Read the Full Video Transcript

Deborah Charych: Good afternoon. I'm Deborah Charych. And I'll be talking to you today about the value of animal models in translating from mouse to human.

So the answer is going to be Yes, they're valuable. But we'll go into a little more detail about that. So at the risk of saying the obvious, we're in the-- my group is particularly in the field of drug development. So we're developing new molecules that don't yet exist. So at the risk of stating the obvious, when you're developing new de novo chemical matter, we don't know yet what it's going to look like in a human.

So absolutely, the mouse models are critical. And the reality is that we're actually always learning from you guys, from you guys as you're developing these drugs in the clinic, the things that we think look good, we reverse translate back to mouse and then use that as a template for our new molecular designs.

So what I thought would be fun to do would be to have a mouse tumor board in a way. So we're going to go through some molecules and we're going to pretend that these are new molecular entities that don't exist yet. And we're going to ask the question, should we progress them further into development, or should we kill it?

So we're going to pretend there's this new agent that was investigated in mice, and we were investigating its clearance properties in two mouse strains. And the kidney retention was negligible one day after injection. And this was in two separate mouse strains. Should we take this molecule forward into development or kill it?

OK, do you know what the molecule is? I guess you can guess. And this, of course, is PSMA-617, which for the property of kidney retention or kidney clearance from a pharmacokinetic perspective, is not too bad in mice, and it's also not too bad in humans. And in fact, it's actually a great benchmark for us when we're developing new molecules. PSMA-617 is a really good molecule. It clears nicely. Doesn't get retained.

It does what it's supposed to do. It exits the body, and it excretes in the urine. And that behavior, while it may not be exact from a numerical perspective between a mouse and a human, the behavior is the same. And so we like to use that as a template.

So the next mouse tumor board, I guess, is a new agent was tested in three different mouse strains. In one mouse strain, the kidney looked quite good. It was clearing nicely. But in the other two mouse strains, it wasn't clearing much at all.

In fact, on the left is the, some of you may recognize this data, but so on the left is the mouse data in the kidneys where a day later, there's still a good 20% ID per g left in the kidneys, but in the other mouse strain on the right, there's only 2.5% left in the kidneys at a later time point.

And now the mouse strain on the right has an intact immune system. So it's tempting to say, well, that's more of a realistic case. Whereas the other two cases on the left side of the slide are mice that are immune compromised. So now we have this sort of mixed situation. So does anybody know what this molecule is and what would you do? Would you move it forward or would you kill it?

All right, well, probably some of you know already, this molecule is actually the preclinical data for PSMA-I&T, which in the clinic technically shows not much difference that I can tell, I'm not a clinician, between that molecule and PSMA-617.

But if we had come across this molecule in the lab and were developing it de novo, and we had this kind of result where it had pretty high kidney retention in two models and one model looked good, probably we would have killed it. That doesn't mean it should have been killed, but I'm glad we have it. But just giving you a flavor of what it's like when you don't know the answer from the clinical data.

So the next mouse tumor boards, we have to decide between two lead development candidates. One of the lead candidates is the blue molecule that's on the left side of the slide. It's the blue. And seems to have pretty high kidney retention. That's a solid line at the top.

The second development candidate in this program is the red molecules on the right. And the red molecules clear nicely through the mouse kidney. So again, we're going to ignore the tumor for just a moment. Which one would you pick for further development, if you had a choice?

Well, we probably pick the one on the right, the red molecule, because that's the one that's clearing. But the answer actually is, in this case, both were chosen to go to human study and to phase 0. Because if you can't decide, you can put both into phase 0 study in humans. And this was actually done recently with a molecule that binds to Nectin-4, which is a bladder cancer target.

And this was a drug developed from Aktis Oncology in collaboration with Pretoria. And they had these two molecules that had high kidney retention and low kidney retention in the mouse.

And in the humans, which is the graph on the right side of the slide, there was also the concordant high kidney retention with the blue molecules and the low kidney retention with the red molecules. So this was a pretty nice example.

Now, again, it doesn't mean that it's numerically going to be the absolute value of that translation. But these guys did a really nice job actually in this case, where they did some really thoughtful allometric scaling, and they actually got pretty close to where the mouse predicted with what the kidney retention or the kidney absorbed dose would be, pretty close to what the human predicted the human kidney absorbed dose would be. So the red molecule on the left is the one in the mouse, and you can see the corresponding one in the human on the right red molecule

So the last case study of whether our mice are relevant is, in this case, we have three novel PSMA agents that are showing excellent tumor volume reductions in mouse models. And this time, we're going to talk about tumor efficacy. And these are different mouse strains. So that's good. And all three agents have extended biological half lives in the blood. And the question is, will this beautiful efficacy that we see in the mouse models, will this translate well into human efficacy?

And the answer is not so much. So on the left, one of those mouse efficacies was from the antibody, the J591. These are all Lutetium, by the way. And from the clinical data, we know that there's an 11% PSA50 response. In the Lutetium PSMA-617, we know we have about a 60-ish percent PSA50. And there's another molecule that's an albumin bound molecule.

Now we talked about albumin binding earlier today. In that case, also a long biological half life in the blood that had a 36% PSA50. And you can see all the different half lives on the bottom, the antibodies 39 hours biological half life versus less than an hour for the PSMA-617. And something like five hours for the albumin bound one, which can have a range.

So why is that? There's probably several reasons. But one of the reasons potentially is the hematological toxicity that I think we've all talked about before is quite high for the antibody and the albumin bound one in humans. Because mice have pretty robust bone marrows. You can dose them up and you can get that beautiful efficacy in the mouse model. But you're flat lining tumors. But it's really hard to do when you hit that dose limiting toxin in human.

So I mean, I would say that similar phenomenon sometimes cannot be observed with antibody drug conjugates and other modalities as well where there's a long biological half life.

So I'll stop there and say that Yes, mice are useful. There are certain things that they're not our friend for and some things that they are our friend for. I think when it comes to PK parameters, like kidney clearance, or blood pool clearance, I think there is if you're trying to rank molecules by our prioritization, the mouse can be extraordinarily helpful.

And in some cases, when we're looking at efficacy and the efficacy is driven by a biological half life exposure in the blood, I think the mouse is less of our friend, because of their robust little bone marrows.

And I think we all know that when it comes to salivary gland, I didn't talk about it, but we had a discussion about it earlier, mouse models don't at all translate, and neither do monkeys. You really have to actually go into a human to see it with the 617 agent. So thank you very much.