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

All right. Good evening, and thanks for sticking with us for our JPMorgan Healthcare Conference. I'm Eric Joseph, senior biotech analyst at the firm. And our next presenting company is Revolution Medicines. And it's my pleasure to welcome CEO, Mark Goldsmith, to tell us a little bit about the company. [Operator Instructions] With that, Mark, thanks again for sharing some of your time with us.

Thank you, Eric, for the opportunity to introduce Revolution Medicines. Slide 2. Revolution Medicines is a clinical-stage precision oncology company focused on large unmet medical needs in RAS-addicted cancers, representing a very significant share of all human cancers. Our science-driven strategy includes targeted therapeutics organized into 2 portfolios: First, our RAS(ON) inhibitors, deriving from an innovative technology platform, are unique compounds that target diverse oncogenic RAS variants through highly differentiated chemical and pharmacologic profiles. We are very pleased to announce today that our first 2 RAS(ON) inhibitor programs have entered development: RMC-6291 targeting the KRAS G12C oncoprotein; and RMC-6236, a multi-RAS inhibitor for a range of diverse RAS proteins that cause cancer. Second, our RAS companion inhibitors directed against key nodes in RAS-signaling circuits are potential backbones of targeted combination treatments to maximize clinical benefit. Throughout 2020, we demonstrated that RMC-4630, our SHP2 inhibitor, exhibits clinical activity and is advancing in a broad clinical program. RMC-5552, our innovative mTORC1-selective inhibitor, has completed its IND-enabling program and is preparing to enter the clinic shortly. And we announced today that RMC-5845, our SOS1-selective inhibitor, has also entered development. Hence, we now have 2 clinical-stage programs and introduced 3 new development candidates that have begun IND-enabling development. Slide 4. A critical foundation of our company is that RAS proteins are the primary cause of many cancers, and these same proteins also induce oncogene addiction by cancer cells and elicit their own drug resistance mechanisms. Our goal is to outsmart RAS-addicted cancers by understanding and then overcoming these fundamental complexities. And our product strategy specifically addresses 3 key features: one, many distinct variants of the RAS family of proteins cause cancer and underlie addiction; two, it's the activated or RAS(ON) form of these proteins that drives oncogenic signaling; and three, complex natural circuits within human cells can be exploited by cancer cells as drug resistance pathways, and these pathways require carefully chosen companion drugs for combination strategies in order to achieve deep and durable clinical responses. Slide 5. The scope of the unmet need in RAS-addicted cancers is enormous, including 30% or more of human cancers in the U.S., causing severe morbidity and mortality. As I mentioned, there are many different variants of RAS that can cause cancer, and each RAS-addicted tumor can be defined by its genetics. Shown here is our estimate of the new RAS-addicted cancers diagnosed each year in the U.S. organized into convenience groups based on these genetics that also provide a road map of which types of RAS inhibitors are needed. For example, at the bottom of the graph is the G12C mutation responsible for 30,000 new cancers each year. Although it is the best-known RAS mutation, it is not the most common as both G12D and G12V are more common. Some are less common, such as G13C, but nonetheless important. In aggregate, some 230,000 new cancers could be served by RAS inhibitors in the U.S. alone each year. These are quite large numbers. Slide 6. We believe that Rev Med is uniquely positioned to serve these enormously important clinical needs, and we have tremendous momentum toward doing so. As a pre-IPO company at last year's JPMorgan conference, we had one asset in clinical development, RMC-4630. Today, our cohesive pipeline includes 5 complementary and compelling development-stage product candidates deriving from our innovation engine and targeting key drivers of cancer, RAS addiction and resistance. Today, I'll focus particularly on RMC-6291 and RMC-6236 as the newest pioneering assets in our emerging RAS(ON) inhibitor portfolio. Slide 8. At the center of RAS control in a cell is the activation cycle in which cell growth signals normally cause the inactive pool of RAS proteins, termed RAS(OFF), to be converted to the active form or RAS(ON) that drives cell growth and survival until the signal is terminated by cycling back to RAS(OFF). The 3 isoforms, KRAS, NRAS and HRAS, are highly related to each other in amino acid sequence and structure. They play similar roles in RAS signaling, and each has 3 mutational hotspots, G12 G13 and Q61. Amino acid substitutions in any of these can cause the protein to be sustained in the RAS(ON) form that causes excessive signaling and drives cancer. So with 3 isoforms, 3 mutational hotspots and many possible amino acid substitutions, there are dozens of cancer-causing variants that need to be addressed. We have found that inhibiting the RAS(ON) form of an oncogenic RAS variant can have dramatic effects on RAS-addicted tumors. Slide 9. Our RAS(ON) inhibitors are highly innovative and sophisticated compounds that enter a cell, bind to an abundant chaperone protein known as cyclophilin A, and then engage a particular RAS(ON) target to form a tri-complex that quickly inactivates RAS signaling by sterically preventing activation of downstream effectors like RAF. We've shown compelling single-agent activity for these inhibitors in preclinical in vivo models of human cancer, examples of which we'll highlight today as well as combination benefits. We've observed preclinical features of these inhibitors that suggest potential clinical benefits in terms of breadth, depth and/or duration of antitumor impact that will deserve to be evaluated in patients. And we've proven the broad reach of our inhibitor collection across a wide array of oncogenic variants. Slide 10. Indeed, Revolution Medicine's outstanding research organization has leveraged our proprietary drug discovery platform to produce a vast collection of novel compounds, some examples of which are shown here, bound to various specific RAS(ON) targets in these color-coded representations of co-crystal structures from our collection. We have potent and cell active RAS inhibitors that inhibit the oncogenic or ON form of their respective targets that reach across every RAS isoform and that inhibit RAS proteins containing all or nearly all cancer-causing mutations and therefore, could address the vast majority of RAS-addicted cancers. As we will demonstrate today, some of our compounds are highly selective for a single or a few targets and some have multi-target activity. Today, we'll introduce you to the exciting first 2 inhibitors from this unique collection that have entered development. Slide 11. RMC-6291, shown in the co-crystal structure on the left, is our first-in-class, highly potent, selective oral tri-complex inhibitor of KRAS G12C (ON) and with excellent drug-like properties. It exhibits subnanomal or potency for suppressing RAS/MAP kinase signaling through phospho ERK and causing growth inhibition in KRAS G12C-driven tumor cell lines and for NRAS G12C-driven cells as well. It's engineered to exhibit very high and specific reactivity to cysteine 12 on the mutated RAS G12C target, to be highly selective for RAS G12C-driven cell lines over RAS independent tumor lines and to have a low-risk profile as determined in an expansive off-target safety panel and cysteinome screen. And it shows good oral bioavailability across small and larger species with low to moderate metabolic clearance. In sum, this profile is highly favorable compared to the current class of KRAS G12C(OFF) inhibitors and is differentiated from others by acting through direct inhibition of KRAS G12C(ON). Slide 12. RMC-6291 displays a highly distinguished cellular signature compared to RAS(OFF) inhibitors. Biochemically, as shown on the left, within 5 minutes of exposure to the compound, virtually all of the cellular KRAS G12C protein has been bound to and covalently modified by 6291, as represented by the upward shift of KRAS G12C protein in gel electrophoresis indicated by the green arrow. Functionally, as shown on the right, 6291 very rapidly terminates RAS signaling as measured by levels of RAS/RAF complexes within the cells, a critical step in driving cancer. In cells exposed to either of 2 leading KRAS G12C(OFF) inhibitors, it takes hours to observe a 50% reduction of RAS/RAF signaling complexes. The time course for these compounds is dictated by availability of the KRAS G12C(OFF) target, which is inherently limited by the turnover rate of RAS(ON) to (OFF) forms. In obvious contrast and consistent with its direct action on KRAS G12C(On) proteins, the inhibition effect is nearly instantaneous in cells exposed to RMC-6291, with virtually all RAS/RAF signaling complexes vacated by 5 minutes. This unique signature highlights a fundamental mechanistic and class-based distinction between the RAS(ON) and RAS(OFF) inhibitors. Slide 13. The striking cellular signature of RMC-6291 is consistent with its impressive antitumor effects in vivo as observed in preclinical xenograft models. In this example, a well-known human lung cancer line bearing a KRAS G12C mutation is engrafted into mice and followed over time for tumor growth or regression. RMC-6291 caused deep and consistent tumor regressions at all 3 doses shown here, 25, 50 or 100 milligrams per kilogram given daily and orally. These effects compare well to the benchmark used in this experiment, MRTX849, at 100 milligrams per kilogram daily, the standard dose used by Mirati despite our dosing RMC-6291 with as low as 1/4 of the MRTX849 dose. All animals in all treatment groups tolerated treatment well. Slide 14. Looking a little more closely at a comparison of 6291 with MRTX849 reveals potentially important differences. The end-of-treatment measurements on the left show that RMC-6291 in all 3 dose groups actually induced deeper regressions than the MRTX849 benchmark. 10 of 10 animals in each group scored as complete regressions, or CRs, defined by a reduction in tumor volume of greater than 80% compared to the starting volume compared to 7 of 10 in the MRTX849 comparator group. Furthermore, after discontinuing treatment and tracking tumor regrowth over time, as shown on the right, it's clear that RMC-6291 induced a more sustained antitumor response compared to MRTX849. Together, these results show the superior antitumor efficacy of RMC-6291 in a preclinical model. Slide 15. In a second example, we tested 6291 and the comparator in LUN092, a patient-derived lung cancer carrying a KRAS G12C mutation engrafted into a mouse. Both 6291 and the MRTX compound given at 100 milligrams per kilogram per day drove regressions in this PDX model, and both are well tolerated. But again, as shown previously in the CDX model, in this G12C dependent patient-derived model, RMC-6291 induced deeper and more consistent responses. Slide 16. Based on these data and additional findings we'll report at a future scientific meeting, RMC-6291 has a best-in-class preclinical profile that we believe predicts a best-in-class clinical profile. It binds to and inhibits RAS(ON). It has subnanomolar potency. It shows dual selectivity for KRAS and NRAS G12C in the ON form. It causes deep and durable antitumor responses in vivo. It's currently in IND-enabling development with IND submission projected for the first half of 2022. And our clinical superiority thesis, which, of course, must be tested in patients, is that RMC-6291 will exhibit advantages in terms of range of sensitive tumor types, response rate, depth and/or duration and monotherapy effects as well as beneficial combinations with RAS companion inhibitors. Slide 17. We're also excited to share with you our second distinguished RAS(ON) inhibitor to enter development. RMC-6236 is a first-in-class potent RAS selective oral tri-complex, multi-RAS(ON) inhibitor that we expect to be useful in multiple RAS-addicted tumor types. Examples of co-crystal structures on the left convey that RMC-6236 drives multiple diverse oncogenic forms of RAS into inhibitory tri-complexes, including KRAS G12V(On) and KRAS G12D(ON). These are the 2 most common mutant RAS drivers of human cancer, together causing 100,000 new cases of cancer each year in the U.S. alone that remain unserved by targeted therapies. RMC-6236 exhibits low nanomolar potencies across a range of RAS-dependent tumor cell lines, yet is highly selective over RAS-independent tumor cells. It also scores as low risk in a broad off-target safety panel. It has attractive drug-like properties, including good oral bioavailability across multiple species. And it has low to moderate metabolic clearance across multiple species. This very attractive compound has entered development. Slide 18. As mentioned, the key feature of 6236 is that it potently inhibits RAS-dependent tumor cells in vitro. Shown here, it exhibits single digit nanomolar potency for growth inhibition across multiple cells carrying specific RAS mutations, such as G12V and G12D mentioned earlier as well as excellent potency against a broader array of other RAS mutants. It's also effective against upstream pathway mutants such as EGF receptor activating mutations that depend on wild-type RAS for their oncogenic effects. But importantly, it shows virtually no pharmacologically relevant impact on cells with downstream pathway mutations that function independently of RAS. Slide 19. These in vitro characteristics suggest that RMC-6236 may be useful for serving numerous and very substantial unmet needs in RAS-addicted cancers. For example, on the left, its most obvious application will be for oncogenic RAS mutations for which a mutant selective inhibitor is unlikely to be produced based on the properties of the mutated amino acid side chains such as the very common KRAS G12V. It should also be useful for oncogenic RAS mutants for which mutation-selective inhibitors are likely to be developed in the future, such as KRAS G12D and G13C or G13D. There are also numerous cancers in which excessive signaling through wild-type RAS is the oncogenic driver. As shown on the right, for example, amplifications of wild-type KRAS are known cancer drivers, and RMC-6236 could be useful in these cases. And finally, tumor cell escape from drugs targeting the RAS pathway often occurs by various mechanisms that amplify signaling through wild-type RAS to cause RAS-mediated adaptive drug resistance, and these should be treatable with RMC-6236 as well. Hence, 6236 has the opportunity to address the majority of unserved RAS-addicted cancers. Slide 20. Let's spend the next few minutes considering the properties of RMC-6236 in vivo. This experiment shows deep and sustained RAS pathway inhibition following a single dose of RMC-6236 administered to mice with xenografts of a human lung cancer line carrying a KRAS G12V mutation and the concurrent meta amplification. Either of 2 low doses, 10 or 25 milligrams per kilogram given once orally, caused inhibition of tumor DUSP6 mRNA, a convenient molecular marker RAS pathway suppression that typically parallels that of phospho ERK. 50% inhibition is evident at either dose level by 1 hour and nearly complete suppression observed thereafter. Notably, a single 25-milligram per kilogram dose caused sustained pathway suppression even out to 72 hours despite the fact that the compound's half-life in mouse plasma is under 3 hours. We believe this prolonged tumor impact is at least partially attributable to a tumor drug depot effect caused by binding of RMC-6236 to high intracellular concentrations of cyclophilin A in tumors. We'll share more insights about this interesting and useful ADME profile at a future scientific meeting. Slide 21. The pharmacodynamic biomarker effects of RMC-6236 also correspond to impressive dose-dependent antitumor effects in xenografts. In the same lung cancer model, 6236 administered orally caused obvious tumor regressions, as shown on the left. The 10-milligram per kilogram group showed 8 out of 10 regressions, as measured at the end of the study indicated on the right, while the 25-milligram per kilogram group showed 10 out of 10 complete regressions. Importantly, both daily doses were also well tolerated, arguably much better than the field may have expected. Of course, at high doses well beyond what is needed for these tumor regressions, ON pathway toxicity is observed as would be expected for a compound that inhibits the RAS family of proteins. Nonetheless, as shown here, 6236 has a clear preclinical safety margin, which is likely attributable to several factors, including at least RAS addiction of tumors compared to normal tissues and the tumor drug depot effect described previously. We'll share with you more detail about factors promoting the therapeutic index at upcoming scientific meetings. Slide 22. Marked antitumor effects were also seen in other human cancer models carrying a KRAS G12V mutation. For example, a pancreatic ductal adenocarcinoma cell line-derived xenograft shown on the left and the colorectal cancer patient-derived xenograft shown on the right both showed significant tumor regressions in response to RMC-6236 and in some cases, complete regressions. These were observed at various low doses of the compound and when administered daily or intermittently. These results clearly show the great promise of 6236 for treating KRAS G12V-bearing cancers. Earlier, I also showed the inhibitory effect of RMC-6236 on KRAS G12D in vitro. And here, we display that antitumor effects in vivo. This is a human pancreatic cancer line carrying a KRAS G12D allele engrafted into mice. Again, 6236 administered orally at either 25 milligrams per kilogram per day or intermittently 50 milligrams per kilogram every other day caused profound antitumor responses as shown on the left. Indeed, at the end of the experiment, all of the animals in both treatment arms showed complete regressions. Again, both treatment regimens were well tolerated. The sum of the evidence is strong, that RMC-6236 causes pronounced antitumor responses in preclinical models of human lung, colorectal and pancreatic cancers caused by multiple notorious RAS variants for which no targeted treatment is available today. Furthermore, these benefits are manifest despite good tolerability. Once again, this exquisite compound can turn RAS addiction into a profound vulnerability for these tumors. Slide 24. Based on these and extended findings, we expect RMC-6236 to be able to serve very broad and compelling clinical opportunities for large unmet needs. It acts via binding and inhibition of RAS(ON) proteins. It has low nanomolar potency. It's selective for the RAS family proteins, and it causes deep and durable antitumor responses in vivo at tolerated exposures. It's currently in IND-enabling development, and we project IND submission in the first half of 2022. Our broad clinical thesis is that numerous RAS genotypes across multiple patient segments will show sensitivity to 6236 alone as well as potentially beneficial combinations with RAS companion inhibitors. Slide 25. We believe that RMC-6291 and RMC-6236 are both groundbreaking compounds that can address a staggering range of RAS-addicted cancers. 6291 is a compelling first-in-class mutant selective RAS(ON) inhibitor for cancers driven by KRAS G12C or NRAS G12C with potential to demonstrate clinical superiority, while 6236 is an unprecedented RAS(ON) inhibitor with potential to demonstrate benefit across a wide range of oncogenic RAS family targets as shown here. We remain deeply committed to systematically and smartly serving numerous patient segments in the most optimal way. In this context, we will continue progressing additional exciting compounds from our mutant selective RAS(ON) inhibitor series. For example, our KRAS G12D selective and G13C selective programs are progressing well in lead optimization and will advance into development as DCs are selected. Scientifically, we can expect that there are trade-offs to consider between compounds with individual mutant selectivity carrying lower ON pathway toxicity risk but higher potential for RAS-mediated adaptive resistance and compounds with RAS family selectivity carrying lower RAS-mediated adaptive resistance potential but higher on-target toxicity risk. We intend to lead the field in using these unique compounds to evaluate these phenomena in humans and determine exactly which profiles actually provide the greatest patient benefits for treating different cancer histologies and genotypes. Over the coming months, we will share more preclinical information about these remarkable compounds at various scientific meetings. Slide 27. Let's briefly review the status of our RAS companion inhibitor portfolio. Starting with RMC-4630, our potent, selective and oral inhibitor of the SHP2 protein, a master regulator of the RAS signaling pathway. In 2020, we reported multiple times on our broad clinical development program conducted under our global partnership with Sanofi. We've reported monotherapy activity in multiple cancers and genotypes and are currently conducting expansion work at the recommended Phase II dosing schedule for monotherapy. We've also reported preliminary activity in combination with a MEK inhibitor in RAS mutant colorectal cancer and are currently conducting expansion work at the recommended Phase II dosing schedule for this combination. We've also reported evidence from paired tumor biopsies supporting enhancement of the tumor immune microenvironment in patients treated with RMC-4630. Slide 28. Our central clinical thesis remains that RMC-4630 should serve as a backbone of rational mechanism-based combination strategy, and this table summarizes the multiple combination strategies we are evaluating. Four clinical experiments are currently underway testing various combination hypotheses, and 2 additional clinical experiments are planned but have not yet begun. These studies involve 5 institutional collaborators, including 4 big pharma partners. We look forward to providing updates as these studies progress. Slide 29. Our second RAS companion inhibitor is RMC-5552, a highly innovative and potent mTORC1-selective inhibitor. I'm excited to report that we have submitted the IND for this compound and are preparing to begin a Phase I monotherapy dose escalation study imminently. Once we have initial monotherapy experience in patients, we expect to test 5552 in combination with RAS inhibitors for patients harboring RAS and mTOR signaling co-mutations. Today, we also briefly introduce our third RAS companion inhibitor, Slide 30, RMC-5845, an attractive, potent and oral inhibitor SOS1, a major switch for RAS(OFF) to RAS(ON). This compound is highly selective for binding and inhibiting SOS1 over SOS2, and it's well-tolerated preclinically. It's intended for select combination therapies for certain genetically defined RAS-dependent cancers. RMC-5845 is now in IND-enabling development, and we will share more information about it at a scientific meeting later in 2021. Slide 31. In summary, despite the menacing COVID-19 pandemic and chaotic macro environment, 2020 has been a spectacularly productive year for Revolution Medicines on behalf of patients, and we are proud of our expansive pipeline and strategy to outsmart RAS-addicted cancers. I've described for you today our first 2 RAS(ON) inhibitors to enter development, RMC-6291 and 6236, both of which have striking profiles and we believe offer very exciting potential benefit on behalf of many cancer patients. We also continued 2 other programs that are in lead optimization and additional RAS(ON) inhibitor research targets behind these. I've also summarized for you the progress and status of our 3 development-stage RAS companion inhibitors. RMC-4630 is progressing through the clinic. RMC-5552 is about to enter the clinic. And our newest development candidate in this companion group, RMC-5845 is now in IND-enabling development. We will continue our tireless commitment to patients and look forward to providing updates as things progress this year and beyond. On behalf of our employees, investigators, patients and partners, thank you very much for your interest in Revolution Medicines.

All right. Great. Thanks, Mark, for that presentation. Picking up on some of the data behind the newly introduced candidates -- the RAS(ON) candidates. I guess the first one here for 6291, you showed some nice contrasts with the RAS(OFF) inhibitors in some cell line and patient-derived xenograft models. We -- on the clinical side, we see some differential activity with the RAS(OFF) inhibitors depending on tumor type, lung cancers probably a little bit better than colon cancer. I guess do you -- or do you see a similarly favorable advantage with the RAS(ON) inhibitor approach compared to the RAS(OFF) inhibitors? Have you looked at that preclinically?

Yes. That's obviously an important area of investigation, and it's ongoing. And I think I alluded to it that we think that is one of the areas in which there could well be clinical differentiation. I think as this year rolls out, we'll show some more scientific data that will address specifically that point. But I think it's fair to say that in each dimension that we've investigated, we have seen data that suggests superiority for the RMC-6291.

Okay. Okay. And with 6236, you talked a little bit about sort of where your comfort level is so far and how it's been mature with the anticipated therapeutic window. I guess are you able to put a little more context around what you're seeing so far, perhaps drawing contrast with other maybe nonmutationally selected but RAS pathway inhibitors like a MEK inhibitor or RMC-4630, for example, that sort of might inform anticipated therapeutic window with that approach that you talked about?

Yes. Well, I think there's no doubt that RMC-6236 can have dramatic regressive effects on tumors in these models without causing any harm to the animals. So -- and I think I showed that in maybe 2 or 3, 3 or 4 different experiments. So that's been a consistent finding. And as you know, with RMC-4630, we can elicit antitumor responses in preclinical models, but there is sort of 4 that we can't get below because we, at some point, start to run into pathway toxicity, and we've managed that in the clinic through creative innovative dosing regimens. But I think although it's always very hard to compare different compounds against different targets with each other, I think it's pretty clear to us that the efficacy we see with the RMC-6236 is quite profound with very good tolerability in this dose range. And when I alluded to ON pathway toxicity, what I was trying to say is that if we push that dose up, we can certainly get to a point where it starts to look like a MEK inhibitor or something like that, but we don't expect that we would have to do that based on the preclinical data. And I think there's real precedent for this, I think, to try to sort of put this in the context of the field. What's most important for getting a useful therapeutic index is that you have ON mutant activity. It's less important that it be selective for the mutant. And I think there's plenty of precedent for that in the field going all the way back to Tarceva, osimertinib and so on. So I think that we are getting activity out of 6236 because of its direct effect on the RAS tumor driver, and there does seem to be a differential effect with regard to nontumor tissues. Now I also mentioned some of the specific reasons for that. I mentioned RAS addiction is one factor, without a doubt. And I think the other factor is this very interesting distribution effect that seems to load up the compound into cause of depot in tumor cells based on high cyclophilin A expression. There are other factors beyond those 2, and I think I alluded to the fact that we'll lay out data to support those in the coming months. Steve Kelsey will be speaking at a RAS summit meeting in February, and I'm sure that he'll use some of his time to address that with even more data.

The -- I guess, how should we be thinking about the bio -- the oral bioavailability of these compounds and just sort of what the preliminary PK are suggesting in terms of their ability to be orally administered compounds? I didn't catch whether you spoke to that in the presentation.

I did very quickly. There's so much to cover. But yes, so we've, of course, tested oral administration across multiple species, and we think we're very much in the range that we would expect to translate into humans. I'm not going to give you human dose projections today, but we don't anticipate any issue. In fact, the oral bioavailability for these compounds is in the range of any other RAS inhibitor that's currently advancing in the clinic or in some cases, even superior. So I just don't think we have an issue or concern about that. Obviously, we have to test it in humans. You just don't know until you administer. But there's nothing about the data we've seen that would suggest we should have any particular concern about these compounds versus any other compound going into oral administration.

I'm sure you're not sort of going to, like, kick back and just -- with just these 2 assets. I am curious with 6236, whether the sort of the ON wild-type RAS activity is a -- is it a liability that sort of couldn't get dialed up, where you couldn't get -- sorry, dial up that activity and still address G12D and G12V? Or is there an advantage to having some of that ON wild-type RAS activity?

Yes, Well, sure. Yes, in the Silicon Valley parlance, is it a bug or is it a feature? And it's probably both in that, as I mentioned, if you have pure mutant selectivity or enough old mutant selectivity, like with RMC-6291, you're pretty sure that you're going to hit the mutant and not hit the wild type. That could be very good for any mutant-driven activity, but it's a liability for what is an increasingly apparent set of mechanisms that involve activation of other alleles be wild-type alleles of RAS that can essentially substitute or fill in when the mutant allele is suppressed. And that's under, obviously, a tremendous drug pressure. That's a very real phenomenon. And so that's a liability of a mutant selective inhibitor like the G12C inhibitors and so on. Flip that around, if we want to block that wild-type K, N or HRAS activity that can substitute, well, essentially, by definition, then you carry the risk for on-target toxicity. And we're really pleased that 6236 has such a good tolerability profile in the preclinical efficacy models that we've run, all of them. That's very encouraging. We don't know for sure what the therapeutic index is in people until we go study it. But as I mentioned, I think the evidence is pretty strong that multiple RAS pathway inhibitors that hit their mutant target but also hit the wild-type protein can experience -- can exhibit an attractive profile that can lead to antitumor benefits and tolerability. We'll just have to see, but we're quite excited about this. This is a very significant compound. It has the potential to treat nearly all, if not all, RAS-driven tumors. So the opportunity is really enormous. It's almost incalculable. We'll see how it behaves in humans. But most importantly, and I did mention this in my talk and would like to emphasize it, we don't want a hubris about this question. This is a really important question. The field has historically bounced back and forth between selectivity and breadth, at least in terms of tyrosine kinases. We've all lived through in decades of that debate. The best way to answer that question is going to be to test it and to test it in human disease. And we have a unique collection of compounds that will allow us to test it, and the answer may differ for different genotypes and even different histotypes within a genotype. The role of wild-type RAS proteins may vary across different patient segments, and we'd like to be the ones to figure that out if we can and to make sure that we have the best product opportunities -- best treatment opportunities for those patients. So we are going to lead the field. We're going to do those studies. We're going to continue to advance our assets until we learn something that suggests we don't need to advance anything further. We'll keep an eye on the data, but we're very committed to this, and we have a highly productive innovation engine that's going to keep putting these out until Steve Kelsey decides that we put out what we need to put out.

That's great. Well, I think we're coming up for time here. So Mark, I want to thank you again for your time this afternoon. And thanks, everyone, for tuning into the webcast. Have a great evening.

Thank you very much.