A lab technician works on investigational COVID-19 treatment drug remdesivir at a facility in Cairo, June 2020. Photo: Reuters/Amr Abdallah Dalsh.
On August 22, the Journal of the American Medical Association (JAMA) published the results of a randomised clinical trial of remdesivir in patients of moderate COVID19. Remdesivir is an antiviral drug that inhibits RNA polymerases, the group of enzymes that is required for viruses – like SARS-CoV-2 – to replicate. It is already approved for use in the US, the European Union, India, Australia and other countries for use with patients who have severe COVID-19. However this study aimed to expand the scope of its use to moderate disease.
In this study, 584 patients treated in the US, Europe and Asia with moderate COVID-19 disease were divided into three treatment arms. The first received a 10-day course of remdesivir, the second received a five-day course, and the third received standard care (antibiotics, steroids and/or other medications). This three-arm trial design also allowed the investigators to determine whether 10 days of remdesivir was more effective than five days.
A seven-point scale was used to determine the severity of COVID-19 disease, ranging from death (1) to ‘not hospitalised’ (7). The same group used this scale in their previous remdesivir trials as well (ACTT-1 and GS-US-540-5773). Those participants included had evidence of lung involvement on CT scans but near-normal lung function, categorised as moderate disease. Patients received either remdesivir or standard treatment from the date of diagnosis, and were compared for clinical severity on day 11 and treatment-related side effects. The researchers conducting the trial also compared effect on mortality, reduction in duration of symptoms and time to hospital discharge.
The study found that on day 11, patients in the five-day remdesivir arm were 65% more likely to achieve better clinical scores than the standard care arm. Using the same yardstick, there was no benefit with 10 days of remdesivir. Also, there was no reduction in hospitalisation, mortality or reduced duration of oxygen requirement. They concluded that patients receiving five days of remdesivir had a significantly different clinical status than those receiving standard therapy, with uncertain clinical importance.
As with previous COVID-19 related clinical trials, there are several red flags we need to address before interpreting the results. Of note: the sponsor of the trial is Gilead Sciences Inc., a major manufacturer of remdesivir. Additionally, the end-point is an unimpressive one: odds of a better clinical status is quite vague relative to other, more clinically meaningful measurements like mortality, hospital stay, duration of oxygen requirement, etc. None of the authors’ own exploratory end-points, such as time to recovery or time to improvement in clinical status (by 1, 2 or 3 points) were significantly different with remdesivir.
Of the patients included, 80-85% had a clinical score of 5 (hospitalised, not requiring supplemental oxygen but requiring ongoing medical care). It is unclear how the same end-point – change in clinical status – that was used by the trial group for their previous trials involving severe COVID-19 patients can be applied to patients with moderate disease.
At the outset, the premise of this trial is questionable at the least. Remdesivir costs Rs 5,000 per dose, implying a cost of Rs 30,000 or Rs 60,000 for a five- or 10-day course respectively. In people with moderate COVID-19, where the mortality was 1%, it is impossible to justify remdesivir’s use without evidence of concrete clinical benefit. There is nothing objectionable about negative trials – which are those trials that fail to achieve their desired end-point. However, the fact that this trial was allowed to report its result as positive, in favour of remdesivir, is highly suspect.
In a nutshell, this trial encapsulates the issues with trials funded by the pharmaceutical industry in the context of the COVID-19 pandemic. By applying an ambiguous end-point to a patient population that may not have needed any treatment other than supportive care or supplemental oxygen, this trial bears all the hallmarks of bad science. Another difficult concept to grasp is why patients receiving 10 days of remdesivir had no difference in clinical status compared to standard treatment, while those receiving five days of treatment did. Considering that all three arms had comparable characteristics and were randomised, the benefit from remdesivir should have been apparent in both arms, not just one.
By consistently allowing poorly designed trials to make their way in to prestigious journals – JAMA is the third highest rated medicine journal in the world – the medical fraternity is doing itself a great disservice. Research can only be considered reliable when it is scientifically sound and can withstand the scrutiny of rigorous peer review. Once trust in science is eroded, the after effects will be felt long after the pandemic has abated.
Dr Narayana Subramaniam is a head and neck surgical oncologist at the Mazumdar Shaw Medical Centre, Bengaluru.