Brian Murphy, CSO of UK digital health company Cumulus Neuroscience, explores how AI tools could transform drug development in the difficult area of cognitive disorders
Cognitive disorders such as dementia are often poorly understood and effective treatment methods are lacking. This is in part because of the difficulties in measuring neurodegeneration, made worse by the fact that many neurodegenerative diseases develop long before symptoms appear and the underlying disease mechanisms are not well understood. In the meantime, doctors evaluating a patient often only see a single snapshot of that person’s symptoms on a good day or bad for them, making diagnosis even more difficult.
This presents a number of challenges for early intervention and drug development for these diseases. When combined with all of these factors, there have unfortunately been few groundbreaking innovations in this area for decades, despite Biogen’s recent approval of Aduhelm.
As a result, there was less chance of a financial return on developing potential treatments – for example, Alzheimer’s drug candidates have one of the highest failure rates of any disease area at 99.6%, which further discourages research and development.
But, given the huge unmet need and size of commercial opportunity, the pharmaceutical industry is still very active in this area, trying to find treatments that can be transformative for patients at any stage of cognitive disorder. To do this, we need to better understand these conditions at the patient level – not just in the clinic, but also on a large scale and in real-life environments.
AI-driven technologies for cognitive health care
Researchers increasingly believe that the key to solving these problems lies in artificial intelligence (AI) technologies. With AI, it becomes much easier to measure the subtle but highly relevant cognitive and behavioral patterns that indicate early neurodegeneration.
At Cumulus Neuroscience, we worked with eight of the world’s largest pharmaceutical companies to select the five behavioral and physiological areas they see most important for cognitive research; We then merged advanced tools for each of these areas into an integrated physiological and digital biomarker platform that provides a comprehensive overview of patient cognitive health.
With a combination of AI algorithms and sophisticated assessment tools, our integrated platform can measure cognitive health much more accurately than ever in a real-world environment and improve the speed and success of clinical trials for new cognitive therapies.
Let’s take speech recognition as an example. Advanced technologies, such as the language analysis tools built into our platform, can now assess subtle changes in language, grammar, and acoustic properties of language that may indicate neurodegeneration. For example, a patient who uses more pronouns, fewer nouns, and more high-frequency words in their language may show signs of semantic impairment with emptier, more vague, and more unspecific language. Some of these changes are incredibly difficult for humans to distinguish – such as acoustic abnormalities such as longer pauses between words and sentences – and this is where AI can play an important role.
The brain is also very adaptable and can mask early damage by finding workarounds, which means that a person’s memory capacity or other cognitive performance can appear unaffected. Because of this, it’s important to look at the underlying brain activity as well. Cumulus has the unique ability to record electroencephalography (EEG or “brain waves”) in patients’ homes while performing playful roles in decision making, learning and emotional intelligence. This rich, multimodal, longitudinal sample of patient performance and brain activity produces large complex data sets that are perfect for advanced AI techniques.
All of the assessment technologies in our platform have been carefully designed for low patient exposure and frequent, repeated use, with remote monitoring and integration with standard tablet devices.
With a technology platform like ours, it is now more than ever possible for pharmaceutical companies to often perform comprehensive cognitive assessments with real-world data across multiple diseases, populations, symptoms and mechanisms of action. Just as large genome databases pioneered the understanding of genetic disorders, having frequent, objective records of how patients’ cognitive conditions change will be invaluable in the future development of new dementia therapies.
With global impact
To really have a significant global impact on cognitive diseases, we need tools that work for everyone, everywhere – and that means we need to focus on affordability, scalability, and big data analytics powered by AI.
Combined with a cloud platform, automated methods, ease of use, and inexpensive, scalable, manufacturable hardware, we are uniquely positioned to capture the data needed to answer the most valuable questions in cognitive research. Cumulus’ platform has been extensively field tested by hundreds of users for thousands of hours in patient and control populations.
The plethora of well-publicized clinical trial failures in cognitive disorders like dementia often suggests that there is little hope for future treatments – but some of the world’s brightest minds focus their efforts on drug discovery in this area, which remains one of the most greatest opportunities in terms of unmet need.
Thanks to the increasing digital transformation, we can now assess drug candidates better and more promptly earlier in development. The ultimate goal of Cumulus is to provide an industry standard metric for registering cognitive impairment; This will allow us to significantly improve the robustness of clinical trials and, hopefully, help offer important new therapies to patients suffering from these life-changing diseases.