Viruses like SARS-CoV-2 can circle the globe with astonishing speed by taking advantage of human networks. The global medical research community couldn’t immediately contain it because it had no comparable network for defense — but we are moving quickly to create one. Employing adaptive clinical trials will help.
The initial response to Covid-19, driven by the best intentions, was to stand up more than 1,000 clinical studies. Many were small, uncontrolled, and unlikely to yield reliable information. This is one reason vaccines and definitive cures continue to elude us.
Even so, researchers have scored some remarkable wins. Less than a month after the first Covid-19 cases surfaced in late December, a research team sequenced the virus. Since then, other teams have identified cellular proteins that bind to it, shared data in open-access journals, and kicked off at least two dozen credible Phase 3 trials for vaccines, antivirals, and other treatments.
More milestones will follow as we expand research networks and exploit major innovations in clinical trial design. One advance in particular — an iterative, long-duration study known as an adaptive platform trial — is helping researchers up their game. The use of adaptive clinical trials allows them to simultaneously test multiple interventions against a single, shared control arm, add and eliminate treatments as the trial progresses, and update the study design as the treatment landscape changes.
In contrast, each traditional standalone Phase 3 trial must recruit patients to its own control group and test just one hypothesis. Even designing in techniques such as subset analysis and adaptation, these are extravagant undertakings that can’t match the supple efficiency and productivity of well-managed adaptive platform studies.
The adaptive clinical trials approach has already led to a series of high-value, actionable research results:
- In April, data from an adaptive platform study called the Adaptive Covid-19 Treatment Trial, or ACTT, from the National Institutes of Health, validated the clinical benefits of remdesivir, an antiviral agent, in patients hospitalized with Covid-19.
- A few weeks later, the similarly-designed Randomised Evaluation of Covid-19 Therapy (RECOVERY) trial in the United Kingdom demonstrated the benefits of dexamethasone, a widely used steroid.
- RECOVERY further showed that the malaria drug hydroxychloroquine provides little help for people with Covid-19, an invaluable contribution in the face of public confusion and pressure to use the medicine. It also showed that a combination of HIV treatments, lopinavir and ritonavir, was ineffective in treating the disease.
Other adaptive platform studies are launching as we write this. A coalition in which we participate, the Covid R&D Alliance, will soon share details about a large-scale adaptive program that drew inspiration from the U.K.’s RECOVERY project and from an initiative launched by the University of California, San Francisco, called I-SPY II.
The I-SPY II teams recently embarked on a Covid-19 trial involving several members of the Covid R&D Alliance. Another pandemic-centered program called REMAP CAP, which had been conceived before the pandemic emerged for just such an eventuality, has enrolled nearly 1,300 patients at 242 sites in 15 countries.
Imagine how the Covid-19 picture might look today if initiatives like these had been staged in advance — anticipating an inevitable pandemic — and networked into a globe-spanning adaptive platform ecosystem.
Such an uber-program would allow researchers on every continent to pool knowledge, data, analytics, materials, and other resources, and adding and deleting trial arms based on interim readouts, real-world data, and insights from the field.
Adaptive platform trials are not new. The methodology was already gaining traction in oncology and other areas before Covid-19 struck. They’re a subset of master protocol trials that Janet Woodcock, the Food and Drug Administration’s long-serving director for drug evaluation and research, and Lisa LaVange explored in a review article in the New England Journal of Medicine in 2017. This May, the FDA’s updated guidance on Covid-19 drug development recognized the potential of these trial designs, and the agency is likely to issue further recommendations.
Because the Covid R&D Alliance sprang from the crucible of this pandemic, its origin story is unlike any other: In the second week of March, with outbreaks raging on three continents, research heads at Amgen, AstraZeneca, Bristol Myers Squibb, GlaxoSmithKline, Johnson & Johnson, Novartis, and Takeda began exchanging information. We were soon joined by R&D heads of other major biopharma companies, as well as leaders across the biotechnology ecosystem.
Struck by the magnitude of the pandemic, our shared responsibility to act, and the scale of resources at our disposal, we communicated with minimal ground rules or administrative red tape. The alliance immediately began screening compounds for antiviral activity, building on work that members such as Novartis already had initiated.
Over the next two months, we put together a centralized screening mechanism for treatment candidates and started evaluating agents not only from our own pipelines but those submitted from the field.
We recognized that data-sharing protocols needed to change. The required summary-level study updates on ClinicalTrials.gov often lag a year behind data collection and analysis. And what researchers seek above all is patient-level data that currently isn’t available there.
Utilizing new analytic tools from the Bill and Melinda Gates Foundation as part of the International Covid-19 Data Research Alliance and Workbench, members of the Covid R&D Alliance and others will be able to look at summary-level statistics with a lag of just seven days. They will also be able to interrogate patient-level information from Covid-19 trials through a data-sharing platform created by TransCelerate, another industry consortium.
Covid-19 has also spurred collaboration around the tidal wave of real-world data from hospital records, insurance claims, and other sources. Real-world data makes it possible to design adaptive trials around a disease that is still new and poorly understood. Acting alone with proprietary analytics, no team could make sense of the data deluge. Through coalitions, though, we can divide up responsibilities coherently and share the analysis.
From the perspective of people afflicted with Covid-19 and their families, this essay paints too rosy a picture. Even as the global research community pushes toward new frontiers, we are saddened over the loss of life that might have been averted, and that continues today.
Viruses leverage animal and human networks. With an adaptive research ecosystem, we could begin to match them in speed, mobility, and evolutionary vigor, and ultimately keep the world safe.
