Once amyloid status is fixed, adding Aβ42/40 to p-tau217 buys little for tau-PET positivity

Question: How do multi-analyte plasma biomarker panels combining p-tau217 with amyloid-beta 40/42 or other markers compare against single-marker p-tau217 for discriminating tau-PET positivity among amyloid-positive individuals?

Answer

The retrieved literature does not contain a head-to-head test of this exact comparison, but the assembled evidence points in a consistent direction: gains from multi-analyte panels are demonstrated almost entirely on the amyloid endpoint, where adding Aβ42/40, GFAP and NfL cuts the inconclusive intermediate zone roughly threefold relative to p-tau217 alone, whereas on the tau endpoint p-tau217 by itself already tracks Braak stage with AUCs of 0.93–0.98 and outperforms companion markers and their ratios. Once a sample is conditioned on amyloid positivity, plasma Aβ42/40 has largely exhausted its dynamic range and adds little information about which A+ individuals have fibrillar tau; the markers that do outperform p-tau217 for tau-PET positivity are tau-specific analytes — endogenously cleaved MTBR-tau, a functional p-tau binding assay, and (in CSF) p-tau205 — not amyloid-ratio additions. Taken together, these papers support a division of labour: panels earn their keep at the amyloid gate and in resolving borderline cases, while conditional tau staging inside A+ remains a single-marker problem whose residual error is more likely to be closed by better tau analytes than by adding Aβ40/42.

Panels are validated against amyloid status; their published gains are indeterminate-zone reductions, not tau discrimination

A multi-analyte algorithmic test combining p-Tau217 with Aβ42/40, GFAP and NfL achieved an AUC of 0.92 with 90% agreement with amyloid PET or CSF, and reduced the intermediate zone threefold to 12% versus p-Tau217 alone [17]. That improvement was quantified on amyloid status, with performance reported as similar across race, ethnicity, sex, age and APOE4 status [17]. The problem the panel solves is real and precedes any tau question: dual-cutoff single-marker p-tau217 strategies leave 19.4% (CSF-defined) to 34.2% (PET-defined) of cases indeterminate [6], and a separate two-cutoff Simoa assay placed 30.9% of samples in the intermediate zone [3]. Broader head-to-head work in memory-clinic samples likewise positions p-tau217 as the best single amyloid classifier, ahead of t-tau and Aβ42/40 [21] and ahead of the other plasma biomarkers assessed [8].

On the tau endpoint, p-tau217 alone already carries the staging signal and companion markers do not improve it

In a cohort with paired amyloid and tau PET, p-tau217 outperformed p-tau181, NfL, BD-tau, Aβ42, Aβ40 and their related ratios, separated Aβ+ from Aβ− with AUC 0.96, and tracked tau-PET Braak stages with AUC 0.93–0.98; using amyloid-derived thresholds it detected medial temporal and neocortical tauopathy with >90% sensitivity and specificity [1]. Amyloid-ratio information appears specifically uninformative about tau topography once A+ is established: among 396 amyloid-positive, tau-positive participants, adjustment for plasma Aβ42/40, amyloid Centiloids or PET tau burden did not reveal subtype modification comparable to that seen with p-tau217, indicating Aβ42/40 did not encode the tau-topography signal [2]. A 220-plex plasma panel anchored against amyloid and tau PET found that brain-derived p-tau217 plus endogenously cleaved MTBR-tau formed the parsimonious backbone of a five-stage framework, with the remaining 197 non-tau proteins contributing stage-associated but not definitional information [7]. Note the tension with amyloid-endpoint work, where the pTau217/Aβ42 ratio achieved the highest accuracy for Aβ+ status (AUC 0.927 vs 0.909 for pTau217 alone) and mitigated renal-function confounding [20] — a benefit that does not obviously transfer to a population already selected as A+.

The analytes that actually outperform p-tau217 for tau-PET positivity are tau-specific, not amyloid-ratio additions

A functional assay detecting pTau binding to recombinant tau reached 94.2% sensitivity and 96.1% specificity for tau-PET positivity (AUC 0.97) and outperformed plasma pTau217 in cognitively unimpaired individuals (PPV 85.9% vs a maximal 57.5% for pTau217 at the 0.256 pg/mL threshold), with the advantage concentrated in early tau-PET stages (Braak 1–3 AUC 0.96 vs 0.74, p=0.003) [4]. Plasma eMTBR-tau was significantly more accurate than plasma p-tau217 in classifying Braak staging and ADNC status at autopsy, retained AUC 0.91–0.95 across blood-to-death intervals up to a decade, and reached up to 94% concordance with tau neuroimaging [5]. In CSF, a staging model incorporating Aβ42/40, p-tau217 and p-tau205 attributed the strongest cortical atrophy, cognitive impairment and incident-dementia risk (HR 6.40) to the final p-tau205-positive stage [9].

The binding constraint on single-marker p-tau217 is early-stage and intermediate-zone performance, not missing amyloid information

Where p-tau217 performs poorly for tau, the failure is at the low end and in the middle of the distribution rather than for want of an amyloid covariate: two p-tau217 assays discriminated tau-PET-positive from tau-PET-negative participants with AUC of only 0.71 each [14], and in cognitively unimpaired ALFA+ participants fluid biomarkers showed low positive predictive values (0.09–0.33) for Braak I/II tau-PET positivity [13]. Plasma p-tau217 and tau PET are described as reflecting partly different aspects of pathology, with p-tau217 becoming abnormal earlier [11], which is precisely why a plasma tau marker cannot be expected to be a tau-PET surrogate at the earliest Braak stages. Even the definition of positivity dominates outcomes: PET-anchored versus distribution-based thresholds classified the tails consistently but disagreed in the intermediate latent class, and the boundary definition influenced p-tau217 positivity more than the analytical platform did [18].

Co-pathology modifies the amyloid-to-tau transition in ways no current plasma panel captures

Among 636 ADNI participants, α-synuclein positivity (19.0%) amplified the amyloid–tau association across all Braak stages (meta-temporal interaction β=0.258, p=0.001), strongest in Braak III–IV, without modifying tau–cognition relationships [10]. Amyloid-positive individuals classified as clinically "vulnerable" relative to their tau burden showed higher frequencies of TDP-43 neurodegeneration patterns and α-synuclein positivity than resilient or canonical groups, and plasma p-tau217 served as a tau proxy in these models [12]. The probability of tau severity also depends jointly on amyloid burden, age and clinical impairment, with a three-way interaction on likelihood of tau severity [19] — a structure that a fixed panel score combining p-tau217 with Aβ42/40 does not encode.

Conflicts and gaps

The most important gap is that no retrieved study reports ΔAUC for tau-PET positivity comparing a multi-analyte panel against p-tau217 alone within an amyloid-positive stratum; panel superiority is established only for amyloid endpoints [17], and tau-endpoint superiority is established only for competing tau analytes [4] [5]. There is a direct directional conflict between the claim that p-tau217 outperforms all companion markers and Aβ-related ratios [1] and the finding that pTau217/Aβ42 is the single most accurate amyloid classifier [20]; the two are reconcilable only if the ratio's benefit is confined to the amyloid question, which no paper tests explicitly. The VeraBIND comparison against p-tau217 depends on the p-tau217 threshold chosen and rests on 145 participants [4]. Beyond discrimination, plasma-biomarker machine-learning models retain AUC across cohorts but lose substantial negative predictive value under transfer (0.831 to 0.644, ADNI→A4) with systematic miscalibration [16], so any claimed panel margin over p-tau217 must survive dataset shift. Finally, absolute p-tau217 values differ across immunoassay and mass-spectrometry platforms even when amyloid prediction is equivalent [15], complicating cross-study comparison of panel-versus-single-marker results.

What would change the answer

A pre-registered head-to-head study restricted to amyloid-positive participants (Centiloid ≥25 or CSF Aβ42/40-defined), with tau PET as reference, is the decisive experiment: fit nested models of p-tau217 alone versus p-tau217 + Aβ42/40 versus p-tau217 + GFAP/NfL versus p-tau217 + eMTBR-tau or MTBR-243 versus p-tau217 + functional pTau binding, and report ΔAUC, continuous NRI/IDI, calibration slope, and intermediate-zone proportions at fixed 95% sensitivity/specificity cutoffs. A longitudinal design predicting incident tau-PET conversion in amyloid-positive cognitively unimpaired individuals would test whether panel information anticipates tau onset earlier than p-tau217 alone. Formal interaction and mediation models treating Centiloid or Aβ42/40 as a modifier of the p-tau217→tau relationship, mirroring the α-synuclein interaction design [10], would establish whether amyloid-ratio terms carry any conditional tau information at all. All of this must be replicated across at least two independent cohorts with harmonized p-tau217 values [15] and evaluated by decision-curve analysis for the concrete use case of tau-PET referral or trial enrichment, since AUC-parity has already been shown to coexist with clinically meaningful calibration loss [16].

References

[1] Bayart, J.-L., Villain, N., Planche, V., Boyer, E., Colmant, L., Le Ber, I., Picard, G., Clot, F., Bombois, S., El-Mazria, H., Dingeo, G., Bahroun, S., Nabeebaccus, F., Huyghe, L., Gerard, T., Quenon, L., Salman, Y., Durand, E., Bedel, A., Auriacombe, S., Lhommel, R., David, C., Kienlen-Campard, P., Ivanoiu, A., Douxfils, J., Levy, R., Lamari, F., Hanseeuw, B. J.. "Robustness of plasma p-Tau217 diagnostic thresholds for Alzheimer's disease across various clinical populations and laboratory environments". medRxiv. 2025-12-22T00:00:00.000Z. URL: https://doi.org/10.64898/2025.12.19.25342412

[2] Zheng, L., Ghanbarian, E., Albarran Morillo, C., Younes, K., Glover, C. M., Rabin, L. A., Grill, J. D., Ewers, M., Sajjadi, S. A., Ezzati, A., the Alzheimer's Disease Neuroimaging Initiative (ADNI),, the Health and Aging Brain (HABS-HD) Study Team,, the PREVENT-AD Research Group,. "Tau-PET subtype modifies the association between plasma p-tau217 and memory in Alzheimer's disease". medRxiv. 2026-09-28T00:00:00.000Z. URL: https://doi.org/10.64898/2026.09.25.26363629

[3] Wilson, D. H., Khare, M., Triana-Baltzer, G., Wolfe, M., Sheehy, P., Copeland, K., Hesterberg, L., Vasko, A.-J., van der Flier, W. M., Verberk, I. M. W., Teunissen, C. E., Miller, M.. "Analytical and Clinical Validation of a High Accuracy Fully Automated Digital Immunoassay for Plasma Phospho-Tau 217 for Clinical Use in Detecting Amyloid Pathology". medRxiv. 2024-11-04T00:00:00.000Z. URL: https://doi.org/10.1101/2024.10.31.24316186

[4] Hanseeuw, B. J., Quenon, L., Bayart, J.-L., Boyer, E., Colmant, L., Salman, Y., Gerard, T., Huyghe, L., Malotaux, V., Kienlen-Campard, P., Blondiaux Pirson, F., Lhommel, R., Dricot, L., Ivanoiu, A., Shamsundar, K., Pak, W., Soldo, J., Iqbal, K.. "Tau pathological binding activity in plasma before the onset of symptomatic Alzheimer s disease". medRxiv. 2026-07-17T00:00:00.000Z. URL: https://doi.org/10.64898/2026.04.03.26350110

[5] Shi, R., Zeng, X., Farinas, M. F., Nafash, M. N., Smirnov, D. S., Kamboh, M. I., Snitz, B., Kofler, J. K., Cohen, A. D., Pascoal, T. A., Ikonomovic, M. D., Zetterberg, H., Lopez, O., Karikari, T. K.. "Plasma eMTBR-tau as a robust biomarker for Alzheimer's disease tangle pathology up to 10 years before death". medRxiv. 2026-10-05T00:00:00.000Z. URL: https://doi.org/10.64898/2026.09.29.26363951

[6] Keshavan, A., Wiltshire, K., Wee, R., Gorostiaga Belio, I., Tucker, K., Hart, M., Lunn, M. P., David, M. C. B., Rizzo, L., Sadeghi-Alavijeh, O., Wilson, P., Gale, D. P., Heslegrave, A. J., Zetterberg, H., Fox, N. C., Malhotra, P. A., Schott, J. M.. "The Alzheimer's Disease Diagnosis and Plasma Phospho-Tau217 (ADAPT) study stage 1: validating clinical cut-points against CSF and amyloid PET". medRxiv. 2025-09-12T00:00:00.000Z. URL: https://doi.org/10.1101/2025.09.08.25335317

[7] Gu, Y., Kim, J. P., Kim, B. H., Lee, H., Lee, J., Kang, H., Yim, S., Shin, D., Park, H. K., Kim, S., Lee, B. H., Lee, J., Kim, H., Zetterberg, H., Chun, M. Y., Na, D. L., Lee, J.-H., Cho, S. H., Kim, B. C., Kim, S. E., Sohn, G., Yun, J., Lim, J.-S., Kim, J., Lee, K., Kim, J., Kim, H. J., Kim, E.-J., Jang, H., Seo, S. W.. "Plasma proteomics reveals stage-dependent biological remodelling in Alzheimer's disease". medRxiv. 2026-09-14T00:00:00.000Z. URL: https://doi.org/10.64898/2026.09.14.26362968

[8] Rudolph, M. D., Sutphen, C. L., Register, T. C., Lockhart, S. N., Rundle, M. M., Hughes, T. M., Bateman, J. R., Solingapuram Sai, K. K., Whitlow, C. T., Craft, S., Mielke, M. M.. "Evaluation of plasma p-tau217 for detecting amyloid pathology in a diverse and heterogeneous community-based cohort". medRxiv. 2025-01-20T00:00:00.000Z. URL: https://doi.org/10.1101/2025.01.20.25320851

[9] Lantero-Rodriguez, J., Janelidze, S., Palmqvist, S., Braun-Wohlfahrt, L. S., Vavra, J., Bali, D., Orduna Dolado, A., Mattsson-Carlgren, N., Stomrud, E., Zetterberg, H., Blennow, K., Hansson, O., Montoliu-Gaya, L., Salvado, G.. "P-TAU205 IS A BIOMARKER LINKED TO TAU-PET ABNORMALITY: A CROSS-SECTIONAL AND LONGITUDINAL STUDY". medRxiv. 2025-10-22T00:00:00.000Z. URL: https://doi.org/10.1101/2025.10.21.25338437

[10] Negida, A., Alzheimer's Disease Neuroimaging Initiative,. "Alpha-synuclein co-pathology amplifies amyloid-driven tau accumulation across Braak stages without modifying tau-cognition associations". bioRxiv. 2026-04-06T00:00:00.000Z. URL: https://doi.org/10.64898/2026.03.31.713304

[11] Insel, P., Mattsson-Carlgren, N., Langford, O., Caruso, V., Leuzy, A., Young, C., Boxer, A., Aisen, P., Sperling, R., Mormino, E., Donohue, M.. "Concurrent longitudinal changes in plasma p-tau217, tau PET, and cognition in preclinical Alzheimer's disease". medRxiv. 2025-03-26T00:00:00.000Z. URL: https://doi.org/10.1101/2025.03.25.25324632

[12] Brown, C. A., Mundada, N. S., Cousins, K. A. Q., Sadeghpour, N., Lyu, X., McGrew, E., Korecka, M., Chen-Plotkin, A., Xie, L., Wisse, L. E. M., Detre, J. A., McMillan, C. T., Lee, E. B., Nasrallah, I. M., Das, S. R., Mechanic-Hamilton, D., Yushkevich, P. A., Shaw, L. M., for the Azlheimer's Disease Neuroimaging Iniative,, Wolk, D. A.. "Tau-Clinical Mismatch Identifies Individuals with Co-Pathology and Predicts Clinical Trajectory". medRxiv. 2025-07-25T00:00:00.000Z. URL: https://doi.org/10.1101/2025.07.25.25332195

[13] Shekari, M., Gonzalez Escalante, A., Mila-Aloma, M., Falcon, C., Lopez-Martos, D., Sanchez-Benavides, G., Brugulat-Serrat, A., Ninerola-Baizan, A., Ashton, N. J., Karikari, T. K., Lantero-Rodriguez, J., Montoliu-Gaya, L., Snellman, A., Day, T. A., Dage, J. L., Ortiz-Romero, P., Tonietto, M., Borroni, E., Klein, G., Kollmorgen, G., Carboni, M., Quijano-Rubio, C., Vanmechelen, E., Minguillon, C., Fauria, K., Perissinotti, A., Molinuevo, J. L., Zetterberg, H., Blennow, K., Grau-Rivera, O., Suarez-Calvet, M., Domingo Gispert, J., ALFA Study,. "Associations of 18F-RO-948 Tau PET with Fluid AD Biomarkers, Centiloid, and Cognition in the Early AD Continuum". medRxiv. 2025-07-11T00:00:00.000Z. URL: https://doi.org/10.1101/2025.07.09.25331011

[14] Sehrawat, A., Zeng, X., Abrahamson, E. E., Deek, R. E., Gogola, A., Kamboh, M. I., Pascoal, T. A., Villemagne, V. L., Lopez, O., Ikonomovic, M. D., Snitz, B. E., Cohen, A. D., Karikari, T. K.. "Pittsburgh plasma p-tau217: classification accuracies for autosomal dominant and sporadic Alzheimer's disease in the community". medRxiv. 2025-05-06T00:00:00.000Z. URL: https://doi.org/10.1101/2025.05.03.25326526

[15] Quadrini, K. J., Aubrecht, J., Vernhes, F., Zeng, X., Johnson, A. D., Snitz, B. E., Cohen, A. D., Tooker, R., Quiceno, M. E., Campbell, B. M., Karikari, T. K., Swanson, C. J.. "Biomarker Convert: a novel approach to transform plasma pTau217 results across immunoassay and mass spectrometry analytical platforms". medRxiv. 2026-10-01T00:00:00.000Z. URL: https://doi.org/10.64898/2026.09.28.26363900

[16] Korni, A., Zandi, E.. "Cross-Cohort Generalizability of Plasma Biomarker Machine Learning Models Reveals Calibration-Driven Degradation in Clinical Utility". medRxiv. 2026-04-13T00:00:00.000Z. URL: https://doi.org/10.64898/2026.04.09.26350514

[17] Wilson, D. H., Copeland, K., Vasko, A.-J., Hesterberg, L., Khare, M., Wolfe, M., Sheehy, P., Verberk, I. M. W., Miller, M., Teunissen, C. E.. "Clinical Performance of LucentAD Complete: A Multi-Analyte Algorithmic Blood Test for Detecting Brain Amyloid with a Scalable, Multiplexed Digital Immunoassay". medRxiv. 2025-04-01T00:00:00.000Z. URL: https://doi.org/10.1101/2025.03.31.25324984

[18] Mavromati, K., Dyer, A. H., Beazer, J. D., Hughes, L., Kennelly, S. P., Quinn, T. J.. "Latent biomarker states underlying disagreement between PET-anchored and distribution-based plasma pTau-217 positivity thresholds". medRxiv. 2026-07-19T00:00:00.000Z. URL: https://doi.org/10.64898/2026.07.17.26358314

[19] Cody, K., Sokolowski, A., Johns, E., Medina Guerra, L., Winer, J., Young, C., Younes, K., Dumitrescu, L., Archer, D., Durant, A., Sathe, A., Koran, M. E., Mez, J., Saykin, A., Toga, A., Cuccaro, M., Tosun, D., Insel, P., Johnson, S., Harrison, T., Hohman, T., Mormino, E.. "Characterizing amyloid and tau PET-based stages across the clinical continuum". medRxiv. 2025-09-29T00:00:00.000Z. URL: https://doi.org/10.1101/2025.09.26.25336751

[20] Lehmann, S., Gabelle, A., Duchiron, M., Busto, G., Morchikh, M., Delaby, C., Hirtz, C., Mondesert, E., Cristol, J.-P., Barnier-Figue, G., Perrein, F., Turpinat, C., Jurici, S., Bennys, K.. "Comparative performance of plasma pTau181/Aβ42, pTau217/Aβ42 ratios, and individual measurements in detecting brain amyloidosis". medRxiv. 2025-05-30T00:00:00.000Z. URL: https://doi.org/10.1101/2024.12.07.24318640

[21] Anastasi, F., Fernandez-Lebrero, A., Ashton, N. J., Ortiz-Romero, P., Torres-Torronteras, J., Gonzalez-Escalante, A., Mila-Aloma, M., Contador, J., Garcia-Escobar, G., Manero-Borras, R. M., Navalpotro-Gomez, I., Sahajan, A., Hao, Q., Zhang, B., Jeromin, A., Le Bastard, N., Nadal, A., Mousavi, T., Kollmorgen, G., Carboni, M., Grau-Rivera, O., Zetterberg, H., del Campo, M., Blennow, K., Puig-Pijoan, A., Suarez-Calvet, M.. "A head-to-head comparison of plasma biomarkers to detect biologically defined Alzheimer in a memory clinic". medRxiv. 2024-10-27T00:00:00.000Z. URL: https://doi.org/10.1101/2024.10.26.24316176

How this survey was made

This survey answers one question: “How do multi-analyte plasma biomarker panels combining p-tau217 with amyloid-beta 40/42 or other markers compare against single-marker p-tau217 for discriminating tau-PET positivity among amyloid-positive individuals?”

We keep our own collection of recent preprints and refresh it every day from bioRxiv, medRxiv, arXiv, PsyArXiv, SocArXiv and ChemRxiv. For each survey, we search that collection for recent work and keep the strongest matches: papers that use the same terms as the question, and papers on the same topic in different words.

The text was written from the titles and abstracts of those papers; the full texts were not read. Statements that draw on a particular paper are marked with a number in brackets. The reference list just above is the 21 papers whose work this text draws on, chosen from the 81 papers those searches returned; the rest of the shortlist is not printed. It is not a complete review of everything published on the question.

Treat this as a starting point and a reading list, not a peer-reviewed review. The searches behind this survey ran on October 22, 2026 and covered work from the previous two years.