I skimmed the original paper, available here. From my quick read, the reasoning goes:
[ol]
[li]The lab studies basic science mechanisms of how protein translation machinery in a cell is hijacked by viruses.[/li][li]Previous work showed that HIV uses a protein called eIF5a, which is involved in protein translation in the host and a lot of related RNA biology.[/li][li]eIF5a requires certain specific modifications to become active.[/li][li]The enzymes that modify eIF5a are inhibited by ciclopirox.[/li][/ol]
Basically the whole journey starts with pouring through the literature for mechanistic clues – their intro alone references 55 papers covering decades of work from dozens of labs. Then there’s lots of failed experiments, very little sleep on the part of a lot of grad students, and eventually, with a lot of dedication and luck, finding a hypothesis that turns out to be correct.
My gut reaction is that ciclopirox might have substantial side effects if you dose it intravenously rather than topically. It is, after all, gumming up a fundamental piece of biochemistry. But maybe it could be added to the standard antiviral cocktails in a way that improves overall efficacy.
It’s a long way from “we’ve got a mechanism that works in a dish” to treating patients. And “long way” can be quantified in years, hundreds of millions of dollars, and 99% failure rates.