Trendspotting: Decentralized Trials, AI, Real-World Data in 2021
January 7, 2021 | It seems a bit bold to be trendspotting the first week after 2020 ended, but we really won’t make any progress if we can’t learn from our past, especially in this remarkable year. Leaders in clinical research are working hard to synthesize what 2020 brought us and apply learnings to 2021.
When we gathered predictions from leaders at Veeva Systems, Oracle, Datavant, IQVIA, Huma, Medable, and Inato, some themes arose. Artificial intelligence has a lasting role to play in clinical research, executives from several groups argued. “In a COVID world in which there is a push to make telemedicine more widely available, automation and AI technologies will serve as a key indicator and communicator throughout the clinical trial process,” said Updesh Dosanjh with IQVIA. And Oracle’s Henry McNamara pointed out that AI has shown to be more effective than humans in safety case intake and site selection. “As more companies lean into decentralized trials, and as the pace of trials and the amount of data that needs to be analyzed continues to increase, AI will emerge as the only way to process this data efficiently,” he said.
While some preferred “siteless” and others used “decentralized,” it is clear that non-traditional clinical trials will gain ground as we remain somewhat uncertain about the pandemic’s trajectory. “The move to remote clinical trials is a big shift, as you can collect a lot more data, a lot more easily and from a lot more people,” said Huma’s Dan Vahdat. “Plus, you can also recruit and screen patients remotely. That means you can finish a trial, from A to Z, in almost half the time.”
Several execs pointed out the opportunities for decentralized trials to reach broader trial participant populations than ever before—and in a way that is more palatable to them. Henry Levy with Veeva Systems anticipates that reducing the patient burden will be a top priority for life sciences companies in 2021. Liz Beatty of Inato agreed. “Trials need to adapt to patients—not the other way around. This means enrolling and treating patients where they are, in their communities. Thanks to COVID, 2021 will be the year where we finally bring clinical trials to the community—but it will take a shift in how sponsors partner with sites,” she predicted.
But decentralized trials won’t only offer an easier participant experience. Medable’s Jena Daniels emphasized that DCTs have an important role to play in increasing trial participation diversity—which starts by involving more people of color in protocol development. “Diversity will be first-order priority at the beginning rather than reverse-engineering racial equality into the trial design as a rescue mission,” forecasts Daniels.
Finally, several contributors flagged the opportunities for real-world data. Datavant’s Jason LaBonte predicted that real-world data driven patient recruitment will substantially cut enrollment time and increase access for trials.
His colleague, Travis May, echoed the diversity concern, highlighted an opportunity for real-world data linking to eliminate trial access barriers for hard-to-reach populations. “By the end of 2021 we will see linking clinical data with real-world data eliminate the access barriers currently faced by disadvantaged and hard-to-reach populations. Linking trial data to real-world data will enable researchers to access patient data and study outcomes while reducing the need for frequent in-person site visits,” May said. “The direct result is that clinical trials will become more accessible to all patient populations.”
Here are the full trends and predictions including additional forecasts for mobile health devices, digital biomarkers, external control arms, telemedicine, regulatory responses, and data science. –the Editors
Henry Levy, general manager, Vault CDMS, site, and patient solutions, Veeva Systems
Paperless, patient-centric trials will become the norm. Getting participants to finish a clinical trial is not easy because of the heavy burden placed on patients. In 2019, dropout rates rose to 19.1% in late-stage studies globally from 15.3% in 2012. COVID disruptions and restrictions accelerated change in how trials are run as the industry looks to end its reliance on paper. To reduce the patient burden and modernize studies, paperless, patient-centric trials will be a top priority for life sciences companies in 2021.
Whether it’s an electronic consent process, remote data collection, or enabling virtual visits in a hybrid trial model, there are numerous ways sponsors, CROs, and sites will reduce the patient burden while improving stakeholder collaboration over the course of a study. Ongoing challenges with enrollment and retention, coupled with the need to reach patients in new geographies, will only accelerate the industry’s shift toward a more paperless, patient-centric approach.
Greater diversity in clinical trials will drive more effective drug development. The lack of diversity in clinical trials has become more apparent over the last year. The industry will focus on driving change in this area because including a more representative population can positively impact study outcomes and society. The opportunities created by virtual trials to expand the reach of studies, which have increased due to the pandemic, have great potential to enhance patient diversity.
Sponsors can help spearhead this increase in diversity by specifically seeking out research sites in underrepresented geographies, while CROs invest in more support for investigators in areas with minority populations. Technology will also play a role in expanding the geographic, demographic, and reach of trials by allowing patients to do check-ups and other appointments virtually.
With a greater commitment to diversity in the coming years, life sciences companies will ensure that their clinical studies more accurately represent the patients who will be using their products. More diversity in trials will be better for the industry—and life-changing for the patient populations it serves.
Travis May, CEO of Datavant
Real-world data linking will eliminate trial access barriers for hard-to-reach populations. Among the most prominent difficulties in the traditional clinical trial model is the recruitment and retention of patients. Across many trials, patients are lost to follow up when they are unable or unwilling to return for subsequent check-ins. This difficulty is compounded for trials that require long periods of follow up or involve traditionally hard-to-reach populations. These populations may include patients living miles from the necessary treatment centers, patients who require assistance to reach a clinical site, or are otherwise inhibited from participating in follow up. The result is that many patients and demographic groups are underrepresented in clinical trials; they are unable to participate at all, or difficult access prevents them from participating in the full extent of the study. These patients experience reduced benefit of the treatment in the trial, and researchers are unable to include their data in study and analysis.
By the end of 2021 we will see linking clinical data with real-world data eliminate the access barriers currently faced by disadvantaged and hard-to-reach populations. Linking trial data to real-world data will enable researchers to access patient data and study outcomes while reducing the need for frequent in-person site visits. Patients who for health, economic, or other reasons would be unable to access specialized clinical trial centers will be able to participate, and their trial data will become available to researchers.
The direct result is that clinical trials will become more accessible to all patient populations. This will have the win-win impact of increasing availability of cutting-edge care to historically disadvantaged populations, as well as enabling broader and deeper research that can more quickly accelerate drugs to market.
Jena Daniels, Director of Research and Head of Patient Advocacy Council, Medable
Greater awareness of social injustices will ensure diversity is no longer an afterthought. As we continue to open our eyes to social injustices across this country, more pharmaceutical companies will engage in real conversation about equality and have another important reason to embrace a decentralized approach to trials. But this is just the start. From New Year’s Day 2021 onward, trial sponsors will not only embrace DCTs to expand access but also involve more people of color in protocol development. Diversity will be first-order priority at the beginning rather than reverse-engineering racial equality into the trial design as a rescue mission.
Study teams, too, will become more diverse so all demographics have better representation. They will bring vital insight into cultural and communications differences that can impact trial design and outcomes. Already, people of color are less inclined to take an approved COVID vaccine even as they are a more vulnerable group, yet this could be overcome with better communication and education.
Expect to see diversity increasingly baked into study protocol design and product development over the next year and beyond. And it will extend beyond people of color to include different genders and ages. We will see patients from all backgrounds involved in the study process from the start. Medable’s Patient Advocacy Council (PAC) has patients ranging in age from 20 to mid-70s, a mix of female and male members, and is expanding to add minorities in early 2021. The PAC informs Medable products and trial design – not on the back end, but at the beginning when they can make the biggest impact.
Ingrid Oakley-Girvan, Ph.D., MPH, SVP of Research and Strategy, Medable
Digital biomarkers could expose ‘simmering’ symptoms before they turn serious. Tech-minded epidemiologists are working to crack the code of digital biomarkers as they could signal—early on—important changes in the trajectory of a disease. Researchers are now able to capture quality data, on a longitudinal scale, and harness that data across groups of patients to establish accurate markers. The fall-out of COVID-19 has accelerated this effort as more companies take a decentralized approach to clinical trials and leverage remote data capture.
While the digital biomarkers that can be captured remotely are often constrained by cost and logistics, there is rich data that can be collected reliably with modern technology. For instance, a daily video that shows progressive changes to a Parkinson’s patient’s shuffle, a wearable device that measures hand tremors continuously, and a digital sleep monitor that tracks REM sleep cycles all inject new factors into the research equation to help determine accurate digital signatures. As critical, decentralized clinical trial (DCT) platforms enable researchers to collect higher quality data from patients in their natural environment when they are most relaxed, and eliminating the “white coat effect.”
As DCT adoption increases, science will start to be able to identify the early “simmering” symptoms that could turn into an adverse event later in the trial. Since the development of symptoms has the potential to jeopardize patient safety, instill fear and panic—patients could drop out of a trial prematurely. Identify these early, however, and the care team may be able to rapidly mitigate symptoms, prevent psychological harm, and avoid losing patient participation—especially if the incident was not caused by the investigative drug but rather, external factors such as dehydration or lack of sleep.
In 2021, the momentum of DCTs and virtual technologies will allow clinicians and researchers to better identify digital biomarkers to help improve care and maintain the integrity of vital clinical trials. It will be the beginning of truly proactive patient care that is mindful of individual situations—the ultimate in long-promised personalized medicine and patient-centricity
Vera Mucaj, Head of Implementations, Datavant
External control arms conferring full trial benefit on all patients will increase five-fold.
The backbone of the randomized controlled trial is the control arm. Despite the critical importance of a control, they can also be an obstacle to swift trial execution. Control arms require twice as many patients to be recruited for any given trial, consuming valuable resources and slowing trial enrollment time. They require that half of those recruited patients be treated with a placebo that has no impact on their disease, depriving the patient of a chance to benefit from potentially life-changing therapy, and researchers of the chance to study patient response and outcomes. When control arms are foregone entirely, as is increasingly common in fast-moving or terminal diseases, the statistical credibility of the study erodes.
In 2021, external control arms created using linked real-world data will more frequently provide a viable and effective alternative. Trial sponsors will be able to draw on real-world datasets describing patient demographic, disease, and outcomes data to assemble and study a cohort of untreated patients without recruiting any of them in real life. Without the need for an internal control arm, trial enrollment can be accelerated and all patients will be able to experience the potential benefits of the drug candidate. By comparing the treatment arm directly to the external control arm, researchers will preserve confidence in the treatment’s material benefit. In 2021, increased prevalence of external control arms will lead to faster trials, continued treatment efficacy, and greater patient benefit.
Jason LaBonte, Chief Strategy Officer, Datavant
Real-world data driven patient recruitment will substantially cut enrollment time and increase access. Patient recruitment is among the most difficult parts of clinical trial execution. Trial sponsors approach the same narrow band of clinical sites that treat the same pool of patients, many of whom are already enrolled, or have historically been enrolled, in similar trials. This makes it difficult for sponsors to identify new pockets of patients who are qualified fits for their trials, and slows down recruitment. This also excludes a large population of potential patients—those who aren’t represented at primary recruitment sites—and prevents them from benefiting from trial enrollment.
In 2021, the growing use of linked real-world datasets will address both of these problems. Linking data from centers of care treating new and historically under-represented populations will enable sponsors to identify and recruit from previously untapped patient pools. Accessing and qualifying patients who are likely not in trials already will drive multiple improvements in trial execution. More patients who meet qualifying criteria will become available for recruitment, making the enrollment process faster and more efficient. Patients who have been historically missed or ignored in trial recruitment will be able to access clinical trials, benefiting from innovative medicine and enabling researchers to learn from a more diverse array of outcomes. The result will be a substantial reduction in trial enrollment time, and a meaningful increase in the accessibility of clinical trials overall.
Updesh Dosanjh, practice leader, technology solutions, IQVIA
Virtual/remote clinical trials with AI-enabled safety monitoring will increase clinical trial participation. Virtual and remote clinical trials are helping to keep many medical advancements up and running during the global pandemic. Remote health care visits are on the rise amidst COVID. Remote visits accounted for more than 40 percent of primary care visits for patients with traditional Medicare, increasing by 400 percent (from a tiny 0.1 percent sliver before the public health emergency at the height of the COVID-19 shutdown).
COVID-19 has accelerated the need for a faster, more accessible clinical trial process. With the introduction of virtual or remote clinical trials, we’ll begin to see the concept of the continuous clinical trial evolve. This will include enhanced patient surveillance, or pharmacovigilance monitoring to ensure drug safety—even after the drug is brought to market.
Pharmaceutical companies will become more data science-oriented with advanced technologies to accomplish this, including automation and artificial intelligence (AI). This will help them continuously and proactively monitor patient safety data. By incorporating these automated technologies into the virtual clinical trial process, organizations can identify adverse event indicators swiftly and efficiently, as early as possible, to ensure the utmost patient safety.
In a COVID world in which there is a push to make telemedicine more widely available, automation and AI technologies will serve as a key indicator and communicator throughout the clinical trial process. As further evidence of the growth here, a recent poll of older adults by the University of Michigan Institute for Healthcare Policy & Innovation found that more than 7 in 10 are interested in using telehealth for follow-ups with their doctor, and nearly 2 out of 3 feel comfortable with video conferences. In addition, the government’s flagship health care program, Medicare, has temporarily waived restrictions predating the smartphone era. Medicare covers more than 60 million people, including those age 65 and older, and younger disabled people.
The benefits of virtual and remote trials go beyond patient safety monitoring. These more digitalized trials also enable bi-directional data sharing, reducing friction between collecting important data points and automating the assessment of a patient’s condition with these more informed and robust sets of insights.
Liz Beatty, CSO of Inato
Continued Focus on Diversity in Clinical Trials: The lack of diversity in clinical trials is finally being recognized as a societal issue. It’s also an issue that creates significant community and global health risks, as diverse patient populations have different responses to certain diseases and pharmaceuticals. Currently, diverse patient populations don't have easy access to the large academic sites and that needs to change. In 2021, there will be a larger systematic effort to diversify participants in clinical trials. More sites and sponsors will begin using data-driven metrics to track and progress against their diversity goals.
Sponsors will continue to look beyond large AMCs to conduct trials: The industry has tried to reverse the rising cost trends by partnering even more with the largest sites. However, reducing the number of sites per trial drives increased site competition, while increasing the number of patients required per site drives patient competition at the site level and limits widespread patient access. In 2021, We anticipate sponsors will continue to look at new engagement models with sites, and different technologies such as a marketplace to more quickly identify naive patients for trials.
Community Comeback: Patients are less able & willing to travel to participate in a trial (COVID + changes in patients’ habits & expectations). Addressing patient participation requires a reimagining of the way sites and sponsors collaborate on clinical trials. Trials need to adapt to patients—not the other way around. This means enrolling and treating patients where they are, in their communities. Thanks to COVID, 2021 will be the year where we finally bring clinical trials to the community—but it will take a shift in how sponsors partner with sites.
Dan Vahdat, CEO and Founder, Huma
Data collection is a huge problem for clinical trials. Right now, almost all of it happens at the research center running the trial and it’s slow and it’s expensive. With AI and technologies like ours at Huma, we can run site-less clinical trials where we can remotely collect the vast majority of data, potentially even 100% of it.
The move to remote clinical trials is a big shift, as you can collect a lot more data, a lot more easily and from a lot more people. Plus, you can also recruit and screen patients remotely. That means you can finish a trial, from A to Z, in almost half the time. You can see the impact in our clinical trial with Bayer and Stanford, where it took 5 weeks from signing to going live and in just one month we had recruited all our patients. We are part of the University of Cambridge’s Fenland study of 12,000 patients, and we got thousands of them onto our platform in a matter of weeks, and more are being signed up all the time. This wouldn’t be possible without remote working.
And then there’s COVID-19. We’ve been involved in national services in the UK, Germany and UAE and now with the world’s largest vaccine campaigns we are getting involved in studies that can only happen at the speed and scale needed because of AI."
Henry McNamara, GM and SVP of Oracle Health Sciences
The shift to decentralized clinical trials and decentralized remote patient monitoring is not going away. The lockdowns of the COVID-19 pandemic had a major impact on clinical trials in 2020, leading to rapid adoption of telemedicine and decentralized trial methods. This will continue in 2021—The industry is not going back to “the old way.” Therefore, sponsors and CROs need to set themselves up to operate successfully in this new model.
The rise in mHealth Data will demand changes to the current eClinical environment. Companies will increasingly turn to mHealth devices to gather data from patients remotely. This increased volume and variety of data coming from these devices will require unified technology platforms capable of collecting, managing, and analyzing it.
Automation will be used to ease the burden on study teams. The pace at which study teams have been working through COVID-19 trials is not sustainable. Despite interest in maintaining this “pandemic speed” of trials, staff simply cannot continue working 20-hour days, seven days a week, technology to streamline repetitive processes, automate trial design, and otherwise reduces the routine workload on professionals.
AI becomes mainstream. Artificial Intelligence (AI) pilot projects among sponsors and CROs in safety case intake and site selection have proven that AI is more effective and accurate than humans in these scenarios. These early successes mean that we’ll soon see AI go mainstream, with great expansion into these applications, as well as many more. As more companies lean into decentralized trials, and as the pace of trials and the amount of data that needs to be analyzed continues to increase, AI will emerge as the only way to process this data efficiently.
Regulators won’t stand in the way of decentralized trials. Regulatory guidance around decentralized clinical trials is widely seen as unclear. Moving forward, as situations require and technology enables a continued shift to decentralized trial methods, regulators will respond and accommodate the adoption of these methods