Digital endpoints widely adopted in pharmaceutical and biotech-sponsored clinical trials

Over 130 pharma and biotech organizations have relied on AI-powered digital biomarkers and sensor-derived clinical outcome assessments (COAs) to capture patient-centered endpoints.

In recent years momentum in drug development strategy has shifted toward digital biomarkers powered by artificial intelligence (AI) and machine learning (ML). More than 130 pharmaceutical and biotechnology organizations capture clinical trial endpoints using connected, sensor-based digital health technologies (DHTs) that use AI/ML to process and analyze the pre-processed data. At the same time, regulatory support is increasing: the Food and Drug Administration (FDA) recently published a paper on AI for drug development that outlines current and future applications, such as using AI to aid in synthesizing multimodal data from several DHTs in a trial.

Why is there such focus on deploying digital endpoints in clinical trials? Well, as obvious as it might sound, clinical trials for drugs that are meant for everyday use by everyday humans haven’t always actually had humans at the center of their attention.

Connected sensor based DHTs capturing digital endpoints provide a variety of advantages, including:

More comprehensive, patient-centered datasets — compared to existing approaches that capture point-in-time data at clinical visits, researchers can collect more representative data from connected sensors that monitor study participants continuously throughout their days
Increased diversity, inclusion, and accessibility — connected sensors make it easier for people to participate in clinical trials, since they don’t require regular visits to research centers
Reduced time and cost — phase cycle times and trial costs are driven down by factors like reduced screen failure rates, fewer protocol amendments, and lower cost of connected sensors compared to traditional instruments in clinical trials

ICON’s Atlas platform provides insights on AI-powered biomarkers and sensor-derived COAs. Part of the platform tracks clinical trials conducted across pharmaceutical, academic, and technology manufacturer organizations. For this analysis, we derived data from ClinicalTrials.gov of trials sponsored by pharmaceutical and biotechnology organizations between 2008 and 2022. The analysis focused on records that reported a manufacturer and/or model of the connected sensor based DHTs deployed. The findings reveal adoption of digital endpoints across trial phase, endpoint position, and therapeutic area.

Specifically, Phase 2 and Phase 4 trials are most represented among pharma-sponsored trials. Phase 2 trials are conducted to determine initial effectiveness of a therapy in humans and may be most prevalent because sponsors introduce DHTs at this point before moving into pivotal Phase 3 trials. Meanwhile, Phase 4 trials are used to evaluate long-term safety and benefits and are conducive to collecting real-world data using DHTs. As more trials are completed and adoption grows even further, we expect an increase in digital endpoints among Phase 3 trials that are submitted to regulatory agencies to confirm efficacy of a therapy.

Pharma most commonly deploys DHTs to capture secondary endpoints that support or extend the understanding of the primary aim of the trial. Still, more than 25 percent of the 1300+ endpoints in our analysis were primary endpoints. Collectively, this indicates that digital endpoints are being used to inform drug development across the pharma industry, signaling a shift toward more patient-centered clinical trials.

Digital endpoints in endocrinology, neurology, and cardiology have garnered the most attention from pharma sponsors to date. DHT use in these areas has historically been dominated by connected sensors such as continuous glucose monitors (CGMs) in endocrine diseases and wearable electrocardiogram (ECG) patches in cardiology. Emerging technology and priorities could change that balance in the coming years — for example, increased adoption of AI-powered vocal biomarkers in psychiatric disorders and smartphone-based quality of life COAs in oncology.

Go deeper on industry trends and best practices

ICON’s Atlas platform enables insights into industry trends across a variety of evidence sources, including peer-reviewed publications, regulatory documentation, and clinical trials spanning pharmaceutical organizations, academic institutions, technology manufacturers, and more. The data in this post were solely derived from ClinicalTrials.gov and show that pharma sponsors are increasing their trust in AI-powered digital biomarkers and sensor-based COAs to capture primary and secondary endpoints. With solutions from ICON, your team can identify even more industry trends to support upcoming clinical trials.

Learn how ICON's solutions can expedite evidence-based decision making at your organization.

Over 130 pharma and biotech organizations have relied on AI-powered digital biomarkers and sensor-derived clinical outcome assessments (COAs) to capture patient-centered endpoints.

Go deeper on industry trends and best practices

Blog authored by Max Gaitan, Outcomes Researcher, ICON. Originally written for HumanFirst, a company acquired by Icon in january 2024.

In this section

Connect with us