24 Nov 2020

As it happens, to find out what is going on in the brain of someone with Alzheimer’s disease, all it takes is a drop of blood. Unimaginable only three years ago, researchers now have plasma biomarkers that correlate with amyloid plaques, neurofibrillary tangles, and neurodegeneration—the three major neuropathological hallmarks of AD. So rapidly has the science progressed that C2N Diagnostics, a company that grew out of basic biomarker research at Randall Bateman’s lab at Washington University, St. Louis, recently gained certification for its Precivity mass-spec Aβ assay. In the U.S., regulators at the Centers for Medicare & Medicaid Services approved the test under the Clinical Laboratory Improvement Amendments protocol. CLIA certification is a prerequisite for any laboratory testing of human samples outside of research settings. The EU gave the test its own CE Mark of approval for in vitro diagnostic medical devices.

Developers of plasma phospho-tau assays are nipping at C2N’s heels. Recent publications and meeting presentations bolster the evidence that upticks in p-tau181, p-tau217, and p-tau231 occur soon after Aβ begins to accumulate in the brain. That’s years before neurofibrillary tangles can be detected by PET, and decades before symptoms. Researchers predict these markers, which also distinguish AD from other forms of dementia, will lead to more robust tests than those based on Aβ, because of their larger dynamic range—phosphorylated fragments of tau climb by as much as 10-fold in the blood as AD progresses, whereas the Aβ42/40 ratio falls by about 15 percent at most.

Nevertheless, researchers applauded C2N’s milestone. “The development of a CLIA-certified blood test to help characterize the presence or absence of amyloid plaques and return information to clinical providers, patients, and families in the clinical setting is an important landmark,” wrote Eric Reiman, Banner Alzheimer’s Institute, Phoenix. “It is my hope that we will see other assays, including plasma p-tau217, follow in the footsteps of this effort, and provide additional value to those in the research, treatment development, and clinical setting,” he added (see comment below). Reiman co-founded ALZPath, a company planning to develop plasma p-tau217 and other blood-based biomarkers for use in research, drug development, and clinical care.

Oskar Hansson, Lund University, Sweden, also welcomed the news. “I think this is a truly exciting development for AD diagnostics!” he wrote to Alzforum. And he offered a cautionary note. “We now need to develop appropriate use criteria to avoid misuse and misinterpretation of new blood tests for diagnosis of Alzheimer’s disease, similar to what has already been done for CSF AD biomarkers and amyloid PET imaging,” he wrote (Jan 2013 conference news). 

To be clear, C2N’s Precivity plasma Aβ test is not intended as a stand-alone diagnostic, at least not yet. It has received breakthrough device designation from the FDA, and C2N is still working toward approval as an in vitro-diagnostic.

For now, C2N hopes the assay will help clinicians and scientists make a correct diagnosis. It is not intended as a screening test. “It is for individuals under care of a physician who present with complaint and evidence of cognitive decline and are being evaluated for dementia,” C2N CEO Joel Braunstein told Alzforum. “If we can boost physician confidence in their diagnosis from 55 or 60 percent to 85 percent or greater, then we have produced a useful diagnostic tool,” he said. That said, the company is marketing the test directly to consumers. It is available in 45 states, D.C., and Puerto Rico, and other states are expected to approve it soon.

In clinics, it remains to be seen how quickly this test will be adopted. Thomas Wisniewski, who oversees the Alzheimer’s Disease Research Center at New York University, told Alzforum that his hospital’s administration has not made a decision yet. Ditto for Vanderbilt University Medical Center, said Angela Jefferson, who heads that ADRC. Some physicians, at least, plan to stick with PET. Andrew Budson, VA Boston Healthcare System, told Alzforum he does not intend to use the plasma test at the VA or the Boston Memory Center. “I could envision a practitioner who does not have access to amyloid scans being more interested in it,” he said.

Braunstein recognizes that C2N’s test directly competes with amyloid PET, but thinks his test has a distinct advantage. “Because it costs considerably less, it is our hope that the perceived clinical value is enhanced,” he said. He thinks that as clinicians weigh their options for patients, the threshold for prescribing a plasma test may be lower than for PET. The list price for Precivity is $1,250, whereas a PET scan typically costs about $3,000 (Aug 2019 conference news). The arrival of Precivity might pressure PET centers to lower the cost of amyloid scans.

Phospho-tau markers are poised to enter that competitive space soon. Hansson thinks their implementation in clinical practice is one to three years behind. Besides boasting a broader range, p-taus can be measured by immunoassay, which means more testing centers can get involved. For now, C2N’s CLIA approval is for a laboratory-developed test, i.e., only C2N’s lab in St. Louis can conduct the test. C2N expects to have other sites certified.

Tau Next?
Researchers led by Jeffrey Dage at Eli Lilly kick-started the plasma p-tau revolution when they developed an immunoassay for p-tau181 (Mielke et al., 2018). A similar immunoassay, developed by Kaj Blennow and colleagues at the University of Gothenburg, Sweden, soon followed. Both use two antibodies, one to capture tau fragments that are phosphorylated at threonine 181, and another to detect them. The assays can measure as little as 0.5 pg/ml of tau in the blood, which is sensitive enough to detect increases in people who are asymptomatic but who test positive for amyloid by PET, CSF, or plasma analysis. The scientists have developed similar immunoassays for p-tau217 and p-tau231.

Within two years, several groups had used these assays to measure p-tau181 in nine cohorts: the UCSF Memory and Aging Center; the Advancing Research and Treatment for Frontotemporal Lobar Degeneration (ARTFL) study; a Phase 1 trial of Lilly’s therapeutic antibody against pyroglutamate; the Swedish BioFINDER study; BioFinder 2; the Arizona Study of Aging and Neurodegenerative Disorders Brain & Body; the University of Gothenburg; the Canadian TRIAD study; and a primary care cohort at McGill University, Montreal (Mar 2020 news; Apr 2020 conference news). 

Taken together, the data indicate that plasma p-tau181 begins to tick up in the blood around the time Aβ begins to accumulate in the brain, and that it correlates with CSF p-tau181 and with tau PET signals. It also predicts tau-based Braak staging. Plasma p-tau181 distinguished Alzheimer’s from other dementias, including tauopathies such as frontotemporal dementia and progressive supranuclear palsy, with AUCs around 0.88. In effect, the isoform acts as a marker of Aβ-dependent tau pathology. A new study now adds a 10th cohort.

In the November 10 EMBO Molecular Medicine online, researchers led by Blennow and José Luis Molinuevo at the Barelonaβeta Research Center, Spain, report findings from the Alzheimer’s and Families cohort prospective study in Barcelona. ALFA+ follows healthy volunteers, 45 and older, who are at higher risk of AD due to their family history. Co-first authors Marc Suárez-Calvet, Thomas Karikari, and colleagues report that 131 ALFA+ volunteers who tested positive for Aβ in the brain by CSF Aβ42 or amyloid PET had more p-tau181 in their plasma than 250 volunteers who were Aβ-negative. In fact, even people who had a marginal PET signal—i.e., too low to be deemed amyloid-positive—had more plasma p-tau181 in their plasma, suggesting the marker rises early in disease. In support of this, they found that plasma p-tau181 began to creep higher just as the CSF Aβ42/40 ratio began to fall.

Beyond amyloid and tangle correlations, researchers are finding that plasma p-tau181 predicts disease progression, at least at the group level. In Adam Boxer’s lab at the University of California, San Francisco, Elisabeth Thijssen and colleagues found that higher plasma p-tau181 correlated with faster cognitive decline over 2½ years (Thijssen et al., 2020). Similarly, researchers in Hansson’s lab found that in BioFinder, people with the highest levels were more likely to develop AD over a period of up to eight years (Janelidze et al., 2020). 

What’s more, data are starting to suggest that this holds true at the individual level, as well. This bodes well for using p-tau to stratify people for clinical trials, or to track treatment. In a manuscript uploaded on medRxiv in July, Lund scientists followed individual BioFinder volunteers who had mild cognitive impairment. Co-first authors Nicholas Cullen and Antoine Leuzy modeled how well permutations of plasma p-tau181, plasma Aβ42/40, and plasma neurofilament light (NfL) predicted outcomes among each of the 148 people. The combination of NfL and p-tau181 performed best. Among a subset of 107 people, it predicted AD over the next four years with an AUC of 0.88. The same combo best predicted cognitive decline among a subset of 118 volunteers.

Cullen and colleagues repeated their analysis for people in ADNI. Again, NfL and p-tau181 did best, predicting progression to AD over four years among 74 people with an AUC of 0.88, and best predicting cognitive decline among 64 volunteers.

How did plasma p-tau and Aβ relate? Plasma Aβ42/40 did not boost the prognostic value of p-tau181. “This is logical, since symptoms in AD are linked to tau pathology, and elevations in tau biomarkers appear to be dependent on Aβ pathology,” wrote the authors. Still, they note that plasma Aβ42/Aβ40 may have added value at the preclinical stage, when the Aβ42/Aβ40 ratio has already become pathological but tau and neurodegeneration markers are still increasing. “I think plasma p-tau217 and p-tau181 are very well-suited for identification of AD in individuals with MCI or dementia, but combinations of several blood-based markers are most probably needed for reliable detection of preclinical AD,” Hansson told Alzforum.

Plasma p-tau181 may also help clinicians diagnose dementia in people with Down’s syndrome, as reported by Juan Fortea from the Hospital of Sant Pau, Barcelona. Working with Alberto Lleó, Fortea and colleagues are tracking various fluid and imaging markers in the Down Alzheimer Barcelona Neuroimaging Initiative (DABNI). People with Down’s syndrome carry an extra copy of the amyloid precursor protein gene; most develop AD by age 60.

At this year’s virtual CTAD meeting, held November 4-7, Fortea showed how plasma p-tau181 levels correlated with age in 366 DABNI volunteers but not in 44 controls. Breaking this down by symptoms cross-sectionally, p-tau181 rose as people worsened from asymptomatic Down’s to prodromal dementia, and to frank Alzheimer’s dementia.

The plasma marker distinguished Aβ-positive from Aβ-negative Down’s volunteers with an AUC of 0.77, weaker than correlations seen in the general population. Fortea believe this is because the latter includes people who will never get AD, whereas for people with Down’s, it is but a question of time. “In our opinion that accounts for the lower sensitivity,” he said.

In DABNI, plasma p-tau181 correlated strongest with CSF p-tau181, but also with CSF NfL and, to a lesser extent, Aβ42. It also tracked with cortical atrophy and hypometabolism as measured by MRI. Fortea thinks p-tau181 could be a screening marker to detect dementia in DS and for clinical trials.

P-tau217 on Top
No sooner had scientists been gripped by the promise of plasma p-tau181 than another phosphorylated isoform grabbed the limelight. Hansson’s group and researchers led by Nicolas Barthélemy at Washington University, St. Louis, found that p-tau217 in the CSF more tightly correlated with plaques and tangles than did p-tau181, that levels in the CSF climbed higher than any other p-tau marker, and that it better distinguished AD from other dementias (Apr 2020 conference news). Suárez-Calvet and colleagues also found that in the ALFA+ cohort, people who tested positive for amyloid based on the CSF Aβ42/40 ratio had more p-tau217 in their CSF than Aβ-negatives.

And what about blood? Same story. In July, Hansson’s group reported that among three different cohorts totaling more than 1,400 volunteers, plasma p-tau217 outperformed plasma p-tau181 in distinguishing AD from controls (Jul 2020 conference news). It also correlated with tangle density in a neuropathology cohort of 34 AD patients and 47 controls. In the October 17 Brain, first author Niklas Mattsson-Carlgren and colleagues in Lund reported that plasma p-tau217 rose faster in people with AD or who had MCI due to AD than in healthy age-matched controls, indicating ongoing phosphorylation of tau early in disease. At CTAD, Dage reported that plasma p-tau217 tracked with neuropathology among 113 subjects in the Washington Heights-Hamilton Heights-Inwood Community Aging Project run by Richard Mayeux at Columbia University Medical Center, New York. The correlations were robust, with an AUC of 0.82, even though the blood samples had been taken an average of three years before death.

How would plasma p-tau217 track with imaging? The scientists correlated the marker with amyloid and tau PET among 490 subjects in BioFinder 2. In the November 9 JAMA Neurology online, first author Shorena Janelidze and colleagues report that people who were Aβ-positive had more p-tau217 in their blood, even if they tested negative for neurofibrillary tangles on tau PET scans. This implies that plasma p-tau217 increases before tangles begin to form. Indeed, using event-based modeling to predict the order of biomarker progression, Janelidze found that plasma and CSF p-tau217 rose before global tau PET turned positive, and even before they saw a positive PET signal in the entorhinal cortex, a region that seeds the earliest accumulation of tangles (see image at left). “The findings by Janelidze et al contribute to an emerging consensus in the field that fluid biomarkers are detecting Aβ-triggered changes in tau phosphorylation and secretion,” wrote Thijssen and Gil Rabinovici, also from UCSF, in a JAMA Neurology editorial. “These changes likely precede the aggregation of hyperphosphorylated tau into paired helical filaments that form neurofibrillary tangles and bind PET ligands,” they add.

For prognosis, too, plasma p-tau217 looks promising. At CTAD, Joana Pereira, Lund University, reported that it predicts longitudinal amyloid accumulation, tau burden, brain atrophy and cognitive decline in early AD. The data come from 190 volunteers in BioFinder, of whom 50 had been diagnosed with AD and 15 with some other form of dementia. All had baseline measures of plasma and CSF markers, plus scans for brain amyloid, tangles, and cortical thickness at baseline and again one and two years later.

Pereria split the cohort into those with high and low baseline levels of plasma Aβ42/40, p-tau181, p-tau217, and NfL. High plasma p-tau217 associated with faster accumulation of amyloid and neurofibrillary tangles and faster cortical and subcortical atrophy than those with low plasma levels. High plasma p-tau181 also predicted faster tangle accumulation. However, in a multivariate analysis that accounted for all four plasma markers, the predictive power of p-tau217 was twice as strong. Only p-tau217 predicted thinning of the temporal cortex.

All told, plasma p-tau217 seems the more robust marker. Reiman believes it will be a theragnostic endpoint in prevention or disease-modifying trials; whether it will track with clinical benefit remains to be seen. At CTAD, Reiman said proof for that will only come once an effective therapy is found.

In the meantime, researchers are looking retrospectively at blood samples stored during prior clinical trials. In his CTAD presentation, Dage showed analysis of data from the Expedition 1 and 2 trials of solanezumab. In the amyloid PET substudy of those trials, plasma p-tau217 identified Aβ-positives with an AUC of 0.94, and correlated with progression. People who had more than 0.43 pg/ml p-tau217 in their plasma decline faster on the CDR-sum-of-boxes dementia rating scale. Could the tau isoform be used as a pharmacodynamic marker for drug effect? Perhaps. In Expedition, plasma p-tau217 rose by about 14 percent per year. Dage calculated that to detect a one-third reduction of that increase at 90 percent power would only require a cohort of 35 people in the placebo and drug arms. Alas, in neither Expedition trial did the drug arm show a reduction in plasma p-tau217.

In the much smaller Phase 1b trial of donanemab, however, there was a hint of a signal. In this trial, monthly 10 mg/Kg and 20 mg/Kg doses of the Aβ immunotherapy shrank brain amyloid over 72 weeks. The lower dose had no effect on the plasma tau marker but the larger dose showed a consistent trend to week 48. While this was not statistically significant, Dage said he did not expect it do be given that there were only five patients in the group. “It does encourage us to look at this as a potential pharmacodynamic biomarker in larger trials,” he said.—Tom Fagan

Comments

1. • Eric M. Reiman
     Arizona Alzheimer's Consortium
   • Posted: 24 Nov 2020

   Congratulations to Randy Bateman and his Washington University colleagues for the initial development of this assay and to Joel Braunstein and his C2N colleagues for doing the work needed to introduce its use in the clinical setting! It is an important accomplishment and, I believe, just the beginning when it comes to the use of blood-based biomarkers in the clinical setting.

   The development of a CLIA-certified blood test to help characterize the presence or absence of amyloid plaques and return information to clinical providers, patients, and families in the clinical setting is an important landmark. It is my hope that we will see other assays, including plasma p-tau217, follow in the footsteps of this effort and provide additional value to those in the research, treatment development, and clinical setting.

   Like any tool, there is more work to do to help to fulfill its promise in the clinical setting, including helping the field find the optimal ways to communicate information, extend the approach to make it widely used and informative in the primary care setting, and leverage the information in ways that will have the greatest value for providers, patients, and families, before and after amyloid-modifying and other treatments become available.

   View all comments by Eric M. Reiman
2. • Oskar Hansson
     Lund University
   • Posted: 24 Nov 2020

   I think this is a truly exciting development for AD diagnostics! We now need to develop appropriate use criteria to avoid misuse and misinterpretation of new blood tests for diagnosis of Alzheimer’s disease, similar to what has already been done for CSF AD biomarkers and amyloid PET imaging. I would be hesitant to use blood-based AD biomarkers outside memory clinics before i) prospective studies have been performed in primary care settings, and ii) primary care physicians have received adequate training in how to interpret the results.

   Plasma p-tau assays are probably one to three years behind when it comes to implementation into the clinical practice, but very intense efforts are ongoing. Personally, I think plasma p-tau217 and p-tau181 are very well-suited for identification of AD in individuals with MCI or dementia, but combinations of several blood-based markers are most probably need for reliable detection of preclinical AD.

   If a disease-modifying therapy will be approved for clinical use, then the need for blood-based biomarkers of detection of prodromal AD and mild AD dementia will increase substantially.

   View all comments by Oskar Hansson
3. • Nicolas Barthélemy
     Washington University School of Medicine
   • Posted: 24 Nov 2020

   Suárez-Calvet et al. report the promising performance of novel p-tau assays in cerebrospinal fluid to detect early changes of tau phosphorylation due to amyloid pathology. They monitored a new combination of antibodies capturing different lengths (N-ter or mid-domain) and phosphorylated residues (181, 217, and 231). Notably, their results confirm early changes of CSF tau phosphorylation in cognitively normal amyloid-positive participants as previously published for p-tau181 and p-tau217. The authors suggest p-tau231 significantly increases at the preclinical stage. Consistently, they report good correlation among p-tau levels measured at T217, T181 and T231.

   Threonine231 is located after the main CSF truncation reported on residue 224 (Cicognola et al., 2019). Capturing CSF p-tau231 with a mid-domain antibody, as performed in this study, would recover a relatively low abundant non-truncated tau fragment including the T231 residue. So, p-tau231 is likely less abundant than p-tau181 and p-tau217 and more difficult to detect without a sensitive assay. Of note, p-tau181, p-tau217, and p-tau231, along with p-tau202, are the phosphorylated species that are most enriched in AD brain tau aggregates (Horie et al., 2020). Though these species are increased in preclinical AD, corresponding brain enrichment could impact their levels in CSF once tau pathology is prominent and potentially cause them to decrease at symptomatic stage.

   These sites are part of the residues monitored by the immunoprecipitation mass spectrometry assay we developed at the Bateman lab that measures tau phosphorylation occupancy, or p-tau/tau ratio (Barthélemy et al., 2019). P-tau231 is currently under investigation in the DIAN cohort and our internal data confirms high correlation of this site with p-tau217. P-tau217 in DIAN becomes abnormal more than 20 years before the emergence of clinical symptoms (Barthélemy et al., 2020). Using our mass spectrometry method, we reported similar phosphorylation changes in a group of residues including T111, T153, T181, T217, and T231 (Wildsmith et al., ADPD conference 2020). It is important to note some other phosphorylated sites, such as S199, S202, or T175, are not significantly hyperphosphorylated in AD CSF tau. This finding supports site-specific modifications of tau phosphorylation and would predict different diagnostic efficacy of a CSF p-tau assay depending on the considered site.

   Finally, I found particularly interesting the greater accuracy observed for N-ter p-tau 181 over mid p-tau181 to detect amyloidosis. This suggests phosphorylation might be different depending on the tau truncated fragment considered.

   References:

   Cicognola C, Brinkmalm G, Wahlgren J, Portelius E, Gobom J, Cullen NC, Hansson O, Parnetti L, Constantinescu R, Wildsmith K, Chen HH, Beach TG, Lashley T, Zetterberg H, Blennow K, Höglund K. Novel tau fragments in cerebrospinal fluid: relation to tangle pathology and cognitive decline in Alzheimer's disease. Acta Neuropathol. 2019 Feb;137(2):279-296. Epub 2018 Dec 13 PubMed.

   Horie K, Barthélemy NR, Mallipeddi N, Li Y, Franklin EE, Perrin RJ, Bateman RJ, Sato C. Regional correlation of biochemical measures of amyloid and tau phosphorylation in the brain. Acta Neuropathol Commun. 2020 Aug 27;8(1):149. PubMed.

   Barthélemy NR, Mallipeddi N, Moiseyev P, Sato C, Bateman RJ. Tau Phosphorylation Rates Measured by Mass Spectrometry Differ in the Intracellular Brain vs. Extracellular Cerebrospinal Fluid Compartments and Are Differentially Affected by Alzheimer's Disease. Front Aging Neurosci. 2019;11:121. Epub 2019 May 21 PubMed.

   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.

   View all comments by Nicolas Barthélemy

Make a Comment

To make a comment you must login or register.

References

News Citations

1. HAI—Amyloid Imaging in the Clinic: New Guidelines and Data 29 Jan 2013
2. Are Aβ Blood Tests Ready for Prime Time? 21 Aug 2019
3. A Phospho-Tau Plasma Assay for Alzheimer’s? 31 Mar 2020
4. New Assay, New Cohorts—Plasma p-Tau181 Looks Even Better 3 Apr 2020
5. 217—The Best Phospho-Tau Marker for Alzheimer’s? 10 Apr 2020
6. Plasma p-Tau217 Set to Transform Alzheimer’s Diagnostics 29 Jul 2020

Therapeutics Citations

1. Donanemab

Paper Citations

1. Mielke MM, Hagen CE, Xu J, Chai X, Vemuri P, Lowe VJ, Airey DC, Knopman DS, Roberts RO, Machulda MM, Jack CR Jr, Petersen RC, Dage JL. Plasma phospho-tau181 increases with Alzheimer's disease clinical severity and is associated with tau- and amyloid-positron emission tomography. Alzheimers Dement. 2018 Aug;14(8):989-997. Epub 2018 Apr 5 PubMed.
2. Thijssen EH, La Joie R, Wolf A, Strom A, Wang P, Iaccarino L, Bourakova V, Cobigo Y, Heuer H, Spina S, VandeVrede L, Chai X, Proctor NK, Airey DC, Shcherbinin S, Duggan Evans C, Sims JR, Zetterberg H, Blennow K, Karydas AM, Teunissen CE, Kramer JH, Grinberg LT, Seeley WW, Rosen H, Boeve BF, Miller BL, Rabinovici GD, Dage JL, Rojas JC, Boxer AL, Advancing Research and Treatment for Frontotemporal Lobar Degeneration (ARTFL) investigators. Diagnostic value of plasma phosphorylated tau181 in Alzheimer's disease and frontotemporal lobar degeneration. Nat Med. 2020 Mar;26(3):387-397. Epub 2020 Mar 2 PubMed.
3. Janelidze S, Mattsson N, Palmqvist S, Smith R, Beach TG, Serrano GE, Chai X, Proctor NK, Eichenlaub U, Zetterberg H, Blennow K, Reiman EM, Stomrud E, Dage JL, Hansson O. Plasma P-tau181 in Alzheimer's disease: relationship to other biomarkers, differential diagnosis, neuropathology and longitudinal progression to Alzheimer's dementia. Nat Med. 2020 Mar;26(3):379-386. Epub 2020 Mar 2 PubMed.

Further Reading

No Available Further Reading