This blog focuses on the enhanced applicability of Digital Biomarkers (real time, continuous, data-driven signals) in some of the top therapeutic areas/conditions. Reasons being, companies are increasingly focusing on devising an effective biomarker strategy for their products as well as the data deluge, management, and integration challenges that they are currently facing.
Let us take a step back and look at the chronology of biomarkers:
It is said that ‘cross-talks’ between a person’s genotype and the environment convey useful information about disease causation, progression, and the accompanying phenotypic variations between individuals. Similarly, instead of looking at ‘signals’ from just the specimen/analytics-based molecular and traditional biomarkers (cross-sectional signals), one needs to broaden the field of vision to also include the longitudinal real-time signals generated from digital health-related technologies and tools to get an integrated view.
While remote monitoring (telemetry) has been in use in trials (ECG) for some time now, the increase in penetration of mobility and the advances in IoT technologies (wearables, biosensors, connected devices) have led to an explosion of digital data sources. The volume of data generated from these sources will supplement the insights derived for a particular disease/condition. In certain conditions (e.g. epilepsy), analyzing and correlating this continuous stream of real-world data (RWD) with existing data sources have helped researchers identify changes/perturbations in parameters before the onset of clinical syndromes. This has proved to be of significant value on clinical management and course of the therapy for such conditions.
Patient-facing devices have been the progenitors of digital biomarkers. This is supplemented with advancements in data sciences/data management techniques and tools (collection, integration, analysis using advanced algorithms), and surge in digital health technologies and data troves. Analytics and visualization will be keycapabilities where reviewers would be able to search and select:
- The entire patient profile (from the time a patient embraces digital devices till date- longitudinal assessment)
- Filter based on time periods of interest (time-trend graph)<
- Predict patterns and trends (simulation models) based on the trajectory of available data (and since this is continuous RWD, the reliability/accuracy of the simulated may be high).
- This will lead to constructing better biomarker panels for disease diagnoses, detection, and identification of profiles of interest (research, patient recruitment, etc.)
- Combining lifestyle and behavioral data with omics, clinical, and other conventional macro measurements
How can an integrated panel comprising both the confirmatory as well as the new set of digital markers help?
- Assessing the stage of a disease/condition including the initial detection which might go unnoticed.
- Example - An integrated biomarker panel-based patient selection for recruitment, will definitely lessen the probability of failure (and shorten the time taken for subsequent phases)
- Real-time monitoring of a subject which might not be captured during anamnesis.
- Example - A patient in a clinic setting might not be able to paint the accurate picture which can be better understood from perturbations in the digital markers
- Getting an integrated patient timeline view chart can help ascertain the disease progression/prognosis and treatment benefits.
- The data collected from these digital devices would be in a subject’s home/natural setting, thus further reducing the risk of bias in assessments. Also, since the scale is continuous, the within/ between subject variability and the baseline data per subject can be better ascertained.
Applications of Conventional versus New Age Biomarkers across Top Therapy Areas
So essentially, it is not a choice between or versus traditional and new-age, digitally-identified markers, rather traditional and digital biomarkers supplementing or complementing each other. Especially for conditions which involve multiple clinical manifestations and/or require a combinatorial approach for diagnosis (multiple organ systems and not-so-well-defined clinical observations/symptoms/scales of measurement). This will definitely have a bearing on the way trials are designed and conducted in future.