Let’s have a look at the case of Unity Biotechnology, because this is a story that won’t get so many headlines. Unity has been investigating a really interesting but high-risk idea. It’s in the anti-aging field, so those two adjectives sort of apply by definition, and it’s the hypothesis that one of the problems is the accumulation of senescent cells. Those are old cells that are not yet dead, but not so alive, either – they’ve lost the ability to divide, and problem other abilities as well, but they’re still just hanging around. In fact, they seem to be doing worse than that – senescent cells secrete various inflammation signals and other molecules that actually seem to impair the function (and even the survival) of the cells around them. They’ve been implicated in a whole list of degenerative diseases, and there have been several studies in animal models that show beneficial effects of outright elimination of such cells on overall health and even lifespan as a whole (see those last few links for references).
Their lead candidate is UBX0101, which is a small-molecule inhibitor of the MDM2/p53 interaction (I’ve been unable to find a structure). That’s a protein-protein target that has received a lot of attention over the years, and compounds that affect it have been reported to disrupt the whole senescent-cell secretion phenotype. It’s a complicated story – compounds that inhibit the MDM2 interaction often do so by binding to p53 and stabilizing it. That also tends to mean activating it as well, because MDM2 is sort of the default braking system for p53, which is at the center of a great big gigantic web of cellular activity. It’s involved in tumor suppression (by monitoring for DNA damage), the cell cycle itself, apoptosis (another option if the damage is too severe), and more, and it certainly seems to have a big role in the done-with-that-cell-cycle-business phenotype of senescent cells. As that last link will show, though, such MDM2 inhibitors were first characterized as caused normal cells to become senescent themselves, but closer examination seems to indicate that this was a reversible effect once you stopped dosing cells with the compounds, and that their effect on natural senescent cells was something else entirely.
So Unity had done a Phase I trial with their compound in patients with osteoarthritis of the knee, which showed that it was well tolerated and actually seemed to show the sort of biomarker response that you’d want to see. Today they reported their Phase II data: nothing. No response at all versus placebo in any arm of the study. The compound’s development has been halted, and Unity’s stock has taken a dreadful beating this morning on the NASDAQ.
There are several things to take away from this. None of them are new, but we all could use to be reminded of them (and people outside the field definitely need to be!) First off, Phase I results are not Phase II results, because they are not designed to read out on efficacy. We all need to keep that in mind during these coronavirus days – we simply *do not have* the most important results for all the vaccine candidates that are in the clinic now, and trying to read the Phase I tea leaves can only take you so far. There is no substitute for a well-designed Phase II study, and every new drug has got to pass through one before you can say anything real about its use in human disease. It’s also worth remembering that Phase III results rarely look better than the Phase II ones – if anything, Phase III tends to expose limitations while it’s confirming that the good parts of the Phase II work were actually valid (if in fact it does that second part at all!) This particular drug is obviously not going to get there, but keep that in mind next time something barely clears a Phase II and then goes on further.
Another take-away is that anti-aging work is really, really hard. Gegen den Tod ist kein Krautlein gewachsen, or if you’d rather take that dose in Latin hexameter, contra vim mortis non crescit herba in hortis. In other words, there is no herb against death. (I came across that one not by poring over collections of Latin sayings, but by spending my youth reading science fiction stories). I continue to think that combating aging is a very exciting area for therapeutic intervention, with potential effects on both overall health and lifespan, but it is an area that’s very likely to hand us a lot of unexpected reverses. Like this one.
There are a lot of places where things could have broken down. From the micro to the macro, they include: maybe this compound isn’t the right one to affect MDM2/p53, but it’s still a valid target. Or maybe that particular interaction isn’t a valid target for senescent cell therapy, for reasons yet to be worked out. Perhaps there’s something about osteoarthritis itself that makes it a less attractive test bed for this approach. Or it could be that senescent cell therapy in humans is just not going to work at all. The only way to understand these things is to do more work, apply more brainpower, and spend a lot more money. The hard way, in other words.