Welcome, everyone. My name is Ben Burnett, Biotech Analyst at Stifel. Happy to host the next session with Revolution Medicines, and pleased to have Mark Goldsmith with us, President and CEO. Mark, thanks for doing this.
Thanks for having us. Appreciate it.
Awesome. Like the rest, this will be a fireside chat format. If you have any questions, shoot me an email directly at burnettb@stifel.com, or there's a chat, and I have that pulled up as well, so we'll work that into the conversation. Mark, maybe to kick it off, just give us a brief introduction of RevMed and kind of frame the near-term event path.
Sure. RevMed is a company focused essentially exclusively on RAS-addicted cancers, mostly solid tumors. Of course, RAS accounts for or is correlated with roughly 30% of all human cancers, so it's a very important set of cancer drivers. Our main focus these days is on our RAS(ON) inhibitor collection. We introduced the concept of drugging RAS in the activated or on form as a biologically distinct and, we think, ultimately largely superior strategy for suppressing the RAS pathway.
We have three compounds that have been working their way through clinical development: RMC-6236, our RAS multi-inhibitor that inhibits essentially all forms of RAS that we have studied, RMC-6291, which is unique in its G12C selective characteristics, but that it is a RAS(ON) inhibitor, and then finally, RMC-9805, also a RAS(ON) inhibitor, but this one targeting KRAS G12D, which is the most common mutant form of RAS that causes human cancer. So we have a lot going on in the clinic. We have some initial data we've shared. I'm sure you'll want to ask about that, and we're excited to have a chance to talk with you.
Excellent. Well, I definitely want to get into the data, but I think maybe the place to start, you just had some data at AACR. Maybe you can kind of talk about that and what are the key takeaways from AACR?
Yeah. So at the AACR, we did have a lot of presentations, both in the podium and as well as at the poster level. I think the most important kind of near-term impactful observations related to RMC-6236. As you know, we presented some initial clinical data on RMC-6236 at ESMO and at the Triple Meeting last fall, where we showed a number of patients on waterfalls, swimmer plots, and so on, and really demonstrated that RMC-6236 is highly active in lung cancer and pancreatic cancers caused by RAS mutations. Here at this meeting, we highlighted a few lessons that came from some specific clinical cases in that study and that hadn't been reported previously by us. We walked through each of these cases to really inform on some specific and important technical areas that defined for us the opportunity going forward.
One of those cases was a pancreatic cancer patient whose cancer carried the KRAS G12D mutation, was treated with RMC-6236. The reason we showed this particular case is because the patient developed a complete response to that treatment, that monotherapy treatment. As you may know, complete responses in pancreatic cancer are rather uncommon, and it does indicate the potential depth of anti-tumor effect that can be achieved by RMC-6236. The patient was treated at 300 mg and continued on 300 mg , was confirmed with that complete response, and has continued on treatment since then. That's not likely to be the most common outcome for patients treated with monotherapy. We don't want anybody to misinterpret what we're communicating here. Nonetheless, it does indicate what is possible for such patients and for it to happen and to be that impactful without being intolerable or unsafe.
We also showed a case of a patient with pancreatic cancer who had a Q61, a KRAS Q61 mutation. This is the first time we've ever shown data outside of the G12 mutation context, so now moving to the Q61 mutational hotspot. This patient showed a partial response and did have some side effects and had some dose reduction, but nonetheless has done well so far on RMC-6236, showing that this compound does hop from one target to the other and can reach across the aisle from the G12 mutations to Q61. There was a third case shown that was a patient with melanoma who also had a Q61 mutation, but this time an NRAS Q61 mutation. So now moving to a wholly different isoform of RAS, the NRAS form, which is common in melanoma.
This patient was treated with RMC-6236, did well, and actually developed a complete response. So this is the second complete response we reported at this meeting. So now we've moved beyond pancreatic cancer to melanoma, and we've moved beyond G12 to Q61. And then the last case is a really remarkable case. I don't think it's going to be a common scenario, but nonetheless quite interesting. This was a colorectal cancer patient who started with a BRAF mutation as their driver for this tumor, was treated appropriately with a BRAF inhibitor, and the patient ultimately progressed. And at the time of progression on the BRAF inhibitor, was found in circulating tumor DNA to have three new RAS mutations that hadn't been identified before. So their escape to the BRAF inhibitor was actually RAS mutations.
This included two G12 mutations and a Q61 mutation, and it included both KRAS and NRAS. So here we have sort of a blossoming of RAS mutations as an escape mechanism by these tumors. The patient enrolled in the RMC-6236 trial and then had a partial response to RMC-6236, again showing the potential breadth of 6236 across multiple mutations, but even multiple mutations within a given patient. What this does, we believe, is to validate the concept that the multiple or the breadth of activity by RMC-6236 may be able to suppress multiple RAS-driven resistance mechanisms in the same patient. That really gives us encouragement that this could be a high-impact compound.
That's incredible. Yeah, that's absolutely excellent. Two CRs, multiple tumor types, multiple mutations, characterizing some of the resistance mutations. One thing, all of these case studies, every single patient was started at 300 mg. Is it looking like this is the dose it will be focused on in phase three, or is that still to be determined?
Well, we announced back in January that we had selected 300 mg daily as our preferred dose within Revolution Medicines, and we opened all of the cohorts for monotherapy study, all of the new cohorts out of which these four cases came at 300 mg. Now, the final dose selection to go forward into pivotal trials really has to be derived from conversation with the FDA. And as we indicated earlier in the year, that's something to happen in the first half of this year. I can't give you any specific update on that, but that engagement with the FDA will lead to a decision about what the starting dose is in a clinical trial. Again, our preference is that 300 mg, based on the data, is both well-tolerated, seems to be generally safe, and has encouraging signs of efficacy.
But that's subject to Project Optimus and all the work we have to do on that question.
Yep, understood. Okay. Maybe we could focus on, so stay with the RMC-6236 and kind of focus on some of the PDAC data that you've shown. I think a lot of us are looking forward to seeing how the durability of that data evolves. I guess what level of durability are you kind of looking forward to sort of have confidence that this could win versus second-line?
So in second-line pancreatic cancer, it's been well-established and I think reestablished every time somebody takes a crack at it that chemotherapy in second-line pancreatic cancer yields a median progression-free survival in the ballpark of three months, maybe three to three and a half months at best. And just that marker has not been crossed by really any studies. Our hope is that we can exceed that. And I think that our intention here is to provide not only a compound that may be well-tolerated and generally safe, but one that also delivers more clinical benefit than chemotherapy, which is obviously very poorly tolerated. And as you can see from that durability number, does not have the kind of clinical impact that we're hoping for. So that's the threshold. That's what we're comparing against in a second-line study. We've not yet disclosed parameters around durability for RMC-6236.
We've been maturing a dataset to get to durability into a stable median PFS estimate. One has to have a mature dataset, so that takes time. That's the sort of thing that we are sharing with the FDA and will be part of final decision-making for the go forward on the pivotal trial, and of course, also part of the dataset that we'll disclose as part of our update, which we indicated would happen publicly in the second half of this year.
Okay. That's great. And I guess including that update would also be updated lung cancer data. Is that right?
So that's a different dataset. So there's basically the same story for each of these. That is, what's the right dose? And what does progression-free survival look like? And what is the sort of ultimate safety and tolerability profile? So each of those has its own presentation to go with it. Each of those is a different discussion with the FDA, and we expect, based on what we indicated in February, that we would provide each of those in the second half of this year. They may not be at the same time, though, I think is the point that I want to make here. I wouldn't necessarily expect that they'll be identical.
Yep. Okay. Okay. Understood. And I think you've also talked about front-line pursuits with both indications, pancreatic cancer and lung cancer. Maybe starting with pancreatic cancer. So I guess, number one, what's the latest thinking kind of around what that front-line kind of opportunity could be? Would it be a monotherapy, or would this be a combination? Yeah, maybe we'll start with there.
It's a great question, and I don't know the answer to that at this point. One of our goals for 2024 is to determine what the real options are for first-line trial and for the treatment regimen. Is this a combination with a standard-of-care chemotherapy, or is it an alternative to standard-of-care chemotherapy? Both of those are quite interesting. It's not that one is inherently superior to the other. They're both interesting. In order to go into a monotherapy strategy, one has to be convinced, based on the second-line data, that this is an ethically justifiable thing to do. Or to combine it with chemo, in order to go down that path, we have to be confident that it's safe to combine RMC-6236 with chemo.
So in each case, we have some gates that we have to get through this year, and then we'll be able to prioritize those two options for a first-line study, which we'll do as quickly as we can. It's conceivable the first-line study ends up being in combo with chemo. It could be as an alternative to chemo, or there could be a study that encompasses both of those options, and that's a multi-arm study because there are different patients who might benefit from one or the other regimen based on their performance status and ability to tolerate chemo.
Okay. And I guess for all of these studies where OS is sort of a relevant piece of information, relevant endpoint, there's always the question of crossover. And I think looking at kind of the G12C space with sotorasib versus adagrasib, CodeBreaK 200 allowed crossover, KRYSTAL-12 did not. Unclear how relevant that was. But I guess, do you have an opinion on that? And I realize you'll go to the FDA and get some feedback here. But how do you get that OS, like a solid read on OS in these settings?
Well, it's really important because it has enormous impact for every other future patient to know what it is that they should expect or could expect from treatment. So we really do need to try to establish early in the course of this program whether or not there is an OS benefit, and if so, to what degree. The crossover risk is high if you allow crossovers before certain points in the trial. And I think that's in part where sotorasib ran into some difficulty because patients were able to crossover earlier. So we would likely design around that issue and have some constraints on when crossover could occur in order to minimize potential impact on an OS readout. But that's all to be worked out.
I mean, we're not there yet to declare the details of that trial design, and we'll be talking about that presumably in the context of data that we disclose following the FDA engagement.
Okay. Okay. Well, good. Looking forward to monotherapy data in PDAC and lung cancer. I want to also ask, you have a RMC-6236 in combination with your G12C inhibitor, RMC-6291. That's dosing I guess, question number one, so this is looking at combination with I see you have a combination with those two, and then each one separately in combination with Pembro. And that combination with Pembro has been difficult for other G12C inhibitors. Maybe starting with this, how do you I guess, what do you know about the combinability of kind of your agents, which obviously have a little bit of different MOA, and their combinability with Pembro?
Right. No, that has been a big issue with the G12C off inhibitors. Nobody really has defined exactly why hepatotoxicity can be seen with the agents as monotherapy and/or why that is synergistic or additive with checkpoint inhibitors. I think the mechanism behind all of that still remains rather obscure. Our presumption is that it has to do with some chemical feature or features of those individual compounds. It is clear that it is a more common side effect with some compounds than with others, which really does suggest it has to do with the chemistry of the compounds. We biologically can't really identify a rationale for why the G12C mutation, per se, would be a factor in that. It seems like it's more about the compounds themselves. With that context, RMC-6236 and RMC-6291 are chemically very distinct from all of the G12C off inhibitors.
Other than sharing the same sort of atoms that can be arranged in different ways to make these different structures, they're just completely different. And they have different mechanisms of action, as you said, but also they're just inherently different. And so I think a priori, there's no assumption that they should have the same side effect any more than any two unrelated compounds would be assumed to have the same characteristics. It is important for us to demonstrate that in humans. And so that is a key gating item for any opportunity to go into front-line non-small cell lung cancer. I think that that door is only open if the combination can be delivered without significant additive toxicity. Preclinically, we didn't see any issues with RMC-6236 or RMC-6291 in combination with checkpoint inhibitors, but that's probably not a particularly strong predictor of what's going to happen in humans.
In humans, the data that we have so far come from our monotherapy study of lung cancer patients, many of whom had recently come off of a checkpoint inhibitor, had recently come off of KEYTRUDA, for example. Among the lung cancer patients for either 6291 or 6236, roughly half of those patients had come off of immunotherapy within 90 days of starting or started RMC-6236 within 90 days of coming off of KEYTRUDA. There is now evidence, published evidence, that that window of potential overlap is a significant window to give you a glimpse of what may happen if you just co-administer them in real time. Based on that, the fact that we saw very little hepatotoxicity in those patients, we saw no significant Grade 3 or greater hepatotoxicity, suggests to us that that might be what we'll see around the corner when we get there.
So that's encouraging. But ultimately, we need concurrent administration data. We are now evaluating patients with that combination. When we have a definitive dataset, we'll be able to share it publicly and let people know. It would be very important if we can combine either or both of these compounds with KEYTRUDA. That would create a really clear door for entering into first-line and could be highly differentiating for either or both of these compounds. The combination of 6236 + 6291, we call that a RAS(ON) inhibitor doublet, grew out of preclinical data, some of which we've shown over the last couple of years. A specific dedicated presentation on that was made at AACR recently by RevMed.
In those experiments, we found that in the settings where monotherapy alone delivered modest impact against lung cancers, when we combined the two agents, we could see significant tumor regressions and very sustained tumor responses. We don't know exactly why that combination delivers that kind of punch, but it may be something relating to the fact that the mutant selective inhibitor delivers a strong mutant selective punch, and the multi-RAS inhibitor provides coverage against the other forms of RAS that can be contributory to the cancer and can help sustain the cancer in the setting of a mutant selective inhibitor. Just as we showed with that CRC patient I mentioned earlier who had multiple RAS mutations, you can imagine trying to preempt that by including RMC-6236 in a regimen that also includes a mutant selective inhibitor.
We don't know if that's the explanation, but it's a very reasonable starting point or framework. So we're now evaluating that combination. If 6236 + 6291 is well tolerated and shows solid anti-tumor efficacy, that could be an exciting path or provide an exciting path to considering a chemo-free regimen in first-line treatment. Two RAS(ON) inhibitors perhaps could replace two chemo agents that are currently standard of care for lung cancer. And if one can do that in combination with the third agent, which would be KEYTRUDA, that would be a really exciting and potentially transformative approach. We're not there yet. We're trying to do the deliberate steps to validate these various combinations and particularly safety before we get there. But that would be a really wonderful place to get to.
That's great. And so sticking with the combination of the G12C inhibitor with the RAS multi, how do you prioritize dosing in this study?
Well, the goal is to get to the highest active monotherapy dose that we typically use for monotherapy. The goal would be to achieve that for both agents in the combination. We start that dose escalation by going down a notch or two dose level with each agent, combine them, make sure that's safe. Then you elevate one. If that's safe, then you can elevate the second one. So it's just a stepwise process. And where we'll get, where we will land, I don't know. But preclinically, we were able to administer full dose of each of the two agents in combination. We didn't have any difficulty doing that. So we are hopeful that that's the same experience we'll have with patients, in which case there won't be any question. Now, fortunately, for both of those compounds, there's actually a fairly wide range of active doses.
Even if we have to knock off one or two rungs on one or the other compound, I think we'd still be working with active doses. Let's play it out, and we'll find out.
Yep. Okay. That's great. And maybe just in the last couple of minutes, I want to turn to the G12D asset, RMC-9805. I think when we think about some of the other G12D drugs out there and some of the chemistries that directly target G12D, we have personally wondered about kind of the bioavailability of those drugs. And so I want to ask, you just presented some preclinical data at AACR kind of around some of these metrics, assessed preclinically. But talk about 9805 and kind of how you see that profile stacking up.
Yeah. All three compounds that we have in the clinic showed good oral bioavailability in the preclinical species. We've shown pharmacokinetics data for both RMC-6236 and RMC-6291 in humans as well, which really confirmed what we had seen preclinically and suggested that these compounds, although they're large, they're natural product-like, they're quite distinct from other compounds people are using, they still have the ability to penetrate in the GI tract and then to maintain in circulation. So we don't think from a platform perspective that we anticipate significant problems. And then for RMC-9805, we likewise saw good oral bioavailability across species. We have been dosing patients. We disclosed back in January just verbally that the compound has shown oral bioavailability in humans, that the pharmacokinetics have been very consistent with what we had seen in the preclinical models and projected into humans. As we increase doses, we're seeing increased exposures.
But we had not yet reached a maximal tolerated dose. We haven't identified a go-forward dose. And that's really where we were back in January. I don't have a new update for that. But I think we're optimistic that we can now treat RMC-9805 just like the other two compounds. They're orally bioavailable, and it's a matter of finding the right dose that's safe, well tolerated, and efficacious.
Excellent. Okay. I may want to squeeze one last question here also about the G12D asset. So you showed some preclinical just again, a preclinical characterization of this asset. But you looked at some tolerability in the form of preclinical tolerability in the form of mice body weight. Just curious, how does that compare with kind of what you saw with the RAS multi? And just kind of trying to get at, would you expect kind of a higher kind of therapeutic tolerability with a mutant selective inhibitor versus a multi inhibitor? Or is it too soon to say?
Well, without commenting on the specific compounds, I just think generally, I know that the paradigm and the assumption is that a mutant selective inhibitor will be better tolerated than a multi-selective inhibitor. I would just point out that RMC-6236, based on the data we've shown so far, appears to be one of the best tolerated and safest RAS inhibitors out there. And it's the only one that is not a mutant selective inhibitor. So I think the paradigm is not quite right. I think that's been disproven. But the underlying concept is that the on-target effects of a RAS inhibitor, if it includes only the mutant, is not likely then to be manifest in normal tissues because the mutant doesn't exist in the normal tissues. So any side effects would be from off-target effects. And that's really what's happening with the G12C off inhibitors that have side effects.
That's really off-target effects. RMC-6236, by design, inhibits the wild-type form of RAS. And that will happen in normal tissues as well as in tumor tissues. So there certainly is the potential for on-target toxicity. And I think it's fair to say that what we've seen in the clinic so far has really all been attributable to on-target toxicity and those things that we would have predicted from the preclinical species. Importantly, though, we've been able to identify a dose range and actually quite a wide dose range in which the tolerability is quite good, the safety is quite good, and the anti-tumor activity is quite encouraging. So I think we have a therapeutic index with RMC-6236. I think that question has been definitively answered. And now what we see with each of the other compounds will just have to come out from the profile. What do we see?
What are the off-target effects? How does that compare?
Okay. Well, excellent. Mark, it's been a great conversation. Thanks so much. I think we're out of time. But thank you.
Thank you. Thanks for having us.