Revolution Medicines, Inc. (RVMD) Earnings Call Transcript & Summary

Good day. My name is Katherine, and I'll be your conference facilitator today. Welcome to the Revolution Medicines First Quarter 2022 Earnings Conference Call. Today's call is being recorded. [Operator Instructions] I would now like to hand the conference over to David Arrington, Revolution Medicines' SVP of Investor Relations & Corporate Affairs for opening remarks. David, you may begin.

Thank you, and welcome, everyone, to our first quarter earnings call. Joining me on today's call are Dr. Mark Goldsmith, Revolution Medicines' Chairman and Chief Executive Officer; Dr. Steve Kelsey, the company's President, Research and Development; and Jack Anders, our SVP of Finance and Principal Accounting Officer. Today, we will be referencing selected slides from our corporate presentation. The complete set is available for you to view and download on revmed.com. As we begin, I would like to note that our presentation will include statements regarding the current beliefs of the company with respect to our business that constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act. These statements are subject to a number of assumptions, risks and uncertainties. Actual results may differ materially from these statements, and except as required by law, the company undertakes no obligation to revise any forward-looking statements. I encourage you to review the legal disclaimer slide of our corporate presentation or the earnings press release as well as all of the company's filings with the SEC concerning these and other matters. With that, I will turn the call over to Dr. Mark Goldsmith, Revolution Medicines' Chairman and Chief Executive Officer.

Thanks, David. Good afternoon, everyone. Thank you all for joining us today. I'll start with a few top line comments. We've made great progress in the first quarter. We continue to advance what we believe is the deepest portfolio of RAS-targeted therapeutics in the field, led by our RAS(ON) inhibitors in development with significant opportunity for patient to impact. Excessive RAS(ON) signaling drives some 30% of all human cancers. Today, I am very pleased to let you know that we have submitted the IND for RMC-6236, our RASMULTI(ON) inhibitor, and expect to dose the first patient in mid-2022. Preparation of the IND for RMC-6291, our KRAS G12C(ON) inhibitor, is also on track with our original guidance, and we expect to dose the first patient in the second half of 2022. These 2 RAS(ON) inhibitors approaching the clinic and an exciting pipeline behind them represent a wave of RAS(ON) inhibitor drug candidates that could address the majority of RAS-addicted cancers that lack effective targeted drugs. Concurrently, we continue clinical evaluation of the class-leading RAS companion inhibitors, RMC-4630 and RMC-5552, that are intended as combination agents with direct RAS inhibitors, including our own RAS(ON) inhibitors to maximize patient benefit. Regarding our development-stage compounds, we are transitioning our communication schedule. Going forward, we plan to focus milestones for development-stage programs on clinical initiation rather than IND submissions. Investors should look to our postings on clinicaltrials.gov for indications that an IND is open and that clinical investigation sites are being activated to enable study initiation. We also plan to communicate after we've begun dosing patients in each program. Slide 5. Revolution Medicines has a deep science-driven pipeline of targeted therapies for RAS-addicted cancers. We have 4 RAS(ON) drug candidates that are supported by robust and growing data sets that have large clinical opportunities with the potential to serve patients with a wide range of RAS-addicted cancers. Further, we expect that our pipeline will continue to grow with highly distinctive new assets deriving from our RAS cancer innovation engine, which should expand our reach to other key oncogenic mutations on RAS proteins. Although RAS-addicted cancers are induced primarily by mutations that cause RAS on proteins to behave as cancer drivers, often these cancers are also supported and maintained by other cellular proteins we call RAS cooperating targets and pathways. We believe it is important scientifically to match our treatment strategies to this biological cooperativity by developing RAS companion inhibitors to suppress the cooperating proteins while deploying RAS(ON) inhibitors to suppress the primary RAS drivers. Lastly, I note that in many instances, we expect drugs of these 2 types may be combined to deliver the greatest clinical benefit. In the next few minutes, I'll highlight examples of 3 specific themes that are important to our strategy. First, RAS(ON) inhibitors demonstrate compelling monotherapy anti-tumor activity in diverse preclinical models of genetically-defined RAS cancers. Second, RAS(ON) inhibitors can be combined with RAS companion inhibitors to improve anti-tumor activity in preclinical models that are less sensitive to monotherapy. And third, RAS(ON) inhibitors as monotherapies reverse the immunosuppressive tumor microenvironment in RAS-driven cancer models and can unlock profound anti-tumor immunity when combined with immune system modulators such as PD-1 checkpoint inhibitors. Now I'll turn to specific comments about our portfolio progress. The first theme is RAS(ON) inhibitors as highly active monotherapy agents preclinically. We have produced a large collection of tri-complex inhibitors targeting diverse oncogenic RAS variants through highly differentiated chemical and pharmacologic profiles. Slide 9. As a first example, RMC-6236 is a potent oral RAS(ON) selective inhibitor with broad potential across cancers driven by a variety of RAS mutations. To date, it has been shown to be active in 3 histotypes including pancreatic, colorectal and non-small cell lung cancer models, and across mutations including KRAS G12D, KRAS G12V and KRAS G12R. Cancer drivers for which patients whose tumors bear these mutations lack targeted therapy options. Slide 12. We are on the path to clinical data now that the IND has been submitted. We expect to announce dosing of the first patient in a monotherapy dose escalation study in mid-2022. And in 2023, we plan to provide evidence of first-in-class single-agent activity for RMC-6236. I also note that RMC-6236 may also be deployed as a RAS companion inhibitor in combination with mutant-selective RAS(ON) inhibitors, something I will say a bit more about later. Slide 14. As another example, RMC-6291 is a potent oral selective covalent inhibitor of KRAS G12C(ON) with a differentiated preclinical profile designed to serve patients with cancers driven by the KRAS G12C mutation, including lung, colorectal and pancreatic cancers. 6291 has demonstrated best-in-class potential for treating KRAS G12C-driven lung cancers, non-small cell lung cancers, based on superior outcomes in a mouse clinical trial with KRAS G12C lung cancer models. Slide 17. Our IND preparation is on track for submission in the first half of 2022, and we anticipate RMC-6291 will be our second RAS(ON) inhibitor program to enter the clinic this year and expect to disclose preliminary evidence of superior activity over the first-generation KRAS G12C(OFF) inhibitors in 2023. Slide 19. As a third example, RMC-9805 is an oral selective covalent inhibitor of KRAS G12D(ON), the primary tumor driver for more than 50,000 new patients annually in the United States, predominantly patients with colorectal, pancreatic or non-small cell lung cancer. RMC-9805 exhibits a highly differentiated profile, and we believe it is one of our most technically sophisticated RAS(ON) inhibitors to date. It uniquely engages the KRAS G12D cancer variant covalently through the oncogenic aspartic acid residue by taking advantage of a proprietary chemical warhead, a bespoke linker and our tri-complex binding modality. These design elements deliver a highly distinctive preclinical profile that includes oral bioavailability and selective and irreversible inhibition of this important cancer target. When administered orally to mice, then grafted with a KRAS G12D tumor, RMC-9805 achieves favorable plasma exposures, dramatically suppresses DUSP6 mRNA, a molecular biomarker of RAS pathway signaling, for over 24 hours due to its irreversible inactivation of the target. Slide 20. We believe RMC-9805 is the first ever drug candidate described that can covalently modify an aspartic acid residue in a targeted protein. Drives deep and durable anti-tumor responses in pancreatic and colorectal cancer models in vivo upon oral dosing, and it is well tolerated. Slide 22. As a fourth example, RMC-8839 is an oral selective covalent inhibitor of KRAS G13C(ON). We believe it is the first compound to directly inhibit KRAS G13C, a target primarily for lung and select colorectal cancer patients who are currently not served by a targeted RAS inhibitor. Slide 24. Lastly, in our pipeline expansion programs, we continue leveraging our RAS innovation engine to identify additional orally bioavailable tri-complex RAS(ON) inhibitors to target RAS variants driving RAS-addicted cancers that are unserved by current targeted drugs. As illustrated on this slide, our tri-complex inhibitors bind to RAS on proteins at a site that provides the opportunity for direct chemical interaction with amino acids at each of the 3 well-recognized mutational hotspots affecting residues G12 G13 or Q61. This binding geometry is leveraged in each of our programs to design compounds that are selective in engaging mutant amino acids at these positions that are responsible for most RAS-addicted cancers. Today, I'll share compelling initial data about RM-043, a representative mutant-selective non-covalent inhibitor of KRAS Q61H(ON) that was shown for the first time at the recent AACR Annual Meeting. RM-043 shows nanomolar activity in cells driven by the KRAS Q61H variant is selective for KRAS Q61H over wild-type RAS and drives deep progressions in a KRAS Q61H xenograft model of lung cancer. To our knowledge, this is the first ever example of a targeted RAS inhibitor directed to an oncogenic RAS variant at the Q61 mutation hotspot. This compound not only represents proof of principle for selective targeting of codon 61 by tri-complex inhibitors, but also demonstrates that this modality can be leveraged to develop highly mutant-selective inhibitors even when covalent bonding is not possible. Slide 26. The second theme and parallel approach that I'll talk about today is that our RAS companion inhibitors in development may be combined with RAS(ON) inhibitors to improve anti-tumor activity in preclinical models that are less sensitive to monotherapy. These RAS companion inhibitors are targeted drugs that suppress cooperating targets and pathways known to work in coordination with RAS cancer drivers to sustain RAS-addicted cancers and in some instances confer drug resistance. We believe that combining best-in-class RAS(ON) inhibitors with best-in-class RAS companion inhibitors offers the greatest chance of pathway suppression and durability of response to deliver the best clinical outcomes. Ultimately, the optimal RAS(ON) and companion inhibitor strategy will likely be disease specific. I will highlight here 2 clinical stage assets that support combination treatment approaches. Slide 28. First, RMC-4630 is a potent, oral small molecule that is designed to selectively inhibit the activity of SHP2, an upstream cellular protein that plays a central role in modulating self-survival and growth by facilitating RAS pathway signaling. Amgen continues its initial evaluation of dosing RMC-4630 in combination with sotorasib in second-line treatment of various KRAS G12C tumors in the U.S. CodeBreaK 101C study, and recently announced it has submitted initial data from the study to a medical congress for late summer. Revolution Medicine's clinical study called RMC-4630-03 is progressing and continues to enroll. This is a global Phase II study of RMC-4630 in combination with sotorasib in patients with advanced non-small cell lung cancer with a KRAS G12C mutation who have failed prior standard therapy and who have not been previously treated with a RAS(ON) inhibitor -- with a RAS inhibitor. We are on track to enroll the study fully this year and have sufficient data by the end of the year to share some of the high-level findings. Slide 29. Second, RMC-6236, the exciting RASMULTI(ON) inhibitor I described earlier, is notable within our broad RAS(ON) inhibitor portfolio because it, in particular, has the potential to be deployed as a RAS companion inhibitor as well as a primary cancer driver targeted agent. In some clinical context, patients may gain maximal clinical benefit from the broad activity of this RASMULTI(ON) inhibitor in combination with the deep and sustained target coverage provided by a mutant-selective RAS(ON) inhibitor such as RMC-6291. Slide 30. At the AACR meeting, we reported that RMC-6236 in combination with RMC-6291 demonstrated enhanced anti-tumor activity in KRAS G12C non-small cell lung cancer and colorectal cancer models that are relatively resistant to single-agent treatments. An example shown on this slide, CRC022, is one such KRAS G12C colorectal cancer model in which either the G12C(ON) inhibitor, RMC-6291, or the RASMULTI(ON) inhibitor, RMC-6236, as a single agent slows tumor growth but fails to induce tumor regressions. In contrast, combining these 2 agents converts the impact into significant tumor regression. Slide 32. Lastly, RMC-5552 continues to advance. This drug candidate is a potent, first-in-class bi-steric mTORC1-selective inhibitor designed to suppress phosphorylation and inactivation of 4EBP1 for cancers with hyperactive mTORC1 signaling, including certain RAS-addicted cancers. We aim to combine RMC-5552 with RAS(ON) inhibitors in patients with cancers harboring RAS and mTORC pathway co-mutations. We are making progress in our ongoing Phase I/Ib clinical trial evaluating RMC-5552 as a monotherapy, and are now focused on dose optimization in preparation for combinations with RAS(ON) inhibitors. Slide 11. Our third theme is unlocking the anti-tumor immune response by targeting RAS cancer drivers within tumors. We have seen examples, multiple examples in which RAS(ON) inhibitors as monotherapies reverse the immune-suppressive tumor microenvironment in RAS-driven cancer models and can unlock profound anti-tumor immunity when combined with immune system modulators such as PD-1 checkpoint inhibitors. In particular, both RMC-6236 and RMC-6291 alone can favorably transform the immune microenvironment in RAS tumors and are highly additive with a checkpoint inhibitor. RMC-6236 favorably transforms the tumor immune microenvironment by modulating both the adaptive and innate immune cells infiltrating these RAS-addicted tumors, and these changes significantly increase the sensitivity of such tumors to immune checkpoint inhibitors. Hence, the combination of RMC-6236 with a checkpoint inhibitor causes profound, durable and even complete anti-tumor responses in some preclinical models in mice with intact immune systems. Slide 16. RMC-6291 is able to modulate the immune microenvironment via tumor intrinsic effects at prime cancer cells for anti-tumor immunity in the presence of a checkpoint inhibitor. Here, we show this combination is also able to drive complete responses in an immunogenic model of KRAS G12C cancer. In summary, these prepared comments have provided an update on our portfolio across 3 core themes that are important to our strategy. First, RAS(ON) inhibitors demonstrate compelling monotherapy anti-tumor activity in diverse preclinical models of genetically defined RAS cancers. Second, RAS(ON) inhibitors can be combined with RAS companion inhibitors to improve anti-tumor activity in preclinical models that are less sensitive to monotherapy. And third, RAS(ON) inhibitors as monotherapies reverse the immune suppressive tumor microenvironment in RAS-driven cancer models and can unlock profound anti-tumor immunity when combined with immune system modulators such as PD-1 checkpoint inhibitors. These concepts, in conjunction with preclinical data sets behind each of the development stage assets in our R&D portfolio, underly our belief that we may be able to serve significant unmet clinical needs for patients with a wide range of RAS-addicted cancers. I'll now turn to Jack Anders, our Senior Vice President of Finance, to report on our financial condition. Jack?

Thank you, Mark. The details of our financial results are in our press release, so I'll focus on a few highlights as shown on Slide 36. We ended the quarter with $519 million in cash and investments. Revenue from our collaboration agreement with Sanofi was $7.6 million in the first quarter of 2022. The decrease in revenue from the prior-year period was due to lower development cost reimbursement from Sanofi. Total operating expenses for the first quarter of 2022 were $65.5 million and increased by 38% over the prior-year period. The increase in operating expenses was largely due to R&D expenses associated with our preclinical portfolio and increased headcount. Net loss for the first quarter of 2022 was $57.6 million or $0.78 per share. Our financial guidance for 2022 remains unchanged, and we continue to expect full year GAAP net loss to be between $260 million and $290 million, which includes estimated non-cash stock-based compensation expense of between $35 million to $40 million. And with that, I'll now turn the call back over to Mark.

I'm proud of the continued excellent execution by our R&D team with support by our broader organization and many partners and collaborators. We expect to have 2 RAS(ON) inhibitors in the clinic this year and to advance 2 additional RAS(ON) programs subsequently. We're pursuing an exciting multi-part approach aiming to outsmart RAS-driven cancers, including both RAS(ON) inhibitors and RAS inhibitors. We believe that Revolution Medicines is in an excellent position to continue aggressively pursuing our mission on behalf of cancer patients whose participation in our clinical studies is deeply appreciated. This concludes our prepared remarks for today. And I'll now turn the call over to the operator for the Q&A session. Operator?

[Operator Instructions] Our first question comes from Marc Frahm with Cowen.

Congrats on all the progress and getting -- starting to get the INDs filed. Maybe just starting with 6036 (sic) [6236] and 6091 (sic) [6291], Mark, if you could describe just kind of what you've submitted in terms of starting dose? And how close is that to what you think might be the active range if the preclinical modeling on exposures is correct? And within 6036 (sic) [6236], how might that dose differ based on the underlying RAS alteration?

Marc, thanks for your questions. I think with regard to the specifics of starting dose, I don't think that's something that we'll be disclosing today. I will say for RMC-6236, obviously, given that it is a RASMULTI inhibitor, and we know at some point it will have effects on normal tissues, we certainly are starting on the lower end to make sure that we creep up on the optimal dosing. But beyond that, I don't really have anything more specific to say today. With regard to whether the dosing will differ across different mutants, maybe I can ask Steve Kelsey, our President of R&D, to comment on that question.

I think -- sorry, Marc. I think it's highly unlikely, and that's certainly not the intent. I don't think there's any, at this stage, any rationale or preclinical justification for believing that the dosing schedule would be different for any given histotype or any given genotype, particularly as a single agent. I think where the dosing of RMC-6236 may also going down the track is that it's given in combination with some -- and then that might require some form of adjustment either to the dose schedule of both. But as a single agent inhibitor of RAS mutations, I don't think so. I think that we're going to explore the optimal dose and schedule for inhibiting RAS, and that will be it.

Okay. And then just brought up combos, and that's obviously part of the longer-term strategy here. I guess, what do you need to see to start opening up those combos? Is it just safety at reasonable exposures and go straight into combos? Do you want to see clinical activity? Do you want to see recommended Phase II dosing schedule for these agents? Just when do you open up that part of the program?

And Marc, is that specifically regarding RMC-6236 and using it as a combination agent, or are you asking about a mutant-selective inhibitor like RMC-6291 and when we would add -- begin adding a RAS companion inhibitor to it?

For either of them.

Okay. Yes. Right now, the strategy for RMC-6291 is to really start combinations as soon as we possibly can. And the minimum amount of information we need is some -- we need preliminary tolerability data, and we also need to make sure that we've got the schedule right. It's not immediately obvious that a once-daily schedule is right for any given RAS inhibitor and as I think has been illustrated admirably in the debate between [indiscernible]. So we need to make sure that we know the PK, and we need to make sure that our schedule is right, and then I think we can start combinations with RMC-6291. And combining it with RMC-6236 is pretty high on the agenda right now, given what we know about the mechanisms of escape from adagrasib. So that partly answers your question about RMC-6236 combinations as well. With regards to using RMC-6236 as a RAS mutant inhibitor in combinations with that, that may not be quite as early on in the program with regards to other inhibitors. I think firstly, when we're looking at combining with other agents that have some form of single agent toxicity profiles themselves, we need to understand a little bit more about the toxicity profile of 6236 before we start combinations. So that may take a little bit longer. But I don't see any real barrier to starting combinations of RMC-6236 with the mutant-selective RAS(ON) inhibitor 6291 as soon as we possibly can.

Okay. Very helpful.

Marc, if I could add a conceptual, sort of another layer on it that would match up with Steve's comments, obviously RMC-6291's entering a more crowded space where there are multiple other KRAS G12C inhibitors, and we indicated that our goal is to differentiate it from those others. And so it makes sense for us early in that program to begin including the best combination agents that make the most sense to us based on preclinical work and mechanisms of understanding. In contrast, RMC-6236 of course is being tested in complete white space in diseases for which there are no targeted therapies available. No RAS-directed targeted therapies for sure. And so there we're looking to figure out what's the optimal use of RMC-6236 and what's the best kind of anti-tumor activity we can obtain. And maybe a related concept, obviously, is that RMC-6236 is sort of a RAS companion inhibitor built into our RAS direct inhibitor. It has the ability not only to target the mutant driver, let's say, KRAS G12D or G12B or G12R, but also other companion signaling activities that can support those primary drivers. And so it's a multi-component in one. So the urgency around getting the combinations with it is not nearly as high.

Our next question comes from Michael Smith (sic) [Schmidt] with Guggenheim.

Maybe just a couple of ones, maybe sort of following up on the comments from just now. I guess on 6291, in terms of the longer-term development strategy and the regulatory path, is there -- do you view it as a monotherapy player? And if so, I guess, would that need to be in a sort of RAS-ad refractory patient population? Or is it mainly viewed as a combination strategy that would sort of differentiate the drug from others?

I think Stephen and I probably both have some things to say about that, so why don't we start with the expert, Steve.

Yes, let's try to -- I think the approval of sotorasib was basically divided by non-small cell lung cancer space into patients that get treated before sotorasib and patients that get treated in that second-line space and then people that get salvage after it's failed. I think that's the most easy conceptual way of thinking about it. And I'm deliberately using non-small cell lung cancer as probably the best example to answer your bigger question. So if we are -- there is an opportunity potentially to use RMC-6291 one as a single agent in the salvage space. It's not our -- it's not the -- our primary strategy for 6291. But there are some mechanistic bases for believing that it would work there in some patients that do fail on sotorasib. Having said that, we all believe the best way of approaching those patients is with some sort of combination that involves a KRAS G12C inhibitor, and we would obviously prefer that G12C inhibitor to be 6291, in combination with something else that's suppressing the [ nonstate ] mutations, of which I think 6236 is probably a good candidate. The most obvious place where you could get 6291 for single agent is that it's demonstrably superior to sotorasib in the patients who are currently getting sotorasib. And I think that that's an obvious opportunity for us. It's the data we've seen in the preclinical, what we call nice clinical trial, if you like, plays out in the clinic. Then there is obviously a huge opportunity there to get RMC-6291 approved as a single agent. It still doesn't mean it's the best treatment for patients. I think we still believe that the best treatment for patients is going to be combining 6291 with some form of companion inhibitor, whether that's 4630 or 6236 or 5552 or something else. We just don't know right now. I think where there's really probably not so much of an opportunity for single agent 6291 is in that first-line space where you're going up against chemotherapy plus pembrolizumab. And that's a pretty tough ask, to be honest, for any single agent, not the least a single agent KRAS G12C inhibitor. So that's probably an obligate combination play, I think, and probably an obligate combination play in combination with a checkpoint inhibitor, which going back to the previous caller's question is a very high priority combination for us to find.

Right. That makes sense. And then on 6236, I guess -- so you're positioning it as a combination drug, to some degree, with something like 6291. And then I guess my question is, what percentage of patients that are treated with a G12C selective inhibitor would, I guess, benefit from the multi-inhibitor addition? If you think about resistance mechanisms, and how would that combination compare to something like a SHP2 inhibitor combination, for example?

Okay. Well, there are a lot of layers to that, Michael. Good comment and questions. If you go back to something Steve said, I just want to emphasize that the preclinical data with RMC-6291 suggests that it alone is superior in a head-to-head comparison with, let's say, a KRAS G12C(OFF) inhibitor in lung cancer model studying KRAS G12C. So that's our -- that is part of our hypothesis going into clinic is that that's what we'll see as well in the clinic. And whether that translates into a registration path or not, you got into some of the nuances in second line versus first line. But that, Steve I think said something that's really worth emphasizing. That's not the same thing as saying that that's necessarily the best way and most durable way to treat a patient. And we know that RAS pathway-driven cancers are extremely resilient. And as you know, we've said this now for several years, it's really important to address those resiliency pathways, those pathways of persistence, and ultimately, those will require combinations. So the answer isn't cleanly, yes, we're only interested in monotherapy or, no, we're only interested in combinations. It's both. And we first have to see a bit about its monotherapy profile in the clinic, and then we'll begin pursuing the combinations. And the combinations may differ depending on the context, as you just asked. The question that I think you asked and I haven't addressed is RMC-6236 a better RAS companion than RMC-4630, a SHP2 to inhibitor. I want to say something about that, and then I want to ask Steve to comment on it. But something I want to say first is we think of RMC-6236 first as an inhibitor of the driver of cancers that are not currently served by G12C inhibitors. So our first goal is to get it into the clinic and see what it does in monotherapy for KRAS G12D, G12B, G12R, et cetera. And we're still excited about that. But we did introduce this second idea that has occurred to a number of people, including us, that there may be benefit in adding it to regimens that are anchored in a RAS mutant-selective inhibitor like RMC-6291. And we think that's a parallel path to pursue. These really aren't competing with each other. They are the 2 -- some of the independent hypotheses. So with that in mind, then, that's teeing up for Dr. Kelsey to comment on RMC-6236 versus RMC-4630 as a companion agent.

I think the answer, the honest answer to your question is that right now, we don't know. There's very good biological rationale and preclinical data to support the use of either of them as a companion or a RAS mutant-selective inhibitor that inhibits KRAS G12C. And I think we are not going to get any further down that road by doing more preclinical studies. I think this is going to be something that has to be tested in the clinic and real world. And we will see in due course which one of them is most appropriate in which circumstance. Don't forget there are at least 3 populations of patients with KRAS G12C mutations that are significant -- that occur with significant frequency to create an unmet medical need. There are patients with lung cancer, colon cancer and then other, which includes pancreatic cancer, and the answer may be different for each of those 3 groups. And if you go back to my sort of lung cancer division, again, I mean the answer may be different depending on whether you're looking at first line, second line or third line and whether the patient has or hasn't previously received a KRAS G12C(OFF) inhibitor. So I think we're just going to have to do the clinical experiment and figure it out. We're privileged to have the wealth of companion inhibitors that we have, and hopefully we will figure out how best to use them for the treatment of any given group or hopefully an individual patient.

I'm sure we've simplified that for you, Michael, now.

Yes. Too many drugs, I guess.

Our next question comes from Christopher Zopf with Goldman Sachs.

I wanted to talk a little bit more about 4630. We have a couple data sets coming up second half of the year. The 2 studies were kind of set up because of limitations and the CodeBreaK study in terms of being able to come up with a strong answer for what we can see with SHP2 inhibition in lung cancer. So I was wondering what you hope we should all learn with the Amgen update towards the late summer. And how might that augment the plans for the next step with 4630? Would the next steps kind of become clear with that data and combining that with what we will show from the 03 study later this year? Or will the next kind of state the development become clear more next year after the second updates?

C.J., thanks for your question Yes. I think with regard to anticipating what Amgen will or will not show this summer, we can't really comment on that, of course. They are the sponsor of the study and will present what they present. So I think it'd be easier to discuss that after they presented something we can discuss what the implications of that would be. With regard to the first part of your question, though, which is relating the CodeBreaK study to the 03 study, they really are complementary and to some degree overlapping. Amgen had a head start with the CodeBreaK study. We entered that collaboration a few years ago, and they've had an opportunity to really learn a lot about dosing tolerability and safety and then to expose patients with various histotypes to RMC-4630 and look for anti-tumor activity. They'll report about those things when they do. We've built on that by creating a dedicated single histotype study, the lung cancer, non-small cell lung cancer, and used the opportunity in a dedicated study to create boundaries or several different cohorts within that, as you know, specifically to help us understand the sensitivity of those tumors with only a KRAS G12C mutation driver and not concurrent mutations elsewhere versus those that carry co-mutations in RAS and RAS-related signaling. Because we do know from the experience over the last couple of years now, something we didn't know at the beginning of all of this, which is that at least for the KRAS G12C(OFF) inhibitors, there are some factors that can create differential sensitivity. And so we needed to build that in, and this was an opportunity to do it since it was not preemptively built into the CodeBreaK study. And then as to how will we use the results of these 2 studies, we'll look at them both in deciding what to do going forward. Our aim is to learn enough that we can design a compelling, credible Phase III study, and we'll take information from either of those sources and both of those sources to help us make those decisions.

Our next question comes from Eric Joseph with JPMorgan.

This is [ John ] on for Eric Joseph. Congrats on the progress. So just coming back to the RAS(ON) 6236, we are wondering how we should be thinking about the patient demographics eligible for the Phase I study, by tumor or mutational backgrounds and histology and perhaps PD-1 status? And how broadly are you looking at the potential activity and pharmacodynamics in the Phase I study on top of safety and PK?

Thanks very much for your question. It's got a few layers to it. I think Dr. Kelsey can address how we're approaching bringing 6236 into the clinic initially and what mechanisms we're using to be able to learn as much as possible as quickly as possible.

Yes. The protocol really allows for a patient with advanced cancer who would be eligible for a Phase 1 trial who also has a mutation in a KRAS G12 position. And we expect the vast majority of those patients in the first instance, purely because of the demographic distribution of those rotations, the majority are going to have non-small cell lung cancer, pancreatic or colorectal cancer. I think that's just the truth. And these will be patients that by necessity, by ethical necessity have failed available therapies that are approved and available to them in the geographic location in which they reside. So I don't think there's anything terribly unusual about the patient population in the early stages of the program. Then of course, as we start to see what we see, as we start to see signals, then it will become a lot easier to refine the program or even open it up to other patients that were not included in the initial stages of the development. And as we go, that could include patients with other mutations that are not at the G12 position. It could include focusing on specific histotypes, or it could include, of course, combinations as well as we've already discussed. So I hope we're not -- we are very keen to give you the impression that in the initial stages of development of this compound, a lot of the general principles in oncology will be there. And that's partly because we want it like that, and it's partly because of the constraints that are placed upon us by take testing a double agent.

And if I could add a couple of things that relate to the disclosures we've made in the past. As Steve said, there principally are 3 major histotypes that will be represented simply because epidemiologically, those are the tumors that are largely represented. We focused on G12 mutations, what we call for convenience G12X, where X is any one of several different substitutions. Really only because in our mouse clinical trial, while we saw that sensitivity across many different genotypes, we just saw greater sensitivity in those G12X mutant-driven cancers. That's not to say that we're not interested in the other ones. It's just that we're elevating those as early as possible to try to help us get the signals as early as possible. Once we see those signals, we have the opportunity, as Steve alluded to, either to broaden the criteria to bring in other genotypes and/or to focus further on places where we're seeing the signal or both. We may be both broadening and focusing at the same time. And the last comment I want to make is to remind you that there is a backfill recruitment mechanism, or filling the slots approach that allows us as we're dose escalating to expand for many below MTD expansions so that we can increase the data sets. And in those settings, in that context, we'll not only get more experience as we're dose escalating, but we also have the opportunity to sample diverse genotypes. So a lot going on there. But at the end of the day, we'd like to first demonstrate that it's a safe and useful agent; and second, identify who's most likely to benefit from it so we can further study them.

Yes, just to wrap up your question about specifically about biomarkers, I think what we tried to imply is that the majority of the focus, the biomarker-based focus of the program is on patient selection, and also obviously we'll be monitoring things like circulating tumor DNA to look for mechanisms of escape. There's not a huge amount of effort going into looking at pharmacodynamic markers of drug activity or hitting a target, because the vast majority of the preclinical data that we generated with RMC-6236 suggests that if you hit the target, then the tumor will shrink. So we're going to be -- I know it sounds a bit trite, but the reality is that we don't see an awful lot of value in figuring out whether or not we're inhibiting the RAS pathway with RMC-6236, because if we do it effectively, then we expect to see the tumor shrink So PD marker of choice is the efficacy marker of choice, which is the CT scan. But that doesn't mean we haven't put a huge amount of energy and resource into looking at genomic basis, [indiscernible] sensitivity and genomic basis are potentially safe.

Our next question comes from Jonathan Chang with SVB Securities.

This is Faisal on for Jonathan. Wanted to ask about RMC-5552. We saw the title in the ASCO program. I wanted to see if you could help set expectations for that data update and what we might see relative to what you disclosed in January.

Thanks for your question. I think Steve can comment on that.

We don't have a formal update on the RMC-5552 program as yet. We've dosed 14 patients across 5 dose levels. We're still doing dose optimization to some extent because we really do feel that we need to get the dose right for this for at least the single-agent expansion, and then obviously at some point in combination with our RAS(ON) portfolio. All we can tell you is essentially what we previously reported, which is we clearly have an active drug at the dose of 6 milligrams weekly. It may be possible to dose it at a dose that's slightly higher than that, but certainly not as high as 12 milligrams weekly. And the toxicity is mainly on target toxicity. In fact, almost exclusively on target toxicity. So we're feeling pretty good about RMC-5552, and we almost certainly will have a further update for you at some point, but we don't have a huge amount of additional data to report right now.

Great. And then a slightly different question. Can you remind us what we know about potential CNS activity of the RAS(ON) inhibitors?

Yes. We don't know a lot about it. It hasn't been a major area of focus. We do know that these compounds in general are beyond rule of 5 molecules, so they're rather large. The starting -- the chemistry that we started with isn't intrinsically expected to deliver compounds that can cross the blood-brain barrier into tumor metastases, but we haven't studied it extensively. Our focus right now is on systemic disease where we feel both within the G12C space and the other mutant space, we still have a long way to go before we can declare victory in terms of systemic disease. But we do acknowledge that CNS disease can be very important in some of these histotypes.

Our final question comes from Ben Burnett with Stifel.

This is Neil Carnahan on for Ben. Novartis had a presentation at AACR showing a response rate in lung cancer patients that was relatively in line with what we've seen for LUMAKRAS and adagrasib. Can you talk about some of the preclinical metrics maybe a bit beyond the mouse model data you've shown that suggests 6291 will be improved relative to this mid-40s response rate we're seeing in lung cancer patients once it reaches the clinic?

Thanks for your question. Not sure I totally understand it. I think you asked do we have more preclinical data other than in mouse xenograft models, and that's really the only setting in which we can do those sorts of studies. We primarily have looked at response rates in our public disclosures from those kinds of studies, but we also look beyond -- we provided data beyond response rates. We talked about both depth of inhibition of tumors and durability. And we reported that we've seen positive impacts across all 3 of those parameters. So while I think response rate is certainly something we can't avoid looking at, and it's the sort of simplest thing to start with, that's really not the whole game, and it's not necessarily the singular endgame for showing clinical superiority.

As there are no more questions in the queue, I'll turn the call back over to Dr. Goldsmith for closing remarks.

Thank you, operator, and thank you to everyone for participating today and for your continued support of Revolution Medicines.

This concludes today's conference call. Thank you for participating. You may now disconnect. Everyone, have a great day.