy NR, Horie K, Sato C, Bateman RJ. Blood plasma phosphorylated-tau isoforms track CNS change in Alzheimer's disease. J Exp Med 2020;217:e20200861.
3. Thijssen EH, La Joie R, Wolf A, et al. Diagnostic value of plasma phosphorylated tau181 in Alzheimer's disease and frontotemporal lobar degeneration. Nat Med 2020;26:387-397.
4. Thijssen EH, presenting author. Comparative diagnostic performance of plasma P-tau217 and P-tau181 in Alzheimer’s disease and frontotemporal lobar degeneration and correlations with [18F]Flortaucipir-PET uptake. Alzheimer’s Association International Conference 2020. https://www.alz.org/aaic/downloads2020/AAIC2020-BloodBiomarkers.pdf
5. Shaw LM, Arias J, Blennow K, et al. Appropriate use criteria for lumbar puncture and cerebrospinal fluid testing in the diagnosis of Alzheimer’s disease. Alzheimers Dement 2018;14:1505-1521.