Scientists Say It's Time to Update ATN Criteria

With the development of sensitive biomarkers that reflect the hallmark pathologies of Alzheimer’s, the field has advanced from a clinical to a biological definition of the disease. Four years ago, the AT(N) biomarker classification system was introduced, where A stands for amyloid, T for tau, and N for neurodegeneration (Apr 2018 news). At the time, either CSF phospho-tau or tau-PET were considered acceptable ways to gauge the “T.” But are these measures really interchangeable? A study published September 9 in Brain says no.

  • Among people with brain amyloid, elevated p-tau217 predicted future tangles.
  • Elevated tau-PET predicted neurodegeneration and cognitive decline.
  • Scientists recommend adding phospho-tau (P) to AT(N) classification.

Scientists led by Rik Ossenkoppele and Oskar Hansson at Lund University, Sweden, investigated the relationships between various fluid, imaging, and cognitive measures among people without dementia. They found that among those who had accumulated amyloid in their brain, elevated CSF p-tau217 predicted a subsequent rise in tau tangle burden, as reported previously (for example, Mattsson-Carlgren et al., 2020). However, CSF p-tau217 did worse at predicting impending neurodegeneration or cognitive decline. In contrast, elevated tangle burden strongly correlated with subsequent brain atrophy and cognitive loss. The findings support the idea that fluid p-tau becomes abnormal before tau-PET does, and that the two markers reflect distinct aspects of the AD pathogenic cascade.

First author Colin Groot and colleagues conducted their study using longitudinal data from the BIOFINDER cohort. Their sample included 195 cognitively unimpaired people and 36 with MCI, all of whom had had tests for fluid and imaging biomarkers, brain atrophy, and cognition at baseline and over a two-year follow-up period. Based on thresholds of abnormality for amyloid-PET (A), CSF p-tau217 (P), and tau-PET (T), the researchers classified these 231 people into four groups: 135 were designated as controls (A-P-T-), 30 as tau-negative (A+P-T-), 48 as tau-discordant (A+P+T-), and 18 as tau-positive (A+P+T+).

What did they find? In short, having elevated CSF p-tau217 at baseline predicted future accumulation of tau tangles as measured by tau-PET. Compared to controls or the A+P-T- group, those in the tau-discordant A+P+T- group had a steeper annual rise in tau tangles, particularly in early Braak stage regions. Those in the tau-positive A+P+T+ group had faster tau accumulation than any of the other groups in both early and late Braak regions.

The two tau biomarkers related differently to other markers of AD. The cortex thinned faster relative to controls among people who were amyloid-positive at baseline. However, elevated CSF p-tau217 alone did not predict further atrophy. Only those in the tau-positive A+P+T+ group saw exacerbated thinning of the entorhinal cortex. Similarly, among those who were amyloid-positive, CSF p-tau217 alone did not predict faster cognitive decline, but having both tau markers be abnormal did.

More broadly, studies conducted over the past few years in DIAN, and other longitudinal research cohorts worldwide, have found much the same thing. In short, tau phosphorylation at residue 217, but also 231 and 181, becomes detectable and then rises soon after Aβ biomarkers turn positive. This happens years before tau PET picks up tangles. For their part, tau PET scans turn positive closer to symptom onset than do CSF and blood p-tau assays. Together, the findings suggest that while elevated blood and CSF p-tau signals that the amyloid-induced AD cascade is in motion, the accumulation of tau tangles heralds the neurodegenerative phase of the disease.

This implies that the tau-discordant A+P+T- group might respond best to tau-targeted therapies, Hansson said. As such, imaging and fluid tau biomarkers would serve distinct purposes and could complement each other in selection for clinical trials and research studies.

Regarding the AT(N) biomarker classification system, Hansson believes it’s time to add a P, for fluid p-tau, before the T. Ultimately, it may be possible to measure the “T” with other fluid markers, Hansson said. For example, recent studies suggest that CSF p-tau205 rises later than p-tau181 or p-tau217, suggesting it might serve as a proxy for abnormal tau-PET.—Jessica Shugart

Comments

    • User Profile Image Kanta Horie
      Washington University School of Medicine
    • Posted:

    This report by Groot and colleagues takes the critical step in the current AT(N) framework to suggest further classification with “P”, phosphorylated tau (p-tau). In this study, the team deeply characterized CSF p-tau217 as a representative of this proposed “P” class. They clarified that the A+P+T- group did not show faster atrophy and cognitive decline than the A+P-T-, while the A+P+T+ group showed faster atrophy and cognitive decline compared to all other groups. That supports p-tau217 as an early stage AD biomarker modulated closely with amyloid, rather than tau pathology followed by neurodegeneration as reported previously (Milà-Alomà et al., 2022).

    Conceptually, I agree with the suggested framework of APT(N); however, I also make a reminder that diverse characteristics of different p-tau species are reported and different statements may be available depending on the specific p-tau species to represent the class of “P.” For instance, Barthélemy and colleagues reported that p-tau217 began with the initial increases in aggregate amyloid-β as early as two decades before the development of tau pathology in dominantly inherited AD, whereas others, like p-tau205, increase with atrophy and hypometabolism closer to symptom onset and tangles formation (Barthélemy et al., 2020). These findings have been validated in sporadic AD cases (Barthélemy et al., AAIC2022), suggesting that it is necessary to define "P" classification more in detail to consider the newly suggested framework of APT(N). 

    Finally, now it is obviously desirable to establish fluid biomarkers to track changes in tau pathology specifically that occur even after AD dementia onset, because the current p-tau and other biomarkers are not sufficient to cover this point.  The ongoing studies to fulfill this unmet need will further refine the APT(N) framework drafted in this study by Groot et al.

    References:

    Barthélemy NR, Li Y, Joseph-Mathurin N, Gordon BA, Hassenstab J, Benzinger TL, Buckles V, Fagan AM, Perrin RJ, Goate AM, Morris JC, Karch CM, Xiong C, Allegri R, Mendez PC, Berman SB, Ikeuchi T, Mori H, Shimada H, Shoji M, Suzuki K, Noble J, Farlow M, Chhatwal J, Graff-Radford NR, Salloway S, Schofield PR, Masters CL, Martins RN, O'Connor A, Fox NC, Levin J, Jucker M, Gabelle A, Lehmann S, Sato C, Bateman RJ, McDade E, Dominantly Inherited Alzheimer Network. A soluble phosphorylated tau signature links tau, amyloid and the evolution of stages of dominantly inherited Alzheimer's disease. Nat Med. 2020 Mar;26(3):398-407. Epub 2020 Mar 11 PubMed.

    Milà-Alomà M, Ashton NJ, Shekari M, Salvadó G, Ortiz-Romero P, Montoliu-Gaya L, Benedet AL, Karikari TK, Lantero-Rodriguez J, Vanmechelen E, Day TA, González-Escalante A, Sánchez-Benavides G, Minguillon C, Fauria K, Molinuevo JL, Dage JL, Zetterberg H, Gispert JD, Suárez-Calvet M, Blennow K. Plasma p-tau231 and p-tau217 as state markers of amyloid-β pathology in preclinical Alzheimer's disease. Nat Med. 2022 Sep;28(9):1797-1801. Epub 2022 Aug 11 PubMed. Correction.

    View all comments by Kanta Horie

  2. The currently described work, along with a number of other publications in the field, clearly shows that changes in p-tau occur well before increases in tau PET. I'm hesitant, though, to suggest the field move toward adding another category. For the most part, there is a very high congruency between p-tau species and amyloid as measured with PET or CSF. Even p-tau205, which elevates later in the disease course than do p-tau231, p-tau181, or p-tau217, seems to be elevated a decade or more before tau PET. I think the only biofluid measure that has come close to capturing increases in tau PET are some of the MTBR species as published by Horie et al., 2021.

    References:

    Horie K, Barthélemy NR, Sato C, Bateman RJ. CSF tau microtubule binding region identifies tau tangle and clinical stages of Alzheimer's disease. Brain. 2021 Mar 3;144(2):515-527. PubMed. Correction.

    View all comments by Brian Gordon

  3. We appreciate this interesting discussion. Based on our data, CSF and plasma concentrations of p-tau217 are not simply a marker of Aβ plaques. In fact, both neuropathology-based and PET imaging-based studies have shown that fluid p-tau217 is related to both plaques and tangles to similar degrees (Mattsson-Carlgren et al., 2021). This is very different from Aβ42/Aβ40 and p-tau231, which both are mainly related to plaque pathology (Salvadó et al., 2022). Further, in human brain tissue, p-tau217 is mainly found in neurofibrillary tangles and neuropil threads, but also in granulovacuolar degeneration bodies and multi-vesicular bodies, and these brain changes correlate strongly with plasma p-tau217 levels in amyloid-positive individuals (Wennström et al., 2022). Together, these studies and others strongly suggest that p-tau217 is reflecting amyloid-induced changes in tau metabolism and aggregation, which occur subsequent to (and likely downstream of) Aβ aggregation.

    From a clinical point of view, it is important to note that amyloid-positive but p-tau217-negative cases (A+P-T-) do not accumulate tau tangle pathology in the coming years as revealed with longitudinal tau-PET imaging (Groot et al., 2022). This is in contrast to those who are both amyloid-positive and p-tau217 positive, but still tau-PET-negative (A+P+T-). These cases do indeed show evidence of tau tangle accumulation in the next couple of years.

    More importantly, in amyloid-positive, cognitively unimpaired cases, p-tau217 is a stronger predictor of cognitive decline over the coming four to six years than is amyloid PET, again indicating that A+P+ individuals are further along in the disease process than A+P- cases (Mattsson-Carlgren et al., unpublished data from BioFINDER-1 and WRAP studies).

    In our hands, CSF concentrations of MTBR-tau243 are indeed very closely related to tau-PET. This marker holds great promise to replace tau-PET in certain situations, when one needs a measure of tau tangle pathology (T). But I am uncertain if this tau biomarker can replace p-tau217 (P) as a marker of amyloid-induced changes in tau metabolism in sporadic preclinical disease, which was our intention with the idea of using APT(N).

    References:

    Mattsson-Carlgren N, Janelidze S, Bateman RJ, Smith R, Stomrud E, Serrano GE, Reiman EM, Palmqvist S, Dage JL, Beach TG, Hansson O. Soluble P-tau217 reflects amyloid and tau pathology and mediates the association of amyloid with tau. EMBO Mol Med. 2021 Jun 7;13(6):e14022. Epub 2021 May 5 PubMed.

    Salvadó G, Ossenkoppele R, Ashton NJ, Beach TG, Serrano GE, Zetterberg H, Mattsson-Carlgren N, Janelidze S, Blennow K, Hansson O. Specific associations between plasma biomarkers and post-mortem amyloid plaque and neurofibrillary tau tangle loads. medRxiv, August 22, 2022. medRxiv

    Wennström M, Janelidze S, Nilsson KP, Netherlands Brain Bank, Serrano GE, Beach TG, Dage JL, Hansson O. Cellular localization of p-tau217 in brain and its association with p-tau217 plasma levels. Acta Neuropathol Commun. 2022 Jan 6;10(1):3. PubMed.

    Groot C, Smith R, Stomrud E, Binette AP, Leuzy A, Wuestefeld A, Wisse LE, Palmqvist S, Mattsson-Carlgren N, Janelidze S, Strandberg O, Ossenkoppele R, Hansson O. Phospho-tau with subthreshold tau-PET predicts increased tau accumulation rates in amyloid-positive individuals. Brain. 2022 Sep 9; PubMed.

    View all comments by Oskar Hansson

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. New Definition of Alzheimer’s Hinges on Biology, Not Symptoms

Paper Citations

  1. Mattsson-Carlgren N, Andersson E, Janelidze S, Ossenkoppele R, Insel P, Strandberg O, Zetterberg H, Rosen HJ, Rabinovici G, Chai X, Blennow K, Dage JL, Stomrud E, Smith R, Palmqvist S, Hansson O. Aβ deposition is associated with increases in soluble and phosphorylated tau that precede a positive Tau PET in Alzheimer's disease. Sci Adv. 2020 Apr;6(16):eaaz2387. Epub 2020 Apr 15 PubMed.

Further Reading

Papers

  1. Reimand J, Collij L, Scheltens P, Bouwman F, Ossenkoppele R, Alzheimer's Disease Neuroimaging Initiative. Association of amyloid-β CSF/PET discordance and tau load 5 years later. Neurology. 2020 Nov 10;95(19):e2648-e2657. Epub 2020 Sep 10 PubMed.

Primary Papers

  1. Groot C, Smith R, Stomrud E, Binette AP, Leuzy A, Wuestefeld A, Wisse LE, Palmqvist S, Mattsson-Carlgren N, Janelidze S, Strandberg O, Ossenkoppele R, Hansson O. Phospho-tau with subthreshold tau-PET predicts increased tau accumulation rates in amyloid-positive individuals. Brain. 2022 Sep 9; PubMed.

Annotate

To make an annotation you must Login or Register.