Markets got very excited (briefly) this week about a study finding a Gilead Sciences drug helped coronavirus patients heal a little more quickly.
But that was all the trial found: remdesivir isn’t the miracle cure that will get us all out of lockdowns tomorrow, unfortunately.
Worse, as Bloomberg's Faye Flam writes, the trial was rushed to get quick FDA approval, without getting helpful information on what kinds of patients it helps or hurts the most; and now that the study is over, we’ve forever lost a chance to help doctors treat virus patients better.
All of which raises a significant number of questions and Acute Care Surgeon (and Asst Professor of Surgery at Wash U.) Mark Hoofnagle warns "I am truly sorry to say, Remdesivir is probably worthless..."
In an excellent Twitter thread, Hoofnagle details what he calls "some fascinating drug company shenanigans."
First, the pre-test probability that an infused, small-molecule inhibitor of a virus would improve mortality in symptomatic patients was already pretty low. Unfortunately, antivirals work poorly in acute disease. This has to do with their mechanism of action, and host response.
Antivirals usually target some aspect of viral replication/assembly/transmission. Remdesivir is a clever pharmacologic prodrug that inhibits a key piece of RNA viruses that mammals don’t have - the RNA-dependent RNA polymerase, and inhibits viral replication.
Unfortunately, by the time you are symptomatic with a virus, you are usually already high/peak viral load. So, when you give an antiviral to someone who is already ill, the damage from the virus is largely done.
It’s there in big numbers and in the cells.
Pick your metaphor. The cat is out of the bag. The damage is done. At this point the host response to virus is activated, and your body is suppressing replication through a variety of mechanisms (which also make you feel terrible).
So how could inhibiting RDRP after the fact help? The answer is, it probably doesnt. It certainly didn’t in this trial - no difference, not even a trend in mortality, but in subgroup analysis maybe shortened disease duration in early/mild disease.
Now, critics of stupid drugs that should never have been stockpiled by govts say, “sounds like Tamiflu!”
Yes. This is the same as Tamiflu, which also maybe shortens flu by a day, but otherwise is a largely useless antiviral (and actually harmful with bad side effect profile).
Fortunately, side effects of remdesivir did not seem severe in this trial with only about 3x as many patients stopping than placebo, some rashes, nothing life threatening.
Where do the Shenanigans come in?
Well, remember how maybe this Chinese trial showed a shortened course in a subset of patients? Like tamiflu? But didn’t change mortality?
Well a month ago the NIAID trial changed their endpoints to remove death and instead look at dz duration.
They changed the destination half way through the race to match the only positive outcome of another trial, that they (or Gilead at least) certainly had a copy of the paper once it was submitted to Lancet.
Shenanigans! Get a broom!
Since NIH remdesivir trial is in the news...— Walid Gellad, MD MPH (@walidgellad) April 29, 2020
was there an explanation about why the primary outcome (now positive) was changed last month to 'time until clinical recovery?' @matthewherper https://t.co/fCTc1EGI1d pic.twitter.com/W1hAACnO1r
This is like declaring a race and then when you realize you’re not going to win, declaring the destination was actually wherever you are standing at the moment.
Then, even more fishy, *the same day* as this Lancet trial is release, Gilead and NIAID claim a “positive trial” and they’ve “shortened the course of the disease significantly”. Notably, the mortality benefit did not reach significance.
By the end of the day, reports that FDA is going to emergently approve remdesivir for treatment of COVID.
Gilead gets what they want. No one will want to be in a control arm in further trials and they will argue all future trials must be non-inferiority.
Before we have the answer whether this drug actually changes anyone’s destiny, it’s going to become the gold standard therapy. We will likely now never know if (the unlikely possibility) it changes mortality.
Absolute genius. You have to salute them. On the day a negative trial of their drug is reported, based on a press release they took over the news cycle, and with some midstream edits to their endpoints their now “positive” trial wins them FDA approval and a halted trial.
It’s an infusion, once symptomatic, you need an admission, a test, etc., really even symptoms are probably too late a goal for such a therapy to work. Prophylaxis (like Gilead’s Truvada/PreP) would be better - but unworkable in its current form.
Either way, a big win for Gilead, but I’m unimpressed with any if the evidence presented so far that this is a game changer.
* * *
How long before Hoofnagle is banned from Twitter?
Hoofnagle is not alone in his skepticism.
Naked Capitalism's Yves Smith exclaims, it is disturbing to watch the push to con the public into seeing remdesivir as the only promising treatment for coronavirus, and points to a new study that found and tested 47 old drugs that might treat the coronavirus: Results show promising leads and a whole new way to fight COVID-19