Among the many drug-discovery lessons that this pandemic is highlighting is the difficulty of meeting the challenges of a new target, a new pathogen, a new disease, or a new mechanism of action. When you look at the history of the drug industry, the typical time for an effective therapy to be developed from a standing start is somewhere around ten years. Once in a while it’s shorter, and it can be a lot longer. You can see this when a new virus appears (such as HIV), or when a new disease process is discovered (either a particular biochemical pathway or the gene for a genetically-driven condition), because those have a definite starting pistol that goes off. For other things, well. . .try this example: thanks to Gilead’s combination therapy (and the work of many other companies along the way!) we can now flat-out cure hepatitis C. For how many decades had people been trying to treat hepatitis C before that was realized? At the far end of the scale, for how long have we been trying to treat (say) Alzheimer’s, with basically nothing being realized at all? Drug discovery is hard. We keep on saying that, sure, but that’s because it keeps on being true.
Now, in the pandemic era, we get to see the process work under time pressure. I’ve talked about this topic in passing in other posts, but I thought it might be good to break it out by itself. The list, in order of rapid deployment, looks something like this:
1. Existing drugs. We don’t really have anything approved against coronavirus infections, so all of these are going to have to be repurposed, and you have to hope that something works. Don’t expect cures – the odds of something working that well when it was developed for some other use entirely are extremely small. Wandering through the shelves of the auto-parts store and throwing things out into the parking lot at random is unlikely to fix your broken lawnmower, for similar reasons. If you’re very lucky, you might find something that can be jammed on with a hammer and aligned with duct tape, and that’s pretty much what drug repurposing is like even at its best. But the good part is that if you find something, it can be deployed more or less immediately, especially as compared to the next options. This is where remdesivir, hydroxychloroquine, azithromycin, falapirivir, ivermectin, basically every single one of the drug ideas you’ve heard about come from – they’re all attempts at repurposing things that weren’t developed with a coronavirus in mind. I’ll have more to say on the general topic of drug repurposing in a post in the next few days.
2. Monoclonal antibodies. These take months (in theory) to years (in practice, so far), and only when you have an absolutely clear shot at a particular protein that you know you want to shut down. That’s what mAbs are good for: they will stick to a protein and jam up its function but good. And that’s the tough part, because all too often we don’t quite know what protein we should be zapping in order to fix a disease. There might be no good candidates, or there might be too many to choose from with no real standouts, or you have strong reasons to believe that blasting your target protein out of the sky will do lots of other things you don’t want, or the only protein targets that we’re certain of are things whose activity we would want to increase, not wipe out, and so on.
The good news for us versus the coronavirus is that we have such a target (the viral Spike protein or some part of it) that we have strong reason to believe would shut things down, and the mAb folks (Regeneron and others) are going at it hammer and tongs as we speak. They’re trying to set the new speed record, so we’ll see how that goes. Keep in mind, a mAb therapy would not confer lasting active immunity (via antibodies of your own), but it would be something targeted and effective that could be given to you if you have a bad nCoV-19 infection, and we don’t have anything of the sort right now. I’ll be doing a post on this stuff soon.
3. Vaccines. I’ve done a couple of whopper blog posts on these recently, so I’ll just refer everyone to those. The current speed record is about five years, for the Ebola vaccine, and as you can see from those earlier post, everyone is pulling out the stops to break that by a wide margin. I have no idea if that will work; vaccines typically have the same sorts of fearsome attrition rates in the clinic that the rest of the industry has. The hope is that if we have a lot of people taking shots at this from a lot of different directions, then something will get through. The number of people trying to develop a vaccine might cancel out the uncertainties. We all get to see if that’s true, lucky us! If everything goes perfectly for someone, first time through, then we might be ready for an emergency use authorization early next year at the very earliest possible date. Beyond that, there’s literally no way of knowing.
The great thing about a truly effective vaccine is that it makes the disease go away. Smallpox went away. Polio has gone away, except in a few places that are in such bad shape, politically, physically, and/or mentally, that the vaccination campaign hasn’t quite made it through. In the developed world, things like rubella, measles, diphtheria and whooping cough have largely gone away, and would pretty much go away forever if everyone got vaccinated, which is why anti-vaccine activism makes so many of us in the business want to scream and kick things. And a solid coronavirus vaccine could do the same: we’ll have to see if such a thing can be developed, because you’ll notice that there no one could ever get that to work against (say) hepatitis C or tuberculosis.
4. New bespoke therapies. As mentioned, not every pathogen can be dealt with by a vaccine (although I have every expectation that this coronavirus can be, really). And not every disease, needless to say, can be solved by raising or injecting antibodies to some protein. So for those occasions, we’re back to good ol’ small molecule drugs or engineered proteins, Huey Lewis “I Want A New Drug” territory, and this is the longest timetable of all. Bringing up such things from scratch is fraught with uncertainty. You have to find a target, and then hope that your target is something that you can find potent chemical matter against, and hope that said chemical matter doesn’t do a bunch of other horrible stuff to people at the same time, and can be dosed in a reasonable fashion at a reasonable interval, all the classic worries of classic drug development. And then you get to go to the clinic and hope that you picked the right flippin’ target in the first place. If you’re lucky, there’s a good enough animal model to help you defuse that question before you go to the massive trouble and expense of clinical trials, and the good news in infectious disease is that such animal models are more likely to exist than they are in most other therapeutic areas.
The bad news is that all these development steps take time, and some of them are pretty incompressible. Finding a good compound takes the time it takes, however long that is, and you have no way of knowing beforehand what that will be. Safety testing takes the time it takes, and although you have a much better idea of how that is, you can’t shorten it and you generally can’t go into humans until it’s done. And so on. The upshot is that this is the longest option of all; getting it all done in five or six years would be extremely impressive. As mentioned above, the standard is more like ten. If we absolutely have to do this for the coronavirus, it’s because the mAbs and the vaccines have failed us. We will be Breaking Glass in Case of Emergency, and that will not be a good thing.
But I really don’t think we’ll end up there. I think the chances for useful recombinant antibody therapies are very good (although manufacturing them on scale will be nontrivial, as we like to say), and I think that the chances for a useful vaccine are also good, especially with as many organizations hammering away on the problem as we have. We may in fact end up with the luxury of arguing over which vaccine is more appropriate, and that’ll be just fine.