Impact case study database

.

Statisticians based at Lancaster University have played a leading role in helping to identify the effectiveness of established and novel treatments as part of the RECOVERY trial and AGILE platform, helping to save lives and mitigate the harmful effects of COVID-19 at a national and global scale. Their rapid design and implementation of efficient Multi-Arm Multi-Stage (MAMS) and sequential dose-finding trials, the production of guidance documents for practitioners and clinicians, and their expert consultation as members of steering committees on major national trials, have resulted in a series of world-leading clinical breakthroughs.

4.1 Impact of Lancaster University statisticians on the statistical approach adopted by RECOVERY trial: RECOVERY is the UK’s leading national clinical trial for COVID-19 treatments in hospitalized patients. It has so far enrolled 23,000 patients, across 176 sites, to test 5 different treatments, making it the largest trial of its kind in the world. Jaki, as the lead statistician on the RECOVERY trial steering committee, was responsible for the design and implementation of the MAMS approach adopted by the trial, which drew on methodologies he devised in [3.1] and [3.2]. As Professor Peter Horby, Chief Investigator of the RECOVERY trial has said of the contribution made by Lancaster University research: “Underpinning this trial has been the design of Professor Thomas Jaki, who is Co-Investigator of the trial and member of the trial steering committee. Building on our earlier collaboration on clinical trial designs for epidemic infectious diseases, Prof. Jaki and I worked closely together during the early stages of the pandemic on the design of clinical trials for Covid-19.” He goes on to describe how Jaki used his research insights in [3.1] and [3.4] in the “initial design of the study and has been integral in subsequent modifications of the platform. This efficient design together with the pragmatism of the study have been key to deliver policy-changing evidence for treatments of COVID-19 in record time” [5.1]. The research methodologies also continue to inform the ongoing development and modification of the RECOVERY trial as it ventures into later phase testing and research: “The trial is currently entering a new phase in which more experimental treatments (rather than re-purposed agents) will be explored within the platform and Prof Jaki is contributing substantially to the design of these evaluations once more. This includes the imminent addition of a paediatric Phase II sub-study with Bayesian methods” [3.1] and [5.1].

4.2 Global adoption of dexamethasone as the first confirmed effective treatment for COVID-19: A major impact of the trial, and a direct consequence of its innovative statistical approach, was the rapid identification of dexamethasone as the first effective treatment for COVID-19 patients – formally announced on 16 June 2020 [5.2]. This breakthrough showed that it was effective in up to one-third of hospitalised cases and has since resulted in the successful treatment of over 20,000 patients in the UK as well as 650,000 worldwide [5.1]. It was heralded by the New York Times as “a major discovery” that has “transformed COVID-19 care worldwide,” and demonstrates “the critical need for randomised trials to separate drugs we hope work from treatments we know work” [5.3a]. Likewise, an article in Nature ascribed the trial’s success to its “short, flexible protocol – just 20 pages long – that laid out the design and data and regulatory requirements, [which] allowed trial arms to be halted or added” in an adaptive and rapid manner [5.3b]. The Director-General of the World Health Organisation (WHO), Dr Tedros Adhanom Ghebreyesus, has also publicly applauded the work of RECOVERY: “This is the first treatment to be shown to reduce mortality in patients with COVID-19 requiring oxygen or ventilation support,” describing it as a “lifesaving scientific breakthrough” [5.3c].

These findings resulted in widespread changes to national and international policy guidelines. On the 2nd September, the WHO announced a change in their guidance on the use of Corticosteroids (such as dexamethasone), stating how “work on this guidance began on 22 June when the RECOVERY trial published a preliminary report on the impact of Corticosteroids” [5.4]. The findings produced by the RECOVERY trial fed into the WHO’s own meta-analysis as part of the SOLIDARITY trial, which looked at data from seven clinical studies investigating the benefits of dexamethasone (the largest of which was RECOVERY) and found the discovery to be robust [5.4]. The European Medical Agency also changed their product information and guidelines in September 2020, following an extensive review of the data provided by RECOVERY, and concluded “that dexamethasone can be considered a treatment option for patients who require oxygen therapy (from supplemental oxygen to mechanical ventilation)” [5.5a-b]. These changes in guidance from leading international bodies has resulted in the adoption of the drug in healthcare systems across the globe.

4.3 Identification of hydroxychloroquine, lopinavir-Ritonavir and azithromycin as ineffective treatments for COVID-19: The MAMS trial methods, developed by Lancaster University statisticians and deployed in RECOVERY, resulted in the quick elimination of two treatments (hydroxychloroquine and lopinavir-ritonavir) that were found to be ineffective in combatting the symptoms associated with COVID-19, even pointing to potential harmful effects for patients in both cases and has more recently disqualified a third (azithromycin). The most significant of these discoveries was hydroxychloroquine, which had been widely described as a potential “miracle cure” on social media and by various public figures, including the former President of the United States. This led to global stockpiling, price hikes, and shortages of the drug, with negative consequences for neighbouring fields of health in which the drug is commonly used; for example, 60% of rheumatologists in Africa reported shortages of the drug, leading some to reduce doses for their patients to prolong supply [5.6a]. Against expectation, the RECOVERY trial’s randomised testing model very rapidly demonstrated hydroxychloroquine’s ineffectiveness, concluding on 5 June 2020, just 78 days from the trial’s start, that “there was no clinical benefit from use of hydroxychloroquine for hospitalised patients with COVID-19” [5.6b]. This finding resulted in the WHO halting its trial of the drug on 17 June [5.8], and the National Institutes of Health following suit on 20 June [5.6c].

Another drug that had been widely cited as a potential cure was lopinavir-ritonavir – an antiviral treatment commonly used to treat HIV/AIDS. The MAMS design for the RECOVERY trial meant that within the same trial it was shown that, as in the case of hydroxychloroquine, there was no clinical benefit associated with its use versus ordinary care [5.7a-b]. This finding had a major global impact, with healthcare services discontinuing its use, and the WHO halting its own trial of the drug at the same time as hydroxychloroquine [5.8]. Following these initial results (achieved in just three months), the Executive Chair of the Medical Research Council, Professor Fiona Watt, applauded the statistical methodologies and efficiency of these trials, stating publicly that “The UK’s RECOVERY trial is the world’s largest randomised trial of potential COVID-19 treatments and has worked with unprecedented speed to start delivering the answers we need.” She goes on to note that “whilst it is disappointing that lopinavir/ritonavir, like hydroxychloroquine, has been found to be ineffective, the earlier findings with dexamethasone were positive. Researchers and health professionals are now focusing their efforts, and patient care, on other promising drugs” [5.7a]. Since then, RECOVERY identified azithromycin as another ineffective drug, a result that has again been lauded by the Chair of the Medical Research Council: “Although it is disappointing that azithromycin isn’t an effective treatment for hospitalised COVID-19 patients, negative results are important so that clinicians can focus patient care on drugs that have been shown to work. This is particularly vital for antibiotics like azithromycin, because inappropriate use of antibiotics contributes to bacteria in the body becoming resistant” [5.9].

4.4 The adoption of Lancaster University’s MAMS and adaptive dose-finding methodologies to the AGILE platform and testing of novel treatments for COVID-19: Alongside the need to test existing treatments for their efficacy in fighting COVID-19, there is also an urgent need to explore the potential of novel ones. Trialling novel treatments brings the additional hurdle of identifying safe and effective dose levels. The AGILE trial platform, established in July 2020, brought together experts from the Universities of Lancaster, Liverpool, and Southampton, to tackle these and other problems associated with the identification and trialling of novel drugs. Drawing on their established expertise in the development of adaptive dose-finding [3.5-7] and MAMS [3.1-4] trials, the Lancaster group, headed by Jaki, Mozgunov, and Titman, were tasked with designing and implementing a novel trial infrastructure capable of simultaneously evaluating treatment candidates to find safe and effective dose levels, and thus accelerate progress toward later phase clinical trials [5.10]. Their designs have proven fundamental to AGILE’s ongoing success, as Professor Saye Khoo, the platform’s Chief Investigator has put it: “Professor Thomas Jaki, Dr Pavel Mozgunov, and Dr Andrew Titman at Lancaster University have led the development of the AGILE trial design.... their innovative approach… has played a critical role in the rapid set up of the AGILE trial platform and its success to date.’ Moreover, their methods have enabled ‘AGILE to carry out extensive national-level trials in an efficient and rapid way. These innovations significantly reduce the time taken to identify safe dose levels and drug effectiveness, resulting in more rapid trials than possible using conventional methods” [5.10].

Trials for four drugs are currently underway with a further four due to commence in 2021 [5.10]. Currently, the trial consists of four novel compounds: EIDD-2801, VAL-083, VIR-7831, and VIR-7832. EIDD-2801 was one of the first drugs to be included as part of the trial and has seen the most success to date. The Lancaster group established the trial design for both phase 1 and 2 in April, the trial was then approved in July and the first patient randomised in phase 1 in September 2020. Thanks chiefly to the efficiency of the statistical methodologies deployed by the Lancaster group, it commenced phase 2 in November with renewed financial backing (just 2.5 months after the randomisation of the first patient in phase 1). The co-founder and CEO of Ridgeback Biotherapeutics, who are the company producing and funding the drug, stated that the rapid testing of EIDD-2801 represents “a tremendous trans-continental effort to bring EIDD-2801 into a Phase 2 efficacy protocol to benefit COVID-19 patients in the U.K. immediately following Phase 1” [5.11]. Trials of VIR-7831 and VIR-7832 have also seen significant progress toward the end of the REF-period, with GlaxoSmithKline and Vir Biotechnology agreeing to invest over GBP3million in late December to accelerate early-phase development as part of the AGILE programme [5.12]. Though trials are ongoing and remain in an early stage, AGILE’s success was recognised by the UK government, when on 16^th December 2020 it was officially designated a “national platform” (alongside RECOVERY), meaning that all early phase trials of novel drugs intending to use NHS resource must now go through the AGILE platform [5.13]. Consequently, two of the UK’s four national platforms are now underpinned by Lancaster University research.