First randomized trial of remdesivir suggests antiviral drug is not associated with significant clinical benefits, more research needed

Treatment with the antiviral drug remdesivir does not speed recovery from COVID-19 compared with placebo in hospitalised patients who are critically ill, according to the first randomised trial of its kind involving 237 adults (aged 18 and older) from ten hospitals in Wuhan, China, published in The Lancet.

Although their study is the first randomised controlled trial (considered the gold standard for evaluating the effectiveness of interventions) of remdesivir, the authors caution that interpretation of their findings is limited because the study was stopped early after they were unable to recruit enough patients due to the steep decline in cases in China*. They conclude that more evidence from ongoing clinical trials is needed to better understand whether remdesivir can provide meaningful clinical benefit.

In the absence of any known treatment for COVID-19, remdesivir is one of a handful of experimental drugs undergoing clinical trials worldwide. It has only been available to patients with COVID-19 on compassionate grounds (the use of unapproved drugs when no other treatment is available). In the trial, all patients received standard care including lopinavir-ritonavir, interferons, and corticosteroids.

“Unfortunately, our trial found that while safe and adequately tolerated, remdesivir did not provide significant benefits over placebo,” says Professor Bin Cao from China-Japan Friendship Hospital and Capital Medical University in China, who led the research. “This is not the outcome we hoped for, but we are mindful that we were only able to enroll 237 of the target 453 patients because the COVID-19 outbreak was brought under control in Wuhan. What’s more, restrictions on bed availability resulted in most patients being enrolled later in the disease course, so we were unable to adequately assess whether earlier treatment with remdesivir might have provided clinical benefit.”

He continues: “Future studies need to determine whether earlier treatment with remdesivir, higher doses, or combination with other antivirals or SARS-CoV-2 neutralising antibodies, might be more effective in those with severe illness.”

Remdesivir was originally developed to treat Ebola, and is designed to interfere with the way a virus reproduces, thereby stopping it from multiplying inside the body. It has been shown to successfully block SARS-CoV-2 from replicating in vitro, and had activity against other coronavirus infections like severe acute respiratory syndrome (SARS), Middle East respiratory syndrome (MERS), and COVID-19 in animal studies. Case studies have also reported benefit in some severely ill patients with COVID-19, but until now, there have been no clinical trials of remdesivir, so its efficacy against COVID-19 in patients is unknown.

In this study, 237 adults hospitalised with severe laboratory-confirmed SARS-CoV-2 infection were enrolled between February 6 and March 12, 2020. To be eligible, patients had to enter the study within 12 days of symptom onset, have pneumonia confirmed by chest imaging, and oxygen saturation of 94% or lower. Participants were randomly assigned to either daily infusions of remdesivir (158 patients; 200mg on day 1 followed by 100mg on days 2-10) or a placebo infusions (79 patients) for 10 days. One patient in the placebo group withdrew before receiving treatment.

Trained investigators measured time to clinical improvement over 28 days using a six-point scale of clinical status ranging from hospital discharge (score=1) to death (score=6). Clinical improvement was defined as at least a two point improvement compared to a patient’s admission status.

No statistically significant difference in time to clinical improvement was noted between the groups (average time to clinical improvement 21 days for remdesivir group vs 23 days placebo group). In a secondary outcome (a planned outcome measure in a trial that is not as important as the primary outcome measure, but is still of interest in evaluating the effect of an intervention), although not statistically significant, in patients treated within 10 days of illness onset, those receiving remdesivir appeared to recover faster than those given placebo (average time to clinical improvement 18 days vs 23 days).

Death within 28 days of randomisation was similar between the groups, with 14% (22/158) patients dying in the remdesivir group compared with 13% (10/78) in the placebo group. However, in secondary outcomes, patients treated with remdesivir within 10 days of illness onset had a lower, but not statistically significant, difference in mortality, with 11% (8/71) patients dying compared with 15% (7/47) receiving placebo.

Similarly, in secondary outcomes, the duration of invasive mechanical ventilation, although not significantly different between the groups, was shorter in remdesivir recipients than placebo recipients (average 7 days vs 15.5 days). No significant differences were noted between the groups in duration of oxygen support, length of hospital stay, or time to discharge or death (table 3).

Furthermore, treatment with remdesivir did not result in significant reductions in viral load (how much SARS-CoV-2 is present in an infected person’s body) or detection of the virus in the upper or lower respiratory tract compared with placebo.

There was no difference in adverse events between the groups (65%; 102/155 remdesivir patients included in the safety analysis vs 64%; 50/78 placebo), and the overall proportion of serious adverse events was lower in patients given remdesivir than placebo (18%; 28/155 vs 26%; 20/78). However, more patients in the remdesivir group discontinued treatment because of adverse events including gastrointestinal symptoms (eg, nausea, vomiting) and cardiopulmonary failure (18 remdesivir group; 12% vs 4 placebo; 5%).

“Participants in our trial were less ill, and treated earlier in their disease course, compared with a recent observational study of patients with COVID-19, who received remdesivir on compassionate grounds. Yet, remdesivir did not provide greater benefits in our study population as expected,” says co-author Professor Ronghui Du from Wuhan Lung Hospital, China. “Ongoing clinical trials of remdesivir, will provide important additional information about whether and under what circumstances it may provide benefit.”

The authors note several limitations of the study, including that stopping early gives insufficient power to detect differences in clinical outcomes, and it did not account for the possible emergence of reduced susceptibility to remdesivir (which would contribute to a reduction of remdesivir effectiveness), because of a lack of data at the time. Finally, they point out that the frequent use of corticosteroids in patients in this study might have aided viral production, as observed in SARS and MERS.

Writing in a linked Comment, Professor John Norrie (who was not involved in the study) from the University of Edinburgh, UK, discusses the challenges, and importance, of conducting randomised trials in epidemics, and writes: “Promising signals from observational data must be rigorously confirmed or refuted in high-quality randomised trials — particularly given that for COVID-19 no proven safe and effective treatments yet exist…That is additionally challenging in a pandemic, and the temptation to lower the threshold of convincing evidence must be resisted, because adopting ineffective and potentially unsafe interventions risks only harm without worthwhile benefit, while making it even harder to undertake trials to find truly effective and safe interventions.”

He continues: “The study was well designed — a double-blind, placebo-controlled, multicentre randomised trial — and well conducted, with high protocol adherence and no loss-to-follow up…An absence of statistical significance in an underpowered trial means that the findings are inconclusive. The particular challenges of delivering pandemic trials underline the importance of data sharing, allowing rapid curation of relevant datasets for individual patient data meta-analyses. With each individual study at heightened risk of being incomplete, pooling data across possibly several underpowered but high-quality studies looks like it will be our best way to obtain robust insights into what works, safely, and on whom. We eagerly await the ongoing trials.”