Doctors flying blind deciding which patients should receive remdesivir

After months of rising death tolls and anxious anticipation of a therapeutic breakthrough, on April 29 we learned of the first two clinical trials of remdesivir, an antiviral agent targeted at COVID-19. Two days later, the Food and Drug Administration issued an emergency use authorization that allows physicians to request use of the drug. But the manner in which the data were presented, and, perhaps more importantly, how they were not, have left physicians and patients with more questions than answers about when, how and to whom this treatment might be given.

The projected demand for the drug has already outstripped supply. Gilead Sciences, the manufacturer, has pledged to donate its currently available stock: enough for approximately 50,000 patients, with up to 90,000 more courses to be produced by July. The most recent reports from the Centers for Disease Control and Prevention estimate 20,000 to 25,000 new hospitalizations per week, meaning Gilead’s supply could be exhausted within weeks.

In deciding who should receive this medication, how does one choose between, say, the 34-year-old health-care worker requiring oxygen support and the 65-year-old liver transplant recipient who has spent the past two weeks on a ventilator? Typically, to determine how to prioritize a limited supply of a high-priority drug, epidemiologists delve into existing studies to help clarify who would get the maximum benefit. In the case of remdesivir, those methods fall short.

The first of the two studies, from Wuhan, China, was peer-reviewed and published April 29 in the Lancet. The investigators enrolled only 236 individuals, just half their intended sample, because a decline in covid-19 cases had reduced the number of eligible participants. The authors reported no significant differences in mortality (14% with remdesivir, 13% with a placebo), in the time to disease improvement (21 days with remdesivir, 23 days with the placebo) or, most concerning for a drug that works by blocking viral replication, in time to clearance of the virus (no difference up to 28 days). Furthermore, the common use in China of immune-suppressing medicines might cloud this study’s interpretation in the United States, where the practice is rare.

The second of the two studies on remdesivir, the Adaptive COVID-19 Treatment Trial, conducted by the National Institute of Allergy and Infectious Diseases, included a much larger number of patients, but also leaves us needing more data. This study was not published in a peer-reviewed journal or posted online in preprint format; instead, a study oversight committee reported preliminary results in an April 29 news release. What did we learn? That patients who received remdesivir had a marginally faster time to recovery than those who got a placebo (11 days vs. 15 days) and a non-statistically significant decreased mortality rate (8% vs. versus 11.6%), and that “more detailed information about the trial results, including more comprehensive data, will be available in a forthcoming report.”

What didn’t we learn? Essentially anything else. Who got the most benefit from the drug – the young? Elderly? Those needing only a touch of oxygen support, or those suffering from the life-threatening pneumonia often seen in the “second wave” of the infection? Did the drug prevent a need for ventilators, or reduce days spent in an intensive care unit?

In the wake of this media bombshell and data tease, we are left with a treatment that the FDA has permitted us to use, and that patients and families will justifiably expect, but with an extremely limited drug supply and no evidence-driven guidance on how to use it. Health-care systems are faced with the daunting task of designing ethical protocols for rationing this agent. Moreover, if remdesivir is now perceived to be a standard of care for COVID-19, studies of other drugs could be viewed as unethical if they do not use it for comparison. But since so little of the drug is available, obtaining it to conduct those studies might not be feasible.

So we close with a plea: Make the data public from the Adaptive COVID-19 Treatment Trial. Doing so would comply with National Institutes of Health policy on federally funded research. It would allow the public, epidemiologists and policymakers to analyze, vet and interpret the trial results. It would promote transparency and consistency in our decisions about how we use this rare and potentially valuable commodity. Simply put, the individual and policy decisions to be made in response to the FDA’s emergency use authorization are too critical to allow the data in support of it to be kept behind closed doors.

For now, physicians, patients and policymakers are left to guess how, when and why remdesivir should be used. Until we can see the data, the first point in the Gilead Sciences notice about the drug remains startlingly true: “It is not yet known if remdesivir is safe and effective for the treatment of COVID-19.”