An analysis of clinical trials of potential treatments for COVID-19
suggests that most clinical trials registered between January and March
this year have significant design flaws that will limit their long-term
usefulness.
Of 201 trials registered with
ClinicalTrials.gov or the
WHO international trial registry
before 26 March this year, a third exclude clinical endpoints, almost
half were designed to recruit 100 patients or less and more than 70%
were open label.
“Because of these weaknesses, many of these studies are likely to
yield only preliminary evidence,” commented Hemalkumar Mehta, PhD, an
assistant professor at Johns Hopkins University Bloomberg School of
Public Health and lead author of the study, which was published in the
journal
BMJ Open.
“Given the urgency of identifying definitive evidence on potential
COVID-19 treatments, this is an instance where we wish we did not have
to say ‘further research is needed’ because of basic trial design
shortcomings and small trials.”
Mehta and co-authors acknowledge the exceptional circumstances of the
pandemic and the need for fast answers to the scientific problems it
poses, but highlight that small, open-label trials without clearly
defined endpoints only have limited statistical use and are subject to
bias.
Instead, they advocate the use of randomized study designs for better
accuracy and long-term benefit to patients, although they recognize
that these trials are often more expensive and require strong scientific
leadership to be successful.
“We understand the urgency of clinical research on COVID-19, but this
is a time when we need rigorous science to inform policy and clinical
decision-making,” explained co-author and G. Caleb Alexander, MD, a
professor at the Bloomberg School of Public Health.
“Any treatment that is ultimately deemed safe and effective via robust trials could potentially be used by millions of people.”
Overall, the analysis showed that of the 201 trials, 87.6% were based
in China (49.8%) or in the USA (37.8%). A total of 92 drugs and
antibody-rich plasma were being tested in the trials, although most of
the drugs were already approved for treatment of diseases or health
problems other than COVID-19 such as infection with other viruses,
malaria, autoimmune diseases or cancer. Only eight products or
combinations of products were new.
Although randomized trials were more common than non-randomized
trials, with 152 having some form of randomization, the majority of the
trials in this group were open label with only 55 trials including at
least single blinding to help minimize investigator bias.
Most of the trials analyzed in this study were registered in February
and March, with registrations rising rapidly between early March and 26
March. Notably, as of the beginning of June, the number of COVID-19
related trials on these registries has risen to more than 2000.
“This study provides early evidence of the benefits of global
registries to characterize urgent clinical trial research questions now
under investigation,” conclude the authors. “Used wisely by active
researchers, these registries can help to identify the most promising
avenues for developing new therapies, avoid unnecessary duplication and
define unanswered questions that inevitably arise from early research.”
Most Early COVID-19 Clinical Trials Poorly Designed
