National Academies Press: OpenBook

The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop (2020)

Chapter: 4 Digital Health Technologies for Recruitment and Safety

« Previous: 3 Digital Health Technologies for Characterizing Disease
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

4

Digital Health Technologies for Recruitment and Safety

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

During the second session of the workshop, the role of digital health technologies (DHTs) in recruitment and safety trials was explored. Christopher Leptak, director of the Regulatory Science Program in the Office of New Drugs at the Center for Drug Evaluation and Research at the U.S. Food and Drug Administration (FDA), provided an overview of how FDA uses drug development tools (DDTs) as part of its drug development programs and discussed how DHTs may offer an opportunity to inform the development of DDTs. He also described FDA’s regulatory approach to defining and determining whether tools are fit-for-purpose as well as the conceptual framework for biomarker acceptance. Yvonne Yu-Feng Chan, senior director of medical affairs for digital medicine at Otsuka Pharmaceutical, explored the use of DHTs for engaging the public in

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

research. She described a large-scale decentralized trial as a case example of how to modernize clinical trials and explained how the study enabled the acceleration, democratization, and standardization of certain research methods. Chris Benko, chief executive officer at Koneksa Health, discussed the roles of DHTs and remote monitoring as part of the drug development process during early-stage clinical trials; he shared findings from the company’s validation studies to evaluate the use of remote technologies for assessments in early-stage clinical studies. He also suggested some potential uses for digital technologies to address the coronavirus disease 2019 (COVID-19) pandemic. Eric Perakslis, Rubenstein Fellow at Duke University, described how layers of data and communication overlay the interaction of clinical care and research. He presented his vision for the structure of a telehealth-based learning health system. The session was moderated by Deven McGraw, chief regulatory officer at the Ciitizen Corporation.

REGULATORY PERSPECTIVE ON DRUG DEVELOPMENT TOOLS

Christopher Leptak, Director, Regulatory Science Program, Office of New Drugs, Center for Drug Evaluation and Research, U.S. Food and Drug Administration

Leptak explained that FDA’s DDT program1 currently operates under a statute that was included in the 21st Century Cures Act2 at the end of 2016 (see Box 4-1). The statute broadly defines a DDT as any material, method, or measure that aids in drug development regulatory review, as determined by the Secretary of Health and Human Services, and calls out two specific types of DDTs: biomarkers and clinical outcome assessments (COAs). In addition to playing a role in the collection of information for biomarkers and COAs, he said, DHTs themselves can serve as standalone, independent tools. In fact, FDA is beginning to explore a pathway for DHTs to lay the groundwork for regulatory discussions, Leptak said.

Components of a Drug Development Tool

Although the definition of a DDT in the statute is beneficial for innovation and flexibility, its breadth can make it difficult to provide generalizable advice or describe a specific approach. To clarify the concept of

___________________

1 For more information on the Drug Development Tool Qualification Programs, see https://www.fda.gov/drugs/development-approval-process-drugs/drug-development-tool-ddt-qualification-programs (accessed June 7, 2020).

2 The full text of the 21st Century Cures Act is available at https://www.congress.gov/bill/114th-congress/house-bill/34 (accessed May 17, 2020).

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

a DDT, Leptak highlighted three tangible components: (1) the concept of value to the drug development context, (2) the measurement of concept or how information is gathered, and (3) the interpretation of concept. In the case of a biomarker, value would be a defined biologic response, an assessment of a physiologic organ function, or a finding on a radiography assessment, Leptak said. Measurement of concept is the domain in which many proposals for DDTs can contribute information. For example, for a disorder that affects movement ability (e.g., muscular dystrophy), the concept of interest is how a patient moves. DDTs, such as sensor arrays or other means of assessing movement, could be used to measure this concept, perhaps in real time in a patient’s home environment. From a regulatory point of view, he said, the most important component is the interpretation of the concept. Beyond simply collecting data, a proposal for a DDT should indicate whether it contributes to improvement for an individual’s daily life—specifically, if the DDT is beneficial to the extent that the change is substantial and of personal value to people, or if the change is so small that it does not improve people’s lives to a large extent. In some respects, the interpretation of concept is subjective and is largely based on what patients perceive to be a meaningful improve-

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

ment. Another consideration for novel DDTs pertains to which parts of the proposal are new and which already exist or could be repurposed: (1) an existing concept with existing measurement, (2) an existing concept with new measurement, (3) a new concept with existing measurement, or (4) a new concept with new measurement.

Integrating Drug Development Tools

DDTs can come to FDA through several pathways, Leptak said. These pathways do not exist in isolation, and in many cases parallel efforts are under way within or between pathways (Daniel et al., 2016). All of the pathways “share common core concepts, are data-driven, and involve regulatory assessment and outcomes based on the available data” (FDA, 2018). One pathway is a direct submission to FDA as part of a pharmaceutical company’s investigational new drug (IND) application. Through this approach, the company can bring forward technological ideas and negotiate with subject matter experts in FDA’s clinical division about the utility of the DDT and how it might be used in a clinical trial setting. Another pathway is scientific community consensus, typically through publication in scientific journals or consensus statements put forth by professional societies. This approach can be useful for hypothesis generation, but in many cases it does not make primary data as readily available to FDA as the IND pathway does. Consequently, the DDTs that come through this pathway do not tend to be as “regulatory ready.” The third pathway is through DDT qualification programs, through which tools are developed independently of a drug program that—if successful—can be used in drug programs. This process generally involves presenting the data to FDA for rereview, he added.

Regulatory Perspective on Digital Health Technologies as Biomarkers

For FDA, a biomarker is a defined characteristic that is measured as an indicator of normal or pathogenic biological processes or as a response to an intervention, Leptak said. In contrast to a COA, a biomarker is not a clinical assessment of how a patient feels, functions, or survives. Biomarker considerations include the reproducibility of data, the adequacy of the analytic device to assess a biomarker’s reliability, and the feasibility of the biomarker should a drug be approved—that is, whether the analytic will be widely available and suitable for integration into clinical practice paradigms.3

___________________

3 The National Institutes of Health–FDA Biomarker Working Group has published a glossary of terminology and uses of biomarkers and endpoints in basic biomedical research,

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

Determining Biomarker Fit-for-Purpose

Biomarkers need to be matched with a specific drug development goal in a way that is supported by data, Leptak said. For example, susceptibility and risk biomarkers are fit-for-purpose for monitoring changes in a person’s normal, non-diseased physiology.4 There is much variability in this concept of normalcy, from variation in a specific person over the course of a day or a lifetime to variation between patients with different characteristics. Understanding this normal variability helps in the assessment of whether a change from normal is beneficial to a given person.

Pathologic changes in the body over time can be used to develop clinical findings about the symptoms of diseases. A different set of biomarkers—diagnostic, monitoring, and prognostic—are fit-for-purpose here. Once a therapeutic intervention is initiated, then pharmacodynamic, predictive, and safety biomarkers are fit-for-purpose for monitoring changes in a person’s physiology. The aim of therapy is to slow or stop the progression of the disabling characteristics of a disease or, in a best-case scenario, reverse the progression to a more normal physiological state. At this point, a response biomarker might be an endpoint in a clinical trial. A small subset of biomarkers that are predictive of clinical benefits might become surrogate endpoints, he added.

Digital Health Technologies as Biomarkers from a Regulatory Perspective

A common question is whether certain DHTs might be considered biomarkers by default or whether they are considered to be another type of drug development tool in addition to COAs and biomarkers, Leptak said. The field is struggling with the use of the term “digital biomarkers,” which took hold early on but may not adequately capture the regulatory distinction between a concept of interest and how it is measured. There are many different types of biomarkers that, in and of themselves, may either be the source material of the biomarker or how the biomarker is measured. In most cases, proposed digital tools are methods of measurement or data collection, and, as such, the tool itself is not the biomarker. It is the concept of interest of the tool that is the primary concern for regulators. Although DHTs for data collection are essential, they do not typically constitute a biomarker from a regulatory perspective. Benko added that

___________________

medical product development, and clinical care. This resource can be found at http://www.ncbi.nlm.nih.gov/books/NBK326791 (accessed May 10, 2020).

4 The concept of “fit-for-purpose” in this context refers to the regulatory acceptability of using a specific tool for a specific purpose in drug development. For more information on FDA’s Fit-for-Purpose Initiative, see https://www.fda.gov/drugs/development-approval-process-drugs/drug-development-tools-fit-purpose-initiative (accessed May 20, 2020).

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

in the experimental space, biomarkers are often developed to satisfy the requirements of large sponsors that have vast amounts of resources at their disposal to make their decisions. Although they may be conscious of FDA’s guidance, these organizations tend to be more concerned with commercializing a DDT or making a case concerning registration. Often, they have a more informal engagement with FDA through type B and type C meetings to explain the work.5 Adhering to the right methods is important, Benko said, but in many cases this work does not necessarily need to be oriented around the more well-known validation frameworks.

Conceptual Framework for Regulatory Acceptance of Digital Health Technologies in Biomarker Development

Leptak discussed a conceptual framework for biomarker development for regulatory acceptance (see Figure 4-1). Within the framework, the process of proposing a novel tool or technology begins with a need statement that is independent of the tool itself. The need statement should specify how current drug development is stymied due to current challenges or barriers—for example, the heterogeneity of the patient population or the lack of patients who have a rare disease. It should also clearly express the targeted need that the tool intends to address and how the tool will help to address it. The next step is to establish the context of use for the tool, such as its outcome in a clinical trial, its benefit for patients of a certain subtype, or its contribution to the better management of safety signals. Many safety concerns relate to off-target effects, so the context of use for some tools might be differentiating the possibility of those effects. Subsequent stages in the framework are to evaluate the benefits and risks to patients compared with the status quo, which informs the stringency of the evidentiary criteria that will be required to gain regulatory approval.

Improving Drug Development Through the Use of Drug Development Tools

Several components of a DDT contribute to the success of drug development and approval, Leptak said. These include the DDT itself, how the DDT is measured, the targeted patient population for which the DDT is indicated to have value, and other elements of the clinical trial design, such as the input. Any of these elements can lead to failure, he said, so it

___________________

5 Type B (e.g., pre-IND meeting) and C meetings (including anything outside of the purview of Type A and Type B meetings) are informal meetings that occur between a sponsor or applicant and FDA staff. A Type A meeting is a formal meeting immediately necessary for a stalled drug development program to proceed.

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Image
FIGURE 4-1 Conceptual framework for biomarker development for regulatory acceptance.
NOTE: COU = context of use; EC = evidentiary criteria.
SOURCES: As presented by Christopher Leptak, March 24, 2020. Originally from FNIH, 2016.

is important to optimize as appropriate and feasible. Designing a clinical trial that introduces a novel tool or technology requires consideration of the current scientific understanding and how the tool would improve on it. This in turn depends in large part on how the science is understood. Because science—especially biology—is subjective in many respects, he said, it is helpful for a trial designer to carefully consider the assumptions that are involved and how the current state of the science is being interpreted. Devoting time to those conversations at the outset can be useful because it allows for learning from negative results in situations in which the trial design or data collection process does not go as predicted.

Addressing Unmet Drug Development Needs

Proposals for DDTs typically include an explication of the unmet drug development need that will be met by the tool, Leptak said. This may include an overview of the current approach used in drug development for the intended population that highlights the challenges and limitations of this approach. Examples of the types of unmet needs the DDT could address include the need to apply new technology or knowledge providing measures of disease severity; the lack of treatments for a specific condition for which a new diagnostic tool could aid in patient identification; the lack of a system for characterizing subtypes of a condition that may exhibit different responses to the same therapy; or identification of toxicity resulting from exposure to an investigational drug. Generally, the proposal will also include descriptions of the nature, severity, and prevalence of the dis-

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

ease or condition and other characteristics of the target population as well as any other justifications for the need to be addressed. It is also useful to describe the added value that the DDT could provide to the current drug development and regulatory review processes and how it might address any other potential public health benefit. There are several safety biomarkers in development within FDA’s biomarker program, he said, referring participants to FDA’s biomarker qualification submissions website (FDA, 2020c). Digital technology may be beneficial for some of them, although not all of those tools will necessarily be novel. The DDT qualification program provides many resources online to support this process, he added.6

ENGAGING THE PUBLIC IN RESEARCH USING MOBILE HEALTH

Yvonne Yu-Feng Chan, Senior Director, Medical Affairs for Digital Medicine, Otsuka Pharmaceutical

To explore how DHTs can help accelerate and democratize research, Chan described experiences and lessons learned while she and her team at the Icahn School of Medicine at Mount Sinai were involved in conducting the Asthma Mobile Health Study, a large-scale decentralized trial for which she was the principal investigator (Chan et al., 2018).

Overcoming Research Barriers to Digital Health Technologies

Clinical studies have been conducted for centuries in a way that is inaccessible for many members of the public, Chan said. Mount Sinai, where Chan previously worked, was one of the five original launch partners for Apple’s ResearchKit,7 a framework that helped address this barrier of bringing research to the masses. The initial pilot for the Asthma Mobile Health Study used the iPhone iOS platform because it was the first such technology available to allow anyone interested in participating to

___________________

6 Resources include a list of qualified biomarkers, available at https://www.fda.gov/Drugs/DevelopmentApprovalProcess/DrugDevelopmentToolsQualificationProgram/BiomarkerQualificationProgram/ucm535383.htm; biomarker qualification submissions, available at https://www.fda.gov/Drugs/DevelopmentApprovalProcess/DrugDevelopmentToolsQualificationProgram/BiomarkerQualificationProgram/ucm535881.htm; a table of surrogate endpoints, available at https://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/ucm613636.htm; a list of qualified COAs, available at https://www.fda.gov/Drugs/DevelopmentApprovalProcess/DrugDevelopmentToolsQualificationProgram/ucm450689.htm; and COA qualification submissions, available at https://www.fda.gov/Drugs/DevelopmentApprovalProcess/DrugDevelopmentToolsQualificationProgram/ucm625989.htm. All resources accessed May 17, 2020.

7 Apple’s ResearchKit is an open-source framework designed to help researchers and developers create apps for medical research. More information is available at https://www.apple.com/researchkit (accessed May 11, 2020).

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

download the app. Participants could register for the study if they met the study criteria, which included having doctor-diagnosed asthma and using prescription medicines; the study had no controls. Despite the relatively stringent recruitment criteria, around 3,000 patients were recruited within 3 days of the launch, and this grew to around 10,000 participants recruited from Ireland, the United Kingdom, and the United States within about 1 year. This is indicative of the feasibility and promise for this type of mobile health recruitment paradigm, she said. This type of strategy is able to surmount historical barriers to participation in research studies, such as geography, work–life challenges, and psychosocial factors.

A common criticism of research strategies using DHTs is that only a single platform is prioritized, which can exclude some participants, Chan said; offering the app through iOS and Android platforms when feasible can help to overcome some of those generalizability issues. Despite using a single platform, the Asthma Mobile Health Study was able to reach patients with severe baseline disease—around 13 percent of participants had a history of intubation, a marker for severe disease. More traditional methodologies would be less likely to reach those patients or others who would be less likely to participate in a study, such as people living outside of large academic hubs and in rural areas. In Chan’s study, 90 percent of the participants recruited lived outside of the New York City metro area.

When conducting a study that exclusively uses a smartphone to collect multidimensional data, Chan said, it is important to demonstrate the feasibility of the approach and to evaluate the validity and usefulness of the data collected. Just because a technology is new does not mean that it intrinsically has value, she said, so she and her colleagues rigorously assessed the various types of patient-generated and patient-reported information that were being collected. To evaluate the quality and validity of the study data, Chan and her colleagues assessed if data collected had similar intervariable correlation as has already been established in the medical literature. For example, it had already been established that men with good baseline asthma control who are tall should have higher peak flow relative to other subgroups of the patient population. Demonstrating these types of correlations in the study data was helpful in demonstrating the value of the study data, she said.

Chan outlined some of the different types of data that can be collected “in the wild” using DHT. In addition to electronic patient-reported outcomes, it is possible to collect geolocation data, environmental data, and data from connected devices. An example of a sub-analysis performed using the study data illustrates the importance of designing studies that appropriately obtain consent from patients that is broad enough to ensure that data collected could be used in a future sub-analysis and for other purposes. For example, the study team was able to analyze data from participants who lived in the affected areas during the 2015 wildfires in

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

Washington State. Information such as environmental data, patients’ self-reported triggers, and patients’ clinical status enabled the researchers to perform analyses to better understand how the disaster had affected the study participants.

DHTs can also make it possible to integrate datasets that have historically been siloed and separated into a common platform, Chan said. In terms of data analysis, mobile health allows researchers to collect prospective, granular data that can facilitate time-series analyses, cluster analyses, and the discrimination of patient subtypes. The current capacity to categorize patients, she said, is relatively crude. Using mobile health data to refine patient subtypes for specific conditions could help lay the groundwork for more personalized treatments.

Digital Health Technologies to Promote Recruitment and Retention

DHTs can promote recruitment and retention in research studies, Chan said. For example, the Asthma Mobile Health Study benefited from the involvement of Apple ResearchKit and its introduction at a large Apple event. Most traditional research efforts do not enjoy such advantages, she said. Some of the traditional methods of recruitment have moderate effectiveness, but a promising new approach is to use social media in recruitment strategies. Different approaches can be used to target and engage specific types of patients, and digital platforms can also be used to reach out to patients more effectively than traditional methods of conducting research via postal mail or phone calls—forms of communication that are no longer an integral part of many patients’ lifestyles today. Relying on those traditional methods can yield a study population that is even less representative. Understanding how to reach customers and patients is an evolving process, and a useful principle is to strive to be where your patients are, she said.

“Retention is the Achilles’ heel of mobile health,” Chan said. To help promote better patient retention, she suggested encouraging health care providers to endorse or advocate for the use of digital tools as well as creating communities or other types of platforms to foster connections among study participants when possible or appropriate. DHTs will never obviate the need for human contact in the clinical and research realms, she said; however, a strategy for combining the two elements might involve letting digital technology do most of the “heavy lifting” of more mundane tasks and using the scarcer resource—human contact—on a strategic and periodic basis. Ritu Kapur, head of biomarkers at Verily Life Sciences, agreed that human connection is an important element of recruitment, retention, and engagement with digital technologies but is often difficult to scale. Learning from the individuals who participate in research about what elements help them feel connected can help inform future technology development, she added.

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

DIGITAL HEALTH TECHNOLOGIES AND REMOTE MONITORING IN DRUG DEVELOPMENT

Chris Benko, Chief Executive Officer, Koneksa Health

DHTs and remote monitoring could be used in early-stage clinical studies to address some of the barriers encountered in traditional clinical study protocols, Benko said. His organization, Koneksa Health, focuses on the development and implementation of digital biomarkers for patient-centric assessments of novel investigational products. Koneksa Health primarily works on products not yet fully proven in terms of safety and efficacy.

Use of Clinical Pharmacology Units in Early-Stage Clinical Trials

Early-stage clinical trials in many therapeutic areas—excluding oncology—are conducted among healthy volunteers with the aims of establishing the pharmacokinetic, pharmacodynamic, and safety profiles of an investigational drug. The healthy volunteers are typically confined to a clinical pharmacology unit (CPU) while the pharmacokinetic, pharmacodynamic, and safety data are collected. The CPU is a protected health care setting that allows volunteers to be evaluated and monitored for any adverse safety effects. The duration of confinement depends on the anticipated safety profile of the drug being investigated.

There are some drawbacks to relying on CPUs, Benko said, and it can be one of the most expensive and rate-limiting components of the drug development process. Furthermore, confinement in the CPU for extended periods of time is inconvenient for study participants, and it may not provide data that are reflective of normal day-to-day activities. Because the drugs are unproven and potentially risky at this point in the development process, it is important to limit the volunteers’ exposure to the drug for no longer than is necessary to answer the research question. Furthermore, little or no safety information about the drug—other than the participant’s memory recall—is available after the participant is discharged from the CPU and in between follow-up visits. This can make it difficult to interpret potential safety findings, he added.

Digital Health Technologies for Assessments in Early-Stage Clinical Trials

Given the disadvantages associated with the reliance on CPUs in early-stage clinical trials, there is growing interest in the potential role of remote technologies for the assessment of pharmacokinetics, pharmaco-

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

dynamics, and the safety of new drugs, Benko said. Two of the sponsors of Koneksa Health, Merck & Co., Inc., and Takeda Pharmaceuticals, have expressed interest in the logistical potential to gather more data without necessarily confining study participants in CPUs and to potentially develop deeper phenotypes or better baselines to understand normal human variability by gathering continuous data outside the clinic. The latter type of data could shed light on factors that may be affected by participants’ real-world activities in a way that could not be captured from participants living in a controlled setting.

The COVID-19 pandemic has given rise to a new set of challenges specific to phase I drug development that are likely to cause substantial disruption to that stage of the process, Benko said. Traditionally, phase I studies require bringing people into health care facilities, which can compromise social distancing and introduce other possible risks that can cast doubt on whether phase I studies should be initiated at all during the pandemic (Upadhaya et al., 2020). Despite the pandemic there will be an impetus to continue the clinical studies of drugs at later stages of development to the extent possible. At that point in the development process, drugs will have already demonstrated lifesaving or significantly health-altering potential. In contrast, most drugs in phase I of development outside of oncology—by definition—do not yet offer an established health benefit to the study participants. In addition, phase I units are being considered as excess capacity for health systems across the world as many become increasingly overburdened in the pandemic response. As a result, CPUs with the capacity to monitor vital signs will not likely be allocated toward studies of new unproven medicines for some time.

DIGITAL HEALTH TECHNOLOGIES AND THE COVID-19 PANDEMIC

In dealing with the COVID-19 pandemic, there are emerging areas of interest related to using DHTs and the remote monitoring of vital signs and potential COVID-19-related symptoms as a proxy for disease incidence prior to confirmation with laboratory testing, Benko said. DHTs such as sensors and digital biomarkers connected to the body could be used to monitor vital signs and symptoms remotely using software; this could be complemented with electronic patient-reported outcomes collected via mobile phones or other devices. In March 2020, FDA released rapid guidance to support the adaptation of clinical trials during the COVID-19 emergency.8 The guidance identifies several types of non-invasive remote

___________________

8 The FDA guidance document is available at https://www.fda.gov/media/136238/download (accessed June 19, 2020).

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

devices that could be helpful in monitoring patients’ body temperature, cardiovascular function, respiration, and pulse oximetry.

These types of remote measures and monitoring technologies could also help facilitate the continuity of other types of clinical studies ongoing in various disease areas that are at risk of being disrupted by the pandemic, Benko said. Shifting conventional clinical assessments into a remote mode could help protect clinical trial participants by reducing the need for participants to visit clinical settings where they could be put at risk, put others at risk, or otherwise place additional burdens on health care systems. For example, pulmonary function tests typically require a person to breathe into a machine during an in-hospital assessment, potentially promoting the spread of disease. Providing a patient with a Bluetooth-enabled individual spirometer, he said, could enable the patient to measure pulmonary function in the home without risking exposure or transmission in a hospital.

Interpreting and Validating Measurements Collected via Remote Monitoring

More work will be needed to better understand, interpret, and validate measurements of vital signs collected through continuous monitoring using remote devices, Benko said. Body temperature appears to be a useful indicator of the progression and severity of COVID-19, so there is interest in the use of continuous remote temperature monitoring through a device such as a patch worn on the chest. Although continuous monitoring can generate rich datasets, the measurements can be challenging to analyze in the context of traditional standards. For example, it can be difficult to interpret measurements and establish alert thresholds for continuous temperature monitoring because of the poor correlation between those measurements and measurements of body temperature taken in body cavities (Izmailova et al., 2019). Regular spot checks with digital thermometers that measure body temperature in the oral cavity have well-established reference ranges and intervals, which makes the measurements relatively simple to interpret. If a person who appears healthy has a very low oral temperature measurement, then the person would typically be asked to repeat the measurement with the thermometer in an adjusted position in the mouth—or some other adjustment—until a reference value is attained that is better aligned with the person’s presentation. However, continuous temperature monitors are more prone to generating aberrant values than traditional methods. Benko and his colleagues looked at data from a single healthy individual using a continuously worn temperature monitoring device outside of a controlled setting, and this demonstrated how variable those measurements can be, with excursions

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

well outside of healthy ranges—from 34°C to more than 38°C. It can be difficult to control for these types of deviations, because the data are affected by a variety of factors, such as ambient temperature, clothing, and physical activity. More normative studies may be required to better understand measurements collected through the continuous monitoring of temperature or other vital signs before they can be integrated into a clinical development program, he added.

To illustrate how data collected using DHTs can be validated, Benko described a well-controlled crossover study that Koneksa Health designed with Merck to examine the potential for DHTs to detect meaningful real-world changes in cardiovascular and vital activity (see Box 4-2). The measurements captured by the wearable technology were concordant with traditional in-clinic approaches, and the technology was able to detect the expected changes in the participants’ vital signs during in-home use. The results of the study, he said, build confidence in the use of DHTs to capture real-world data on vital signs with sufficient sensitivity to detect the kinds

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

of treatment effects of concern with respect to safety. Vital signs can also be interpreted in other meaningful ways, he added. Significant changes in vital signs often indicate other changes in functional status. In settings such as oncology, for example, changes in vital signs can serve as a dynamic predictor of hospitalization or other decompensation—or the deterioration of an organ or organ system to maintain adequate physiological function.

Incorporating Digital Health Technologies into Oncology Research and Care

DHTs do not play a prominent role in oncology research and care, Benko said, although a recent study found activity level to be a significant predictor of hospitalization for patients with locally advanced non-small-cell lung cancer (Ohri et al., 2019). However, oncology drug developers in general are still resistant to adopting endpoints beyond progression-free and overall survival, which Benko said is short-sighted. To differentiate long-term benefit and long-term survival outcomes of targeted therapies from traditional chemotherapy and radiation, especially relative to cost, a patient’s functional status needs to be considered, he added. Measuring real-world components of functionality, such as activity, satisfaction, and sleep, are meaningful for those types of long-term analyses. However, drug development teams at pharmaceutical companies tend to focus on short-term milestones. An often-neglected consideration is that, over the long term, payers will have to choose between different therapies that may have been developed using different endpoints. In many cases the differentiation between those traditional endpoints either has not been established or is not compelling enough to justify the cost of one treatment over the other. Furthermore, there has not yet been a market force to drive this, he said. Benko highlighted another study that looked at patient satisfaction and activity among people with myeloma over the course of a long-term therapy (Chari et al., 2019), but he said that this type of work remains uncommon. Expanding the body of research on functional status in a way that uses DHTs for data collection will require addressing a substantial change-management problem within the oncology therapeutics leadership at many of the major drug development companies.

DEPLOYING DIGITAL HEALTH TECHNOLOGIES AT THE INTERSECTION OF CLINICAL CARE AND RESEARCH

Eric Perakslis, Rubenstein Fellow, Duke University

The use of DHTs should begin with the idea that necessity is the mother of invention, Perakslis said. “If you bring the right problem, you

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

are going to find a reason to bring the right technology to it,” he said. A focus of his own work is the dichotomy between data used for research and data used for care. Clinicians are often asked to work with bad data, as evidenced by those who report struggling with electronic health records or being unable to find a comprehensive history on a patient during a case review. If data are not considered to be good enough for research, they should not be considered good enough for clinical care, he said, adding that the opposite also holds true. Although data for research and data for care have different purposes and functions, he questioned how different the two types of data actually need to be.

Envisioning a Telehealth-Based Learning Health System

Perakslis described his vision of how a telehealth-based learning health system might function during an infectious disease outbreak, based on his own experience in the field during the previous two Ebola virus disease outbreaks (see Figure 4-2). In this vision, there is a complex network of data, learning, and communication that intersect across the domains of clinical care and research. Each clinical interaction should be able to support cycles of learning and clinical research. Within this system, for example, a family would receive a telehealth visit in which they are guided through layers of data collection supported by standard case definitions, standard protocols, and trusted sources for information; the information would then be fed back into the provision of clinical advice and care. During an outbreak, Perakslis said, the most important conversation to be had is to reassure people at home who are wondering if they are taking the right steps to protect their health and the health of their loved ones. However, these ideal system capacities for collecting data remotely, picking up samples, monitoring people at home, and propagating data in a reasonable way have not yet been achieved in real-world systems. Some of the individual elements and connections are functional and may connect with each other in a given system, but the entire enterprise does not function as a whole. “If you think big, you can always act small,” he said. “If you are thinking small, you are not going to trip and fall into big.” Those leading the federal-level response to COVID-19 should “think big” in their efforts to address the deep-rooted systemic problems that the pandemic has exposed.

Supporting Patients and Ensuring Data Privacy Outside of Traditional Clinical Settings

To more effectively incorporate DHTs into clinical care and research, Perakslis suggested working with community health workers (e.g., nurses

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Image
FIGURE 4-2 Telehealth-based learning health system during an infectious disease outbreak.
NOTE: CDC = Centers for Disease Control and Prevention; Dev = development; Dx = diagnostic; NHS = United Kingdom National Health Service; R&D = research and development; RWE = real-world evidence; WHO = World Health Organization.
SOURCE: As presented by Eric Perakslis, March 24, 2020.
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

or clinical trial managers), who play a critical role in delivering health care within the “last mile.” He emphasized that efforts to engage community health workers should ensure that there are fewer protocol deviations, the appropriate data are being propagated, the drugs are reaching the sites where they are needed, and people understand how to appropriately use DHTs. Community health workers may serve an important role in the implementation of “click-and-mortar” solutions, which mix the old paradigm of using brick-and-mortar sites with new DHT applications to fill gaps in clinical care and research. Another important consideration for the use of DHTs is data privacy, he added. He suggested that a proactive approach to data privacy could help ensure the security of data as increasing volumes of information are collected and integrated across multiple layers of the health care system. Starting with “privacy by design” to obtain appropriate consent and communicate clearly is important, he said.

Ensuring data standards and data validity when collecting data outside of traditional clinical settings requires high-level coordination, standardization, and organization, Perakslis said. During the Ebola outbreak in West Africa, the World Health Organization (WHO) served as the coordinator to support Guinea, Liberia, and Sierra Leone. WHO established a single case definition and a single set of triage forms. The degree of standardization that was achieved across the countries during the outbreak is something that many hospitals in the same U.S. cities have difficulty achieving, he said. Strong organization up front is also critical during a crisis such as an infectious disease outbreak. For example, it would be unsafe to recruit older populations with mild disease into a clinical trial by bringing them into a clinical setting. In those situations, a community health worker could visit patients in their homes to introduce the trial and lead them through the consent forms. Simple “click-and-mortar” solutions can enable the collection of clean data on the front end, even if there is variability downstream at the patient level. A single case report form could be developed for every COVID-19 patient, or a standard set of 10 questions could be asked at the beginning of every telehealth session. Crisis settings require swift and strong leadership to take control and ensure that the right data are collected from the outset, he added. This involves convening the right experts to rapidly develop standardized data collection protocols and propagate them downstream appropriately. All of the necessary technologies already exist, he said, but deploying them effectively depends on strong organization.

How, McGraw asked, can researchers ensure that privacy is not sacrificed in efforts to aggressively pursue data collection that could help to understand and halt the COVID-19 pandemic? The first priority should be to act appropriately without making avoidable mistakes, Perakslis said. For example, the appropriate patient consent and institutional review

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×

board approval should be obtained early so that valuable data are not lost. Every COVID-19 test result could be immediately tokenized so the data can be shared in a way that preserves privacy up front, he said. This is an example of existing technology that is simple, inexpensive, and widely available. It is important to take advantage of every opportunity for data collection, even in the midst of a crisis, he said. Each clinical interaction offers a unique and irreplaceable opportunity to capture information from that patient; the technology to collect data rapidly, securely, and comprehensively is already available and should be employed to its full extent.

DISCUSSION

Streamlining Regulatory Approval During Crises

The pharmaceutical industry, a workshop participant said, is entirely dependent on the rules and regulations established by governing bodies, while other industries have seen a paradigm shift in which consumers and manufacturers are compelling regulators to change more swiftly. Given that the COVID-19 pandemic demands rapid action, he asked, how could industry demonstrate the value of DHTs and drive change at an accelerated pace? The actions taken by regulators to expedite the pace of approval might continue long after this pandemic is under control, another participant suggested. The crisis has created a situation in which clinical providers and patients may be more likely to try out DHTs, such as telemedicine, Chan said, which could potentially drive demand and adoption of DHTs over the longer term. As telemedicine has taken off, she added, the Centers for Medicare & Medicaid Services and some payers have adapted by changing their approach for reimbursement. Perakslis suggested thinking about how DHT development and incorporation into the standard of care might differ in crisis versus non-crisis environments. For example, he pointed out that in non-crisis situations, it might be typical for safety and risk to be evaluated through the course of clinical research. However, in an accelerated-approval or crisis scenario, there may be a shift toward more risk assessment in post-market settings. Experience based on previous outbreaks has indicated that crises can be used to spur forward momentum when it comes to technology development and implementation, he added.

Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 41
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 42
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 43
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 44
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 45
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 46
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 47
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 48
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 49
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 50
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 51
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 52
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 53
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 54
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 55
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 56
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 57
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 58
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 59
Suggested Citation:"4 Digital Health Technologies for Recruitment and Safety." National Academies of Sciences, Engineering, and Medicine. 2020. The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/25850.
×
Page 60
Next: 5 Digital Health Technologies for Pivotal Trials »
The Role of Digital Health Technologies in Drug Development: Proceedings of a Workshop Get This Book
×
Buy Paperback | $65.00 Buy Ebook | $54.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

On March 24, 2020, a 1-day public workshop titled The Role of Digital Health Technologies in Drug Development was convened by the National Academies of Sciences, Engineering, and Medicine. This workshop builds on prior efforts to explore how virtual clinical trials facilitated by digital health technologies (DHTs) might change the landscape of drug development. To explore the challenges and opportunities in using DHTs for improving the probability of success in drug R&D, enabling better patient care, and improving precision medicine, the workshop featured presentations and panel discussions on the integration of DHTs across all phases of drug development. Throughout the workshop, participants considered how DHTs could be applied to achieve the greatest impact—and perhaps even change the face of how clinical trials are conducted—in ways that are also ethical, equitable, safe, and effective. This publication summarizes the presentations and discussions from the workshop.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!