The clinical trial enterprise has been due for a reckoning for a long time. The good news is that a fundamental re-thinking of how clinical trials are done – one that could bring lasting positive changes – has started to happen.
I believe there is good reason to be optimistic that some long-overdue changes will be made permanent after the pandemic.
Consider the unprecedented mobilization of academia, industry, and government. As of May 11, there are more than 144 active trials of therapeutic agents with another 457 development programs for therapeutic agents in the planning stages for COVID-19 treatments, according to the U.S. Food and Drug Administration (FDA). Guidance by FDA and others to help improve COVID treatment trials are crafted and made publicly available in weeks rather than the usual process that can take years.
The sense of urgency is palpable. It’s visible how quickly sponsors, investigators and regulators are moving through every step of the process – including development of protocols, drafting of informed consent forms, IRB approvals, training at clinical sites, all the way through enrollment of subjects.
We’ve seen regulators providing detailed feedback on proposed clinical trial designs in days rather than weeks. IRB approvals, which can notoriously sometimes take weeks or months for the most minor modifications to study protocols, are now sometimes coming in as little as 24 to 72 hours.
One of the most inspirational examples I’ve seen, has been with the RECOVERY master protocol. That study, intended to evaluate a handful of different therapeutics for COVID-19, went from draft protocol to first patient enrolled in a lightning-fast nine days in the UK.
In the US, similar stories are emerging, COVID-19 hospital-based studies are moving from initial conception to first dose in patients within a week or two rather than the average 3+ months needed to get through ethics review and contract processes at most academic, VA and hospital-based sites.
A new sense of urgency
This rapid pace of research in times of COVID-19 is new, for obvious reasons. The FDA was charged, via the Kefauver-Harris Amendment of 1962, with the responsibility to ensure that each treatment approved for medical use in the US, be shown to be both safe and effective.
Randomized clinical trials are the best form of study for understanding benefits and risks of medical products, in light of that legislation. But they aren’t perfect. Randomized controlled trials have become prohibitively complex, expensive and slow to start and finish. The US, in particular, is a highly litigious environment, and institutions are quite mindful of that when signing on to study protocol contracts and informed consent forms. Those costs, and risks, act like sand in the gears of researchers and patients seeking answers to important scientific questions.
So what was the typical process for starting a clinical trial, pre-pandemic? An investigator or team of investigators would draft a written protocol (and supporting documents) which formed the basis (like a recipe book) for a clinical trial. Once written, several committees were consulted, and several levels of approvals to start the trial had to be obtained: for example, hospital committees and institutional review boards (ethics committees), who are tasked with protecting the welfare, rights, and privacy of participants in clinical trials.
All of this is done in collaboration with the sponsoring entity — a company, a government agency, a foundation, or some combination of all three. There are also regulatory bodies — sometimes multiple regulatory bodies — that weigh in on whether the study design is valid, or could use some revisions.
Before you know it, a series of well-intended steps turned into a straitjacket of red tape that reined in creative thought and slowed down the enterprise to the point where it could actually undermine the mission of the trial — which is supposed to seek a valid answer to an important medical question.
Luckily, for COVID-19 treatment trials, some of these processes have gone at unseen speed. Sadly, for most other trials in the pre-COVID-19 era, this sense of urgency — and resulting rapid advancement — was lacking. That meant patients with cancer or genetic diseases or other deadly or disabling illnesses simply had to wait much longer than they should have because of a dysfunctional process.
This requires high quality clinical trials
Now that we have speed, it is even more critical to do the right trials, the ones that can get us to reliable answers for treatment options. Speeding up the process only to do “small uninformative trials” — those that are not designed to get us to reliable answers while keeping participants safe — is not progress.
At the Clinical Trials Transformation Initiative (CTTI), where we live and breathe improving clinical trials, we’ve created resources specifically for COVID-19 treatment trials. Previously, we also developed recommendations to design high quality protocols. You can accomplish this by engaging everyone including patients as equal partners in clinical trial design, and gathering feedback on symptoms that are important to their quality of life, but which don’t always get factored into a study’s primary or secondary endpoints.
We recommend incorporating digital health technologies for endpoint capture into trials, and conducting more decentralized clinical trials where the participants can stay at home, so they can enter a study without having to worry about the risk of entering a building that could be a COVID-19 hot zone. Sometimes, information can be simply gathered from home; other times a telemedicine visit can serve as a substitute for an in-person visit, while keeping the study on track.
Before COVID-19, barriers to change were mostly cultural (“This is how we’ve always done it.”) and risk aversion (“What bad thing might happen if we change anything?”)
The status quo is generally considered the default “safe” option.
No more excuses, trials can and need to be done differently
COVID-19 has shown us we cannot run trials in this way, and consider that the “safe” option. An FDA guidance, created for trials in progress during COVID-19 was released only weeks into the pandemic. Everyone recognized they must rise to the challenge, and that means discarding a lot of time and process that wasn’t actually all that necessary. Because of shelter in place regulations, active trials had to be stripped of non-critical elements and slimmed down to what matters that what could be captured in alternate ways answering the most important research question at hand, minimizing burden on participants and research sites. Other ongoing trials were halted.
With 1,000 to 2,000 people in the US dying every day and the very real prospect of a second wave in the fall, we need to keep moving at this breathtaking speed to match the medical need for treatments and vaccines. Simultaneously, we need to adapt medical practice in real-time. New treatment guidelines should be based not on anecdotes or a case series of similar anecdotes, but on methodical and rigorous medical evidence that is the gold standard.
We need to design large, well-powered and streamlined randomized, controlled trials that are feasible and simple for health care workers to conduct, now for COVID-19 trials as well as future trials in other disease areas. We need to identify multiple sites rapidly, and we need to test multiple potential treatments at once rather than one at a time. How one can go from single therapy to cocktail therapies will also need to be considered, as was the case with HIV in the 1980s.
Focus on what matters
Fortunately, a framework exists to aid in designing high-quality clinical trials. It’s best known in the industry as quality by design or QbD. The key principles of QbD are to focus on things that matter. Quality does not have to slow things down. By focusing on quality in trial design, urgently needed efficiencies are created. Those efficiencies can help to rapidly launch protocols so trials can begin accruing patients and minimize extra steps and contact between people.
Specific to COVID-19 treatment trials, a recent FDA guidance document outlined what efficacy endpoints could be considered, what matters, and how they vary depending on the severity of the disease. We should focus on getting the answer to those data points — and those data points alone — unless there is a very good reason. The endpoints are objectively measurable and meaningful — reduction of hospitalization rate for an out-patient participants with mild to moderate symptoms of COVID-19, or time on a ventilator or all-cause mortality for trials of more severe patients. Trials should be powered to answer those questions — no more, no less.
Drugs in earlier development may need to be tested in smaller-than-usual proof of concept trials. It really is a question of matching the design of your trial with the purpose of your trial. This is not the time to overload clinicians or burden participants with excessive procedures and requirements that are not directly tied to the endpoints. Doctors and nurses now have their hands full, and simply cannot deal with excess data collection.
A single center trial, no matter how good the idea, will not get us reliable data that can be applied more broadly. This is the time to collaborate to not do trials the way we have always done them because it is comfortable, and because change is hard.
Uncommon circumstances require a new way of thinking.
We can create additional efficiencies by rethinking ways to collect informed consent electronically and run trial data embedded in e-health records. Randomization of treatments to a standard of care is the only way to determine benefit of the studied treatment over the standard of care.
There is power in collaboration
We also need to create a site identification system in a nimble, systemic approach that leverages existing networks such as hospital systems, physician groups and NIH trial networks, rather than identifying one site at the time. Currently, sponsors of COVID-19 treatments and vaccines are seeking trial sites in hotspots, which can often be moving targets. We must be positioned to respond to the pandemic in real time and not wait for it to come to a site near you.
When considering the urgent need for therapies and the number of drugs that could be tested, there has been great interest in using master protocols to simultaneously test multiple products to get answers as quickly as possible and not have to worry about picking a “winner” in your single treatment one-off trial. Master protocols have been used to speed the development of products for diseases such as cancer, and trial experts believe they can be useful for pandemics, including this one.
Sponsors should consider master protocols that enable multiple drugs to be tested against a control arm. One international preparedness trial in pneumonia that existed prior to the pandemic, REMAP-CAP, is using a novel “learning while doing” approach. The trial, developed by researchers worldwide including at the University of Pittsburgh School of Medicine, was able to offer an elegant solution by adding a COVID-19 arm to the protocol to determine whether to quickly adopt new “off-label” therapies during a pandemic, such as the anti-malarial drug hydroxychloroquine, or wait until they are tested in longer clinical trials. In addition, it uses an efficient machine-learning model and is embedded in the electronic health record.
While there are many promising trials underway for COVID-19, they are not all high-quality and expected to provide reliable results. There are many small trials of just a few patients, enrolled at a single site, that aren’t randomized with a control group. These trials are cheap and relatively easy for a lone wolf investigator to start up, but they are ultimately a waste of everyone’s time and effort. They will not provide the steadfast scientific answers we need to help a large percentage of the population.
On the other hand, there are many worthy master protocol trials that are testing potential COVID-19 therapies. What we need is coordination between these master protocols. They should do the following:
- Enable alignment of endpoints. Are we looking for COVID-19 treatments to provide a mortality benefit, to reduce hospitalization time, or something else? We should have a consensus.
- Ensure treatments for all types of COVID-19 patients – exposed, mild, moderate and severe disease – are being tested.
- Share process and data. Hoarding raw patient data at individual sites, in proprietary EHR systems that don’t interoperate, has got to go. Not sharing protocols, consent forms and other critical documents will stymie progress.
- Compare and, if needed, combine results across trials, as in meta-analysis.
We cannot go back and lose this progress
We have the opportunity to reconsider and optimize our whole evidence generation system and we should. Some of these changes ought to become the new standard procedures of the future.
Our challenge now is to transfer some of that same speed, that same critical thinking, we have been forced to do in this emergency, to the next set of non-COVID-19 trials.
Once these fundamental changes make trials leaner and meaner, we should then be better poised after this pandemic to tackle some of the most pressing global public health challenges.
If we’re smart, history will look back and see the beginnings of a meaningful scientific revolution.
Pamela Tenaerts, MD, MBA, is the executive director of the Clinical Trials Transformation Initiative, a public-private partnership involving all the parties as equal partners in clinical research (patients, academia, government, investigators, sponsors, technology, IRBs, etc), co-founded by Duke University and the U.S. Food and Drug Administration to develop and drive adoption of practices that will increase the quality and efficiency of clinical trials.
Views expressed in this publication do not necessarily reflect the views of CTTI nor do they represent the views or policies of FDA or HHS.