Last week researchers released the first results from the UK Biobank Imaging Study, a massive effort that ultimately aims to scan the brains of 100,000 people and use the data in conjunction with detailed health information to investigate disease progression during aging. The findings from their first 5,000 subjects offer an early peek at an enormous data set that includes a treasure trove of health information from magnetic resonance imaging (MRI) scans and other measures. The study is one of several projects worldwide taking a population-level approach to better understand diseases, and is part of an ongoing movement in neuroscience toward global, collaborative brain research.
For many of the diseases that afflict us as we age, from diabetes to dementia, we have no real way of knowing that they are on their way until symptoms begin to reveal themselves. It is likely, however, that there are subtle signs in our bodies much earlier. That’s where the UK Biobank Imaging Study comes in. “The idea is, we’re characterizing people in great detail before they actually have begun to accumulate the health problems that people have in old age,” explains Karla Miller, lead author of the study. “What we’re aiming to do is get a glimpse of the various kinds of markers that we might be able to identify that then presage problems later on.”
This detailed characterization includes extensive surveys about demographics and lifestyle factors, gene expression analysis and in-depth medical tests. Subjects ranging in age from 40 to 69 also undergo six different kinds of MRI scans that capture brain anatomy, microstructure, function and connectivity, along with scans of their other organ systems. Scans are done at three dedicated imaging centers distributed across the U.K. The initial results were published last week in Nature Neuroscience. “To a large degree what we were trying to do in this paper was not to ask any really specific question,” Miller says, “but to demonstrate to people, here's the power of this amazing and very unique data resource”
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
The value of a data set 100,000 strong is that the participants will inevitably develop a wide array of diseases as they age, which researchers will be able to track through the U.K.’s National Health Service. They can then examine disease progression in the context of other health information recorded in the study, potentially revealing important trends, connections or early biomarkers for disease.
The first batch of data contains tens of thousands of significant correlations between various health measures. For example, the preliminary results revealed a link between increased alcohol consumption and signs of injury to brain connections, along with another link between tobacco intake and imaging signals associated with increased iron deposits in the brain.
Miller points out, however, that simple correlations, even highly significant ones, cannot provide a complete picture of an individual’s health. “A lot of the time, if what you find is that you have a correlation between connectivity in one brain region and a specific aspect of an individual's lifestyle, it's hard to know what to make of that,” she says.
Miller and her colleagues confronted this challenge by using trends consistent across the population in the brain-imaging data as well as in information about health and lifestyle to create composite measures that incorporate multiple factors simultaneously. They found a number of independent patterns of aging that, according to Miller, “are described by aspects of an individual's lifestyle [and] also have an imprint on the brain.”
The researchers are hopeful that further unraveling such complex relationships will eventually allow them to use health measures from the study to predict who will go on to develop a particular disease or who may respond well to a specific intervention.
David Van Essen, principal investigator on the Human Connectome Project (which used MRI scans to create a highly detailed human brain map, released earlier this year), is enthusiastic about the study but echoes Miller’s hesitancy about reading too much into these early results. “They have presented an intriguing potpourri of observations and correlations that pique one's interest,” says Van Essen, who was not involved in the research, “but they have also very thoughtfully included much needed caveats and warnings about the complexities of the data, the complexities of the relationships that we want to decipher, and the issues and potential concerns about overinterpretation or misinterpretation.”
Miller hopes to synthesize the UK Biobank’s research efforts with other ongoing population-scale brain imaging studies such as the Maastricht Study in the Netherlands as well as the Rhineland and German National Cohort (GNC) studies, and even fine-scale brain studies like the Human Connectome Project.
Wolfgang Ahrens, science director of the GNC Association, says that pooling population studies to further increase the number of subjects will be particularly valuable for studying rare diseases such as testicular cancer or amyotrophic lateral sclerosis. He calls the UK Biobank Imaging Study “the ignition” for the GNC, a testament to the increasingly collaborative nature of brain research. “This kind of research now is international,” he says.
The imaging study was designed with exactly this idea in mind: Like the Human Connectome Project, the Allen Brain Atlas and others, its entire data set is completely open-access. Scientists anywhere can access it for their own research, whether for direct analysis or designing follow-up studies.
The UK Biobank Imaging study is “setting the stage for exploring a wide range of brain-related characteristics to behavior and characteristics that will be predictive of early markers of disease,” Van Essen says. “In essence, we’re looking now at the tip of a large iceberg of important data.”