NEW YORK – The Global Alzheimer's Platform (GAP) has completed a first-of-its-kind cross-sectional study comparing the performance of Alzheimer's diagnostic assays. Top-line results are expected in July of next year, with a second, longitudinal study to be completed by the end of 2024.
The researchers conducting the study also plan to eventually bolster their initial findings with genomic and proteomic profiles of study participants in an attempt to identify new biomarkers or multifaceted biomarker panels to improve Alzheimer's diagnosis.
The Bio-Hermes study recruited approximately 1,000 people and tested the performance of a variety of blood-based and digital Alzheimer's diagnostic assays against both each other and the "gold standard" assays of positron emission tomography (PET) and cerebrospinal fluid (CSF) taken via lumbar puncture, both of which largely scan for the tau and beta amyloid proteins implicated in Alzheimer's characteristic plaques.
Both gold standard assays have certain drawbacks that drive the need for more biomarker tests.
Beta amyloid plaque PET images, for instance, are both very expensive and not ubiquitously available in the United States, limiting their accessibility to all patients.
CSF collection, meanwhile, is physically demanding and contraindicated in cases of hypertension, which is more prevalent among Latino and African American populations, leaving it applied unevenly across ethnic backgrounds.
In response to these drawbacks, and to advance Alzheimer's diagnostic abilities more generally, a variety of blood-based and digital biomarker assays have been developed, such as C2N's PrecivityAD assay and Diadem's AlzoSure Predict. Until now, however, these tools had not been systematically compared to each other and to the gold standard tests.
"We wanted to be one of the first ones to be able to offer the field data on who performed better under what circumstances," said John Dwyer, president of the GAP Foundation.
Dwyer also noted that because of the healthcare inequity inherent in lumbar CSF collection, this test was used as a comparator in only a "small subset" of cases.
Carlos Cruchaga, a professor of psychiatry at Washington University who studies the biology of neurodegenerative disease, said that "biomarker cutoffs can [vary] depending on sex and race," and that therefore, systematically evaluating the performance of Alzheimer's tests will help move the field toward more personalized care.
One study, for example, found that the reliance on disproportionately White research populations has resulted in biomarkers that perform inconsistently across different ethnic backgrounds, raising the odds of misdiagnosis in non-White patients.
Addressing that disparity, approximately 22 percent of the Bio-Hermes study population were African American or Latino, with a further 2 percent of Asian or Native American ancestry.
The GAP Foundation selected study sites that committed to outreach in those communities and also partnered with social media influencers to further boost the study's visibility, while keeping enrollment open until reaching the target 20 percent representation from African American and Latino participants.
"It's the most diverse [Alzheimer's] biomarker study in history, based on our review of the literature," Dwyer said.
Blood tests in the study were compared to both PET scans and to each other for their sensitivity, specificity, and overall accuracy in detecting beta amyloid, as measured by correlations to each other and to beta amyloid PET images. Digital tests for cognitive impairment were compared to more traditional paper-and-pencil cognitive tests.
The GAP Foundation expects to have top-line results ready for the next Alzheimer's Association International Conference in July of next year.
"We expect to have top-line data lock around April 30," Dwyer said, at which point "all partners will have access to the data and be able to start cranking on the analytic process."
The results of the study will likely be of great interest to numerous test developers in the growing Alzheimer's diagnostics space. In addition to Diadem and C2N, for instance, Biogen, Fujirebio, and Roche all have assays on the market.
In addition to the head-to-head comparisons, the GAP Foundation and its partners plan to accompany diagnostic test results with genomic and proteomic profiles of study participants.
"We're going to get the whole genome and put it right up against every participant in the study," Dwyer said, "so that as you get these blood tests, you're also looking at who responded to the blood test as positive and what was their genetic or genomic profile," and whether and how those profiles correlate with diagnostic test results.
SomaLogic and the lab of Allan Levey at Emory University will conduct independent proteomic analyses of all 1,000 study participants, while Eli Lilly will develop genomic profiles.
SomaLogic said via email that it is working under a fee-for-service contract with the GAP Foundation and has already completed 400 samples using its SomaScan platform, which can measure up to 7,000 proteins from about 55 µl of blood.
Levey's lab is using both the SomaScan and mass spec methods to look for plasma-based biomarkers that can add value to existing biomarkers, such as through predicting disease progression and/or adding to the sensitivity and specificity of other existing biomarkers. His work with the Bio-Hermes study builds from previous research using post-mortem brain samples to understand how the proteome changes across the Alzheimer's disease spectrum.
"The vast majority of people with [Alzheimer's] symptoms have mixed pathologies," he said. "Our strategy is to really broaden the perspective beyond amyloid, tau, and neurofibrillary tangles."
Lilly has been working to develop both Alzheimer's diagnostic tests and therapies. The company recently inked an $11 million deal with Quanterix to further develop Alzheimer's assays, initially focusing on the p-tau217 protein, a blood-based biomarker that has shown some encouraging signs as a possible early detection marker.
Cruchaga said that most proteomic and genomic studies of Alzheimer's currently focus on APOE, a major genetic risk factor in Alzheimer's. While genetics may likely provide an individual's "background risk" for Alzheimer's, Cruchaga explained, "there is the need to have other more dynamic markers like proteins or transcript levels."
Several studies, he said, are currently seeking to leverage genomics for Alzheimer's biomarkers although these are largely in early stages.
"But they are promising," Cruchaga said, "and [they] can be scaled up very quickly."
Scalability, Cruchaga explained, could prove a key factor in determining which tests gain market space. Those based on mass spectrometry, for instance, frequently provide high degrees of accuracy but can be challenging to scale up.
Dwyer said, however, that while the GAP Foundation is sensitive to a test's scalability and plans to include some discussion of that, along with each test's current availability, in the study results, scalability was not a determining factor in choosing which tests to include.
"You have to [assess] accuracy first," he said, explaining that it was more important to focus on metrics such as sensitivity and specificity first, then worry about how scalable a test is — or can be made — later.
The GAP Foundation hopes to soon launch Bio-Hermes Two, which will evaluate test performances longitudinally, complementing the cross-sectional design of the current study. Although there are few specifics yet, the second study will include tests for tau protein and the foundation hopes to complete it by approximately December 2024.