GAP President John Dwyer spoke to Alzforum about the Bio-Hermes study.
With disease-modifying treatments for Alzheimer’s disease potentially on the cusp of approval, there will be pressing demand for quick, simple ways to screen patients and determine if they have amyloid plaque in their brains. The current gold-standard diagnostics used in research—amyloid PET scans and cerebrospinal fluid biomarkers—are expensive, invasive, or not readily available to many patients. With that in mind, the Global Alzheimer’s Platform has launched Bio-Hermes, a collaborative study with 10 different technology companies that will assess how accurately low-cost, rapid blood-based, and digital biomarker tests detect amyloid accumulation and cognitive decline.
- Bio-Hermes study will assess digital and blood-based biomarkers.
- Results will be compared to amyloid PET and traditional cognitive tests.
- The goal is to find rapid, low-cost biomarkers for trial and clinical use.
Besides informing clinical practice, GAP researchers hope that quick and accurate screening methods will speed clinical trial recruitment and reduce the cost of screen failures. Alzheimer’s disease trials typically have the highest screen fail rates and longest recruitment times of any disease. The Bio-Hermes study will enroll 1,000 people at different stages of Alzheimer’s, with a particular emphasis on recruiting diverse populations. All data will be made freely available to the public.
John Dwyer at GAP believes the time is right for this because of advances in blood-based and digital markers. “The science around this has improved so much in the last couple of years,” he told Alzforum.
“It is a fantastic initiative, and great to see that so many companies are collaborating to move the science forward,” said Oskar Hansson at Lund University, Sweden. Hansson believes the biggest contribution from Bio-Hermes may be the data on new digital and behavior analysis tests, which have not been studied in existing cohorts like ADNI and AIBL. Other studies exist to compare fluid biomarkers, but not digital and behavior markers.
In one such ongoing effort, called NeuroToolKit, Swedish academics collaborated with Roche to develop a standard panel of CSF biomarkers that are used in the same way in numerous clinical cohorts (Dec 2019 conference news; Milà-Alomà et al., 2020; Van Hulle et al., 2021). Additional biomarkers are added to this panel as they emerge from basic research and small studies, to be compared to the established markers.
Bio-Hermes, named for the Greek messenger god, will compare a diverse array of biomarker tests developed by its partner companies. These include several blood tests, with Quanterix providing a ptau181 assessment, AbbVie providing ptau231, and Eli Lilly, ptau217. The study will also compare plasma A?42/A?40 measures from both Lilly and C2N Diagnostics, and will examine plasma NfL as well.
In the digital testing realm, Cognivue will contribute a cognitive test, Aural Analytics a speech assessment, and Linus Health its digital clock test (Rentz et al., 2021). Linus Health is also providing several other assessments, including measures of gait, eye movement, reaction time, and word recall. Canadian company Retispec will test its retinal scanning technology. Other technology partners are Merck and the U.K. neuroimaging firm IXICO. The trial is funded by the Alzheimer’s Drug Discovery Foundation.
Researchers will administer the tests over the course of three study visits to a clinic—no continuous monitoring via wearables. They will compare the results to standard cognitive and functional tests—MMSE, FAQ, RVLT, GDS—as well as to florbetapir amyloid PET. Scans will be read at a central facility to reduce site-to-site variability.
The trial will recruit 400 cognitively healthy participants, 300 with mild cognitive impairment, and 300 with mild AD. At least 20 percent will be African American or Latino. Diana Kerwin at Kerwin Medical Center, Dallas, the study’s primary investigator, noted that this will help ensure that the findings apply to the general population. Some evidence suggests that biomarker cutoffs may vary by race (Jan 2019 news; Apr 2021 news). Trial sites will be located in areas that enable outreach to a diverse population. So far, the study includes nine sites in Florida, Illinois, Kansas, New York, and Texas, but more sites may be added.
Participants also consent to full genotyping. These data will allow researchers to determine if biomarker cutoffs are different in people who carry specific risk factors.
Because the various tests assess such different modalities, combining them could provide new information, noted David Bates at Linus Health. “We believe that by combining tests we could add specificity and be able to distinguish between Alzheimer’s disease, frontotemporal dementia, and dementia with Lewy bodies,” he told Alzforum.
Dwyer expects primary data collection to wrap up in about a year, with results available to the public around the end of 2023. Data will be accessible on the Alzheimer’s Disease Data Initiative website.
Originally posted by Alzforum on April 24, 2021.