How Digital Biomarkers Enable Patient-Centric Clinical Trials
Digital biomarkers have the potential to transform clinical trials with objective and continuous real-time data.
Complete the form below to unlock access to ALL audio articles.
From smartphones to wearables, digital devices have become a key aspect of everyday life. These technologies can collect an array of data about a person, such as their fitness level and mood. Harnessing these devices in clinical trials to capture digital biomarkers has the potential to enable more adaptive, patient-centric trials.
The increasing amount and variety of data collected by digital devices, alongside advancements in the miniaturization of electronics, have created new possibilities for integrating digital biomarkers into clinical trials.1 These measures have several advantages over traditional clinical outcome assessments and can be useful in trials where accurate and timely data collection is crucial for assessing investigational treatments.
What are digital biomarkers?
Digital biomarkers are objective, quantifiable indicators of health, disease or treatment response collected and measured by digital devices such as smartphones, wearables and implants. Examples of digital biomarkers include heart rate variability from a smart watch and recorded speech patterns to track mood.2, 3
“Digital biomarkers can be integrated into clinical trials in several ways: as exploratory endpoints to understand physiological correlates of disease activity or treatment response, as stratification tools to identify high-risk participants or even as surrogate endpoints of treatment response,” Dr. Robert Hirten, associate professor of medicine, artificial intelligence and human health at the Icahn School of Medicine at Mount Sinai, told Technology Networks.
Digital biomarkers can be used in tandem with traditional trial measures, such as self-reports or clinical interviews. This enables the collection of objective, continuous data reflecting participants' real-world experiences, which complements standard trial endpoints. “This can result in a more precise assessment of how a treatment affects daily functioning and behavior,” Dr. Nick Jacobson, associate professor in biomedical data science, psychiatry and computer science at the Geisel School of Medicine at Dartmouth, told Technology Networks.
“It's possible that digital biomarkers could detect treatment responses or subtle changes more quickly or sensitively than conventional methods, potentially leading to more efficient trial designs,” Jacobson continued. “They also provide valuable information on how a treatment impacts everyday life outside the clinic setting.”
Digital biomarkers of mental wellbeing and gut health
Many areas of medicine could benefit from the implementation of digital biomarkers. In clinical trials for neurological disorders, cognitive and motor functions are traditionally assessed with specific tests, clinical interviews or through self-reporting of symptoms. Digital biomarkers can provide a non-invasive and objective measurement that more accurately represents how a patient is responding to a treatment.4
Smartphones and wearables offer significant advantages to the field of mental health. “They allow for continuous, objective monitoring of behavior and physiology using passive sensors like accelerometers and GPS right in people's daily lives,” explained Jacobson. This data stream allows for the development of digital biomarkers that “can assist in predicting and detecting mental health conditions such as anxiety and depression earlier than might otherwise be possible.”
Beyond clinical trials, wearables and sensors that collect digital biomarkers offer a range of benefits for monitoring and treating mental health conditions.
“For monitoring, these tools provide a highly detailed, real-time view of symptom fluctuations, offering insights that are difficult to capture through infrequent clinic visits or memory-based self-reports,” Jacobson said. “When it comes to treatment, this continuous data stream supports the creation of personalized, just-in-time adaptive interventions. Such systems can identify moments of increased need or risk and deliver targeted support right when it could be most helpful.”
Jacobson and colleagues have developed a generative AI therapy chatbot called Therabot. This is designed to provide real-time support for the many people who lack regular or immediate access to a mental health professional.5
“Digital tools, including smartphone applications and AI-driven platforms like the Therabot system developed in my lab, can deliver evidence-based assessment and therapeutic interventions at scale. This reaches individuals who might otherwise struggle to access traditional care due to cost, location or stigma,” stated Jacobson.
The use of wearables to collect digital biomarkers has applications for a range of diseases beyond mental health conditions. “Wearable technology offers a potentially new approach to managing IBD [inflammatory bowel disease] by enabling continuous, non-invasive monitoring of patients in their daily lives,” explained Hirten.
In a recent study, Hirten and colleagues evaluated how several physiological metrics are associated with IBD flares.6 They found that circadian patterns of heart rate variability identify such inflammatory instances. In addition, changes in these metrics can identify and precede flares of IBD by up to seven weeks.
“Although further research is needed, our hope is that this can give patients and their doctors a critical window to intervene early,” said Hirten. “Beyond flare prediction, wearables hold promise in being able to monitor treatment response, identifying ongoing symptoms and inflammation – all outside the traditional confines of the clinic.”
Challenges in integrating digital biomarkers in clinical trials
As outlined in the examples above, the potential of digital biomarkers to advance many aspects of human health is immense. However, there are still challenges that need to be overcome for them to see more regular use in clinical trials.
One of the primary concerns is around data privacy, given the sensitive nature of the data collected. This is in addition to other challenges that still need to be evaluated, “Including ensuring data quality, validating digital measures against gold standards and navigating evolving regulatory frameworks,” said Hirten.
AI and machine learning have the potential to help overcome some of these challenges by improving data quality and integrity.
“AI and machine learning are central to extracting meaningful insights from the high-volume data generated by wearables. These tools can identify subtle patterns and temporal trends that are not discernible through traditional analysis,” Hirten explained.
Proteomics: Tools, Techniques and Translational Potential
Proteomics continues to evolve rapidly, offering powerful tools to unravel complex biological systems. This eBook explores key methodologies and innovations reshaping proteomics, from foundational techniques to AI-enhanced data analysis.View eBook
By learning an individual’s unique physiological baseline and deviations from it, AI can enhance the accuracy and clinical relevance of data collected using wearable devices.
“AI also drives personalization; my research, supported by the National Institute of Mental Health, indicates that deep learning models tailored to an individual's unique data streams often provide the most accurate predictions of symptom changes, paving the way for highly personalized mental healthcare,” stated Jacobson.
“AI can also help by integrating data from multiple sources, like sensors and self-reports, to build a more complete picture of an individual's mental health.”
Given the data security and regulatory challenges that come with integrating digital biomarkers in clinical trials, Hirten recommends that “A thoughtful, methodologically rigorous approach is essential to unlock their full potential.”
Future outlooks for digital biomarkers in clinical trials
Digital biomarkers offer substantial value in clinical trials by providing objective, continuous and real-time data on participants. “In the future, I believe digital biomarkers will become integral to the design and execution of clinical trials. They will enable more adaptive, patient-centric trials—where real-world data informs eligibility, dosing and endpoints in real time,” said Hirten.
Continued advancements in AI and machine learning are set to further enhance the accuracy of digital biomarkers. Moreover, digital biomarkers could improve patient engagement by integrating other digital technologies such as remote monitoring and AI chatbots.7 “We’ll also see digital biomarkers supporting broader inclusion by potentially reducing the need for frequent site visits, which is especially valuable for participants in remote or underserved areas,” Hirten concluded.
“As regulatory standards evolve and validation frameworks mature, digital biomarkers may transition from exploratory tools to accepted clinical endpoints, helping us bring more precise and timely therapies to patients with chronic diseases.”
