Alzheimer’s and other neurodegenerative diseases are ultimately responsible for around 100,000 deaths each year in the U.S. alone, according to statistics from the National Institutes of Health. But for the most part, researchers have struggled to come up with simple, reliable ways to catch those diseases early, predict how they’ll unfold, or treat them. As a result, 57 million people around the world—and ten million more diagnosed each year—will still struggle with the memory loss, cognitive challenges, and even personality changes that come with neurodegeneration.
Fortunately, a new effort, the Global Neurodegeneration Proteomics Consortium (GNPC), is working to develop predictors for Alzheimer’s, Parkinson’s, and related conditions. Launched in 2023, the consortium aims to gather and sift through vast amounts of data to find proteins in blood and other biofluids that could signal the beginnings of neurodegenerative disease and perhaps identify new treatments.
Essential to the project is collaboration. GNPC brings together researchers—and protein data—from 23 public and private institutions, including the Iqbal Farrukh and Asad Jamal Stanford Alzheimer’s Disease Research Center, and produces a “harmonized dataset” comprising tens of thousands of samples in one place.
That data is already helping uncover new insights into neurodegenerative disease, said Tony Wyss-Coray, director of the Knight Initiative for Brain Resilience at the Wu Tsai Neurosciences Institute, and a leading member of the consortium. “It’s a paradigm shift,” Wyss-Coray said.
An overview of the GNPC data was published in Nature Medicine July 15 along with a collection of “deep dive” papers by consortium member labs delving into specific aspects of the data. In their companion paper, Wyss-Coray’s team analyzed GNPC data to produce an unprecedented look at how the aging blood-brain barrier alters the balance of proteins present in the human brain with age—a potential factor in the brain’s resilience or vulnerability to age-related cognitive decline and dementia.
Silent and complex
Among the challenges facing doctors looking to diagnose neurodegenerative disease is that there are few indicators it’s coming. Right now, doctors rely on symptoms such as memory loss and confusion—and testing to rule out other conditions. But neurodegeneration typically begins years or even decades before any outward symptoms appear, so that by the time a diagnosis comes, it may be too late to stop, let alone reverse, any underlying damage.
Ideally, Wyss-Coray said, doctors could perform a simple blood test that could, in conjunction with a standard checkup, catch Alzheimer’s and other diseases early enough to head off the worst. “Twenty years ago, most neurologists would not collect blood from their patients, because they wouldn’t see the point—it’s a brain disease, right?”
But Wyss-Coray pursued the idea anyway, and as early as 2007, Wyss-Coray’s lab showed there were molecular signatures of brain health in blood. Still, researchers at the time didn’t have the techniques to make it useful in the real world of healthcare.
Much has changed in the last two decades. Mirroring similar trends in genetics, scientists developed the field of proteomics—essentially, a set of tools to map the concentrations of thousands of proteins in the body in both health and disease. With those tools in hand, researchers could, in theory at least, find proteomic signatures of coming neurodegenerative disease and perhaps identify treatments—but a more practical roadblock remained.
A coalition for precision
The underlying problem is data: To make progress in the field, researchers need a lot of data, and while there were pockets of relevant data spread out around the world, scientists lacked a way to put it all together into one large, accessible dataset. Without that one big dataset, they couldn’t draw the kinds of precise conclusions they’d need to make a clinically useful blood test.
Enter GNPC, which undertook the challenge of gathering clinical and proteomic data from its 23 partner institutions—among them private hospitals, the U.S. Department of Health and Human Services, and, with support from the Knight Initiative, Stanford’s Alzheimer’s Disease Research Center. Because each institution gathered slightly different kinds of data, GNPC researchers then had to “harmonize” or standardize the data so that the different sources were compatible with each other.
That effort has resulted so far in 35,000 biofluid samples—blood plasma or serum or cerebrospinal fluid (CSF)—taken from 19,000 people. That’s finally enough data, all in one place, that researchers can start to uncover meaningful patterns, Wyss-Coray said.
The data have already led Wyss-Coray’s lab to some novel results they couldn’t have achieved otherwise. In particular, the lab studied how premature aging in individual organs—including the brain, but also kidneys, liver, and others—might predict brain health. First, they used data on blood plasma proteins to estimate the “biological age”(as opposed to the chronological age) of the brain, liver, kidneys, and other organs in about 3,000 of the people in the GNPC database.
Then, they compared those biological ages to data on the neurological health of the people in the study, with intriguing results. Among other things, the lab found that prematurely aged arteries, liver, and intestines were linked to an increased risk of Alzheimer’s disease, while prematurely aged muscles were linked to an increased risk of Parkinson’s disease.
Looking deeper
Researchers working as part of the GNPC project have uncovered new links between specific proteins and Alzheimer’s and Parkinson’s disease, as well as links between those proteins and a genetic risk factor for Alzheimer’s called APOE e4. In addition to helping identify potential treatments, results like those could help them better understand the genetic and molecular processes that drive neurodegenerative diseases, the GNPC team wrote in their overview paper.
In their “deep-dive” companion paper, Wyss-Coray and his lab took a closer look at how aging affects the brain’s filtration system, called the blood-brain barrier or brain barrier system, which controls what sorts of molecules can get in and out of the brain. Researchers know that this system degrades as people age, and that a leakier barrier may accompany neurodegeneration.
The trouble is, tools to measure that leakiness are limited, in turn limiting researchers’ understanding of the blood-brain barrier’s role in aging and neurological disease, said first author Amelia Farinas, a graduate student in Wyss-Coray’s lab.
To address that gap, Farinas, Wyss-Coray, and colleagues turned to data on 2,171 people from the GNPC and other sources. For each person and each of 2,304 proteins, they computed the ratio of protein levels in the CSF versus blood plasma—one of the largest-scale investigations yet into how the brain barrier system changes over time in both healthy aging and disease.
Their initial findings suggest notable age-related differences in the selectivity of the brain barrier and the proteins it allows into the brain. In particular, the CSF-to-plasma ratio of about a third of the proteins the team examined increased with age even in healthy people. That suggests that the brain’s filtration system is in fact degrading with age—but it might not be a bad thing. In fact, certain proteins leaking into the brain appeared to have a protective effect against disease.
“Since leakage of blood proteins into the brain is typically considered harmful, it was surprising to find that increased brain levels of some of these proteins may actually be beneficial for cognitive function and brain health,” Farinas said.
Although more work remains to be done, the team writes that their findings will help identify both healthy patterns of brain aging and perhaps identify potential targets for drugs to treat neurodegenerative disease.
More data, more discoveries
Wyss-Coray envisions GNPC as enabling a wide range of new research with even more detailed looks at the data. “These studies create new hypotheses. You start asking, what are these proteins doing? And then you model it in mice, and you may find a new drug target,” Wyss-Coray said.
Gathering more data will bring that concept even closer to reality. “If you can look at 20,000 proteins relatively cheaply” using blood tests and other relatively non-invasive means, “imagine what type of information that provides,” Wyss-Coray said.
Adding more people to the dataset will also help. “If you imagine you have now a million people that you have profiled—and there’s easily a million blood samples out there—then it could lead to personalized medicine,” which until now has been “a pipe dream,” Wyss-Coray said.
“This GNPC initiative shows we can work together, and if we work together, we will make discoveries,” Wyss-Coray said. “You will not make discoveries with a couple hundred samples, but if you put them together and get 40,000 samples—or a million samples—you can start making discoveries.”
