Rules And Roles Of Decentralized Clinical Trials Coming Into View

By Deborah Borfitz

March 1, 2022 | The current regulatory landscape for decentralized clinical trials (DCTs), and a trio of use cases focused on medical devices and in vitro diagnostic (IVD) products, were among a bevy of presentations on the decentralized study model at the recent Summit for Clinical Research Ops Executives (SCOPE) in Orlando, Florida. Adoption of DCTs accelerated exponentially when the pandemic hit a long two years ago and health authorities around the globe seem to recognize both the promise and perils, according to Scott Askin, global program regulatory director for innovation at Novartis.   

The main concerns of health authorities are that DCTs be compliant with the ICH E6 Good Clinical Practice (GCP) Guideline ensuring human subject protection and high-quality data, Askin says. Importantly, study investigators need to maintain their oversight responsibilities even when home nursing is involved. 

The overall value of DCTs won’t be established for at least a couple of years once enough of them have been subjected to full regulatory scrutiny, he notes. 

Both the Danish Medicines Agency and Swedish Medical Products Agency have programs supporting implementation of DCTs, Askin continues. A pilot program last year in Sweden offered free scientific advice and shared learnings, and the knowledge exchange is expected to continue. 

Industry’s focus has been on patient perspectives and terminology, he adds, while the two main collaboratives—the Clinical Trials Transformation Initiative and TransCelerate—have been working on modernizing studies in general to improve their quality and efficiency.

Denmark released the world’s first DCT guidance on May 4, 2021, which was updated in September, says Askin. In another first, Denmark’s ethics committee released its own guidance a few weeks ago. In Switzerland, meanwhile, a position paper was jointly released last year by its health authority and ethics committee. 

The expectation is that all of this will feed into European guidance to be published in the first quarter of 2022, “any day now,” Askin says. In the U.S., DCTs are on the list of proposed guidances of the Food and Drug Administration (FDA) for this year and industry is eager for the input. 

Convergence on DCT regulations is needed and, ideally, will be harmonized, says Askin. Smaller nations typically adopt the guidances of bigger countries, he adds, and South American countries, as a rule, wait on the lead of the FDA. Neither China nor Canada have released DCT-specific guidance but have talked about the benefits of DCTs. 

The Trials@Home consortium is defining best practices for the conduct of DCTs across Europe where approaches are varied but some common themes have emerged among larger nations, says Askin. With no common standard, decentralized trials on a global scale are difficult. 

The significance of the DCT model became clear when the pandemic struck and ushered in a lot of temporary, COVID-related guidances that many would like to see implemented over the long term, he says. 

To address barriers to scaling DCTs, Askin suggests a flexible approach and taking “baby steps” to demonstrate their value via implementation. “We need to be GCP-compliant even if that is not always well defined [for decentralized studies] and ensure investigators demonstrate oversight, especially with patient safety in mind.” 

Additionally, he says, digital endpoints (e.g., sensor-generated data and electronic patient-reported outcomes or ePROs) collected outside of a clinical setting need to show comparability to similar measurements taken in a more controlled and predictable environment. 

Some study participants may need to visit a physical site, more as a matter of preference than regulation, says Askin, pointing to the necessity of hybrid trials. “We have not flipped the switch on regulatory acceptance,” he concludes, adding that communication channels are needed with both ethics committees and health authorities. 

Weighing the Options    

Accelerating adoption of decentralized and fully virtual trials was the topic of a presentation by Jane Hart, senior director of clinical affairs at Exact Sciences, best known as developer of the at-home colon cancer screening test Cologuard. Hart has over 25 years of experience in clinical research, the last seven in medical devices and now IVDs specifically. 

Medical devices and IVDs tend to be a tad behind pharma in the way studies are conducted, Hart says. Companies, including Pfizer, led the way on the decentralized front nearly two decades ago, she adds. Exact Sciences, for its part, is now embracing DCTs and risk-based monitoring in a big way.

DCTs present opportunities to address the limitations of the traditional, resource-intensive study approach—notably, recruitment that is limited by geography and the ability to get patients to sites, says Hart. Subjects not previously exposed to research can be identified, potentially improving trial diversity, and required procedures and assessments can be more efficiently done by a team of local providers. 

“It does not have to be an all-or-nothing situation,” she says, noting that decentralized components can be incorporated only where they make sense. The upsides of DCTs include the reduced cost of remote data collection, increased access to patients, simplified site management of one “meta-site,” decreased study burden for patients, and the ability to engage patients virtually to improve retention.

On the flipside are the challenges: increased reliance on patient-reported data, a potentially higher screen failure rate, and the required infrastructure to get started that can be harder for smaller companies, Hart continues. DCTs also introduce some data-related privacy and compliance issues that need to be considered, together with means of engaging medical staff and patients that don’t come off as “completely robotic.”

On the lengthy list of enabling technologies and processes are the virtual physician investigator (PI), digital and self-reported prescreening tools, remote electronic medical record review, eConsent, ePROs and clinical outcomes via a tablet or web portal, telemedicine visits with follow-up by a healthcare professional, data capture tools, remote and home health nurses, wearable devices, a subject payment system, and centralized monitoring (including risk-based monitoring). Many of these are useful in the IVD space for trials involving a single visit and specimen collection, Hart says, but less so for long-term oncology, cardiac and wound care studies. 

She offers an equally long checklist of considerations for determining if the decentralized approach is suitable for a specific study. For example, is data mining needed to identify patients with a disease or will it be enrolling healthy volunteers? Does eligibility need to be verified by a physician? Does data collection need to be patient-reported or investigator-reported? Is documentation needed in medical records to support that data? How will specimen collection be done? Will data streams for the study be integrated or separate? 

Risks And Rewards

The first two case studies presented by Hart were DCTs conducted by 3M, where she worked as vice president of global clinical development until early last year. The first was evaluating the choice of different screenings and used eConsent on a tablet, and electronic screening and ePRO either on patients’ smartphone or a web browser. Patients could view educational videos on their device to learn about the trial, were randomized into one of three arms, and took an electronic follow-up survey at one year. “It was pretty straightforward,” she says. 

In the second instance, a hybrid decentralized model was used, Hart continues. The study had a central virtual PI with a U.S. site network vendor, targeted 8,000 diverse subjects, and maintained the doctor-patient relationship at sites. “It was scalable for us and quite successful in the U.S.,” she says, noting that the vendor identified patients (who then reached out to in-network sites) and took care of much of the collection and tracking of data. 

Exact Sciences conducted a completely virtual trial under a single virtual PI that enrolled 5,000 colonoscopy-naïve subjects via a national ad campaign, she shares. Participants had blood draws taken by remote home health staff and study nurses were available for support through a call center.

Recruitment happened via the social media sites Facebook and TicTok, inviting people to log onto a patient platform to complete a questionnaire. Being full virtual, patient flow was “fairly complex,” says Hart.

One vendor had responsibility for digital recruitment and patient-facing ads and set up the screening questionnaire and sent out email notifications confirming eligibility with next-step instructions, she says. Through an app on their phone, individuals could self-enroll with eConsent to be sent a study kit and the home health group was notified and assigned a nurse to participants based on their ZIP code to do the mobile phlebotomy. 

Patients received half of their promised compensation when they consented to be in the study, and a second payment was triggered once their sample was collected and shipment was confirmed, explains Hart. 

“The takeaway is there’s a way to do [a fully virtual trial],” she says. “It’s the largest study we’re done with this model.” 

In all three DCT instances benefits were realized, says Hart, including the rapid enrollment of 3,000 subjects in under three months at the end of 2021. Data cleaning timelines were also notably faster.

Exact Sciences is particularly proud that it succeeded in increasing diversity with the virtual trial, she adds, with a 40% jump in the enrollment of Black participants. Proportionally, they represented 14% of all study subjects—well above the 12.4% they constitute in the U.S. population at large. 

“We didn’t meet the census proportion for Asians and Pacific Islanders, but we doubled [their representation] and came very close,” says Hart. 

Her offered recommendations are to ensure a clear understanding of the role of vendor partners—which in this case included the virtual PI having site responsibilities—and build that into contract language. “That piece of the contract took longer than anticipated.” 

When selecting vendors, she continues, it’s especially important that they’re aware of data security and cybersecurity risks and have prevention and mitigation strategies in place. The major issue here is how to ensure patients are real and not bots designed to enrich nefarious individuals, particularly if compensation is being offered for participation.

The virtual trial at Exact Sciences initially screened a few of the same patients, says Hart, a problem that was quickly identified and mitigated with the conditional payments and by capturing the IP server and email addresses of people entering data to flag any duplicates.