Digital Technology: The Future Is Bright

It is not that innovative technology in measuring medical outcomes with a digital or electronic means is a particularly new thing. In the early 1960s the Holter monitor for measuring ECG heart activity became available for use. At first these devices were not digital, nor were they even portable. But today a Holter monitor is clearly a digital device that can be used to monitor your ECG, but also can be used to record ECG endpoints. Is the Holter monitor the first medical digital device? If not, it is in the running. What is certain, however, is that it is not the last. About 2012 one of the authors of this editorial was a statistician on a study in which historically the endpoint of interest was gathered with a paper diary in which subjects noted each time an event happened, the analysis endpoint being the number of times the event happened over 24 hours. There was a company that we became aware of that could develop a device for patients in which they could record their events in a device. This made sense. We worked with them and we had an electronic means to collect these data and transmit them automatically into a database. Now, there became new concerns, such as what do you do when a patient says they have had 70,000 events in a day? These ended up, however, being issues that one could fix with intelligent design. Time goes by and one starts seeing Fit Bits being all the rage. People are recording steps and vital signs and other things to keep track of their own health. It does not take long to recognize that if an individual can do this with a phone then we ought to be able to collect similar sorts of information for clinical trials this easily. Why is this important? Well, it could mean less or shorter visits for patients during a clinical trial. The possibility of continuous 24-hour monitoring became more feasible. With more data, could we require fewer subjects? All kinds of possibilities became apparent. But more excitingly is that nonapparent ideas that could revolutionize medicine could take place. Improved data quality and improved assessment of clinical outcome can have profound implications for clinical research—increased accuracy, reduced need for clinical site visits, shift from treatment to prevention (due to longitudinal data monitoring), streamlined clinical decision-making, etc. However, these promises necessarily come with challenges—more data does not always mean better data, and data analysis must be done properly so that the results are convincing. Digital technologies are powerful tools that hold much promise in clinical research and development. However, the success of their implementation will require care: well-trained and qualified individuals with necessary medical and statistical expertise. Discussions among stakeholders in academia, industry, and health authorities, and development of new statistical methodologies to design and analyze clinical trials utilizing wearable devices and digital medicine products, will be inevitable for making progress in this direction. To some extent, the artificial pancreas notions that seem to have a lot of promise for those with Type 1 diabetes will most likely be powered by digital technology. With a digital device like Dexcom1 for monitoring of blood glucose along with an insulin pump which can now deliver insulin electronically, we see not only the possibility of better endpoints but better healthcare delivery. Will it take long to bring an implantable artificial kidney2 for testing in clinical trials for chronic kidney disease? If this device shows favorable risk–benefit properties for such a high unmet medical need, it could bring tremendous help and improve quality of life for many patients who are on hemodialysis. Will we see drug–digital combinations in the near future? On November 13th 2017, the US Food and Drug Administration (FDA) announced approval of Abilify MyCite for the treatment of schizophrenia, bipolar I disorder, and as an adjunctive treatment of adults with major depressive disorder.3 Abilify MyCite contains an ingestible event marker sensor embedded into oral Abilify tablets. The sensor tracks if patients have ingested their medication and communicates this information to the patient and healthcare provider, thereby providing an objective summary of drug ingestion over time and offering a new way of monitoring patients. This is perhaps the first example of when two previously approved products (psychiatric drug Abilify and the ingestible sensor) are combined to produce a novel drug–digital medicine. Will there be more? The articles in this special issue attempt to reveal the potential of intelligent digital devices in clinical research and development. Izmailova et al.4 discuss various applications of wearable technologies in clinical trials as well as corresponding promises and challenges. The authors highlight the utility and use of wearables in many therapeutic areas while emphasizing the need for more scientifically rigorous experiments before wearable technologies can be widely adopted. Rogozinska5 provides nice examples of how the use of digital health technologies can potentially enhance productivity of drug development by increasing speed, quality, and product commercial value, and reducing cost and workload. The described regulatory, scientific, and operational challenges can also be viewed as opportunities for pharmaceutical companies to improve their business performance. Walton et al.6 describe a new type of clinical trial design, the Micro-Randomized Trial (MRT), for development of adaptive mobile health interventions. In this design, treatment options for each individual are selected based on carefully defined randomization algorithms, applied 100s or even 1,000s of times per individual during a study. The MRT is an example of an adaptive trial design to create an optimal treatment strategy for further testing in large randomized comparative studies. Koydemir and Ozkan7 describe innovative ways of using smartphones as biomedical sensors and platforms for collection, display, and/or sharing biomedical data acquired by wearable devices. The authors cleverly argue that smartphones can democratize advanced measurement and can potentially create transformative opportunities in various fields, including but not limited to medicine and education. Byrom et al.8 give a thorough overview of brain monitoring digital devices in neuroscience clinical research. The scope of applications is very broad and includes studies of pain, Alzheimer's disease, traumatic brain injury, epilepsy, sleep, etc. While data from many of these technologies cannot replace clinical endpoints, the authors provide a good rationale and justification for implementing them as exploratory tools in clinical trials. The high-frequency data can provide valuable insights into brain functioning, cognition, patterns of sleep, etc. which can also aid interpretation of the clinical outcomes. Finally, Sverdlov et al.9 discuss digital therapeutics—a new treatment modality in which digital systems can be used to treat medical conditions. It is argued that digital therapeutic technologies have broad applications, including cognitive behavioral therapies, gaming consoles for postdischarge rehabilitation, hybrid closed-loop systems, and many more. The authors also present business models and regulatory considerations for developing digital therapeutics, and provide some recent examples of digital therapeutics that have gained regulatory approval. Digital technology is in its infancy. There will be a lot of failure. The natural evolution of humans took millions (or billions?) of years, recycling imperfections or redundancies. Today, science and technology have reached a state when discoveries and advances in medicine and healthcare will be made at much faster speeds. It does not take a rocket scientist (or a clinical pharmacologist) to see that this technology will be revolutionary and we imagine that will change much of what is done in clinical trials and will have a major impact on how medicine is practiced in the future. No funding was received for this work. The authors declare no competing interests for this work.