Precision Medicine News

Using Predictive, Data Analytics to Expand Knowledge of Alzheimer’s

Brigham and Women’s Hospital researchers are investigating the cognitive decline in individuals due to Alzheimer’s disease with predictive and data analytics.

predictive analytics data analytics

Source: Getty Images

By Erin McNemar, MPA

- Using predictive and data analytics, Brigham and Women’s Hospital researchers developed a resource to better understand the cognitive decline in individuals due to Alzheimer’s disease (AD).

The team created induced pluripotent stem cell (iPSC) lines from over 50 different individuals with longitudinal clinical data, quantitative neuropathology data, and rich genetic and molecular profiling of brain tissue available.

Researchers then demonstrated the use of this resource through a series of studies that turned humans stem cells into brain stem cells. The team then analyzed molecular pathways active in the living neurons and identified specific forms of amyloid beta-protein (Aβ) and tau protein associated with cognitive decline and Alzheimer’s disease.

“We are finding that different genetic backgrounds in humans generate different profiles of Aβ and tau,” Tracy Young-Pearse, PhD, of the Brigham’s Division of Neurology, said in a press release.

“Those stem cell-derived neuronal profiles have a predictive nature, in terms of the clinical outcome of an individual’s Alzheimer’s disease. This large set of human cell lines from AD and cognitively normal people provide the scientific community with a powerful experimental system for untangling why some people develop AD and others do not."

READ MORE: App Uses Artificial Intelligence Algorithm to Predict Alzheimer’s

The study featured a cross-institutional collaboration that followed a cohort of people from the Religious Orders Study (ROS) and the Rush Memory and Aging Project (MAP) at Rush University. Individuals that entered the ROS and MAP studies had no history of Alzheimer’s or any other neurological diagnosis, but over one-third of them developed Alzheimer’s over time.

Stem cell lines were generated from blood samples of over 50 participants, who lived to an average of 90 years old. While the subjects were still alive, detailed clinical records and full genome sequencing were collected from each person. After their death, their brains were studied and compared to their cultured brain cells.

The teams measured Aβ and tau generated by the stem cell-derived neurons and discovered that certain Aβ and tau species were linked to levels of plaque and tangle deposition in the brain as well as cognitive decline.

“If you look across cell culture samples from 50 people, you can predict from the Aβ and tau profiles some features of the cognitive status of that person: their rate of cognitive decline and whether they developed AD, which is absolutely remarkable,” Young-Pearse said.

The authors recognized that since ROS and MAP participants were primarily White, their observations may not be applicable to diverse populations. However, the research team has begun the development of 50 new iPSC lines that will be derived from people of color.

READ MORE: Using SDOH Data to Boost Population Health Management in Alzheimer’s

“This is a resource for the whole community, for scientists around the world to use,” said Young-Pearse. “We already have around 50 labs, all around the world, starting to use these cells in their studies.”

Additionally, the researchers also investigated if the stem cell-derived neurons could be used to predict whether drugs like aducanumab, a new Alzheimer’s drug recently approved by the Food and Drug Administration, will be more effective in specific groups of Alzheimer’s patients.

“The new Alzheimer’s drug is an antibody that recognizes Aβ and clears it from the brain. Our work shows that different genetic backgrounds of human neurons produce different profiles of Aβ,” said Young-Pearse.

“What our system provides is a platform to test who might be responsive to different AD therapeutics, for example, anti-Aβ and anti-tau immunotherapeutics. This will be important because neurons from different individuals produce different profiles of Aβ and tau, and different antibody cocktails may be more effective against one profile over another.”

The research team wants to convey to the public that Alzheimer’s is not typically the result of a single genetic mutation. Instead, different sets of genes contribute to risk across different people. The primary causes of Alzheimer’s can be varied; therefore, a personalized medicine approach is necessary for treating the disease.

READ MORE: Deep Learning, Genomic Data May Help Predict Alzheimer’s Disease

“There’s a perception that this disease is one uniform disease that shows up and progresses the same way for everyone,” Young-Pearse said.

“It’s important to understand that Alzheimer’s is actually quite heterogeneous in its underlying causes, ages of onset and disease course. This is the first time we have a system in place to study living human brain cells from many people to understand better why some develop AD in a very specific way and others are resistant to the disease.”