Palmqvist S, Janelidze S, Stomrud E, Zetterberg H, Karl J, Zink K, Bittner T, Mattsson N, Eichenlaub U, Blennow K, Hansson O. Performance of Fully Automated Plasma Assays as Screening Tests for Alzheimer Disease-Related β-Amyloid Status. JAMA Neurol. 2019 Jun 24; PubMed.

AlzBiomarker Database

Meta-Analysis

Curated Study Data

Biomarker
(Source)		 Cohort
(N)		 Measurement
Mean ± SD		 Method;
Assay Name;
Manufacturer		 Diagnostic
Criteria
Aβ40
(CSF)		 AD
(94)		 17.9 ± 6.4
ng/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 1984; 1993
Aβ40
(CSF)		 CTRL-
CNC
(34)		 18.3 ± 6.7
ng/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics
Aβ40
(Plasma)		 AD
(94)		 0.437 ± 0.106
ng/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 1984; 1993
Aβ40
(Plasma)		 CTRL-
CNC
(34)		 0.439 ± 0.102
ng/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics
Aβ42
(CSF)		 AD
(94)		 672 ± 335
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 1984; 1993
Aβ42
(CSF)		 CTRL-
CNC
(34)		 1133 ± 410
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics
Aβ42
(Plasma)		 AD
(94)		 26.1 ± 6.5
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 1984; 1993
Aβ42
(Plasma)		 CTRL-
CNC
(34)		 30.1 ± 6.5
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics
NFL
(CSF)		 AD
(64)		 2002 ± 1835
pg/mL		 ELISA;
NF-light;
UmanDiagnostics AB		 McKhann et al., 2011
NFL
(CSF)		 CTRL-
CNC
(366)		 918 ± 490
pg/mL		 ELISA;
NF-light;
UmanDiagnostics AB
tau-p181
(CSF)		 AD
(64)		 36.3 ± 16.3
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 2011
tau-p181
(CSF)		 CTRL-
CNC
(366)		 17.5 ± 5.3
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics
tau-total
(CSF)		 AD
(64)		 384 ± 143
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 2011
tau-total
(CSF)		 AD
(94)		 365 ± 159
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 1984; 1993
tau-total
(CSF)		 CTRL-
CNC
(34)		 230 ± 113
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics
tau-total
(CSF)		 CTRL-
CNC
(366)		 209 ± 62
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics
tau-total
(Plasma)		 AD
(64)		 16.7 ± 6
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 2011
tau-total
(Plasma)		 AD
(94)		 15.3 ± 4.5
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics		 McKhann et al., 1984; 1993
tau-total
(Plasma)		 CTRL-
CNC
(34)		 13.8 ± 4
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics
tau-total
(Plasma)		 CTRL-
CNC
(366)		 16.6 ± 4.7
pg/mL		 Fully Automated Immunoassay;
Elecsys;
Roche Diagnostics

Recommends

Please login to recommend the paper.

Comments

  1. Drawing closer: Alzheimer’s blood test for primary care

    This study of Palmqvist et al. showed with fully automated immunoassays that the plasma Aβ1-42/Aβ1-40 ratio can predict amyloid plaque load (amyloidopathy) in the brain (BioFinder study). In addition, at the upcoming AAIC meeting, Inge Verberk from Amsterdam University Medical Center will present a follow-up study of a previously published Simoa immunoassay approach (Verberk et al., 2018) using newly developed AMYBLOOD Simoa assays. She will report comparable clinical performance using a different patient cohort and technology. The Simoa assays to be presented by Charlotte Teunissen's group at Amsterdam UMC are developed in close collaboration with ADx NeuroSciences.

    The potential value of the plasma Aβ42/Aβ40 ratio has been increasingly recognized over the last several years by:

    • Performing studies focused on a specific context of use (~Aβ-PET imaging);
    • The availability of a reference method (FDA-approved Aβ-PET imaging with visual interpretation);
    • Better selection and characterization of subjects for inclusion in the study (CSF biomarker profile, Aβ-PET imaging);
    • Use of larger sample sizes and integration of qualification, followed by validation cohorts;
    • Use of technology platforms with better precision (automation, random access) and lower analytical sensitivity.

    Before a plasma Aβ immunoassay can be used to rule out the need for a costly Aβ-PET scan, the test should achieve a high sensitivity (>85 percent) and a high negative predictive value for a specific clinical context, for instance for pharma trials.

    Suggestions for the use of the plasma Aβ42/Aβ40 ratio were already published more than a decade ago based on classical ELISAs and supported recently by the more labor-intensive mass-spectrometry technology (Nakamura et al., 2018; Ovod et al., 2017). However, with their current design, it seems immunoassays cannot reach the same diagnostic accuracy for Aβ1-42/Aβ1-40 as mass spectrometry, pointing to the need to (i) extend the algorithm for blood testing by integration of other proteins or protein isoforms, (ii) have a better understanding of the characteristics of monoclonal antibodies that are used in the assay design, (iii) improve analytical performance of immunoassays, and (iv) generate standard operating procedures for collection and storage of blood samples.

    Several obvious plasma biomarkers (e.g. Neurofilament Light, tau, BACE1 protein, YKL-40) (Vergallo et al., 2019) are not able to fill that clinical accuracy gap between mass spectrometry-based studies and immunoassay data (Feb 2019 news). However, ongoing longitudinal studies in (pre)clinical study cohorts, such as from the SCIENCe project, the pre-insight AD cohort (Verberk et al., 2018; Vergallo et al., 2019), or the AIBL study cohort, can potentially confirm and validate that a simple plasma test might help to identify a stage of AD before MCI and thus might aid in the setup of new clinical and prevention trials.

    It is important that biology and assay performance are more closely linked to each other. Precision Qualified Assays (PQAs) will provide a solution in the future by combining clearly defined analytical performance requirements of an assay with the observed effects in patients (see also Biomarker Qualification: Evidentiary Framework). This requires a high(er) level of standardization of biomarker assays than done now. Not only are extensive standardizations needed at the level of the lab, the sample, and the assay, but also the biological variation or biological differences emerging for a specific context of use need to be taken into account.

    There is a need for: 

    • More detailed reporting of study results, including characteristics (e.g. selectivity, specificity) of antibodies used in each assay (which protein isoform is detected?);
    • Documented performance characteristics of assays and validation approaches (see also Points to Consider Document: Scientific and Regulatory Considerations for the Analytical Validation of Assays Used in the Qualification of Biomarkers in Biological Matrices); 
    • A better understanding of confounding factors that have an impact on Aβ concentrations in samples (e.g. exercise, BMI, medication) or a better understanding of the mechanism of clearance of Aβ in peripheral fluids;
    • Comparison of clinical performance when using other Aβ isoforms, including but not limited to Aβ34, Aβ-3, … or using other proteins (e.g. total tau, NfL);
    • Open sharing of results obtained in the larger patient cohorts with different assay formats;
    • Standard operating procedures for sample collection and storage. Since Aβ1-42 in plasma is not as stable as in the CSF, more critical review of the sample collection and storage protocol is required. A specific sample type (e.g. EDTA, plasma, serum) can give you an identical or a different outcome;
    • Harmonization and generation of certified reference materials. The field will gain by initiation of harmonization of plasma measurements by production of certified reference materials using plasma samples. The round-robin study on 70 EDTA-plasma samples, organized by Professor Kaj Blennow at the University of Gothenburg, Sweden, and including different technology platforms, will improve our current understanding of how changes in assay design might affect the result. The latter study is more designed to verify correlations in concentration between analytes and is less focused on clinical application of the assays. It is a very good first step in bringing the results of the plasma studies to another level. Such insights are essential and will further streamline efforts to use the different biomarkers in different stages of evaluation of prevention and treatment options. The field will progress faster if the learnings from past experiences in CSF will be integrated in the workflow. All stakeholders will have to work more closely together in a shorter period of time to make the assays available for implementation in a routine clinical environment;
    • Integration of technology platforms with better multiplexing possibilities, low sensitivity, and more easily upscaled for possible worldwide use.

    References:

    Nakamura A, Kaneko N, Villemagne VL, Kato T, Doecke J, Doré V, Fowler C, Li QX, Martins R, Rowe C, Tomita T, Matsuzaki K, Ishii K, Ishii K, Arahata Y, Iwamoto S, Ito K, Tanaka K, Masters CL, Yanagisawa K. High performance plasma amyloid-β biomarkers for Alzheimer's disease. Nature. 2018 Feb 8;554(7691):249-254. Epub 2018 Jan 31 PubMed.

    Ovod V, Ramsey KN, Mawuenyega KG, Bollinger JG, Hicks T, Schneider T, Sullivan M, Paumier K, Holtzman DM, Morris JC, Benzinger T, Fagan AM, Patterson BW, Bateman RJ. Amyloid β concentrations and stable isotope labeling kinetics of human plasma specific to central nervous system amyloidosis. Alzheimers Dement. 2017 Aug;13(8):841-849. Epub 2017 Jul 19 PubMed.

    Vergallo A, Mégret L, Lista S, Cavedo E, Zetterberg H, Blennow K, Vanmechelen E, De Vos A, Habert MO, Potier MC, Dubois B, Neri C, Hampel H, INSIGHT-preAD study group, Alzheimer Precision Medicine Initiative (APMI). Plasma amyloid β 40/42 ratio predicts cerebral amyloidosis in cognitively normal individuals at risk for Alzheimer's disease. Alzheimers Dement. 2019 Jun;15(6):764-775. Epub 2019 May 18 PubMed.

    Verberk IM, Slot RE, Verfaillie SC, Heijst H, Prins ND, van Berckel BN, Scheltens P, Teunissen CE, van der Flier WM. Plasma Amyloid as Prescreener for the Earliest Alzheimer Pathological Changes. Ann Neurol. 2018 Nov;84(5):648-658. Epub 2018 Oct 4 PubMed.

    View all comments by Eugeen Vanmechelen

Make a Comment

To make a comment you must login or register.

This paper appears in the following:

News

  1. Drawing Closer: Alzheimer’s Blood Test for Primary Care
  2. Why Bother With Round Robins on Blood Tests? Q&A with Kaj Blennow

AlzBiomarker

  1. AD vs CTRL : Aβ40 (CSF)
  2. AD vs CTRL : Aβ40 (Plasma and Serum)
  3. CTRL vs AD : Aβ40 (CSF)
  4. CTRL vs AD : Aβ40 (Plasma and Serum)
  5. AD vs CTRL : Aβ42 (CSF)
  6. AD vs CTRL : Aβ42 (Plasma and Serum)
  7. CTRL vs AD : Aβ42 (CSF)
  8. CTRL vs AD : Aβ42 (Plasma and Serum)
  9. AD vs CTRL : NFL (CSF)
  10. CTRL vs AD : NFL (CSF)
  11. AD vs CTRL : tau-p181 (CSF)
  12. CTRL vs AD : tau-p181 (CSF)
  13. AD vs CTRL : tau-total (CSF)
  14. AD vs CTRL : tau-total (Plasma and Serum)
  15. CTRL vs AD : tau-total (CSF)
  16. CTRL vs AD : tau-total (Plasma and Serum)