I guess you have never heard of Viagra.

>This drug has multiple activities, one of which is to alkalise the phagolysosome, which hampers the low-pH-dependent steps of viral replication, including fusion and uncoating. Other mechanisms of antiviral activity are poorly explained.

Which sounds like limited evidence.

[0]: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7134866/

You would be surprised how many drugs have an unclear mechanism of action. If you think we know everything in Biology you might want to revise your judgment.

Furthermore, most chemical engineering/pharmacy students learn that with today's regulations, acetaminophen would've never been approved. Do we think that's a good thing? I don't.

That said, I still very much agree with the parent poster's view.

Once they get hits, they run more big assays, winnowing down likely tox issues or off-target effects, then try them in animal models, then eventually get to Phase I.

https://www.sciencedirect.com/science/article/pii/S016635422...

In general small molecule drugs often hit multiple different pathways. It's not reasonable to put them into neat little buckets.

I suppose what this underscores is: biology is immensely complex and first-principles thinking can only get you so far. Eventually, you need to test things with a big RCT to know what's going on.

[0] https://www.cancerresearchuk.org/about-cancer/cancer-in-gene...

Here is a list of 68 drugs that are known to work, but apparently we don't know how or why:

https://en.wikipedia.org/wiki/Category:Drugs_with_unknown_me...

I mean, it sometimes works, and obviously it's a very attractive idea (no need to wait on slow trials; drug is already assessed for safety) but precisely because it's such an attractive idea people seem to be overly credulous about it.

There is a ton of politically driven disinformation around. Brazilian president Bolsonaro pushed both as treatments while, we now know, government representatives were stalling direct purchases from labs while negotiating vaccines with intermediaries in order to profit from sales to the government.

Which was the excuse they gave when they designed actual drug trials that gave patients lethal doses and ignored the testimony of doctors who found it useful at sub-lethal doses. And therefore a drug used by tens of millions of people a year(if not more) for the past five decades was declared too dangerous to use.

Addendum: I wound up arguing with someone that thus-and-such a trial was actually using what was considered an overdose in 2017 and they would not look at the entry in the PDR.