Some of the largest drugmakers tackling Alzheimer’s disease and its related dementias are banding together to ensure their pursuits make the best use of the truly massive amounts of digital data that can be gathered from patients as the companies look to track their progress.
In a collaboration being led by the nonprofit Digital Medicine Society, Big Pharma players such as Biogen, Eisai, Eli Lilly and Merck plan to establish a core set of digital measurements that can be applied in their clinical trials. They will be joined by researchers from Boston University and Oregon Health & Science University as well as the Alzheimer's Drug Discovery Foundation.
The group aims to select and develop the most effective tech-enabled biomarkers—built on data collected from wearable devices or tests that may involve analyzing voice recordings or tracing hand movements—that can help determine whether prospective treatments are actually working to slow the cognitive declines associated with neurodegenerative conditions.
By agreeing collectively to quantify the course of the disease in a standardized and patient-centered way—and at a point before the real competition begins—the drug developers ultimately hope to increase each of their chances of getting treatments through the regulatory approval process and to patients faster.
But before they do, they’ll have to work their way through a smorgasbord of digital data types, sensor readings and cognitive tests.
“It's frankly alarming to see the lack of industry consensus around what ‘good’ looks like when it comes to selecting and deploying a measure,” Jennifer Goldsack, CEO of the Digital Medicine Society, said in an interview.
The society currently maintains a library of the digital endpoints that have been applied in various industry-sponsored studies: 302 different measurements employed by over 90 companies at last count, spanning everything from changes in a six-minute walking test to the number of recorded coughs per week to a person’s ability to rhythmically tap their finger on a smartphone screen.
“There’s over 70 different measures of physical activity, and there’s over 50 different measures of sleep,” Goldsack said. “The sad truth that led us to this particular project is that despite the enormous unmet need for high-resolution, high-value measures that can be captured frequently over time, only three of the endpoints in the library are being used for Alzheimer’s.
“There’s an enormous amount of work going on, but we’re not seeing them being selected and deployed in one of the therapeutic areas where we believe there is the greatest ability to do good work,” she said.
In addition, very little progress has been made in understanding what are the most meaningful aspects of health outside of simply cognition.
“Digital measures offer enormous promise to bolster collective understanding of Alzheimer’s disease and related dementias,” Jian Yang, Eli Lilly’s associate vice president of digital health, said in a statement. “These measures can deepen our knowledge of disease progression, define new disease phenotypes and support earlier diagnosis—all critically important insights for a condition where inaccurate and delayed diagnosis is common and current standard of care does not target the underlying plaque pathology.”
But at the same time, certain digital measurements have failed to perform accurately when applied across a diverse population. With its biopharma and academic partners, the society will look to set down metrics that can be used consistently and internationally.
“It's kind of like a wild wild west in science right now,” said Rhoda Au, Ph.D., a professor of anatomy and neurobiology at Boston University. “In the digital realm, there's not a lot of precedent and there’s no standardization.”
Beyond clinical trials, Au hopes this collaboration and the efficient use of large datasets can help set the stage for a precision medicine attitude toward Alzheimer’s disease, instead of a one-size-fits-all approach—something that may have success in defining clinical subtypes such as what’s been shown in cancer while also making pre-competitive data easily accessible between developers like in the fight against COVID.
“Right now we do not use cognitive testing correctly in any clinical research study,” Au said in an interview. “Our cognitive capabilities are complex and vacillate from moment to moment. In the afternoon you can have a different cognitive level than you do in the morning, versus the evening and in how you think from day to day.”
“If you think about all those fluctuations and all the factors that influence your cognitive capabilities—we just boil it down to a test score,” she said. “We keep using blunt instrument tools to measure something that’s far more dynamic and far more rich.”
Over recent years, the idea of using complex digital biomarkers has evolved from a mere curiosity to a potentially valuable asset within a drug development program, as more biopharma companies have begun exploring how to wield different technologies.
“We need to make the same investments in capital and time that we've done for other biomarkers,” Au said. “If we think about how much work has been put into developing PET amyloid tracers and tau tracers, and now blood-based biomarkers, we literally need the same concerted effort around digital biomarkers.”
The Digital Medicine Society aims to announce a core set of digital clinical measures within the next two years, after completing literature reviews, mixed methods studies and technology benchmarking. Currently, the society estimates that Alzheimer’s disease and its related dementias affect 47 million people worldwide, with the number of cases expected to double every 20 years.
“This is something that is moving incredibly quickly. I do not want to wait until there are 40 different digital measures being used before we offer this up,” said Goldsack.
“We have an obligation to the science, and we also have an ethical obligation—I don't want patients around the world being lost in validation studies of measures that don't matter.”