A trial run

War is a popular metaphor in times of crisis, and it was used liberally in 2020 as Covid- 19 wreaked destruction on communities and economies. As conflicts so often do, the first pandemic of the digital age inspired strategic and technological developments everywhere, but few are likely to have as great an impact as those affecting clinical trials. Indeed, history may judge Covid-19 as having ushered in a brave new world of research, in which treatments and vaccines can evolve at unprecedented – and often bewildering – speeds.

In the first half of 2020 alone, more than 1,000 clinical trials focusing on Covid-19 were registered worldwide. The overwhelming majority fell by the wayside, their good intentions punctured by poor designs. There have, however, been successes in both vaccine development and in the identification of therapeutics that have played an important role in curbing death rates during the so-called second wave. University of Warwick professor Dr Ayfer Ali, who specialises in the patenting and licensing of new technologies in the pharmaceutical sector, says adaptive clinical trials have made a telling difference.

As the name suggests, adaptive trials permit changes to be made to trial components throughout the study. Unlike the more traditional approach, whereby data is analysed at the end of a study, adaptive trials are designed to facilitate ongoing learning. As the trial progresses, data is assessed and used to decide whether elements such as sample sizes, allocation ratios and eligibility criteria should be altered in line with predetermined protocols.

Ali cites the University of Oxford-based Recovery trial as a prime example of best practice. It combines a clearly communicated and detailed protocol with transparent results, and has focused research efforts on one well-designed trial instead of hundreds of smaller, lower quality studies.

That’s all the more remarkable given the speed at which the Recovery trial was set up – it went from conception to launch in just nine days. From there, a digital system that was accessible 24 hours a day was used to recruit more than 10,000 patients in 176 hospitals in barely two months. The protocol was simplified and highly focused, while the consent process was optimised and data collection streamlined.

Central to its success, however, has been its adaptive design, allowing for multiple treatments to be studied concurrently under the same framework using comparative end points. This maximises time and resources, and supports straightforward comparisons between therapeutics. Treatments that fail to meet expectations can be removed from the trial, while new options can be added and tested without the time, effort and expense of setting up a new protocol. For promising treatments already in the trial, variables can be altered to more accurately pinpoint subgroups with the most favourable responses. Adaptive design also allows for a smoother transition from phases I to III, essentially facilitating one long trial rather than three separate studies.

“Recovery’s adaptive trial design in terms of multiple arms and interim analyses has allowed the abandoning of inefficient drugs and the quick communication of successful treatments, such as dexamethasone,” says Ali. “It also allows for the addition of new therapeutics as our knowledge of the disease develops.”

Such models provide opportunities to improve therapeutic outcomes, but Ali warns that not all adaptive designs are created equal. “We need to be careful not to encourage adaptive trials where the adaptations are ad-hoc and designed to serve a specific agenda,” she stresses. “That can create not just confusion about the quality of the evidence, but also distrust in the results.”

Unity in diversity

Head of clinical development at the Coalition for Epidemic Preparedness Innovations (CEPI) Dr Jakob Cramer agrees adaptive trials, as well as other efforts to combine a diverse range of approaches in one orderly strategy, have been crucial to tackling Covid-19. “Transmission of SARS-CoV-2 is constantly changing, so it has been important to maintain flexibility in conducting clinical trials and analysing data as they accrue,” he says. “Having seen adaptive trial design in action in response to Covid-19, I predict this type of design is likely to be executed more widely in response to other emerging infectious disease outbreaks in future.”

He’s not alone in making that prediction. The use of more flexible designs has indisputably borne fruit, as have decisions to back different types of vaccines. The timeline from the sequencing of the SARS-CoV-2 virus to initial phase III vaccine results was barely ten months – 309 days – and mRNA vaccine technology has finally broken through. “These novel platforms are likely to be used more frequently in the future to accelerate response efforts,” says Cramer.

Meanwhile, dozens of therapeutics have been analysed and either approved or discarded in time to make a tangible difference in a rapidly developing situation. Ali is particularly heartened by the innovations in quickly repositioning pre-existing drugs. “One thing that Covid-19 has brought to the forefront is drug repurposing and specifically the testing of drugs that are used off-label in properly designed clinical trials,” she says. “It would be great if those efforts continued.”

Equally, advances in adaptive study design are allowing earlier access to entirely new treatments, says Dr Lucy Vereshchagina, vice-president of science and regulatory advocacy at Pharmaceutical Research and Manufacturers of America (PhRMA). “Complex clinical trial designs and innovative drug development tools enable studies to evolve as more information is gathered, creating efficiencies that can streamline drug development time,” she explains. In particular, she highlights how adaptive trials can be combined with expanded clinical data analysis techniques to monitor treatment response times, making clinical development nimbler and more costeffective by concentrating effort on optimising the most promising therapeutics.

Ongoing data analysis and trial reviews not only save time but also have direct benefits for patients in the trial population. Firstly, the ability to eliminate ineffective therapies as soon as sufficient data is available minimises the time they waste not receiving the best care. Earlier clinical analysis can also mean a quicker understanding of patient response and fewer patients being randomised on less effective doses or treatments.

Continuing to standardise the assessment of data from trials would also be beneficial, adds Cramer. One collaborative approach has seen CEPI establish a centralised network of seven clinical sample testing laboratories to assess and compare the immunological responses generated by multiple Covid-19 vaccine candidates. The network is designed to minimise variations that can result from the use of multiple labs and, therefore, help unify the way vaccine candidates are evaluated.

This has the potential to further speed up outbreak responses, explains Cramer. “Use of the network is open to all Covid-19 vaccine developers – providing this global service should help to provide a fast and easy way to generate the data needed to guide regulatory decisions.”

An integrated, global response

Cramer says efficiency and flexibility on the manufacturing front has also paid dividends, with vaccine developers and manufacturers working alongside researchers and regulators rather than in the traditional sequential format. While this carries financial risk, accelerated manufacturing has resulted in millions of vaccine doses being ready for distribution from the moment they were approved. Much of this was made possible thanks to initiatives like WHO’s Covax Facility and the US government’s Operation Warp Speed programme, although even the latter’s $10bn price tag is relatively small in comparison with other federal projects. Cramer says the crisis has laid bare the need for investment in a robust manufacturing sector that can be upscaled to work in tandem with more efficient clinical trials and regulatory outcomes.

Several areas in the manufacturing sphere could be modernised based on learnings from Covid-19, agrees Vereshchagina. She cites alternatives to regulatory site visits, including virtual inspections, as one way of leveraging further flexibility to create efficiencies and enhance oversight.

Equally, regulatory collaboration has also been key throughout the Covid-19 crisis, says Cramer. Preexisting regulator networks that included the FDA, EMA, MHRA and Health Canada came into their own as the virus spread, allowing for significant data-sharing and professional dialogue that intensified as the pandemic wore on, notably under the umbrella of the International Coalition of Medicines Regulatory Authorities (ICMRA). “From the very start of the pandemic, regulators around the world have strived to coordinate their opinions and guidelines,” says Cramer. If it continues, this increased collaborative effort should accelerate regulatory approvals, not only for Covid-19 vaccines but also for future medicine development.

Even so, the Organisation for Economic Cooperation and Development’s Covid-19 policy response unit says there is still room for improvement, arguing that the existing diversity of national regulations has slowed research and approvals. It’s pushing for the adoption of a standardised risk-based approach to clinical trials that would reduce the need to file regulatory applications in multiple markets, thus further streamlining the approvals process.

In a similar vein, Vereshchagina says there could be benefits to using technology to further accelerate regulation via a scalable global digital framework. “Data collected using digital technology may support marketing applications in multiple regulatory jurisdictions,” she explains, “so it is important to align on global regulatory approaches for the development, validation and usage of digital technologies throughout the R&D process.”

For all of that potential, Ali says the industry needs to ensure Covid-19 successes are not undermined by a loss of public confidence in the clinical trials process – and that should be a factor in any increase in the number of adaptive trials approved. Systemic changes are needed at the regulatory level to safeguard transparency, including strict requirements relating to the publication of protocols, adaptations and results. “When results are not shared, trials are repeated, and resources and precious knowledge are wasted,” Ali explains. “This knowledge – especially of negative results – can prevent [unnecessary] investment in similar treatments.” Regulators need to enforce transparency, she says, while researchers need better training in adaptive trial design to achieve evidence that is meaningful and usable.

Like Cramer and Ali, Vereshchagina stresses that regulators need to learn from Covid-19 by evaluating their responses and identifying the practices that are most relevant to future drug development. “Incorporating lessons learned during the pandemic is an ideal opportunity for regulators and the industry to align on regulatory approaches that best benefit public health,” she says.