Right, so let's get started with the next fireside chat here, the company's Revolution Medicines. With me today, I have Mark Goldsmith, CEO. Mark, welcome, and thanks for joining us.
Thank you.
So I think most folks are familiar with the story, so I'm just gonna dive right into questions, starting with RMC-6236, which is your multi-RAS inhibitor, your most advanced RAS inhibitor right now in the clinic. And I think many of us have followed sort of the data disclosures around ESMO last year, where you reported initial response data in lung cancer and in pancreatic cancer as well. So remind us, you know, sort of the key takeaways from that disclosure and where you are now in terms of working towards identifying a recommended phase II dose.
Yeah. Thanks very much. Appreciate the chance to be here. RMC-6236 is proving to be a very exciting compound in both lung and pancreatic cancer. We reported both tolerability and safety for RMC-6236, as well as activity in lung and pancreatic cancer over a dose range of 160-300 or 160-400 mg/day . It's a well-tolerated drug. We talked about dose intensity being around 90% at all of these doses, which indicates very, you know, infrequent dose reductions and certainly even less frequent discontinuations. The most common side effect is rash. Maybe you're gonna talk about that in a little bit.
In terms of activity, we've seen significant anti-tumor activity across KRAS mutant, particularly G12, G12X mutant, pancreatic cancer and KRAS G12X mutant lung cancer, non-small cell lung cancer. We reported objective response rates of about 30-upper 30% for lung cancer at that point in time, although we commented that it was an immature data set, and about 20% for second-line pancreatic cancer, which, again, we commented was a relatively immature data set.
Right. And, you know, I think at the time there was some discussion, I think there were some unconfirmed responses included at ESMO, which I believe most of them have been confirmed, since. Could you just confirm that, and
Yeah
and then I have a follow-up.
Yeah. So, the confirmation rate is very high for the unconfirmed PRs, and for sure, since ESMO, many of the PRs that were unconfirmed have confirmed. There were a couple of patients who progressed, who, of course, will not ever confirm, but those who were still on study continued on study and have confirmed. Since then, we have continued to observe data with longer observation times, as well as a heavier weighting to higher doses, which we expected both would pull those response rates up, and they have done so. So now in the aggregate, across the 160-300 mg range, we're seeing objective response rates in the 40s for lung cancer and in the mid- to high 20s in pancreatic cancer.
And specifically, just at the 300 mg dose, just splitting that out from all the lower doses, it's even higher for each of their respective cohorts. So clearly, we're now seeing some dose dependency, as we expected to see, and we're into pretty compelling numbers.
Right.
As I mentioned, dose intensity has been high, which means patients are staying on drug for a long time.
Right. And, so I thought the safety was quite good. I know there was some rash, but generally, you know, not a lot of DLTs, if any. And so, you know, I think you've sort of zeroed in on a 300 mg dose as kind of one of the go-forward doses. Why not go higher? Why not go beyond 400 mgs?
Yeah. We did test a few patients at 400 mg, which, you know, in retrospect, might not have been exactly the right increment, but that's about a 1/3 increase from 300, and so if it was perfectly dose proportional, we would have had a 1/3 increase in exposure. Actually, we had a little bit less than a 1/3 increase in exposure, and so now you're in the range where, while it's mathematically distinct on average from the 300 mg group, on a population basis, there's just too much overlap to distinguish those. So we didn't feel that we had any justification for continuing the 400 mg, and either we really needed to go up to a much higher number, like, say, 600 mg, if we were gonna keep pushing that, or just live with the 300 mg.
The 300 dose is a pretty darn good dose. We're seeing a lot of activity across many tumors. We're seeing durability, at least as measured informally. We don't yet have a mature PFS number, but it's, and it's well-tolerated. So given that, and given the urgency to move forward with pancreatic cancer treatment, and I would say an urgency that we experience from patients themselves. Patients or their advocates are frequently in contact with us, trying to get access to RMC-6236, and there are only so many slots we can make available in a dose escalation trial. So we've made the judgment that we're there now. We have sufficient data to make a decision.
We can go to the FDA with our multiple doses, compare those, show them why we think the starting dose ought to be what we think it ought to be. They'll have an opinion about this, and the smoke and mirrors and all of the stuff that they do to get to an answer will happen, and then we'll come to agreement about a starting dose, and that will be the go-forward dose for the pivotal trials.
Yeah. So what additional data will you be able to present this year from the phase I experience, perhaps some of the dose-finding work?
Yeah. I think the next set of data that we'll present to the public, to investors, will be likely in the second half of this year, and it should be a more sort of definitive, more mature data set that will be the data set upon which we're making our own decision about whether to proceed to pivotal trials. So let's separate lung and pancreatic cancer, treat them separately, because, of course, each of those is a separate decision, a separate trial. In each case, what we need to see is a stable PFS number that materially is superior to standard of care for whatever that indication is.
With that information, plus the tolerability and safety we already know, we would be in a very strong position to make a decision about launching a pivotal trial, and we'd like to share that with the public after we've made that decision. And we expect if we don't proceed, everybody will understand why we're not proceeding, and if we do proceed, everybody will say, "Well, that makes a lot of sense. That's exciting." And my view today, with what I know, and it's some tea leaf reading, is that we're gonna launch at least one and probably two clinical pivotal trials in second-line pancreatic and non-small cell lung cancer later in the year.
Right.
I hope to be able to share the reason for that.
I think you've talked about likely pursuing randomized studies for those two.
Yeah
initial indications. You know, any additional thoughts? You know, why not try and accelerate it in a single-arm study?
Yeah
perhaps in lung cancer? Others have done it.
Well, yeah, others have done it. Let's talk about others who have done it. I mean, so Sotorasib received an accelerated approval, and the availability of Sotorasib as an approved drug actually compromised their randomized pivotal trial and got a major spanking from, you know, from the FDA. I don't think that what we wanna do is to forgo a high-quality randomized trial in order to do a single-arm accelerated trial. I think that would be a mistake. In pancreatic cancer, in particular, there is no accelerated approval path based on objective response rates. It would be based on PFS, which is also the data that will drive a full approval. So there's really not an accelerated path there. The program is accelerated because these PFS numbers are all so short.
It's already intrinsically accelerated, and there's no reason not to go to them with PFS data and even, you know, possibly some OS data, for full approval at that point in time. So I think for pancreatic cancer, it's pretty clear what we need to do. There's some complexities that we'll consider as to what groups of patients to include in the core group and whether to have nests that go beyond the core population.
Mm-hmm.
But those are decisions we can make, you know, over the next next number of months and then proceed with that randomized controlled trial, and I think the FDA will be very interested in that study. Lung cancer, you know, the argument is fairly similar that although there was a good single-arm accelerated path in the past, that pretty much doesn't exist anymore.
Mm.
I mean, to a large degree. The FDA has now signaled that they expect you to have enrolled your full approval trial by the time you seek, accelerated approval, and if you've already done that, you're pretty close to having the data.
Right.
From our point of view, the most robust way for us to proceed is to design full approval, randomized controlled trials, to launch those trials, and if there's some way to get something earlier out of them, we will, but otherwise, they'll play out.
Right. And then, yeah, I know you have outlined possible phase III trial designs in your—
Mm-hmm
you know, in your updates recently. You know, so remind us, what do we know about the activity of RMC-6236 in the various different KRAS mutations, and is there a reason to believe it may, it should not work in all of them? Why wouldn't it work in all of them?
Yeah. In biochemical studies, RMC-6236 is active against every form of RAS we've ever tested, every wild type, K, N, and H RAS, and every mutant form that we have in our testing panel, which is, I think, probably all of them or pretty close to all of them. It binds to and inhibits all of those. They're not all identical because they are somewhat different target molecules, but they are pretty similar. In vitro, that's not exactly what we see, and we think this has to do with the biology of different RAS mutations and how they drive cancer. But in vitro, when we do inhibition studies, that is, in cell culture experiments
Mm
we see a range of responses across different genotypes, and the group we can most distinguish is the G12X mutations that tend to have much higher response and sensitivity, if you will, than do the other mutant cell lines. So again, I think this is probably due to how those different mutants are linked into the biology of cancer, of carcinogenesis, not so much to whether a compound is inhibiting it, but whether or not that effect translates as well. So that's really the question in patients now is, if we test patients who have an NRAS tumor driver instead of KRAS, will we see impact by 6236? If we see Q61 or G13 mutations, will they be as sensitive? And we're testing that now.
We've opened cohorts that are allowing us to evaluate 6236 across different RAS forms and different mutations, and by the time we make our decision to finalize our pivotal trials, we will know whether to incorporate those patients, and if we're incorporating them, are they just part of the core group, or do they need to be nested in some sort of isolated, statistically isolated.
Right
fashion?
Okay, understood. And then colorectal cancer is another histology where KRAS is, you know, well-known as a driver in some of the patients. So what are your plans in colorectal and yeah, any disclosures coming out from that?
Yeah. So, absolutely, you know, the top three major epithelial cancers caused by RAS are lung, pancreatic, and colorectal cancer, and we've really only talked about two of those because that's what our investigators have chosen to prioritize in the studies. And my interpretation of that, since they could have put anybody they wanted, and frankly, I expected they would flood the study with colorectal cancer patients because they have so many of them, they didn't end up doing that. After the first few colorectal cancer patients, they started seeing responses in pancreatic and lung cancer, and they just doubled down on it, and they started prioritizing within their own practices, those patients, which the benefit to us is we got pretty good data sets in lung and pancreatic, but we have no data set in colorectal cancer.
Right.
The great news is, we've already now launched a cohort that includes colorectal cancer patients. We've done it. Now that we've selected a dose, at least informally, we've selected a 300 mg dose, we can give those patients the full dose instead of dosing them at 80 mg or 120 mg, and let's find out what happens there.
Right, that makes sense. And then what work do you have ongoing to potentially enable a move into first-line settings in the sort of key indications in the future?
Yeah. It's a terrifically important area. To really dominate in a particular indication, first line, control first line makes a big difference. So that's where we need to get. And it's also where we tend to have would tend to have the highest response rates and durability of response, obviously. So we are anxious to get to first line. This year, our first priority is to get those second-line pivotal trials open. But our second priority is to figure out how to move into first line and to start that as soon as we can. And that likely involves combinations. I won't say with absolute certainty that involves combinations because there's the possibility of some monotherapy approaches in first line, but I would, you know, a betting person would just have to assume that it's gonna be combinations.
This year alone, we already have three different combinations underway with patients being actively treated. One is with 6236 plus Keytruda. That's going to be an essential combination for entering first-line lung cancer. The second is RMC-6291 with Keytruda, also an essential component of first-line therapy for G12C patients. And third, we have a combination of RMC-6236 and 6291, the first ever combination study of two RAS inhibitors in the same patient, and for sure, the first combination of two RAS(ON) inhibitors. So it's really an amazing study. It's a pretty bold thing to do.
The preclinical work suggested that that combination was just significantly more active than either agent alone, and the evidence for that came from tumor models in which either agent alone as monotherapy was active but didn't really crush the tumors, didn't regress them. But when we put the two compounds together, even if we lowered the doses, the two together drove regressions. And so that's a really exciting kind of thing, and one can imagine moving into front line with a combination where you're bringing really the best possible antitumor effect. That still has to work with PD-1 in lung cancer, so you'd, you'd have a triplet, perhaps, in that setting, but not necessarily in other indications. The last area that I'll just allude to here is combining 6236 with chemo.
One of the strategies for getting into front-line pancreatic cancer would be to add 6236 to existing chemotherapy, standard of care chemotherapy. I don't know if that's the best approach, but it's one we have to consider. It certainly is ethically the easiest thing to do because you're not withholding anything from those front-line patients. We just have to establish that 6236 does not change the tolerability of chemo, which is already not well tolerated. You hear how I put that. I didn't say that the chemo changes the tolerability of 6236, the 6236 changes the tolerability of chemo. We just need to establish that. That's a study we'll do this year, but we don't yet have patients enrolled.
Right. Okay. And then, you know, you, I think you did not allow enrollment of patients with G12C mutations in the 6236 study.
Right.
You know, what is your expectation for how the drug would do in those patients?
Yeah.
I think some people are struggling on, you know, sort of figuring out where 6291 could end up relative to 6236. I know you mentioned the
Yeah
combination in that subset, but it seems the drugs are somewhat redundant, assuming that 6236 works really well in G12C as well.
Yeah, I don't love the word "redundant" because redundant implies that, that there's no benefit to putting them together. But I would say there's an overlap in their spectrum. You know, that might be spin, but it's a conceptual point. And this is gonna be true for all of our mutant selective inhibitors. Since RMC-6236 is active against pretty much everything, there's always gonna be this overlap. And we will have to determine on an empirical basis whether putting those two together with 6236 delivers added punch, you know, or not. And we're gonna have to do that for 6291 and 9805 and all these other ones. The first one we'll get the answer to is with 6291.
Your question, though, was, I went off on a little bit of a tangent here. Your question was about 6291-- Oh, G12C and 6236.
Right.
Right. We did exclude G12C from the 6236 program because we knew there was a lot available for G12C patients and that the likelihood they'd be enrolled would be low anyway, which is, I think, true, even if we had opened it, none of them would have gone on.
Right.
Preclinically, 6236 is one of the best G12C inhibitors out there, okay? So I, I think it would be very competitive with a G12C selective inhibitor. We are now allowing G12C into the 6236 studies. There's a cohort that allows those in the, in this other, other RAS mutant cohort. We'll see how many patients we get. I think we only need to see a few, and we're likely to see responses, in which case we'll check the box and say it's just as active. But it's hard to get G12C patients into studies because there are about 1,500 studies of RAS inhibitors in G12C patients now, and everybody's sort of competing for the same patients. And so you really need to have something that kick-starts it.
You need to have enough evidence to attract patients and investigators to wanna do it, but to get the evidence, you need to attract some patients to do it. So it's a very slow start. Once the patients are in, it might turn the corner. In the meantime, we're studying 6291 as a G12C inhibitor, so you could argue we're even competing with our own-
Right
6236. So I think it's a little bit unclear how G12C will play out for 6236. We're not gonna hold up a lung cancer study for it, and if we have to exclude those patients from it, we will, or if we have to put them in a separate nest, we will. It's a pretty crowded space, and I don't think second-line G12C lung cancer is where it's all gonna be at for us. So we'll just have to see. We're trying to sort it out based on data.
Right. And then maybe a question on 9805, your G12D inhibitor. There's a few other molecules in early development stages, but, you know, remind us just where you are with that molecule-
Yeah
and also how you think about the future there.
Yeah, yeah. Our RMC-9805 is a pretty amazing compound. It is another Tri-Complex inhibitor, so it follows the same general mechanism as all of our other inhibitors. So it's got a pretty high probability of success based on that. Preclinically, it was clearly shown to be orally bioavailable, which made it the first orally bioavailable G12D inhibitor. And now I can tell you in patients, I confirmed in January that we are seeing good oral bioavailability, so good dose-dependent increases in exposure. I can also let you know that it is behaving very well from a safety and tolerability point of view. It's been well tolerated. We've been able to dose escalate. We haven't seen any DLTs, and we haven't reached MTD yet. So it's a pretty exciting compound.
I would say everything else in the field has to chase it, although it doesn't have a huge head start. It has some head start, and the last feature of it that's, I think, really notable is that it is a covalent modifier of the aspartic acid in KRAS in RAS G12D tumors. There is no drug in the history of mankind that is a covalent modifier of aspartic acid. It's never been done before, and it has to do with the chemistry of binding covalently to aspartic acid versus other targets. And our team figured out a way to do it, which was really super clever. It was based on our tricomplex modality, and gives us a covalent G12D selective inhibitor.
Finally, I think by the end of this year, we'll be able to provide a more full update that includes some activity data.
Right. So maybe then, summarizing all this, what are sort of the key data disclosures on your KRAS or on your RAS program this year?
Yeah. So number one is gonna be the data supporting the launch of either or both of those pivotal trials that I mentioned, monotherapy, RMC-6236 data. And really, the one number everybody's looking for is the PFS number, and when we have a mature number and can put it in context, you know, we'll provide that as part of that picture. That's by far the most important thing for us this year. I think behind that would be 6236 in other tumor types and in other genotypes. 6291 in combination with 6236 is a really important paradigm. As I mentioned, it's the first-ever combination of that sort.
I think 6291 and 6236 plus Keytruda are gonna be really important from a safety point of view to establish that they do or don't cause hepatotoxicity. We project that it will not, but we need to see whether that happens. And then finally, 9805 monotherapy activity.
Yeah.
It's a long list of data flow coming out-
Right
Over the year, most of it in the second half.
Good. And then you're obviously well-capitalized following the sort of the EQRx transaction, and you have a lot of moving parts, a lot of studies sort of ramping up. And, you know, in that context, you know, how far or do you think you can do it alone all the way, or are you possibly considering partnerships on this, down the road?
Right. So we, we closed the year with about $1.85 billion in capital. Seems like a lot of money, and still until you start designing multiple, pivotal trials, then it doesn't seem like so much money. But we, we now have the ability to make our decisions to do what we think is best to do for patients and to maximize the value of the assets, so I think we're in a really strong position. In terms of where does that take us? Well, it's clearly gonna take us to our first couple of pivotal trials, no question about that. But, you know, our biggest challenge now is that everything's being successful.
You know, I joked last year that investors had switched from being worried we'd be unsuccessful and that they were wasting their money to being worried we'd be so successful we'd have to raise a lot of capital. And we kind of dealt with that a little bit with the EQRx deal. I think that was clearly meant to sort of move the needle in a big way. You know, but that will change over time, of course, as that capital gets invested. Going it alone versus partnerships, I think of it in a really different way. I think of it as, what will a partner bring us that we must have? And today, I can't think of anything.
Over time, as we're getting more committed ex US, particularly for global pivotal trials and other things to happen down the line, we could use some help in development outside the US. We can do all of that on our own. It's just a matter of money, and we have the ability to do it. Actually, we're running some ex US studies today, so we can do that. But of course, if there's an established footprint where they really understand the and have the relationships with investigators and so on, that could help us. So that's possible. Alternatively, we don't partner outside the US until NDA, at which point we would maximize the value of a deal. We'd have a bidding war over it, but they wouldn't have helped us at all in development. I don't know which is gonna happen.
We don't really have a strong view of that. And companies are very interested in RevMed, very active. They're poking around all the time, trying to get in under the hood. We don't really let them in under the hood. So we're not making it easy for partners to partner with us, but at some point, I suspect it will happen. And I think of it less around runway and much more around capabilities.
Yeah, that makes sense. Great. Well, with that, I think we'll wrap up. Mark, thank you so much for your time today.
Thank you.