This is a preprint version of an article submitted for publication in the Medical Journal of Australia. Changes may be made before final publication. Click here for the PDF version. Suggested citation: Seidler AL, Aberoumand M, Williams JG, Tan A, Hunter KE. The landscape of COVID-19 trials in Australia. Med J Aust 2021; https://www.mja.com.au/journal/2021/landscape-covid-19-trials-australia [Preprint, 4 May 2021].
Research response in Australia has been rapid, but better coordination is imperative.
This perspective explores the landscape of COVID-19 trials recruiting in Australia using trial registry data. We identified 56 trials addressing treatment and prevention of COVID-19, and 12 trials addressing indirect effects of the pandemic. Whilst there was substantial innovation in drug development and digital health, the potential of innovation in methodology, such as adaptive trials, remains largely untapped. Data sharing intentions were low, sample sizes were too small to detect differences in clinical outcomes, such as mortality, and lack of core outcome collection precludes evidence synthesis. Infrastructure for innovation would support coordination of research efforts, and reduce research waste.
The coronavirus disease 2019 (COVID-19) pandemic has seen clinical trials launched at unprecedented speed in unprecedented numbers.(1) Whilst this is a positive development, the rapidity of trial launches and the unpredictable nature of the pandemic bring challenges for the conduct of trials and evidence synthesis. Duplication of effort is a risk and many trials alone are underpowered to find statistically significant effects for clinically important outcomes, including mortality.(1) Additionally, the ‘hard-to-predict’ waves of the pandemic may hinder recruitment due to declining cases, or pose challenges to starting trials quickly in emerging hotspots.(2) Recruitment has been a particular issue in Australia due to low case numbers compared with many other countries. In Australia, funds were made available rapidly to support research addressing the pandemic, but little is known about how effectively these funds have been utilised to drive the global agenda of preventing, diagnosing and treating COVID-19. We aimed to derive an understanding of the current landscape of clinical trials addressing the COVID-19 pandemic in Australia, and to what extent Australian researchers have responded to global need for coordination and collaboration.
We systematically searched the Australian New Zealand Clinical Trials Registry (ANZCTR) and ClinicalTrials.gov from 1st January to 16th November 2020, as these sources capture ~95% of registered trials recruiting in Australia.(3) We included all interventional studies addressing prevention, diagnosis and treatment of COVID-19 as COVID-19 trials, and all trials on indirect pandemic effects (e.g. lockdown measures, anxiety) as COVID-19 related trials. Observational studies were excluded. We analysed number and size of trials, additional recruitment countries, funding, trial purpose, study design, data sharing plans, and availability of core outcomes (mortality, respiratory failure, multiorgan failure, shortness of breath, recovery; Supplementary Table 1) [available in PDF].(4)
Impressive research scale-up
Research scale-up in Australia during the COVID-19 pandemic has been impressive. Of 1,637 studies registered, 1,174 were interventional studies with a recruitment site in Australia. Of these, 56 were COVID-19 trials, targeting 33,757 participants (Figure 1) [available in PDF]. Their characteristics are summarised in Table 1 [available in PDF], with detailed information in Supplementary Tables 2a-c [available in PDF]. Four of the trials (7%) were completed, the remainder were recruiting (n = 26, 46%), not yet recruiting (n = 24, 43%) or withdrawn (n=2, 4%, withdrawal reasons in Supplementary Tables 2a-c) [available in PDF]. The majority (46 trials, 82%) recruited only in Australia, whilst 10 trials (16%) recruited in Australia and internationally. Most (40 trials, 71%) had no commercial sponsor, and were funded by government or not-for-profit sources. Only 7 trials (12%) included populations at high-risk of poor outcomes from COVID-19 such as those with co-morbidities (e.g. cancer, cardiovascular disease, chronic kidney disease).
Nineteen (35%) were prevention trials. We identified 10 (18%) vaccine trials, of those, 2 (20% of vaccine trials) repurposed existing vaccines for COVID-19 prevention. Eight trials (80% of vaccine trials) investigated efficacy using a COVID-19-specific antigen. Thirty-four (62%) were treatment trials, 22 of those were (39%) drug trials. A broad array of drug categories was investigated, including but not limited to: immunosuppressants, immunostimulants, stem cell therapies, antivirals and anti-inflammatories. Merits, risks and proposed solutions of included trials are summarised in Table 2.
We identified 12 COVID-19 related trials (Table 1, Supplementary Table 3 - available in PDF), and the majority (n = 11, 92%) addressed mental health issues related to uncertainty and isolation during the pandemic.
Fast-track procedures may have impacted scientific rigor and research prioritisation
The flipside to the rapid emergence of trials is the haste with which funding, development and implementation happened - leading to concerns about research waste and prioritisation, and ethical/scientific rigor.(5) Adding concern is the fact that no full, publicly available protocols were identified for any of the included trials. Most organisations did not have fast-track procedures in place at the start of the pandemic.(5) The development of publicly available, transparent procedures and standards for all stages of clinical trials (e.g. development, funding, ethics, conduct, dissemination) should be an important lesson from the current pandemic. Such standards must balance the urgency of advancing knowledge with the retention of ethical and scientific rigor.
The majority (89%) of included trials tested pharmaceutical drugs or devices, except for six trials: three telehealth applications for COVID-monitoring/ rehabilitation, one lifestyle intervention, and one intervention on patient positioning during oxygenation. There were no trials on public health communication, community transmission prevention or long COVID symptoms, pointing to omissions in research prioritisation. Additionally, extensive media coverage and public opinion may have influenced prioritisation of interventions that were not particularly promising.(1, 6) For instance, many simultaneous trials emerging on (hydroxy)chloroquine (six in Australia alone) may have put too many patients unnecessarily at risk.
Innovation in drug development and digital health, but potential of innovative study designs remains untapped
Australian COVID-19 trials tested a range of innovative interventions, such as vaccine techniques (nanoparticles and the delivery of genetic material in two vaccine trials) and digital health solutions for home-monitoring of mild COVID-19. Trialists demonstrated adaptability to transmission risks for in-person contact through innovation in trial conduct, with digital recruitment and delivery modes such as video-calls or smartphone applications. Yet, innovation in trial design was lacking. We identified only two trials utilising adaptive methods (Bayesian designs), which can respond to the rapidly changing landscape of treatment options, and thus deliver results more efficiently.
Trials often underpowered for clinical outcomes
The median target sample size was small (150, Q1-Q3=33-395), meaning that individually trials were likely underpowered to detect differences in clinically important outcomes.(4) For instance, to detect a 30% relative reduction in mortality (similar to that observed for corticosteroids(7)) with 80% power, one would need a sample size of 4,424 in a hospitalised population (with baseline mortality rates around 7% (8)) (Supplementary Figure 1 - available in PDF). None of the identified treatment trials are sufficiently powered to detect such a difference in mortality, and with low case numbers in Australia it seems unlikely that a single trial could obtain such large sample sizes.
Limited collection of core outcomes precludes evidence synthesis
Evidence synthesis in the form of meta-analysis across trials is critical to obtain sufficient power to detect differences in core outcomes, or subgroups of participants, particularly when individual trials are underpowered. Core outcome sets were agreed early in the pandemic, and are evolving.(4) Of the 34 COVID-19 treatment trials in Australia, the proportion assessing each core outcome was low (Figure 2 - available in PDF). For instance, only 53% (18 trials) assessed mortality, and 18% (6 trials) assessed shortness of breath, whereas 63% (21 trials) assessed respiratory failure. Only one trial included all core outcomes, and 10 trials (29%) included none. Thus, it will be impossible to synthesise results or make important comparisons for many of the trials.
Data sharing intentions low
The International Committee of Medical Journal Editors (ICMJE) declared data sharing an ethical obligation,(9) to honour the risk trial participants take by increasing the likelihood that their participation results in useful findings.(2, 9) Since the pandemic began, there have been several high-profile calls for collaboration and data sharing across studies, to enable more complex analyses and reliable effect estimates than would be obtained by simple combination of aggregate data.(2, 10) These calls seem to pass largely unheard among trialists in Australia, with 80% (41 trials) indicating they are not planning to share data (Table 3 - available in PDF). Whilst these declarations at trial outset may be conservative and investigators may decide to share data later, they are still concerning. Frequently mentioned barriers to data sharing include a lack of understanding of the relevance, lack of resources to prepare data, insufficient academic recognition, and concerns about participant privacy, ethical approval and data misuse.(11) Structural support by funding bodies, research institutions, ethical committees, and journal editors are needed to address barriers and facilitate data sharing.(11) This could include a recognition system for collaboration and data sharing, and standardised moderated processes for data sharing, following FAIR (Findable, Accessible, Interoperable, Reusable) principles.(12) To date, no recognised FAIR data repository is available in Australia.(12)
Opportunities for strategic coordination and collaboration
As the COVID-19 pandemic evolves, clinical and societal need for research evidence will continue. There may be shifts in research focus as our understanding of COVID-19 grows, perhaps to ‘long COVID’ or other sequelae. Co-ordinating research efforts is a cost-effective, more reliable and timely way of achieving larger sample sizes and thus more impactful research evidence. Prospective meta-analyses and other next generation systematic review approaches provide suitable frameworks to coordinate such collaborative research efforts, and to align on key elements of study design, such as core outcomes.(13, 14) Internationally, researchers have begun applying these frameworks to the pandemic,(2, 10) including an influential prospective meta-analysis evaluating corticosteroid treatment for COVID-19.(7)
In Australia, the COVID-19 pandemic has led to rapid changes in some processes including fast-tracked funding, ethical approvals, trial registration and publication.(15) Yet, too little has happened in creating infrastructure and funding for rapid collaboration, advanced adaptive methodologies and data sharing. In future, with adequate funding for technological innovation, clinical trial registries may play a key role in automatically connecting similar trials and facilitating collaboration. The COVID-19 pandemic presents a unique opportunity to improve collaborative infrastructure and methodologies, and advance future research across all health areas.
- Davis JS, Ferreira D, Denholm JT, Tong SY. Clinical trials for the prevention and treatment of COVID‐19: current state of play. Med J Aust. 2020; 213: 86-93. 10.5694/mja2.50673.
- Petkova E, Antman EM, Troxel AB. Pooling data from individual clinical trials in the COVID-19 era. JAMA. 2020; 324: 543-5. 10.1001/jama.2020.13042.
- Askie L, Hunter K, Berber S, Langford A, Tan-Koay A, Vu T, et al. The Clinical Trials Landscape in Australia 2006-2015. Sydney: Australian New Zealand Clinical Trials Registry, 2017. https://www.anzctr.org.au/docs/ClinicalTrialsInAustralia2006-2015.pdf (Accessed April 2021).
- Tong A, Elliott JH, Azevedo LC, Baumgart A, Bersten A, Cervantes L, et al. Core outcomes set for trials in people with coronavirus disease 2019. J Crit Care Med. 2020; 48: 1622-35. 10.1097/CCM.0000000000004585.
- Bahans C, Leymarie S, Malauzat D, Girard M, Demiot C. Ethical considerations of the dynamics of clinical trials in an epidemic context: Studies on COVID-19. Ethics Med Public Health. 2021; 16: 100621. 10.1016/j.jemep.2020.100621.
- Saag MS. Misguided Use of Hydroxychloroquine for COVID-19: The Infusion of Politics into Science. JAMA. 2020; 324: 2161-2. 10.1001/jama.2020.22389.
- The WHO Rapid Evidence Appraisal for COVID-19 Therapies Working Group. Association Between Administration of Systemic Corticosteroids and Mortality Among Critically Ill Patients With COVID-19: A Meta-analysis. JAMA. 2020; 324: 1330-41. 10.1001/jama.2020.17023.
- Dennis JM, McGovern AP, Vollmer SJ, Mateen BA. Improving Survival of Critical Care Patients With Coronavirus Disease 2019 in England: A National Cohort Study, March to June 2020. J Crit Care Med. 2020; 49: 209-14. 10.1097/ccm.0000000000004747.
- Taichman DB, Sahni P, Pinborg A, Peiperl L, Laine C, James A, et al. Data sharing statements for clinical trials: a requirement of the International Committee of Medical Journal Editors. Ann Intern Med. 2017; 167: 63-5. 10.7326/M17-1028.
- Ma Z, Liu J, Pan Q. Overwhelming COVID-19 Clinical Trials: Call for Prospective Meta-Analyses. Trends Pharmacol Sci. 2020; 41: 501-3. 10.1016/j.tips.2020.05.002.
- Tan A, Askie L, Hunter K, Barba A, Seidler A. Willingness to share individual participant data, and barriers and facilitators to data sharing: a retrospective cohort study and cross-sectional survey. 2020; 9 Suppl 1. 10.1002/14651858.CD202001.
- The International Consortium of Investigators for Fairness in Trial Data Sharing. Toward Fairness in Data Sharing. N Engl J Med. 2016; 375: 405-7. 10.1056/NEJMp1605654.
- Seidler AL, Hunter KE, Cheyne S, Berlin JA, Ghersi D, Askie LM. Prospective meta‐analyses and Cochrane's role in embracing next‐generation methodologies. Cochrane Database Syst Rev. 2020. 10.1002/14651858.ED000145.
- Seidler AL, Hunter KE, Cheyne S, Ghersi D, Berlin JA, Askie L. A guide to prospective meta-analysis. BMJ. 2019; 367: l5342. 10.1136/bmj.l5342.
- Whitmore KA, Laupland KB, Vincent CM, Edwards FA, Reade MC. Changes in medical scientific publication associated with the COVID‐19 pandemic. Med J Aust. 2020; 213: 496-9. 10.5694/mja2.50855.
Publication of your online response is subject to the Medical Journal of Australia's editorial discretion. You will be notified by email within five working days should your response be accepted.