Now that we've cleared away a bit of the underbrush, let's say you think that Eye of Newt Toe of Frog (Eontof) is potentially therapeutically useful against Creeping Crud (CC), and you want to test it. You face a whole lot of considerations. One is that you're going to need funding, which means you need to persuade somebody -- either the National Institutes of Health or a pharmaceutical company, most likely -- to invest in your idea. They're going to want to know that there's a reasonable chance of success with Eontof, and in the case of the pharmaceutical company that they can make money off of it, which brings in other problems we'll put aside for now.
You basically have three possible kinds of argument for why Eontof is worth testing. The first is biological plausibility: you understand, or have a hypothesis about, the etiology of CC and there is a biologically plausible argument for why Eontof will interrupt the disease process. If there is a so-called animal model for CC, that is rats or rabbits or guinea pigs or whatever can have a similar disease, you can test Eontof in animals and see if it works. Don't get too excited though, what works in rats often doesn't work in humans. Finally, doctors might already be giving Eontof to people for some other reason and they notice purely by serendipity that it seems to be effective against CC.
How you and your potential funder interpret all this, and what your next steps are, depend on the nature of CC, which I might have discussed first but I have to take all this in some order so this is the one I chose. Infectious diseases are usually self-limiting, in other words the immune system kicks in and kicks the pathogen out, but the ones that worry us the most and are likely to attract funding for clinical trials will sometimes go on to produce serious disease that threatens long term harm or death. Examples are mostly famous -- plague, smallpox, influenza, Covid-19 . . .
Before you undertake a clinical trial you need to specify your endpoints, in other words you need to state exactly what your hypotheses are about the effect Eontof will have. Presumably these might be a smaller percentage of people who experience severe disease -- which you have to define -- a lower mortality rate, obviously, and maybe faster recovery on average -- which you also have to define. It's a lot more complicated, however, because the risks of CC vary according to people's pre-existing characteristics -- maybe age, sex, socioeconomic status, overall state of health, but also maybe things you don't know about or can't readily measure such as genetics, prior exposure, diet, whatever. Another problem if you're relying on people telling you how they feel to define and measure recovery is the so-called placebo effect. If you give people a completely inert pill, that they think might do some good, they'll be more likely to report that they feel better even if there is no measurable biological reason for them to say so.
The problem of endpoints is different if CC is, let's say a chronic or progressive disease that is rarely self-limiting. (Even cancer and maybe some auto-immune diseases will occasionally spontaneously remit so I don't want to say never.) Then your endpoints might be slower progression or symptom relief rather than cure. The issue may purely be risk - the probability of a heart attack, let's say. But it's still the case that you need to be concerned about other "confounding" factors such as age, sex, other health conditions etc., and the placebo effect.
We'll design our clinical trial next time. But for now, just keep in mind that the people who touted hydroxychloroquine and ivermectin for Covid-19 didn't
No comments:
Post a Comment