On April 29th Anthony Fauci announced the National Institute of Allergy and Infectious Diseases, an institute he runs, had completed a study of the antiviral remdesivir for COVID-19. The drug reduced time to recovery from 15 to 11 days, he said, a breakthrough proving “a drug can block this virus.” Now people are clamoring for $1,000-a-pill remdesivir.
While the results were preliminary, unpublished, and unconfirmed by peer review, Fauci felt an obligation, he said, to announce them immediately. Indeed, he explained, remdesivir trials “now have a new standard,” a call for researchers everywhere to consider halting any studies, and simply use the drug as routine care.
Let us set aside whether research should ever be stopped based on a single group’s (unpublished) findings. Let us also set aside an equally rigorous trial—confirmed by peer review, published, and released the same day as Fauci’s press briefing—found remdesivir not only didn’t improve clinical outcomes, it had zero impact on viral load (ostensibly the drug’s mechanism of action).
Instead, let us focus on something Fauci stressed: “The primary endpoint was the time to recovery.” This was strange: a redundant, technical statement. Results touted to the press, as any researcher knows, must always be primary. Anything else would be ethically dubious, potentially misleading.
Unfortunately, the trial registry information, data which must be entered before and during the trial’s actual execution, shows Fauci’s briefing was more than just misleading. On April 16th, just days before halting the trial, the researchers changed their listed primary outcome. This is a red flag in research. Some press reports took a charitable tone, calling the maneuver “unusual but not unheard of,” or saying “there can be good reasons.” They are mistaken.
The primary endpoint the researchers originally selected for their trial was a scale of worst-to-best outcomes: death, then need for mechanical ventilation, then six more levels ending with fully recovered. This choice, made in the planning stages, was the project’s defining step—the trial’s entry criteria, size, data collection, and dozens of other elements, were tailored to it. This is the nature of primary outcomes: they are pivotal, studies are built around them.
Therefore they didn’t ‘change’ the primary outcome. The trial was essentially finished when they listed a new one. Rather, they changed what they were calling the primary outcome.
In other words they shot an arrow and then, after it landed, painted their bullseye.
Fauci is a career researcher and knows altering a primary outcome after a trial has begun or is complete is among the most recognized forms of fraud in research reporting. Moreover, a spokesperson responded to inquiries explaining the change this way: early in the trial, while still blinded to their data, the researchers decided to switch outcomes because they realized the original outcome might miss “meaningful treatment effects.”
Well, duh. Choosing any primary outcome means potentially missing other effects. Research is hard. You set a goal and design your trial to reach for it. This is the beating heart of the scientific method. You can’t move the goalposts. That’s not science.
Moreover, the explanation is daft. We are in the midst of a deadly pandemic, with ventilator shortages looming. Since when is ‘time to recovery’ more meaningful than death, and need for ventilation?
There are other problems with their explanation. For one, the NIAID trial used something called an “adaptive” design, which, by definition, means researchers are closely watching their results. But even if study leaders were somehow shielded from these, according to the registry, they increased the target trial size at least three times before April 16th, adding hundreds of new subjects. This typically happens when a drug is failing, and researchers attempt to salvage a study by setting a new, lower bar for success (bigger studies can detect smaller differences). It also effectively unblinds study personnel to the fact that a drug is failing.
It gets worse: the researchers intended to record ‘time to recovery’ all along, but just not as a primary outcome. In fact, the registry shows they originally had three different ways of measuring it (secondary outcomes #28, #29, and #30). But when they turned this into their primary outcome, they chose an entirely new definition and a new way of calculating it.
First, they moved the goalposts, then they changed the size and shape of the goal.
Whatever the explanation, the trial they conducted was tailored to a different outcome than the one they reported. In the research world, that is cause for alarm. Now, with no paper or publication, the researchers are claiming both the original and new outcomes demonstrate remdesivir has significant benefits. We should be reassured, they say. The ends justify the means.
Something smells wrong. As we eagerly await full publication, the physician community should bear this in mind: with millions of lives in the balance (and potentially billions of dollars), there is good cause to take a very close look at Fauci’s study—the only one in history to find a benefit with remdesivir.
Daniel Hopkins is a physician.
Image credit: Shutterstock.com