Vanderbilt University Medical Center has been awarded a five-year, $31.7 million grant by the National Institute on Aging, part of the National Institutes of Health (NIH), to harmonize research data gathered on human subjects in scores of disparate studies of Alzheimer’s disease and related dementias (ADRD).
ADRD is studied from various angles, and from one human research cohort to the next the data are collected in different ways and at different scales, with many data points conforming to ad hoc definitions.
Starting with data from more than 30 research cohorts, the new project will pool these data using data harmonization principles that are well established.
This will produce a large-scale, racially diverse, standardized set of transparently defined data that will support machine learning and open new windows into the genetic basis of ADRD and Alzheimer’s resiliency.
“Machine learning and large-scale genomics hold incredible promise for moving the field of ADRD research forward,” said Timothy Hohman, Ph.D., associate professor of Neurology and one of three principal investigators for the project. “The resource we’ll build together stands to deliver the required multimodal data, at the required scale, for the first time.”
Alzheimer’s is a complex disease with a pronounced genetic component, its estimated heritability coming in at 60 to 80%. Research data types encompassed by the project range from clinical information to genomics, cognitive performance, neuroimaging, biomarker data (currently derived from cerebrospinal fluid analysis), and autopsy neuropathology data.
“Our Alzheimer’s disease research team at VUMC is unique in our multimodal focus, which has led to many exciting collaborations to try and integrate diverse data types into our pipelines,” Hohman said. “These collaborations left us well-positioned, when this opportunity came, to pull in the right teams to do the work, including our co-leads at the University of Miami and the University of Southern California.”
Per NIH data sharing policies, the harmonized data will be available to qualified researchers from far and wide, primarily via established, secure computing resources supported by the National Institute on Aging.
The project will dovetail with the NIH’s previously established Alzheimer’s Disease Sequencing Project (ADSP), with sequencing here referring to genomic sequencing.
The new grant establishes the ADSP Phenotype Harmonization Consortium (phenotype: the set of observable characteristics of an individual resulting from the interaction of genotype and the environment). The consortium includes 13 sites located across the country.
VUMC will serve as the primary site, coordinating data activities and working with experts across the country to harmonize cognitive, biomarker, and neuroimaging data. VUMC investigators on the project hail from Neurology, Biostatistics, Electrical Engineering, the Vanderbilt Genetics Institute, and the Vanderbilt Memory and Alzheimer’s Center.
The project’s other two principal investigators are at the University of Southern California, where brain images will be housed, and the University of Miami, where investigators will focus on coordination with ADSP’s diverse workgroups and will collaborate with experts around the country to harmonize autopsy neuropathology data.
According to the Alzheimer’s Association, this year in the U.S., direct medical and long-term care costs for ADRD will reach $355 billion, with cases expected to double by 2050.
Meanwhile, an estimated 30% of people age 65 and older are Alzheimer’s resilient, having all the underlying pathology of Alzheimer’s but never developing any of the clinical symptoms.
The research is supported by the National Institutes of Health (U24AG074855).