Remdesivir: Good news at last?

Beyond the curve: Dr. Peter Lurie's COVID-19 blog

It’s been almost two months since the World Health Organization declared COVID-19 to be a global pandemic. And since then, it feels like all we’ve had is a daily torrent of bad news.

But this week, for the first time, a ray of hope has shone through the fog of despair: preliminary results from a trial of the antiviral drug remdesivir suggest benefits in treating a disease that has already taken the lives of almost a quarter of a million people worldwide.

Here’s what we know: The study was funded by the National Institute of Allergy and Infectious Diseases (NIAID), part of the US National Institutes of Health. The trial enrolled 1063 patients, although it’s not clear on how many we have data yet. Patients were randomly assigned to intravenous remdesivir for up to ten days or to an intravenous placebo. Neither the patients nor the researchers knew what the patients had received. The drug had previously been tested without success during the Ebola epidemic, but has shown promise in animal models against MERS and SARS, other coronaviruses. It has not been approved in the US for any indication.

The main outcome of interest in the study was the time to recovery, and that’s where the benefit was demonstrated. Remdesivir-treated patients recovered after 11 days, compared to 15 days in the placebo control group. Mortality results also favored remdesivir (8.0% for remdesivir compared to 11.6% for placebo), but, unlike the recovery results, this difference was not statistically significant. It was close, though. The standard cutoff for statistical significance is having a so-called p-value less than 0.05; the p-value here was 0.059. Additional follow-up of these patients and others in the trial on which we don’t yet have outcome data could change these results in favor or against the drug. But the finding for recovery time appears to be reasonably robust. The Food and Drug Administration reacted at lightning pace, issuing an Emergency Use Authorization in just two days. This will allow use of the drug in hospitals for patients with severe disease, even as additional data are collected.

Unfortunately, that’s about all we know, and that comes mostly from a NIAID press release and press coverage of NIAID Director Anthony Fauci’s public comments. But the results have not been peer-reviewed, the standard procedure for scientific studies, nor have they been replicated by other scientists, something typically required for a drug to be accepted into clinical practice. Fortunately, almost a dozen more clinical trials with a variety of designs are still under way. All this makes Dr. Fauci’s statement that “This will be the standard of care,” seem a tad premature.

One thing we do know is that this drug is not the proverbial magic bullet. If real, its benefits are modest, though certainly welcome. We need considerably more detail from NIAID: the numbers of patients, any withdrawals from the study, adverse events, deviations from blinding, effects on other outcomes, etc. At a minimum, the preliminary results could be published on NIH’s own website, Clinicaltrials.gov, which is a repository of ongoing clinical trials, and submitted to a medical journal forthwith. Until that happens, color us encouraged but not yet fully convinced.

One more thing. These data were generated the old-fashioned way: by painstaking randomized, controlled trials, in this case funded by the US government. This is the way to generate data, not by hyping anecdotes and uncontrolled trials, the way the President and many of his supporters did for chloroquine and hydroxychloroquine.