The impossible made possible: the RECOVERY trial

Stock image of virus

It feels quite difficult to sit and write anything sensible about leadership in mental health clinical trials right now. I have spent most of the last year trying to keep my clinical trials alive, and most of them remain on life support. Perhaps crisis management, or firefighting would be a more useful title. It has been a dismal 2020 for mental health research. However, I do think there are reasons to be optimistic and hopeful for the future.

I have been involved in the setup and delivery of Covid-19 research over the last year, and the spectacular success of these studies, and of the RECOVERY (Randomised evaluation of COVID-19 therapy) platform trial for treatments for inpatients with Covid-19 platform trial in particular, has made me realise what is possible, and gives me hope for the delivery of clinical trials going forwards.

So what exactly did RECOVERY deliver that was so ground breaking? Firstly, the trial busted bureaucracy - It was 9 days from protocol drafted to first patient recruited, and an even more astonishing 91 days from protocol drafted to change in clinical practice across the world with the finding that dexamethasone reduces death in inpatients by a third. As one Medical Director told me ‘This is the first time that I have seen the research that we have been involved in be directly implemented into practice and save lives’. That direct link between undertaking clinical research and saving lives has really cut through. Research is no longer a ‘nice to have’ option for the NHS, it is absolutely vital.

Secondly the research design was simple and pragmatic - almost everyone with Covid-19 was eligible, there were no lengthy screening assessments, co-enrolment in other studies was also possible. Randomisation and data collection has also been straightforward and online. At the time of writing it has recruited 26,990 patients in 176 sites. Importantly Recovery has been delivered by clinicians rather than researchers, with no requirement for additional qualifications, or lengthy study-level training to take part. The best hospitals have recruited the majority of their patients with Covid-19 into Recovery trial, and these high recruiting sites have been District General Hospitals with clinicians recruiting alongside their clinical practice, rather than Teaching Hospitals of academic centres. Recovery has set the bar for what is achievable in the NHS with research truly embedded into clinical practice. And this places us as world leaders in terms of potential for delivering large scale trials.

Thirdly, I think RECOVERY has demonstrated the value of NIHR and the Clinical Research Network, working together to turbocharge the setup and delivery of a trial right across England. We (almost literally) dropped everything else to focus entirely on the delivery of this, and other urgent public health studies. We now need to take these lessons forward into every other research area.

One key lesson that I’ve learnt is that rapid trial set up is possible! Whilst of course the approvals of Covid-19 research studies have been exceptionally quick, and this timeline I not sustainable, I still think that current processes can be dramatically approved. Why should the standard time for set up of trials go back to being measured in months and years? If a study is approved by the MHRA and HRA then the governance documents should not require separate scrutiny from each NHS Trust. A trial that I am running took 18 months for site initiation at some sites following HRA approval. This is a waste of resources and effort for everyone involved. There are many trials that offer life-saving treatments for patients. They should be given the same level of urgency in set up as Covid-19 studies. HRA is currently running a fast-track review pilot, where they will give an ethical review for global clinical trials and phase I trials for any disease area in 15 days, which is excellent news.

The other key lesson that I have learnt is that the implementation of research findings into practice is also possible (in less than 17 years, which is the oft cited time delay between a research finding and change in clinical practice). RECOVERY managed to leap right across the translational gap which was astonishing to see. The study design certainly facilitated this- they trialled medications that were widely available and already in use. And the study population was generalizable, it was therefore straightforward to implement findings because the patient population in the trial and in clinical practice were the same. Why should our trials be so difficult to recruit to that only a small, unusual group of people with the disorder in question be eligible to take part?

So, I think that now is the time for the rapid scaling up of our ambition in mental health clinical trials. Clinical research is an embedded part of one part of the NHS right now, and we need to keep it there, and make sure this also extends to mental health services. Wouldn’t it be wonderful if every person with a new diagnosis of a mental illness or dementia was similarly offered the opportunity to take part in a clinical trial to improve their outcomes; indeed if that was the expectation of a high quality clinical service, and that we accelerated the discovery of new treatments for our disorders as well.