Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Sure, there's a few. But 3 rodent studies isn't exactly enough evidence for a layperson to worry about, either. It's not even much of a signal for scientists in that area of research.

Ivermectin is pretty safe for people to use regardless of whether or not they have parasites, so sure, do the human RCTs. Maybe we'll get lucky and have another tool in our anti-cancer toolbox.

But trying to extrapolate out that it's reasonable for people to take it for cancer based on the current evidence is premature, at best.



My point remains (and that I learned) is efficacy in rodent studies is more than just "in vitro" or quackery.

>layperson to worry about, either.

We're not talking about laypeople without money nor access to pursue. Adams had money, access, and desperation.

Society would love to put this to bed, but pharma typically avoids funding RCTs for out-of-patent/cheap drugs so we may never get the answer.


Projecting human outcomes from rodent studies is 100% quackery


That's not true given FDA more or less requires rodent testing in pre-human trials.

You didn't know that?


They are a required step along the way to human trials.

But over 90% of drugs that show promise in rodents flunk out in human trials.

Something working in rodents is an indicator that it might be worth doing testing to see if it works and is safe for humans. But if you bet against it panning out, you'd still be right the overwhelming majority of the time.

The only thing you can project from rodent trial success is that it is worth continuing to study. It should not guide any human usage at that point.


OK.

I now understand you were being sensational with your prior language:

> Projecting human outcomes from rodent studies is 100% quackery


No. It's still absurd to project human outcomes from rodent studies.

90-95% of them don't pan out! And that's of the ones that progress from rodent studies to human trials. The actual number is even higher, but more difficult to track.

Surely you can see how it would be absurd to extrapolate success from something with, at best, a 5-10% chance of panning out? And panning out as in being approved - lots of things that are approved have less than 100% success rate, particularly in this area.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: