Why COVID master protocols haven’t been productive in the U.S., and how to fix it
The U.S. needs to prioritize master protocols for COVID and beyond
The U.S. needs to prioritize master protocols to generate data for COVID and beyond.
If the U.S. hopes to generate substantial, meaningful data on therapies for COVID-19 and future pandemics, it will need to overhaul the clinical trial ecosystem to prioritize master protocols.
At least 10 master protocols for COVID-19 are up and running in the U.S., but nearly a year into the pandemic most have yet to produce actionable data.
The problem isn’t the trial design. There’s no doubt master protocols are the fastest, most efficient path to data that can inform clinical practice — the U.K.’s RECOVERY trial is an example of how master protocols are supposed to work.
With over 31,000 participants and counting, and 177 active sites throughout the U.K., the study has produced data on four therapies that have informed global authorization and approval decisions.
But trials run in other regions — especially the U.S. — haven’t been able to replicate the success due to slow site set-up and problems with patient recruitment.
During a webinar hosted by the Clinical Trials Transformation Initiative this month, Janet Woodcock, Robert Califf and other key opinion leaders joined lead investigators on COVID master protocol studies to outline the challenges of running the multisite studies and map out solutions.
Woodcock, who serves as acting director of FDA, stepped away from her post as director of FDA’s Center for Drug Evaluation Research last year to lead the therapeutic aspects of Operation Warp Speed. Califf, a former FDA commissioner, is currently head of clinical policy and strategy for Verily Life Sciences LLC and Google Health.
In the short-term, sponsors are streamlining contracts to accelerate new trial site recruitment, providing staffing resources and decentralizing trials from major academic medical center. But a long-term fix will require setting up systems that prioritize adaptive platform trials and others likely to produce meaningful data.
Major challenges of master protocols
The problem in the U.S. comes down to its clinical trial ecosystem.
There are too many clinical studies, most of which are poorly designed and won’t produce useful data, and they’re diverting resources away from the high-quality trials.
“One of the things we’re also seeing in the U.S. and potentially worldwide is a shortage of supplies and ancillary support of trials,” said Woodcock. She called out big delays in getting lab kits for studies and competition for personnel at trial sites.
Among the 558 interventional trials for COVID-19 treatment and prevention with sites in the U.S., just 5% are sufficiently powered to answer their intended question, according to an FDA analysis referenced by Woodcock and Califf during the webinar.
“Many of the studies we saw were not randomized. Given the variability of the disease and the size of the treatment effect we’re seeing, there’s very little hope of yielding actionable information,” Woodcock said.
Most of the properly designed studies are master protocols.
There has been an uptick in the number of master protocols for COVID-19 over the last six months, with at least five new studies starting enrollment, but most of the master protocols haven’t enrolled enough patients to meet their potential.
Woodcock added that there are few examples of any trial type with adequately powered arms enrolling well in the U.S. “Although these aspire to adequate power, they did not achieve it in a timely manner.”
“If we had more focus and discipline in how we set up the trial infrastructure globally, we would be a lot further along in treating this disease.”
The eight master protocols included in an analysis referenced by Califf have activated an average of 26% of their target number of sites in the U.S., and have enrolled an average of 20% of their intended participants as of the second week in January.
Clinical research is highly concentrated at major medical centers in the U.S., which deepens the competition for trial participants and resources.
“We also need more sites recruited into these meaningful platform efforts, and that means non-academic sites. These are sites that are not motivated by academic publications but clinicians who are very busy, especially now, who are motivated by impact on their patients,” Mark McClellan, founding director of the Duke-Margolis Center for Health Policy, said at the meeting. “Here in the U.S., the largest supply of COVID patients in the world, at most a tiny fraction of 1% of those patients are participating in meaningful clinical trials.”
The resource scarcity is compounded by a healthcare culture in the U.S. that doesn’t place high value on clinical research and a fragmented network of hospitals and healthcare providers that makes setting up sites challenging.
Each trial site requires independent contract negotiations, personnel training and a regulatory green light.
While RECOVERY, REMAP-CAP and other international master protocols studies were created or adapted to COVID in the weeks following the pandemic declaration, most U.S. studies took six months or more to get off the ground. That includes studies that were in the planning stages as the pandemic began.
“If we had more focus and discipline in how we set up the trial infrastructure globally, we would be a lot further along in treating this disease,” said Woodcock.
Steps to a solution
Standardizing contracts and simplifying protocols are the first steps to improving master protocol execution, and many sponsors are already implementing them.
Contract negotiations can be a rate-limiting step to onboarding trial sites, so the RECOVERY trial uses a standard, take-it-or-leave -it contract to gather sites quickly, said Martin Landry, University of Oxford professor and co-chief investigator of the RECOVERY trial. Several other study leaders said they use a similar contract strategy.
Laura Esserman, PI on the I-SPY COVID trial and a professor at the University of California, San Francisco, added that using a single IRB can greatly streamline processes in a complex, multicenter study.
Another step sponsors are taking is finding ways to accelerate personnel training at new sites.
For example, NIH’s ACTIV-2 is providing additional staffing resources, including mobile units of trained physicians. For the outpatient study, the institute also has a contracted home health agency to minimize in-person visits and the drain on hospital resources, said ACTIV-2 study leader Manizhe Payton, who directs the Office of Clinical Oversight at the National Institute of Allergy and Infectious Diseases’ Division of AIDS.
An area where sponsors are just starting to make progress is minimizing a trial’s burden on physicians and patients.
“Our health systems know that what they’re good at high-quality supportive care. Any stretching we do to help answer new questions about novel therapeutics is going to have to play well with that core mission,” said Samuel Brown, a physician at Intermountain Health Center.
That’s one of the factors behind RECOVERY’s success, said Landry.
“A simple and practical design makes it as close as you can to be as easy to randomize a patient as it is to just apply a drug and hope it works,” he said. Simple endpoints that give a clear result are also key to easy integration into clinical care, he added.
The REMAP-CAP study, which has been particularly effective in starting U.S. sites, also relies on seamless integration into standard clinical care.
“We have to address the somewhat hostile environment in many academic health centers toward large multicenter trials.”
Derek Angus, who heads the REMAP-CAP trial in the U.S., told the webinar that the study’s protocol is embedded into the electronic health records of one hospital system, which has allowed the trial to run from some sites that have never done clinical research before.
“You ask the patient one thing, which is: would you be interested in hearing about treatment options? At that point the whole thing is then automated virtually for all steps of enrollment,” he said.
REMAP-CAP moved from zero to 30 U.S. sites “in rapid succession,” said Angus, who chairs Critical Care Medicine at University of Pittsburg Medical Center.
Embedding protocols is a good solution to the lack of experience and resources for clinical research at community medical centers, but it comes with high costs that aren’t always in the budget. Brown said it can cost $100,000 to develop and maintain the platform IP for an embedded trial.
Tweaks to the execution of individual trials are a start, but creating a step change in the U.S. will probably require developing a system that allows hospitals to prioritize high-quality studies while changing the perception of clinical research at major medical centers.
Even if trial sponsors make their way into new sites, they’ll still be competing for resources with poorly designed studies. That’s especially true while clinical research is still centralized to major medical centers.
What the U.S. needs is a way to help hospitals triage clinical trial opportunities to focus on those most likely to change standard of care.
Adrian Hernandez, director of outcomes and health services research at Duke Clinical Research Institute, proposed a policy similar to the Defense Production Act, which allows federal agencies to prioritize contracts for medical supplies, for example, as a way to set clinical trial priorities.
Landry called on every site enrolling COVID master protocols to close one trial that’s not adequately powered and not randomized to boost enrollment in high-quality COVID-19 treatment trials.
Dan Cooper, chief of the pediatric pulmonary division at the University of California, Irvine, called for better education on the importance of clinical research.
“We have to address the somewhat hostile environment in many academic health centers toward large multicenter trials,” he said. “Neither medical students nor nursing students really get a sense of the critical nature of clinical research as part of their roles of taking care of patients.”
“If we had as a core of training for all of our healthcare providers the idea that research is essential, we would be much better positioned,” he added.
The problems with the clinical trial ecosystem have become apparent through the COVID-19 pandemic, but Califf told attendees they affect clinical research broadly.
“I don’t think this is just about the pandemic. I think the same can be said for the common chronic disease — an overwhelming tsunami that is building up is going to hit us when we finally do turn the corner of the pandemic,” he said.