The full data from the remdesivir trial by the National Institutes of Health (NIH) has finally been released, weeks after preliminary results were shared via press release. Those early results generated a great deal of hype and a world wide stampede for the drug. But the release of the full data tells a different story.
The conclusion of the NIH study, published in The New England Journal of Medicine, states:
“These preliminary findings support the use of remdesivir for patients who are hospitalized with Covid-19 and require supplemental oxygen therapy. However, given high mortality despite the use of remdesivir, it is clear that treatment with an antiviral drug alone is not likely to be sufficient. Future strategies should evaluate antiviral agents in combination with other therapeutic approaches or combinations of antiviral agents to continue to improve patient outcomes in Covid-19.”
There are a few problems this paragraph highlights.
First, by opening with a reference to the data “preliminary findings” the authors are acknowledging the fact that this is not a definitive study. How could it be with such scant data? Ideally, this NIH trial and other remdesivir trials would continue to gather evidence to support the findings. But the NIH controversially ended their large trial of the drug and Gilead has suspended smaller trials. More study of the drug will still be conducted through the WHO Solidarity Trial and the Inserm DisCoVeRy Trial, so there is at least a glimmer of hope that we will eventually have robust and transparent data that shows the true impact of the drug on viral load.
Second, the patients who benefit from the drug are those “who are hospitalized with Covid-19 and require supplemental oxygen therapy.” Why must patients be hospitalized to receive the drug? Because the drug has to be administered intravenously, through an in-dwelling catheter. This means that patients cannot take the drug from their homes. The data, preliminary though it is, shows that the drug only helps those with moderate Covid-19. If you view the severity of Covid-19 on an eight point scale, with one being those not hospitalized and 8 being dead, the drug only helps those in groups 4 (hospitalized, but no need for extra oxygen) and 5 (hospitalized and needing oxygen).
In other words, the drug is only useful for those who are likely to recover anyway, with or without remdesivir. The data here and in another published study shows the drug has no measurable effect in patients with serious disease. We also don’t have any proof that the drug prevented progression of the disease from serious to lethal.
The next section contains a more ominous caveat and a dubious proposition. The caveat is included in the sentence that says “given the high mortality rate despite the use of remdesivir,” which suggests that patients who received a high dose of remdesivir intravenously for ten days continued to die at the same rate as the placebo group. This is troubling as it calls into question whether the enthusiasm for Gilead is justifiable in any sense, if the same number of people who fall ill with the disease will continue to die.
The dubious proposition is in the section that says “it is clear that treatment with an antiviral drug alone is not likely to be sufficient. Future strategies should evaluate antiviral agents in combination with other therapeutic approaches or combinations of antiviral agents to continue to improve patient outcomes in Covid-19.”
How can the authors be so sure that an antiviral drug alone won’t work against Covid-19 when antiviral cocktails can cure those with hepatitis C and treat those with HIV/AIDS? And why are they not explicitly suggesting remdesivir as part of a future strategy to evaluate antivirals with other therapeutic approaches?
As far as conclusions go, this one does not give us much to hang our hats on. A deeper dive into the trial design, changed and truncated midcourse as it was, shows that “time to discharge” is one of the weakest criteria by which to gauge the success of a drug therapy. Rigorous criteria include reductions in mortality or, at a minimum for antivirals, reduction of viral load before and during treatment. That is how you know an antiviral drug is working as it is supposed to work. Apparently preventing death and measurement reduction of viral load was not part of the trial design. Is a skeptic allowed to ask why? Perhaps because the previously published study found no such effect. Is remdesivir the “fuzzy wuzzy drug” like the bear with no hair, an antiviral that has no antiviral effect in people?
Doctors and patients both have the right and the need to know how well the drug works and who it works for. Does this drug save lives and does it reduce viral loads? Impossible to know even with the full data from the NIH study which ended too soon.
The need for urgency does not justify incomplete and poorly designed trials. Doctors and patients are hungry for clear, definite answers. NIH and Gilead know how to design and to present such trials that give comfort and guidance to practicing physicians. To do less is a dereliction duty.