Forecasting an 80% chance of an effective anti COVID-19 drug (probably Remdesivir)post by AllAmericanBreakfast · 2020-03-15T19:21:31.187Z · LW · GW · 6 comments
I'm gathering info on anti-COVID-19 drug efforts. Here is what I've compiled so far. I've combined the number of trials in various phases with big data on expected success rates for drugs in those phases to forecast an 80% probability that at least one drug currently in phase 1-3 testing will be approved. Because trial efforts are unusual, I've used the following categories in my model:
0: pre-trial (not factored into probability of success)
1: new drug in early testing
2: approved drug being tested against COVID-19
3: official phase 3 trial
This is just the product of a morning's research, so if anybody finds better information, I'd appreciate it if you'd link a source so I can include it.
Remdesivir is the drug we'll know about soonest - one expert is quoted as saying in a couple months, so around May 2020. Other analysts say they're optimistic that production can be scaled up quickly. It treats the pneumonia of severe cases, so it might cut down on mortality. Other treatments are also being tested.
The CDC projects that "In the coming months, most of the U.S. population will be exposed to this virus." If the virus grows exponentially, the relative timing of a Remdesivir rollout and a COVID-19 explosion could matter hugely in terms of mortality.
Remdesivir has been administered via compassionate use permission already for at least 15 patients who weren't part of the trial (which is recruiting volunteers now). So although we won't know its effectiveness for a while, if it is effective, it can potentially begin saving lives immediately via compassionate use permission.
Comments sorted by top scores.