In-Depth: The rise of the digital clinical trial

By Dave Muoio

Editor's note: Since publication, some of this story's quotes have been edited for clarity.

With billions of dollars and countless hours of human effort spent each year validating new drugs, devices, and medical interventions, it’s little surprise that the healthcare industry is turning to technology to streamline the process. From participant recruitment to data collection to adherence, clinical trials are rife with logistical challenges similar to those already being addressed by health tech’s major innovators — and trial sponsors have taken notice.
“How can you use software, hardware, coaching mechanisms, a combination of the above, to promote adherence … or make previously subjective measures objective?” Heather Bell, SVP, global head of digital and analytics at Sanofi, told MobiHealthNews. “Take something that used to be a patient-reported outcome — now I can simply track it using a wearable, ... increasing my datasets and reducing the number of patients I need in order to get significance, thereby shrinking my trial and making it less expensive and faster to conduct.”
But despite the ubiquity of connected devices and tech disruptions across other industries, Bell admitted that healthcare has so far been slow to incorporate these advances in its clinical trials. To some — like Joe Dustin, principal of mobile health at Medidata, a New York-based company that offers cloud storage, management software, and data analytics services for clinical trials — the number of clinical trials that currently rely on antiquated, paper-based data collection is “pitiful”, considering what’s possible. Fortunately, though, he said he sees an industry that is beginning to turn around.
“Sadly, the technology, in many cases, is not the limiting factor when it comes to speed and efficiency and better processing clinical trials,” he told MobiHealthNews. “A good portion data is still collected on paper and entered into a system somewhere. Most of that paper process is still what they call the electronic case report form, and that’s what an [electronic data capture (EDC)] system captures. That’s moving faster to e-source, where companies are now using electronic means to capture that data the first time as opposed to putting it in paper binders … and that’s clinic data, right? You also have labs, diary data, you now have wearable and mobile health data coming from sensors on the body, you have genomic data, you just have this deluge of data that’s becoming more digital now. And with the coming of digital, the size quantity of this data is exponential.”
The long road to the fully remote trial
Pharmaceutical companies have looked at supplementing steps of the clinical trial with digital approaches for a number of years, with varying levels of success. In 2012, a number of major pharmas formed the nonprofit collaboration TransCelerate BioPharma specifically to investigate how digital technology could be leveraged to improve clinical trial efficiency. The companies made individual moves as well, with Pfizer, for example, launching its web-based clinical trial outreach platform Pfizer Link around the same time. Trial sponsors and contract research organizations (CROs) also began to gradually ramp up adoption of electronic informed consent and EDC systems to collect electronic patient-reported outcomes [ePRO] over the past few years, but Dustin said that these implementations have more often been exception rather than the norm.
“I’ve been in a clinical trial myself, and that was in 2010 … I wanted to see what it was like. [Since] I’d been doing this for work, I understand all the technology,” he recalled. “The study was run completely on paper. No technology, no diary, they didn’t even have EDC, it was all giant paper binders. I was like, ‘Are you kidding me?’ And this is me coming from a company that makes this technology. … If I was going to do that now and I didn’t get any kind of technology, there’s no way I’d want to be involved. I’d want to make sure that I’d do better, and be able to track myself and be participatory.”
But while a number of digital tools have gradually gained momentum, it's the site-less, fully digital trial that has long stood as the industry’s golden goose. In 2011, Pfizer and Exco InTouch launched the REMOTE trial, the first major attempt to conduct a randomized clinical trial that enrolled patients online and allowed them to participate entirely from their home via drug deliveries and a mobile app. A year later, Pfizer discontinued the trial, citing a lack of enrollments despite their success in drawing patients to the study website through Facebook and Craigslist. Pfizer said that they would resume their digital efforts with another version of the trial in 2013, although this promised mulligan never emerged.
"For us, we took away a series of modules that worked and one that didn’t," Craig Lipset, head of clinical innovation at Pfizer, said during a 2015 conference. "That’s not meant to imply that online medication recruitment is a failure. In this particular protocol going after that particular patient population it did fail for us, but the other components did work. So our priority has been to take those components that did work, whether it’s electronic consent or other components, and scale those up. Because we’ve got 200-some clinical trials ongoing every day of the week, so how can we leverage what did work?"
While this approach may have been premature in 2011, subsequent attempts at a site-less digital trial appear to validate Pfizer’s aspirations. In 2015, Sanofi and Scotland-based eClinicalHealth announced their VERKKO phase 4 trial, an investigation of Mendor’s wireless, cloud-connected glucometer that enrolled 60 participants through Facebook and completed all necessary check-ins through eClinicalHealth’s Clinpal system. In October of this year, Los Angeles-based Science 37 also boasted the success of its Network Oriented Research Assistant (NORA) platform, a telemedicine-like system that allows for real-time video chat, electronic data collection, and electronic consent. According to the company, a phase 2b study of AOBiome Therapeutics’ acne treatment marks the first completion of an interventional, randomized, placebo-controlled trial conducted entirely without a central trial site.
“There were no CROs involved, there were 11 physicians involved covering 10 US states, and we did it in a much faster time frame than is needed in that kind of trial,” Science 37 cofounder and CEO Dr. Noah Craft told MobiHealthNews. “That demonstration that this can be done — enrolling patients through Facebook and Google and taking care of them through telemedicine — to us it was a huge milestone. It lent a lot of credibility to our work.”
Sanofi’s Bell shared Craft’s enthusiasm for site-less trials. By removing the barrier to entry, she said, pharmaceuticals are treated to a wider selection of potential participants and, subsequently, more diverse data.
“A lot of the other interventions you would pursue in the trial space, a lot of them still work from the assumption of a fixed trial site, and the patient has to go to the fixed trial site in order to participate, and that immediately presents you with a set of constraints around who’s within driving distance and who has the time to be available,” she said. “What Science 37 does is liberate you from that geographical constraint, and says anyone anywhere who is motivated and engaged can participate in a trail from the comfort of their own home, and just opens up all kinds of possibilities for a much more diverse trial population in all senses of the word — ethnically diverse, geographically diverse, economically diverse.”
Site-less trials can also be distributed through consumers' mobile devices, as Apple has proved with its ResearchKit platform. Launched in 2015, the open-source platform helps researchers build medical apps and more easily recruit patients for clinical trials and other research projects.
“One way to think about ResearchKit is as the beginning of a pipeline that will lead to more apps that are screening, diagnostic, management and treatment apps,” Dr. Bud Tribble, Vice President of Software Technology at Apple told MobiHealthNews in 2015. "In fact, it is a necessary, essential first step to figure out what is needed to develop these apps -- what works and what doesn’t -- before you move into, ultimately, clinical study apps and clearance to use them for diagnosis and treatment.”
Dr. Belinda Tan, Science 37’s chief medical officer and cofounder, explained that the concepts behind site-less trials and her company’s platform have been around for at least a decade. However, she and Craft initially hesitated to pursue their remote platform and site-less trials as a whole due to telemedicine’s lack of mainstream acknowledgement.
“When we started this company back in 2014, it was actually quite a different environment in terms of people’s acceptance — consumers as well as payers and healthcare providers — in using telemedicine to provide care,” she told MobiHealthNews. “That was actually a really huge step forward compared to probably back in 2010, [where it was] only just getting started in areas with the [Department of Veterans Affairs] and more limited types of healthcare settings.”
The culture shift isn’t limited to telemedicine alone. Dustin, from Medidata, said that interest in digital tools for clinical trials has rapidly accelerated among study sponsors and CROs alike, but that much of the movement also sprouts from patients’ reliance on their mobile devices.
“The idea of bringing your own device to engage patients in a digitally-connected clinical trial is not crazy talk anymore,” he said. “If you look at the stats of Google alone, 70 to 80 percent of the [developed] world is penetrated with some smartphone, and then in the third world there are more smartphones than toothbrushes. … This is just normal to them, and they also demand something better than paper when they want to get involved in a clinical trial.”
Patient-driven innovations
The focus on the patient’s experience during a clinical trial is a core tenant of recent digital efforts, Dustin continued. Traditionally, clinical trials would ask patients to fill out excessive paperwork or journals while regularly traveling to and from a clinical site for check-ins, an ordeal that leads many to miss their dosages or drop out of the study. Offering these technologies instead can greatly improve the quality of trial data, he said, but only if the technology itself is designed with users in mind.
“One of the cores of patient-centricity is not only designing the protocol to be patient-centric … but that the app itself, the user experience, is tested with patients and accepted by patients,” Dustin said. “It just makes it brainless and easy and engaging. User experience is the biggest change here, we’re actually investing time and money that will not only improve  outcomes, but keep people in the trial longer, keep them engaged, prevent dropouts and get better data.”
Craft echoed Dustin’s patient-focused sentiment.
“We built our platform around that idea — that if you make clinical trials really easy and convenient for patients, you can help them enroll from home,” he said. “Once you make that shift philosophically to the patient, then you can use technologies, sensors, devices, … and other ways to communicate healthcare data digitally.”
Convenience comes in several forms, though, and trial sponsors are considering a number of different approaches to improving participants’ experiences. For instance, Nathan Treloar, president and COO at Orbita, said that his company’s voice-based home health assistant was tapped to provide support for rheumatoid arthritis patients who would find it difficult to complete therapy assessments. By redesigning the standardized RAPID3 Survey as an Amazon voice skill and pairing it with a blood pressure monitor, he said, his company was able to reduce the complexity of the survey and remove participants’ reluctancy to complete them.
“The whole point was they wanted to improve the accuracy and timeliness of data collection in clinical trials, so they can collect better quality data and hopefully move clinical trial to completion faster,” Treloar said.
Pillo, makers of a personal home robot designed to improve medication adherence, offers pharmaceuticals a more hardware-oriented opportunity. While Pillo’s cofounder and COO James Wyman said that the product was originally designed as a user-friendly, straight-to-consumer product, his team was surprised to receive a number of inquiries from pharmaceuticals and CROs about implementing Pillo as an adherence tool in their phase 3 and phase 4 trials. While the device on its own doesn’t completely remove the need for a central study site, he explained that its facial recognition, time-stamping, and digital interface can act as a support tool located directly inside the patient’s home.
“The phone as an interface is an amazing tool and it’s very convenient — sometimes too convenient, you can just swipe a notification away left or right,” Wyman told MobiHealthNews. “You see the growth of the Echo and these in-home devices, and I think there is a role for the hardware devices, especially when it comes to the most important part of these trials: if people are taking their medicine correctly.”
Maybe one of the longest standing inconveniences for patients has been the difficulties in finding and enrolling in a trial to begin with, Munther Baara, senior director of development business technology and head of new clinical paradigm at Pfizer, told MobiHealthNews.
“If you want to go on vacation, it probably takes you a couple of minutes … to find the best airline,” he said. “Now think about your health, which is the most important thing as an individual. You go online and try to search for a clinical trial because you couldn’t find an approved drug in the market. … It’ll probably take you weeks.”
Baara, who also represents TransCelerate BioPharma, said that while government registries like are important to maintain, they can be extremely difficult for uninitiated patients to navigate, and frequently do not provide clear trial information. To help patients and facilitate more enrollments, TransCelerate recently proposed an alternative registry — what they call the Clinical Trial Registry of the Future — that would offer users an easy to navigate database. This service would include information about registration requirements, time commitment estimates, study design, and other information.
“We’re not trying to create another group or organization, we want to utilize existing government registries and basically enhance the quality and put a new face on it so it becomes really user friendly to the patients,” Baara said. “If you think about what’s important to the patients, they want to easily come in and say ‘Here is the disease I have, here is where I am located, tell me if I have any options.’… Based on this, it gives them a list of the different studies. It’s basically an interactive framework to narrow down the options.”
The benefits of this system and other patient-friendly recruitment efforts — such as XpertDox’s clinical trial search engine, XpertTrial  — serve patients and pharmaceutical companies alike. Faster enrollment shortens the time and money spent by trial sponsors, and every month an effective therapy is kept from approval is another that patients spend without treatment.
“We take a very long time as an industry to develop drugs, and if you can halve the recruitment time and if you can increase retention because it’s easier to participate, that’s a huge win for us and for the patients on the other side,” Bell said.
Digital trials today and tomorrow
The industry’s adoption of digital-enabled trials has been slower than many would like, an effect which Dustin attributes to the risk-averse nature of pharmaceuticals and trial sponsors. Still, both he and Bell said that several of the early concerns about these techniques’ clinical reliability are starting to fade away as reliability and potential becomes more clear.
“I think any time you try things for the first time you have a concern that things might not work. There’s a reason you go and run pilots before you scale things: It’s a learning process for you, it’s a learning process for whoever you’re working with externally, it’s a learning process for the trial sites you may be interfacing with,” Bell said. “We have a lot of the building blocks in place … and I think we also realize that you cannot sit on the sidelines on this stuff forever. You need to get in and tackle the execution challenges, and you need to be prepared to work through the operational change management issues.”
A clear sign of the changing tides comes from the FDA, which revised its stance on software and mobile health technology to be more inclusive in late 2016. This decision, Craft and Tan said, enabled digitally-powered clinical trial platforms like theirs to more easily gain recognition so long as their methodology and safety measures are up to code.
“The questions they ask are mostly around ‘Are you sure that patients are safe?’ and ‘Are you sure that the data you collect is going to tell you if the drug works or not?’" Craft said in regard to the FDA and trial sponsors. "We take this extremely seriously, and you’re under a microscope when you’re trying to change how science is done. ... Each company that hires us and each committee on the FDA asks the appropriate questions around safety and data integrity and whatnot, and we try to answer them as best we can.”
Outside of increased adoption of EDC systems, end-to-end platforms, and new patient-focused interfaces, Dustin and Bell said that that over the next few years they anticipate an uptick in clinical trials employing wearable sensors. Unlike previous studies that almost exclusively used cheap consumer-grade devices, these specifically will begin employing medical-grade sensors, and will target endpoints that would otherwise be impossible without continuous monitoring.
“All they were doing [five years ago] was testing the feasibility of the device in general — will patients wear it, how is the data that comes off the thing, what kind of analysis — no real questions,” Dustin said. “Now, I’m seeing specific goals of trying to look for new digital endpoints in certain therapeutic areas. Mostly in IT groups in bigger pharmas … they are looking for these types of signals that ... detect some signal that's specific. Could be in cardiology, could be in some kind of heart failure, could be in Parkinson’s, a lot of uptick in neurology, … dermatology, [and] even in oncology and pain studies.”
Dustin said that such studies are likely to remain uncommon well into the 2020s, but that a more immediate change will likely come with how the various digital services supporting trials are packaged.
“Sponsors, pharmas, they call it the mClinical vision — it’s one app to rule them all,” he said. “All these different features ... should all be in one app, because what patient wants four apps and three different sensors in one trial? Or in life?”
Regardless of which direction the industry takes, Bell cautioned that any major changes to the status quo shouldn’t be for the sake of change itself — especially when it comes to medicine.
“You don’t want to be technology-driven on it. You don’t want to be ‘Ooh, ahh, this is flashy technology’ — you need to think about the clinical or engagement problem that you’re trying to solve,” Bell said. “It’s a constantly iterative dialogue. You need to see what the technology can do that may make you think of problems in a different way, but equally you need to be driven by the issue, the business problem, the clinical development problem … and then try to bring the appropriate technology to it.”