Enthusiasm for remdesivir as the new standard of care for COVID-19 has only grown since the release of preliminary results from the National Institute of Allergy and Infectious Diseases (NIAID) trial. But something is deeply wrong with the way the preliminary results were rolled out. Despite the hype, a vigorous examination of remdesivir studies suggests that something is amiss with how the drug is working.

Remdesivir is an antiviral drug that is designed to target the center of SARS-CoV-2 and prevent it from replicating. Any positive effect should be apparent in a patient’s viral load. But a controlled clinical trial on remdesivir conducted in China and published in the Lancet suggests there is no such drop in viral load. Among the 237 patients in the study — 158 receiving remdesivir and 79 controls — there was no difference between the groups in viral load reduction. 

This means that if remdesivir is working, it is not because of its antiviral effect. 

Many have critiqued the size of the China trial, as too small to prove anything conclusively. But when it comes to antivirals, very few patients are needed to show whether a drug inhibits replication or not. The first effective treatment for HIV/AIDS — also an antiviral — was proven to work in a group of just 36 patients: 18 given the drug, 18 controls. 

Though laboratory tests were conducted as part of the NIAID study, reduction in viral load does not appear to be an integral part of the study. If it was included, that data should have been released along with the preliminary results. If it wasn’t included, we need to know why — reduction in viral load is a widely agreed upon marker of success for antiviral drugs.

Were we to have access to the full data from the NIAID trial, we might be able to determine what impact remdesivir has on viral load and whether those effects have any impact on how effectively the drug is working. But so far, all we have are those preliminary results. 

The makers of remdesivir recently secured FDA approval for their drug on the basis of the preliminary results. Now they are securing licensing agreements to ramp up global supply and determining how much they will charge for the drug — all without data from the largest controlled clinical trial on the effectiveness of the drug having been released. 

I would like to believe — as we all would — that there is nothing to be wary of in the way events are unfolding. But I can’t help but urge caution on remdesivir until the key question is answered: Where are the NIAID results? 

This article originally appeared on Forbes.