Derivation and evaluation of signatures using plasma β‐amyloid and pTau‐181 for brain amyloid‐β detection. (20th December 2022)

• Record Type: Journal Article
• Authors: Devanarayan, Viswanath; Dhadda, Shobha; Koyama, Akihiko; Swanson, Chad J; Hampel, Harald; Irizarry, Michael C; Kramer, Lynn D
• Is Part Of: Alzheimer's & dementia. Volume 18(2022)Supplement 5
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-20
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.067811
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Detection of brain amyloid‐b (Ab) pathology using blood‐based tests would be considerably more convenient for patients, in addition to being a faster and more cost‐effective method for screening and monitoring patients in clinical trials. Using patient screening data from internal clinical trials, we derive signatures for brain Ab detection using plasma Ab42, Ab40, and pTau‐181, and evaluate their performance accuracy in an independent dataset. Method: Ab42 and Ab40 were measured using immunoprecipitation coupled with LC‐MS/MS by C2N diagnostics in plasma samples from 513 patients at screening and pTau‐181 was measured in plasma samples from 398 patients (n=273 overlap) using Simoa Advantage V2 assay kit (immunoassay) by Quanterix Corp. Over 90% of these patients had mild cognitive impairment (MCI), and approximately 50% had brain Ab pathology based on Florbetapir‐PET visual read. Signatures using these plasma markers for detecting brain Ab were derived using Bayesian elastic‐net, resampled lasso, and gradient boosting models. Additional signatures were derived for assessing the added value of ApoE4 status, cognitive function scores and brain volumetric measures from magnetic resonance imaging (MRI). Performance of these signatures was assessed via 10‐fold cross‐validation within the training set, and further tested in an independent cohort from ADNI (Alzheimer's disease neuroimaging initiative). Result: Using currently available data, signatures derived using plasma Ab42/Ab40 ratio achieved 76% overall accuracy for detecting brain Ab. Similar accuracy was achieved using plasma pTau‐181 alone, which improved to 78% when including Ab42/Ab40, to 80% when adding ApoE4, and to 82% when adding cognitive scores. Inclusion of MRI brain volumetric measures did not improve the accuracy. When testing in the independent ADNI cohort, 75.7% accuracy was achieved by signatures based on plasma Ab42/Ab40 ratio. Signatures using plasma pTau‐181 alone achieved 75% accuracy, which improved to 78% when combined with Ab42 to Ab40 ratio. Conclusion: These results demonstrate the potential of blood‐based amyloid/tau signatures as promising patient screening tools for detecting brain amyloid burden in clinical trials and patient care. Results reported here will be refined further prior to final presentation using impending additional data.