Written evidence submitted by Professors Martin Landray, Peter Horby, and Richard Haynes (CLL0047)

 

This submission addresses the following term of reference:

Lessons learnt on the development of treatments and vaccines.

 

INTRODUCTION AND CONTEXT

  1. Randomised clinical trials play a critical role in establishing which of the many treatments believed to have a potential benefit actually do work. Furthermore, clinical trials provide information about the size of any benefits to patients and the NHS (e.g. death, duration of hospital stay), in whom they work best, and any safety issues.
  2. At the start of the pandemic, the World Health Organization and major drug regulators emphasised the need for a relatively small number of large, randomised trials comparing the effects of possible treatments with usual care alone. They highlighted the chaotic situation in previous epidemics (e.g. SARS and Ebola) in which a multitude of small trials provided no meaningful new knowledge, or where large quantities of unproven treatments were given to patients outside the context of clinical trials. Consequently, despite the expenditure of substantial financial and healthcare resources, no definite evidence was gained on whether patients had been helped or harmed by the treatments they had been given. These events also did not lead to new knowledge about how to treat patients in future epidemics or advance the development of new treatments.
  3. With a few exceptions, the medical world has repeated mistakes of the past during this pandemic: hundreds of small and uninformative trials and observational studies; hundreds of thousands of patients given unproven treatments; and little improvement in knowledge about how best to manage patients in this epidemic.
  4. The UK took a very different approach and has been widely lauded for doing so. In particular, the Randomised Evaluation of COVid-19 thERapY (RECOVERY) trial has delivered vital knowledge to improve the outcomes of patients hospitalised with severe COVID-19. As the world’s largest randomised clinical trial of treatments for patients hospitalised with COVID-19, the RECOVERY trial has received praise for its inclusive approach, sound design and efficient delivery, and for the quality of the results it has produced. These have changed clinical treatment of COVID-19 worldwide and, as highlighted by the Prime Minister in his address to the United Nations General Assembly (26 September), already saved many thousands of lives.
  5. Coordinated by clinical trialists at the University of Oxford, the RECOVERY trial has been a whole country effort, involving every acute NHS hospital in the UK, thousands of NHS clinicians, nurses, and other staff, and the support of national research infrastructure (including National Institute for Health Research, NIHR, NHS DigiTrials, and the NHS in all four nations). The leadership of the Chief Medical Officers has been critical to positioning the trial as a key part of NHS clinical care activities. The trial was funded by a grant from UK Research & Innovation and NIHR, but just as important is the long-term core support for both the University of Oxford’s clinical trials unit that coordinates the trial and for the broader health research ecosystem.
  6. None of this would be possible without the participation of the patients: over 19,000 have taken part to date. Overall, around one in ten patients admitted to an NHS hospital with COVID-19 have taken part in the trial. The trial is deliberately inclusive: the youngest participant was less than 6 months old and the oldest over 100 years, one-third are women, one sixth are of Black, Asian or Minority Ethnic background.
  7. Whenever discussing the pandemic, attention focusses on the numbers, the medical advances, the operational challenges and successes. However, it is essential to reflect that behind these numbers are a collection of individual human stories involving patients, their families, friends, loved ones, and communities. The RECOVERY trial team will be ever thankful for the contribution of those who have taken part so that future patients may benefit from the knowledge gained.
  8. The following text summarises the RECOVERY trial’s progress to date and proposes several recommendations to improve future pandemic preparedness and the management of clinical trials investigating treatments for other common conditions. These are based on the key challenges and achievements of the RECOVERY trial.

 

 

 

 

 

SUMMARY OF RECOVERY TRIAL PROGRESS TO DATE

  1. The RECOVERY trial was rapidly launched in March 2020, at a time when there were no known effective treatments for COVID-19. A candidate vaccine treatment remained at least six months in the future, hence there was an urgent need to evaluate whether any existing drugs could effectively reduce deaths from COVID-19.
  2. RECOVERY is coordinated by the Nuffield Department of Population Health and the Nuffield Department of Medicine, both at the University of Oxford. These departments bring world-leading expertise in emerging viral infections, pandemic response and large-scale, randomised clinical trials.
  3. The trial is taking place at 176 hospital sites across the UK. Patients recruited into RECOVERY receive either standard hospital care alone or standard care with the addition of at least one of the treatments under investigation. Within the first 100 days, the RECOVERY trial produced three results that changed clinical practice. These demonstrated that:
    1. Dexamethasone reduced death in patients with severe COVID-19 by up to a third. This was the first, and so far the only, treatment shown to save lives in this condition. The result was announced at 13.00 on 16 June and just four hours later the Chief Medical Officers wrote to all NHS hospitals recommending this become standard of care. It is now recommended for patients with severe COVID-19 worldwide, and costs just £5 per course.
    2. Hydroxychloroquine (usually a treatment for malaria), which had been very prominently promoted internationally, was of no significant benefit in patients hospitalised with COVID-19. The US Food and Drug Administration removed the previous emergency use authorisation for hydroxychloroquine ten days after the announcement of the RECOVERY result.
    3. Lopinavir-ritonavir (commonly used to treat HIV), which had been recommended in many clinical guidelines around the world, was of no significant benefit for patients hospitalised with COVID-19. It is no longer used for COVID-19.

It should be noted that all three results were contrary to prevailing expectations, emphasising the critical role that adequately large randomised controlled trials play in differentiating treatments hoped to work from those with rigorous evidence that they do work.

  1. Over time, new treatments have been added to the trial protocol, including:
    1. Azithromycin (a commonly used antibiotic but used here for its anti-inflammatory properties);
    2. Tocilizumab (an anti-inflammatory treatment given by injection);
    3. Convalescent plasma (collected from donors who have recovered from COVID-19 and containing antibodies against the SARS-CoV-2 virus);
    4. REGEN-COV2 (an investigational anti-viral antibody combination produced by Regeneron Pharmaceuticals);
    5. Aspirin (used here for its antiplatelet properties which may reduce the risk of complications from thrombosis [i.e. blood clotting] in COVID-19);
    6. Colchicine (an anti-inflammatory drug usually used for gout and pericarditis).
  2. The RECOVERY trial has shown that a nationally-endorsed platform trial can have a transformative effect on the pace of discovery and improvements in care. Importantly, it was designed to minimise the burden on frontline NHS staff and patients, enabling high-quality evidence to be collected even during peak periods of hospital admission. The trial has shown that the UK’s ability to deliver trials of this kind is unrivalled worldwide.
  3. The RECOVERY trial has now been extended internationally in response to demand from other countries. It is likely that Vietnam, Indonesia and Nepal will be the first to join.

 

SUMMARY OF RECOMMENDATIONS

  1. Recommendation 1 – Accelerated trial authorisations:

The experience of delivering timely, high quality approvals achieved during the pandemic should act as a spur to deliver rapid, coordinated, consistent, and proportionate review in ‘normal’ times. For RECOVERY, these approvals included the Medicines and Healthcare products Regulatory Agency (MHRA), Research Ethics Committee and Health Research Authority (HRA). Nevertheless, this streamlining and coordination of approvals should extend to other relevant processes and organisations (including data access and NHS Trust R&D).

  1. Recommendation 2 – Sustainable clinical trial research infrastructure:

Running national-scale trials in both emergency and common conditions requires specific expertise. Long-term investment is required in clinical trial design and delivery expertise (e.g. through establishment of a national trials unit specialised in population-scale trials), NHS data curation (e.g. NHS DigiTrials), and research nurses and associated staffing (NIHR Clinical Research Network).

 

 

  1. Recommendation 3 – Training, education & continuing professional development:

The Royal Colleges should ensure an understanding of clinical trials is integrated into medical and nursing training, and that participating in trials is recognised by medical profession regulators (e.g. the General Medical Council, the Nursing and Midwifery Council) as a key component of good professional practice.

  1. Recommendation 4 – Establishing clinical trials as a core component of high-quality NHS care:

The NHS should recognise randomised clinical trials as a component of routine clinical care (not an optional add-on). This should be recognised in job plans and remuneration of individuals. Contribution to clinical trials activities should be a Board-level responsibility for all NHS organisations (acute hospital trusts and equivalent organisations) and a marker of the quality of the clinical service provided.

  1. Recommendation 5: Adaptive trial design as a favoured approach:

In situations of a new pandemic disease, clinical trials should use an adaptive, platform design in order to respond rapidly to new therapies and evidence changes.

  1. Recommendation 6 – Safe, efficient, and timely use of routine NHS data:

The NHS collects vast quantities of clinical data (e.g. primary and secondary care, death registries, clinical audits, laboratory testing) but substantial investment is required to curate these disparate sources. This should provide a set of efficient data services and associated proportionate information governance controls so that the data can be best used to support clinical trials in the interest of improved patient care and public health.

  1. Recommendation 7 – Transparency, trust, and community engagement:

It is imperative that clinical research is transparent and open to scrutiny. Clinical trials must be entered on public registries, with full protocols (and any subsequent modifications) and trial enrolment numbers made publicly available. 

The NHS and life sciences sector should drive forward a programme of patient and public involvement and engagement to widen understanding of the role clinical trials play in developing new treatments.

  1. Recommendation 8 – Enlightened approaches to trial regulation:

The current regulations and guidelines (so-called Good Clinical Practice guidelines) that govern clinical trials should be reviewed and significantly revised so that they best serve the interests of patients and the public, are focussed on the underpinning scientific and ethical principles of clinical trials, and are fit for the opportunities provided by advances in communications and data management technology and the health challenges of the coming years (including future epidemics, the burden of common chronic disease, and the threat of drug resistant infections).

The Good Clinical Trials Collaborative (hosted by Wellcome and supported by the Gates Foundation and the African Academy of Sciences) provides a framework for this activity. This may provide a suitable, broad-based forum to support such a transformation.

 

ADDITIONAL PERSPECTIVES FROM THE RECOVERY TRIAL EXPERIENCE

The process of approving and launching a large-scale clinical trial can be greatly accelerated (Recommendations 1 and 2)

  1. A clinical trial the size of RECOVERY would normally take many months to set up and possibly years to deliver data. RECOVERY was launched within nine days of its conception, with over 10,000 patients recruited in just two months. This rapid launch and roll-out across the UK was essential in order to recruit patients at a time when hospital admissions were rising fast. This meant that before three months had passed, the trial had already delivered two significant findings, including the dexamethasone result.
  2. The RECOVERY trial has demonstrated that the normal process of launching a clinical trial can be significantly streamlined and simplified. The key factors that enabled its rapid launch included:

a.   Rapid decision making for funding to ensure that preparations could be made once it became clear the UK would be affected by the pandemic.

b.   The involvement of experienced trials units with core ‘infrastructure’ funding from the NIHR and Medical Research Council (MRC), which had staff with necessary skills and experience who could be rapidly re-deployed onto RECOVERY. There was no time to recruit new staff and without such funding, staff with decades of experience in running trials would not have been available.

c.    Expedited review by MHRA, HRA and the Research Ethics Committee. The usual timelines of 30 days for first opinion were not compatible with the speed required, but all these agencies recognised this and provided timely, risk-proportionate opinions.

d.   The Department of Health and Social Care (DHSC)’s purchase of a stockpile of drugs including some RECOVERY interventions. DHSC also collaborated with Public Health England and RECOVERY teams to distribute them to sites.

e.   Support from senior leadership including the Chief Medical Officers, the NHS and NIHR.

Collaboration with the NHS allows effective, nationwide testing of candidate treatments (Recommendations 3 and 4)

  1. RECOVERY is conducted in partnership with 176 NHS hospital sites across the UK. Any patient with COVID-19 who is admitted to one of these sites is eligible to participate in the trial. Collaborating with the NHS has brought significant benefits for RECOVERY:

a.      The ability to scale up rapidly, by simultaneously launching across all participating hospitals.

b.   Access to a large pool of patients eligible for recruitment.

c.    A broad geographical spread, accounting for regional variations in patient population and standards of care, ensuring that results are applicable to the national population. This also accounts for differences in COVID-19 incidence between regions, so that the overall pool of potential patients for recruitment remains large, even if certain areas have low hospital admission rates.

d.   Access to high-quality routine patient data stored within the databases held by the central NHS data custodian: NHS Digital for England; the SAIL Databank for Wales; and Public Health Scotland and the National Records of Scotland.

e.   Support from the NIHR Clinical Research Network (CRN) to provide local infrastructure and identify teams of clinicians, research nurses and support staff to deliver RECOVERY locally. CRN staff assisted with (i) identification and invitation of potential participants; (ii) obtaining consent from and randomising participants; (iii) completing case report forms to provide the minimal dataset required for the trial. The CRN also assisted the trial team by rapidly identifying principal investigators at each trust who could then lead their local team with knowledge of local procedures to ensure the trial worked locally as well as at a national scale.

f.      Rapid dissemination of RECOVERY-related communications to clinicians, including updates, guidance and training materials.

g.   Equity of access to participation in the trial for all hospitalised patients in the UK.

  1. The RECOVERY trial coordinating team has explored how clinicians helping to facilitate the trial at each hospital site may be thanked and acknowledged for their work. For example, on 13 November, the RECOVERY trial joined the NIHR Associate Principal Investigator Scheme, a mentorship scheme for junior doctors and nurses. Over six months, the scheme teaches key skills and concepts in delivering clinical trials, and results in a formal qualification endorsed by the NIHR and the medical Royal Colleges.
  2. The NHS was a key factor in RECOVERY’s success in scaling up so quickly and recruiting thousands of patients to date. The trial sought to bring clinical trials back into the domain of clinical care, so that recruitment into research was seen as an important component of good clinical care, not an optional extra.

Trials need to have an adaptive design from the start, to allow treatments to be added as new evidence and/or therapies become available (Recommendation 5)

  1. From the outset, RECOVERY was designed to be adaptive, so that new recruitment arms could be easily added in light of newly-developed therapies or emerging clinical evidence.
    1. REGEN-COV2, a monoclonal antibody therapy developed by Regeneron Pharmaceuticals, was added to RECOVERY on 14 September. This was the first treatment specifically designed against COVID-19 to be investigated by the trial.
    2. Aspirin was added to the trial on 6 November. This was in response to evidence that COVID-19 patients are at higher risk of blood clots forming in their blood vessels. As an antiplatelet agent, aspirin may protect against clots forming in COVID-19 patients.
    3. Colchicine, an anti-inflammatory commonly used to treat gout, was added to the RECOVERY trial on 27 November. This was based on emerging evidence that colchicine may inhibit inflammatory receptors associated with severe COVID-19.

The burden on patients and front-line healthcare workers can be minimised through linkage to routinely collected patient data (Recommendation 6)

  1. Given the large pressure the first COVID-19 wave placed on the NHS, RECOVERY was designed to not impose an additional burden on frontline staff. The recruitment process was made to be as simple as possible, with staff only having to provide essential information (such as the treatments the patient was receiving when they joined the trial e.g. oxygen). Following this, the trial’s data linkage team links each recruited patient with their NHS record in the national database. This enables their progress to be tracked over time (even if they are transferred to another hospital), including whether they required ventilation, and if they made a recovery or not.
  2. Besides reducing the burden on healthcare workers, linking patients with their NHS record allows the trial to access high-quality data, including previous hospital admissions and prescribed medication. This will help to identify underlying health conditions, to find out which groups of patients might benefit most from the treatments being tested. Since patient data are continually added to their record, this also enables assessment of the long-term effects of the treatments on health outcomes (such as later lung problems or kidney disease).
  3. Routinely collected data provides complete follow-up for outcomes so reduces any bias that could be introduced by incomplete follow-up, and removes the need to follow up with patients directly.
  4. Routinely collected data means long-term follow-up is feasible and affordable. This allows the long-term effects of treatments to be studied and the full health economic effects to be understood.
  5. The University of Oxford has world-renowned systems for keeping confidential data secure. Data is pseudonymised prior to linkage, so that the individuals taking part in the trial cannot be identified.

There is significant potential to increase the evidence base for COVID-19 treatments through improving the rate at which patients are recruited to clinical trials (Recommendation 7)

  1. Despite the RECOVERY trial receiving widespread positive media coverage, recruitment of eligible patients at hospital sites is typically low. During both the winter/spring wave and the second wave in the autumn, about 10% of patients testing positive for SARS-CoV-2 (the coronavirus which causes COVID-19) admitted to hospital with COVID-19 were recruited into RECOVERY. It must be recognised that many patients with a positive test will have been screened as part of their hospital admission for other conditions (e.g. routine surgery). Nonetheless, there is considerable variation in the proportion of patients enrolled from different (and sometimes neighbouring) hospitals, ranging from <2% to >40%. This suggests that significant organisational or systematic factors are in play.
  2. An audit exercise carried out by NIHR investigated the reasons why potential participants were not being recruited to the RECOVERY trial. This found that, for some hospital trusts, lack of staff capacity or staff being unaware of the trial had meant that potentially eligible patients had not been approached. Of the patients who were approached in the audit, almost a third (30%) declined to take part in RECOVERY, including those who wished to see if they improved with the current standard of care. This might be due to differences in the way the study was presented and explained or to substantive differences in the views of particular patient groups.
  3. Factors that appear to have contributed to higher patient recruitment rates at certain hospital sites include highly motivated local leaders; one-to-one training for clinicians in approaching patients and explaining the study; embedding RECOVERY within the routine admissions procedure; and effective communications between hospital wards and research teams.
  4. It is hoped that RECOVERY’s participation in the Associate PI Scheme will improve recruitment rates, by training clinicians to approach patients and explain the study.

Transparency and information sharing must be maintained, even during international emergency situations (Recommendation 7)

  1. All of the key documents relating to RECOVERY are published on its dedicated website: https://www.recoverytrial.net/. These include the study protocols, results papers, training materials for clinicians and minutes from the meetings of the Independent Data Monitoring Committee. This transparency has proved crucial to generate trust and address those who have questioned RECOVERY’s design and methods. 
  2. When there was sufficient evidence to demonstrate that dexamethasone significantly reduced death in severely ill COVID-19 patients, the RECOVERY team recognised that this discovery needed to be announced as soon as possible, to allow it to be immediately incorporated into standard clinical practice. This meant that the usual process of disseminating new discoveries initially through a peer-reviewed medical journal paper was unsuitable. Instead, the result was announced in the form of a press release and media briefings. The Chief Medical Officer was briefed on the findings, and the UK Government immediately authorised the NHS to use the world’s first coronavirus treatment proven to reduce the risk of death; this meant that the treatment was adopted into UK clinical practice hours after the initial announcement in order to save lives without any delay. The WHO also rapidly changed their guidance and the European Medicines Agency released guidance and invited dexamethasone manufacturers to seek updates to their authorisation so they could market their brand for use in COVID-19. Although the RECOVERY trial leaders received some criticism for this course of action, it was felt to be in the best interests of public health. The results were published as a pre-print shortly afterwards, then a fully peer-reviewed paper several weeks later.
  3. Similarly, the negative results for lopinavir-ritonavir and hydroxychloroquine were announced initially as press releases. This was to allow health systems across the world to focus on other candidate therapies and to halt the use of ineffective treatments, since both drugs were being used widely as treatments for COVID-19 without any supporting evidence.

 

 

 

NOTES

 

  1. RECOVERY is supported by a grant to the University of Oxford from UK Research and Innovation/National Institute for Health Research (NIHR) and by core funding provided by NIHR Oxford Biomedical Research Centre, Wellcome, the Bill and Melinda Gates Foundation, the former Department for International Development, Health Data Research UK, the Medical Research Council Population Health Research Unit, and NIHR Clinical Trials Unit Support Funding.
  2. The RECOVERY trial involves many thousands of doctors, nurses, pharmacists, and research administrators across the whole of the UK, supported by staff at the NIHR Clinical Research Network, NHS DigiTrials, Public Health England, Department of Health & Social Care, the Intensive Care National Audit & Research Centre, Public Health Scotland, the Secure Anonymised Information Linkage at University of Swansea, and the NHS in England, Scotland, Wales and Northern Ireland.

Nov 2020