Digital Endpoints Must Meet Patient Needs
Multi-Stakeholder Effort Required – One Size Does Not Fit All
Bray Patrick-Lake
Evidation Health
Christine Manta
Elektra Labs
Jennifer Goldsack
Digital Medicine Society
T

echnological advances are speeding the development of digital measures and endpoints throughout drug development and across therapeutic areas and patient populations. While regulators encourage stakeholders to pursue patient-centric approaches, few actionable guidances or standards exist.

Digital measures of health offer enormous promise to transform how we define health and disease, and in turn improve health outcomes, economics, and equity. However, it is expensive and time-consuming to develop digital measures of health for use in healthcare and clinical research. Disjointed and undisciplined digital measure development that takes a signal chasing approach rather than measuring what matters most to patients can exacerbate the inadequacies of poor assessments in use today. It ultimately costs us the early promise of digital measures to improve the lives of patients in meaningful ways.

Technology should not determine our healthcare quality; it should enable it. The following key principles can help us ensure that digital measures of health are meaningful to patients and result in clinical care and treatments that deliver better patient outcomes.

Principle 1: Engage Patients Continuously Throughout Development

Patient engagement in the development and selection of digital measures is not the same as usability testing of the measurement technologies. It is further not accomplished with a single transaction, such as a focus group. Rather, diverse and representative patients and care partners should be intentionally, actively, and continuously integrated in every process for selecting and developing digital measures and endpoints from the start and throughout development. This will likely require sponsors, developers, clinicians, and researchers to reevaluate existing processes and procedures to involve patients as partners. Existing patient engagement principles, guidance, and toolkits are available from multi-stakeholder groups such as Clinical Trials Transformation Initiative (CTTI), Patient-Centered Outcomes Research Institute (PCORI) , Meharry Vanderbilt Alliance, and FDA’s Voice of the Patient Reports.

Principle 2: Search for Meaningfulness

Broadly, patients want to improve their condition, slow down progression, and/or prevent disease onset. A “meaningful” measure of health, whether digital or traditional, is one that helps clinicians and researchers–as well as patients and care partners themselves–detect a change in what matters to patients in their everyday lives. For example, patients want to walk safely and independently each day. They may want to walk to the bus stop, around the yard, or to a grocery store; they do not want to fall.

Developers and researchers can group these desires into a Meaningful Aspect of Health (MAH) category such as the “ability to perform ambulatory activities.” The clinician or researcher then aims to achieve a positive change in this patient-reported measure when evaluating a proposed treatment or intervention. When this information is provided back to patients and their care partners, they can also use this information to better manage their condition.

By comparison, the 6-minute walk test is an existing ambulatory measure in wide use across multiple conditions, from chronic cardiovascular, pulmonary, and central nervous system diseases to post-operative patients in rehabilitation. Criticisms of this measure abound, from clinicians and patients alike. Correlation with clinical endpoints, such as mortality, is inconsistent. It lacks specificity in some patient populations, and it systematically excludes non-ambulatory members of other patient populations, such as Duchenne’s muscular dystrophy patients. Yet, the evaluation is inexpensive and requires little training, so it remains widely adopted by professional associations and as a clinical trial endpoint. A digital version of this sub-par measure would score low on “meaningfulness.”

Patient advocacy groups are leading the way on defining clinically meaningful outcomes for patients to guide industry, clinicians, and other stakeholders. Us Against Alzheimer’s (UsA2), for example, is actively researching and publishing “first-of-their kind findings on outcomes from dementia treatments and therapies that matter most to individuals with Alzheimer’s and those who care for them.” Top-rated outcomes were mostly associated with emotional wellbeing, including, for patients, “taking medications correctly; not feeling down or depressed; able to stay safe (e.g., remember to turn off appliances or running water, not wandering, not being taken advantage of); using the bathroom alone; and not feeling like a burden to others.” Care partners also ranked correct medication administration and depression at the top. Additionally, they want their loved ones to feel less “anxious, worried, or stressed,” feel “like they have a sense of purpose (self-worth),” and be less “irritable, frustrated, or agitated.” UsA2 maintains an online research community of more than 8,000 patients, care partners, and others interested in brain health to support this research and collaboration.

Similarly, 253 adult patients participating in CreakyJoints’ ArthritisPower registry with ankylosing spondylitis, fibromyalgia syndrome, osteoarthritis, osteoporosis, psoriatic arthritis, rheumatoid arthritis, and systemic lupus erythematosus were asked to select between three and 10 patient-reported outcome (PRO) symptom measures they felt were important to digitally track for their condition via the ArthritisPower app. The symptoms that rheumatology patients prioritized for longitudinal tracking using a smartphone app were fatigue, physical function, pain, and morning joint stiffness. PROs ranked as most important for tracking were PROMIS Fatigue, Physical Function, Pain Intensity, Pain Interference, Duration of Morning Joint Stiffness, and Sleep Disturbance. This study lays an important foundation for the selection of digital endpoints of importance to patients.

Principle 3: Be Intentional, One Size Does Not Fit All

Consider the audience of patients and care partners that stand to benefit from a given intervention and seek meaning from diverse members of these individual populations. Many patients desire the “ability to perform hand-dependent activities.” But this desire may mean very different things for different patient populations. A Parkinson’s Disease population might define this ability through a concept of “non-purposeful upper limb movement,” while a rheumatoid arthritis population might define it from a “joint pain intensity” perspective. Determining the appropriate outcome concept for the population then leads to the selection of the meaningful outcome measure in clinical care or the meaningful endpoint in clinical research.

The challenge for digital measures and instruments is to establish trust among stakeholders that the results are reliable and can support regulator review and approval and/or clinical decision-making. Regulators recognize the critical need for consensus-building and collaboration among multiple stakeholders.

For example, a public workshop hosted by the Foundation for the National Institutes of Health (FNIH) Biomarkers Consortium in 2020 explored the application of remote digital monitoring for medical product development. Several use cases were presented that demonstrate the challenges and methodical work required to identify, specify, and develop digital measures and data collection methods. One success story is the 95th centile stride velocity measure (SV95C), which “can be seen as a disease response or pharmacodynamic marker of treatment in ambulatory patients diagnosed with different neuromuscular diseases, including Duchenne’s Muscular Dystrophy.” A wearable sensor attached to a wrist or ankle generated data about movement. The data could then be correlated with the established 6-minute walk test with a high degree of reliability, particularly in natural, everyday settings.

This case study is a good example of early and continuous patient input in the development and use of digital technologies to support clinical development of medical products for rare diseases, with few centers available to conduct studies. Ultimately, the European Medicines Agency’s acceptance of SV95C as a secondary endpoint in pivotal or exploratory drug therapeutic studies for regulatory purposes when measured by a valid and suitable wearable device offers promise for the regulatory submissions that may follow, particularly for other outcome measures related to gait speed or limb movement.

Next Steps

The Digital Medicine Society (DiMe) offers The Playbook and other collaborative resources for “developing & deploying digital clinical measures across clinical research, clinical care, and population health surveillance.” The Playbook describes a detailed four-step framework representing the synthesis of best practices “for selecting and developing measurements in research and care that are meaningful for patients:”

  1. Meaningful Aspect of Health (MAH)
  2. Concept of Interest (COI)
  3. Outcome to be measured
  4. Endpoint (exclusive to research)

Associations, such as DIA (Drug Information Association) and the Critical Path Institute, offer virtual and in-person events for knowledge-sharing and networking, such as DIA’s Digital Technology in Clinical Trials virtual event in October 2021. The Critical Path Institute offers multiple tools and events related to developing and applying Patient Reported Outcome measures.

Conclusion

As digital health innovation accelerates, we have the opportunity and obligation to build a digital future of health measurement that effectively and efficiently serves patients, regulators, and clinical decision makers. Digital measures of health require significant investments of time and resources through selection, development, and validation. The time is now to partner with diverse patients to systematically develop reliable, meaningful measures and then to incorporate these measures in clinical care and clinical research.
References available upon request. For any questions, please contact Bray Patrick-Lake, bpatricklake@evidation.com.