Please login to the form below

Digital intelligence blog

Pharma insight on digital marketing, social media, mobile apps, online video, websites and interactive healthcare tools

Digital Trials

A framework for success

Digital

There is increasing interest in digital trials, that is the use of mHealth and mobile technology to capture insights outside the traditional clinical setting. The drivers are clear – spiralling costs, questions with respect to existing endpoints and inability to recruit or retain subjects. The pharmaceutical industry spent over four times more on research and development in 2015 than in 1995, with no corresponding increase in the number of drugs approved by the Food and Drug Administration (FDA). An analysis of trials in 2011 found that 19% of trials closed or terminated failed to meet accrual goals (85% of expected enrolment) or were terminated early due to insufficient accrual. In a recent survey carried out by The Centre for Information and Study on Clinical Research Participation (CISCRP), ‘too many study visits’ was the reason given by 24% of participants who decided not to take part in a clinical trial, while ‘travel burden’ was cited by 79% of those surveyed in a recent survey of caregivers on the barriers for clinical research participation.

Digital trials are seen as a vital component to providing a new path, a way of enabling patient- centric trial design and capturing clinical signals that are more responsive to change than traditional outcome assessments. By collecting continuous data streams as individuals go about their daily lives, mHealth technology can capture physiological data and patient insights outside the traditional trial site. This enables patients to remain in their own home while their biometric data ‘travels’, reducing or eliminating the need for patients to personally travel to the site. For those therapeutic areas where site visits are a requirement due to specialist equipment or therapies being required, the use of digital technology has the potential to reduce the length of the site visit, which could be of particular benefit in oncology studies.

Although mHealth technology is continuing to evolve (for example, we have seen the recent FDA clearance of the Apple smart watch to capture ECG), it is now possible to collect a vast array of physiological data including vital signs such as heart rate, respiration rate, oxygen saturation, continuous glucose monitoring, sleep and activity data, and using advanced analytics to monitor patients in their own home outside the hospital environment. The operationalisation of digital trials is not without challenges. Ensuring that data from digital trials are sufficiently robust, reliable and of sufficient quality and that there is minimal data loss is essential. The success of a digital trial requires careful consideration to be focused on three main areas: the patients, the device and the data.

Patients

It is self-evident that if patients are unwilling or unable to use digital technology then the probability of a successful outcome from a digital study is low. Devices should have a minimal impact on patients as they go about their daily lives. It is essential that the devices and sensors capture the study concept
of interest and are sensitive to the minimal clinically meaningful difference for a specific therapeutic area and population, but also equally important is the need to capture insights that have value and meaning for the patient. It is critical to adequately assess patient acceptance of the technology prior to its use in the study. While it is generally accepted that the use of devices and sensors can help create more patient-centric studies, if not carefully managed the use of digital technology can add complexity and place significant burden on patients. Devices selected need to be easy to use and have a minimal impact on patients. A single Study app that provides seamless connectivity to the devices needs to be designed with the user experience in mind and should add value to the patient experience while part of the study. A Bring Your Own Device (BYOD) model is preferable so patients do not have to deal with multiple mobile devices. Increasingly the concept of data exchange as a means of engagement is being considered – with the app acting as an interface between the individual and the trial, and including features that support patients.

Device

It is essential that the study objectives are clear prior to decisions being made about the selection of the devices and sensors. Identifying devices that are fit for purpose is key, and should be driven by the objectives, trial design and patient population. Both medical devices and other grades technology can play a role in digital studies as long as there is a clear rationale for the device selection and sufficient evidence is available to support the use in that specific patient population. This will be a critical issue when presenting the CSR to the regulators. A device that works well in one population may not be readily transferrable to another patient population. For example, a wrist-worn activity monitor may have a high degree of sensitivity and specificity to track activity levels in a very mobile population, but would have significantly reduced sensitivity for an elderly population that uses walking aids. From a regulatory perspective the device needs to be able to capture and transfer the required data in accordance with local privacy and security regulations. For example the ICF (Informed Consent Form) needs to correctly reflect the type of data collected by the device and the purpose for which the data will be used, and the data needs to be pseudonymised, which may present challenges for some of the consumer grade devices.

Data

The emergence of digital platforms that have the ability to ingest and analyse vast quantities of data are transforming the digitalisation of the physiological signal. In an ICON survey on ‘Improving Pharma R&D’, Big Data, AI/Predictive Analytics and Mhealth were cited by respondents as the disruptive technology trends which will have the greatest impact on clinical trial operations. Although endpoints from wearables such as Actigraphy have been used as primary and secondary endpoints in previous trials, the majority of the digital data is currently used to support exploratory endpoints. There has been a recent example of the development of a ‘hard’ digital endpoint for COPD, and although currently use case is limited and restricted, it is anticipated that, as more validation studies are conducted, more hard digital endpoints will be developed and qualified for specific therapeutic areas. The future potential lies in the combination and correlation of data from multiple sensors and devices with electronic patient-reported outcome (ePRO) data. Combining this data with ePRO and other physiological data such as heart rates and respiration rates can create a big, complex data set that is beyond the capacity of standard electronic data capture systems (EDCs). Therefore, when selecting a digital platform, the ability to scale and ingest high frequency data sets is important. In order to generate insights and digital biomarkers, data scientists who are skilled experts in advanced analytics are needed. The platform should enable the data scientists’ work by exposing the appropriate tools to process the data and run machine-learning algorithms.

The recent Clinical Trials Transformation Initiative (CTTI) report recommended that a short feasibility study is carried out prior to implementation of the technology in a large study. This approach allows for the de-risking of the larger trial. These studies can be carried out in a relatively short period of time depending on the study population. In a recent COPD simulation, the study protocol, IRB, sensor selection, app development, recruitment, data generation and analysis were all undertaken within a five-month period. The study objective was to assess the feasibility of capturing biometric data that had potential as a digital biomarker to monitor respiratory patients. The patient interface with all of the devices and the ePRO was via a single study app to reduce patient burden. The devices were selected based on the endpoints that had been identified from the literature that were all linked to COPD exacerbations.

Summary

Declining research and development (R&D) efficiency is one of the biggest challenges the pharmaceutical industry is facing today. The traditional approach of three discrete, fixed-trial phases designed for testing mass-market drugs often is not viable in today’s increasingly competitive, value-based therapeutic markets. It lacks the flexibility, analytic power and speed required to develop complex new therapies targeting smaller and often heterogeneous patient populations. As a consequence, clinical trials are changing. Digital disruption in the form of new wearables, sensors and medical devices enable pharmaceutical and medical device companies to generate new types of data sets. Artificial Intelligence and machine learning can generate new insights and digital biomarkers that have
the potential to be more clinically responsive to change. The successful roll-out of study requires a flexible operational model and experienced
study teams that understand device logistics and management, and that have robust digital platforms to train and engage patients and site staff. Skilled data management teams that can process data sourced from diverse systems can be daunting. Experienced patient engagement teams are an essential part of ensuring successful digital studies.