Good day, and thank you for standing by. Welcome to the Revolution Medicines Q4 Year-End 2022 Earnings Conference Call. At this time, all participants are in a listen-only mode. After the speakers' presentation, there will be a question-and-answer session. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, Peg Horn, Chief Operating Officer. Please go ahead.

Thank you, and welcome, everyone, to the fourth quarter and full-year 2022 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 of Research and Development; and Jack Anders, our Chief Financial Officer. 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 constitutes 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 or update 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. During this presentation, we will be referring to a number of slides from our corporate presentation. The entire presentation was posted to our website immediately prior to this call. With that, I will turn the call over to Dr. Mark Goldsmith, Revolution Medicines Chairman and Chief Executive Officer. Mark?

Good afternoon, and thank you for joining us. Today, I'll provide an update on our company progress. Steve Kelsey will provide additional information about our RMC-6236 clinical program, and Jack Anders will provide highlights of our financial results. Revolution Medicines is advancing our pipeline of groundbreaking RAS(ON) Inhibitors and RAS Companion Inhibitors on behalf of patients with a wide range of RAS-addicted cancers, setting up an exciting and data-rich year. We have advanced into clinical development, the first two drug candidates from this highly innovative RAS(ON) inhibitor portfolio. RMC-6236 is a groundbreaking RAS inhibitor with a potential to treat all or nearly all RAS cancer patients due to its highly differentiated mechanism of action facilitating broad activity across major RAS variants. RMC-6291 is the vanguard of our mutant selective RAS(ON) inhibitor portfolio, which also includes our oral and covalent KRASG12D inhibitor expected to enter the clinic mid-year and a second wave of inhibitors designed to treat a range of RAS-mutant cancers. Together, the two Phase I programs will provide key insights into the potential of each exciting drug candidate and initial information to validate our tri-complex RAS(ON) inhibitor platform as a whole. RMC-6236 and 6291 have been well-behaved so far in the dose escalation portions of the RMC-6236-001 and RMC-6291-001 clinical studies, respectively. Both compounds have exhibited oral bioavailability, leading to increasing exposure levels with increasing dose consistent with our preclinical projections. They have both been generally well tolerated. We've cleared several dose levels for each compound with a once daily dosing schedule and we have not yet reached a maximal tolerated dose or defined a recommended Phase II dose for either compound. Today, we will expand further on the findings related to RMC-6236 in particularly. Momentarily, Steve Kelsey will share with you additional information coming from the RMC-6236-001 trial.…

Thank you, Mark. As we have stated previously based on extensive preclinical studies of RMC-6236, our working assumption is that the on-target effects of inhibiting wild-type RAS in normal tissues will ultimately determine the maximum tolerated dose in people. We have consistently represented that we believe RAS-mediated toxicities will be predictable, manageable, monitorable, and reversible. We have also presented preclinical data suggesting that the slower clearance of RMC-6236 from tumors compared with normal tissues may enhance the therapeutic index for RMC-6236 and contribute to achieving meaningful anti-tumor activity for tolerated exposures. Meaningful clinical activity includes both reduction in tumor size and durable inhibition of tumor growth, represented most frequently as progression-free survival in clinical trials. In the clinical program to date, patients have been treated in five dose cohorts ranging from 10 milligrams daily to 120 milligrams daily. To calibrate you, based on measured drug exposures in these patients, the 10 and 20 milligram doses fall at the lowest end of exposures we evaluated pre-clinically and were associated with some degree of tumor growth inhibition in xenograft models. Drug exposures seen in patients treated at 40, 80 and 120 milligrams daily are similar to those associated with more significant anti-tumor effects in preclinical studies. These included dose-dependent tumor regressions and delay in time to tumor growth in in vivo studies of several RAS-mutant cancer models. But with these doses in patients, we haven't yet reached the exposure levels achieved at the dose we studied the most pre-clinically, that is 25 milligrams per kilogram daily. 36 patients have been evaluated for initial safety and tolerability in the clinical trial so far. As shown in the table of drug-related adverse events on Slide 11, treatment across all dose cohorts has been generally well-tolerated. Some patients have exhibited predicted on-target normal tissue effects,…

Thank you, Steve. The collection of data just presented is, to our knowledge, the first-ever clinical data to be presented on any inhibitor designed to target the (ON) state of RAS, any RAS-targeted therapy used to treat a patient with a tumor bearing the KRASG12D mutation or any oral RAS(ON) inhibitor used to treat a patient with lung or pancreatic cancer bearing KRASG12D or various other non-G12C mutations. While initial data from the patients described today are quite encouraging overall in terms of tolerability and anti-tumor activity in multiple tumor types with multiple RAS genotypes and the case report shows a dramatic effect in KRASG12D pancreatic cancer at a tolerated dose, we will need additional data to project response rates or durability in pancreatic, lung or other cancers and across different RAS mutations at the recommended Phase II dose and schedule. We show these early data today to illustrate what maybe possible with a RAS(ON) inhibitor as represented by the boldest compound in our collection with arguably the greatest priority therapeutic index risk profile, and we will continue to develop more clinical information. These findings are quite encouraging about the ability of a rationally designed tri-complex RAS(ON) inhibitor to be taken orally, for it to exhibit drug-like pharmacokinetics and for the tri-complex mechanism of action to drive inhibition of RAS(ON) proteins and confer promising anti-tumor benefit in patients without being accompanied by extraordinary toxicities. We confirm our intention to provide a more detailed public update on 6236 midyear, including further follow-up from patients described today and information we expect to collect from additional patients in the dose escalation phase of the study. Let's briefly touch on RMC-6291, our mutant-selective KRASG12C(ON) inhibitor, for which we also have some early experience that allows us to build on the earlier comments about general…

Thank you, Mark. As shown on Slide 34, we ended the year with $645 million in cash and investments, which is expected to fund planned operations through 2024 based on our current operating plan. Revenue from our collaboration agreement on SHP2 inhibitors with Sanofi was $15.3 million in the fourth quarter of 2022 compared to $9.5 million in the fourth quarter of 2021. The increase in revenue was due to a $7.6 million non-cash adjustment related to the acceleration of revenue resulting from the termination of the Sanofi agreement, which has an effective date in June 2023. As a result of the termination of the agreement, we adjusted our estimates of the accounting transaction price an estimated percentage of completion of work performed to date, which resulted in a cumulative catch-up adjustment that increased collaboration revenue in the quarter. Collaboration revenue for full-year 2022 was $35.4 million. Total operating expenses for the fourth quarter of 2022 increased to $77 million largely driven by R&D expenses, which totaled $66.1 million. Total operating expenses for full-year 2022 increased to $293.7 million with R&D expenses increasing to $253.1 million. The increase in total operating expenses in 2022 was primarily due to the advancement of RMC-6236 and RMC-6291 in two clinical trials as well as an increase in personnel-related expenses related to additional headcount, an increase in research expenses associated with the company's preclinical research portfolio and an increase in stock-based compensation. Net loss for the fourth quarter of 2022 was $56.5 million or $0.63 per share. For full-year 2022, net loss was $248.7 million or $3.08 per share. Turning to financial guidance for 2023. We expect full-year GAAP net loss to be between $335 million and $365 million, which includes estimated non-cash stock-based compensation expense of $40 million to $50 million. The increase in expected GAAP net loss for 2023 is a result of increased expenses associated with the advancement of our RAS(ON) portfolio and a decrease in expected collaboration revenue. And with that, I'll now turn the call back over to Mark.

Thank you, Jack. We are highly encouraged with the progress of our rich development pipeline, particularly by the body of clinical evidence developed so far that provides initial validation of our RAS(ON) inhibitor platform and as described here, promising early evidence that RMC-6236 can be dosed in patients to induce significant anti-tumor activity without unacceptable side effects. We look forward to sharing further information on our collection of exciting clinical and preclinical assets throughout the year. We deeply appreciate the support of our patients, clinical investigators, scientific and business collaborators, advisers and shareholders and, of course, the tireless efforts of RevMed employees in pursuit of our mission to outsmart RAS-addicted cancer. This concludes our prepared remarks for today. And I'll now turn the call over to the operator for the question-and-answer session.

[Operator Instructions] Our first question comes from Jonathan Chang with SVB Securities. Your line is now open.

Hi guys. This is Faisal Khurshid on for Jonathan. I wanted to ask, your report on those 12 patients who were treated at 40 mgs and above with lung cancer and pancreatic cancer, what's the denominator on that number? Like how many patients were treated there, as in like how many were excluded for being non-evaluable or how many were excluded for having another tumor type?

Steve, do you want to comment on it?

Sure. We didn't report on the patients treated at 10 and 20 basically because there’s really was no evidence from the preclinical models that we were going to be anywhere near the exposures required to see anything worth reporting, to be honest with you. The actual data is restricted largely to reductions in ctDNA. And we haven't reported on the colorectal cancer patients at all because it's simply a different disease, and there is no precedent right now for a RAS inhibitor being effective as a single agent in BRAF mutant colorectal cancer. So it is quite likely that we will compile information on the colorectal cancer patients as it reaches critical mass or the view to putting together a more comprehensive strategy for the treatment of patients with BRAF mutant colorectal cancer RMC-6236. So essentially, the data that we're showing you right now is the data that we're showing you, and all of the rest of the data will be in the more formal update in the middle of the year.

[Operator Instructions] Our next question comes from Michael Schmidt with Guggenheim Partners. Your line is now open.

Hey, guys. Thanks for taking my questions. So it sounds like the preclinical data has been pre-predictive as it relates to dosing ranges and some of the dose response, et cetera. Based on that, I guess, Mark, what is your prediction what the MTD might be in this study? And how much more room to increase dose do you think you'll have on the Phase I study here?

Yes. Hi Michael, thanks for your question. I think we just don't know how steep the curve will be in humans. I think that's the point of the experiment right now. So we're clearly below what's likely to be a recommended Phase II dose. But I think my sense is from all the discussions we've had internally that it's the sort of thing that you'll know it when you see it. I don't know if Steve wants to add any more color to it. But I would just want to confirm, I think that the preclinical data have been very predictive of what we've seen in humans. And we try to convey that sort of in a soft way back in January, but now we're providing concrete data to support that statement. Steve, any more...

No. I mean other than that, as you can see from the adverse event table that we've included the tolerability profile right now, even at 120 milligrams daily, it's pretty benign. I think Mark is correct in that at some point, we are going to be hitting wild-type RAS in all tissues at a point which will make the compound or the dose intolerable. But it's really hard to predict how high we can go there. Just to be clear, I mean the – we dosed mice at 25 milligrams per kilogram daily routinely. And that's because if we increase to 40 milligrams kilogram daily, we saw some intolerability. Right now, we're dosing way below the varying equivalent of 25 mgs per kilogram daily. So we think that's a reasonable amount of headroom left to go, but it's very difficult to say exactly how much.

Yes. But it sounds like you have some additional upside on the dosing. And then, I guess, when we think about read through of this data to your mutant-selective KRAS inhibitors where, I guess, one would expect less on-target toxicities, how should we think about that in terms of the dosing ranges? And I guess, how much wider would one expect the therapeutic index to be as it relates to that?

Yes. Well, we fully expect the therapeutic index to be wider with the new selective inhibitors or by definition, if they're – if it's not wider, if they're having wild-type RAS effects, then they're not mutant selective. So there will be a wider therapeutic index. I think as we talked about before, I mean, ultimately, the issue isn't therapeutic index. It's how high can you dose and what's the absolute impact at that dose level. But it will all converge around 1.0, as you have talked about before, because that's the point of MTD dosing. But I think just to go back to the big picture, we're obviously quite encouraged by a significant level of anti-tumor activity that we've seen at as Steve described a very well-tolerated dose range. And although this is below the 25 mgs per kg that we used as our standard dosing in mice, it's not below dose levels that we know should have anti-tumor activity. So it's actually quite consistent with our collection of preclinical data. So we're quite encouraged about it. But I think the mutant-selective inhibitors, one would expect to be – have much more flexibility around dosing. And we think of RMC-6236 as sort of the highest-risk molecule from that perspective. And today's information, in our view, substantially lowers our calculation around that risk profile.

Yes. Great. And then could you – when I was looking at the waterfall chart on Slide 12, could you just comment on response kinetics? Did they all – did you see these responses early at the first scan? Did they happen later? Did they deepen over time? How should we think about that?

Yes. Well, there is a row across the bottom that says most recent scan, and that was designed to give you the information of what the bar represents. And you'll see that it varies. Some of them are cycle three, day one, which would be the first post-treatment scan, and others are at cycle five day one. I think the case that Steve described is pretty informative there where from the first scan, which would have been cycle three, day one, it was an aggregate 17% tumor lung reduction by RECIST criteria. Although the scan looked a little more impressive than that, that's what it was. But by cycle five day one, it's pretty clear that the tumor is nearly obliterated across all three of those lesions. And so that score is RECIST in the 70% range, just probably in terms of tumor content is far more than a 70% reduction. That's what we have. That's what we're seeing so far. As we get to higher doses, possibly, it will take fewer cycles to see that. Who knows? But it might also just be the kinetics of the tumors. That's the sort of thing we need more information on to be able to give a more precise answer.

Right. And just in terms of the study conduct going forward, do you plan to enrich for certain tumor types, perhaps lung cancer and PDAC? Or are you still continuing with the all-comers approach at this point?

Well, I think as we've described it publicly and I think we'll stay with this is that the backfill patients really give us the greatest opportunity to put finger on a scale, if you will. And clearly, we're showing some information that we think is exciting and the investigators think is exciting. So I would imagine you'll see more of some of these patients in the backfills. There's always this tension, and I know you and I have also talked about this before. There's a tension between wanting to sample across tumor types and genotypes because that's important information too versus wanting to get a deeper data set in a smaller number of tumor types of genotypes. I think in a sense, it's sort of playing out that way we could get immunology with predicted. So we end up with a larger data set with G12D and a pancreatic cancer, not surprisingly. But we'll continue enrolling patients across these potentially with some enrichment by the investigators.

Yes. All right. That makes sense. All right. Well, thank you for taking my questions and congrats on the data update today.

Thank you.

[Operator Instructions] Our next question comes from Marc Frahm with Cowen. Your line is now open.

Hi. Thanks for taking the questions and congrats on the data and seeing that 6236 is active in the clinic. Maybe just a follow-up on one of the earlier questions to start. Steve, I think there's 20 patients dosed at 40 milligrams and above in the AE table and then there's 12 in the waterfall plot. Can you just explain those incremental patients? It sounds like maybe a few of them are colorectal cancer patients that you just don't have enough to talk about monotherapy with. And then are there more patients who are just on study, but haven't been scanned yet versus people who may have dropped out before the first scan?

Yes. There's far more patients on study that have not yet been evaluated for efficacy than there are patients with a colorectal counseling study. I was a little – I was deliberately vague about the other histotypes, but that's because we will provide an update later in the year, which addresses that. But the reality is that the discrepancy is due to the speed at which the study is enrolling and the fact that patients have to be on study for six weeks before they get their first scan. So if you would model a whole bunch of patients within the last six weeks, they are evaluable for safety, but not for efficacy.

Yes. Well, and that's the key, Marc, is that they're on the table for adverse events or they're considered as part of that calculation once they start dosing. But they need to be on study for six weeks before you get a scan.

Okay. That's helpful. And then maybe just to the point of the preclinical modeling being pretty predictive. I guess, one, given the exposures you're seeing so far, I guess, what dose level do you think might be equivalent to that 25 milligrams you tested mostly? And then also kind of related to that, is there any sign pre-clinically that G12D alterations might be maybe modestly more sensitive than the other alterations and makes sense that they would start responding first? Or do you think this is just kind of the noise in the system of seeing a lot of different tumor types and a lot of different mutations?

Yes. On the second question, not so much. I mean, when we – I think we shared data on the G12X versus non-G12X in various presentations, not necessarily corporate, but at some of the scientific meetings we've given a further breakdown. And without any doubt, the G12X as a group showed enrichment for higher sensitivity on average. But then you start to get in fairly small numbers, trying to compare D to B to A to F. In some cases, we only have a couple of models of one particular genotype right now there. So I don't think we can say that today, and I don't think we would conclude that D is more sensitive in the waterfall. It's just not enough. Nine out of the – sorry, eight out of 12 are D in the waterfall plot and three are B and one is A. So it's just really no way to determine sensitivity for that yet. We need to collect more information. Your first question was…

What place will we get to? What dose will we get to...?

Yes. Or how...

What if – given the exposures you've gained – you've got a fair amount of clinical exposure data now to start building the kind of the human model, not just the extrapolation from animals. But based on that, kind of where does that 25-milligram animal dose fall on? Is that 240? Is it 400 milligrams?

Higher than 120, I think, is what we shared today. We'll give you that information. It's not 10 miles away. But I think we'd rather give that to you in the context of showing PK curves and by then, we'll have more information. It just seems potentially misleading to sort of give you a projection that's not showing you the data to go with it. So I think we'll stay with where we are.

Okay. That's very helpful. And then I completely understand your points in your prepared remarks about it's pretty early – it's too early to be talking about specific response rates and efficacy in individual tumor types or genotypes. Do you think you'll be in a position to start saying that for some tumor types and genotypes in the middle of the year at that update?

We hope so. I mean that's the ambition, that's the hope. Of course, it's a matter of opinion. But adjusting 10 out of 12 patients at 40 milligrams above showing some degree of tumor volume reduction with some clear evidence that more time on drug gives you a greater chance that actually hitting up your threshold is quite encouraging. But I just – it's hard to say. And I feel like anything we say here probably will not be rewarded with anything positive. So in the spirit of don’t break in it – not to break in to jail, I think we'll just stay with where we are.

Okay. Fair enough. Thank you, and congrats again.

[Operator Instructions] Our next question comes from Eric Joseph with JPMorgan. Your line is now open.

Hi, good evening. Thanks for taking the questions. On the 20-milligram cohort with it being larger than the others, I just want to confirm that that's a result of backfilling and that there wasn't necessarily some kind of a safety event that triggered further expansion of that cohort? And then maybe just secondly, can you just sort of state where you are right now in terms of further enrollment of the 120-milligram cohort? And if you look – if you do, in fact, proceed through further dose escalation, would you kind of move through 40-milligram increments as well?

So Steve, I guess the first question…

Eric, it's a combination of two things. One is there were a few patients that were backfilled after the escalation to 40 milligram. But we also built in a food effect study at that dose level, and so six of the patients were in a food effect study. We don't have the results yet from the food effect study, but the patients have been enrolled at that dose level and have been evaluated for safety, which is why the denominator for that dose level is somewhat larger. The 100-milligram dose level is fully enrolled, and it will read out. We'll know shortly whether or not we can dose-escalate to the next dose level. And that dose level, by the way, is determined by conversation between us as the sponsor and the investigators. So I'm not really in a position at the moment to tell you what that dose level will be. But it will be some sort of similar increment, 160 or 180, something in that range.

Okay. Got it. And I guess, if we're just trying to prepare ourselves for the scope of the readout that you're planning for mid-year, I guess, any sort of rough guidance in terms of additional patients you would expect to have accrued to the trial? And I guess, additional follow-up that you'd want to have for the evaluable patients that you're describing here today?

Well, there's clearly going to be additional follow-up on all these patients. I mean you'll get more information on whether or not – well, firstly, there's a bunch of patients that have been enrolled that haven't yet been evaluated for efficacy. So we'll have that. There will be more durability data, hopefully, to the patients that are currently on study. As Mark said, the investigators are beginning to lean into the selection of patients for this study, and it's heavily biased towards non-small cell lung cancer and pancreatic cancer. And the epidemiology of that means that around half of them are going to have a G12D mutation, and probably a third of them are going to have a G12B mutation, and the rest will be a scattering of other things. So I would expect there to be a substantially larger database predominantly focused on pancreatic cancer and lung cancer with G12D and G12B being the major mutational forms that we report on.

Yes. I would just say, Eric, we have a discussion about when we do this midyear update. Midyear isn't the date or a time. It's just kind of the time of year. It's a season. And so obviously, we're giving you some information now which wasn't necessarily contemplated previously. So we're going to cut some more information. We're going to try to make the next update a meaningful update incremental to this, but there's no sort of prescribed formula, which is why asking us more about what we'll have. We'll have more patients and we'll have more follow-up. That's for sure. So...

Mark, with that comment, I take it that you're not necessarily targeting presentation at one of the other major scientific meeting? Perhaps there's some expectation that you guys will present at ASCO?

Well, as I pointed out previously, we could not link it specifically to a scientific meeting or not. We've left that unspoken. If we do a scientific meeting, we may still have a corporate session. And if we don't do a scientific meeting, we'll probably have the corporate session and then do a scientific meeting after that. So sort of all of the above are in play here.

Okay, great. Understood. Thanks for taking the questions, guys. Congratulations.

[Operator Instructions] Our next question comes from Chris Shibutani with Goldman Sachs. Your line is now open.

Hi, everyone. This is Charlie on for Chris. Thank you so much for taking our questions, and congratulations on the data thus far. Just real quick from us. Regarding the 120 mg dose cohort, just wondering if you can give us a sense of how long these patients have been at this dose level and being dosed at this level? I understand that it's probably at a lower – on the lower end of the range of duration of therapy that you guys gave us some color on there. But just trying to get a sense of how long they've been on drug. And also if you could give us a sense of the kinetics with the appearance of adverse events such as rash. Like how long does it typically take for the rash to show up? And is there the potential for the severity of rash to increase over time in subsequent cycles? Thank you very much.

Let me address the second question first because it's a more concrete question for which we actually have data. The duration – the time to onset of rash is between one week and two weeks into dosing, depending on the dose being used. So the higher doses, it tends to start sooner but not usually before one week of dosing has been completed. The severity does not increase over time. In fact, if anything, either stays the same or improves. We have had reports of patients whose rash has got better with just persistent dosing. And we have reports of patients whose rash remains unchanged and is easily manageable with cream, just like putting up with local treatment. These rashes are very much like EGFR associates rashes in that they don't – they occur in different parts of the body in different people. Sometimes it's just the face, sometimes it's just the arms or the trunk. But nobody seems to think that it's terribly debilitating. We've actually only had one patient who had to stop for a couple of days and then restart. And that wasn't actually because of rash. It turns out it was because of an associated urticarial itch that was associated with it that seem to accompany the rash. So I think it's – that's pretty much all I can tell you at the moment.

And then in terms of the duration of 120 milligrams, I think the only thing we can fully tell you is that the three patients who are on that waterfall plot, you can see that the bar that's shown there is associated with the first scan.

Got it. Thank you so much. And if I could just sneak one – yes, understood. That makes sense. Thank you. And if I could just sneak in one more. Just based on the safety profile that you're seeing thus far, are you still considering the potential for intermittent dosing of the future? Or has this given you more confidence for the potential for just the once daily? Thank you so much for taking our questions.

Yes. Thank you, Charlie. Appreciate it. I think we're very much where we were in January when we commented on this that we're feeling quite encouraged by the once-daily dosing. All the data we're showing you here is based on once-daily dosing. We've gotten away with it quite well with very good tolerability and safety profile and clear evidence of antitumor activity. So it's not looking to us as if intermittent dosing is on a critical path to defining our first recommended Phase II dose schedule for monotherapy. With that said, we also indicated in January, and I think we'll stand by this, we're likely to test an intermittent dosing schedule. It's not our highest priority at the moment. And to the extent that we can move forward with monotherapy on a daily basis, we want to do that, but we would want to determine at some point whether we're leaving anything on the table in terms of potential clinical benefit. But so far, it doesn't appear to be so.

[Operator Instructions] Our next question comes from Ben Burnett with Stifel. Your line is now open.

Hey. Thank you very much. I wanted to ask about the patient that you just mentioned, so the one patient that stopped treatment due to an itch, but then it was restarted. Can you comment if – were they restarted at the original dose? And also, how long do they pause dosing?

I think I'm just going with memory with that, so forgive me is what I say subsequently turns actually slightly inaccurate. But my recollection is it was only for a couple of days that they actually stopped dosing. And then when they restarted, they started at a dose level below the one that they had originally been on. And that patient remains on. So...

Okay. Okay. That's helpful. And then just going back to – so can you comment on what the median follow-up time is in the adverse event table overall?

The median duration of – it's really short. I mean the median is really short. I would have thought right – again, it's – we haven't actually fallen and calculated it, but I would be surprised if it was in excess of a couple of months.

Okay.

The record we have calculated it.

Yes. No, we haven’t no, we really haven’t calculated, but it's really short. And I think the data that we're showing you is deliberately a very preliminary snapshot of what we're seeing and not something that we could necessarily claim will persist over time.

That's helpful. Okay.

And also say that in the 40- to 120-milligram group with pancreatic or lung cancer, they're all still on so.

Yes. No, all the patients that have been evaluated for efficacy are all on study, the lung, pancreatic cancer patients that we evaluated are still all on study.

Okay. That's great. And then I also wanted to ask, going back to that intermittent dosing conversation. Is there – can you frame the level of toxicity that's tolerable? And is it likely to be different in different indications? Like could there be a scenario wherein you may have intermittent dosing for CRC or for certain tumor types but not others?

Not so much because of difference is the tolerability, but there could be differences in sensitivity across tumor types. And so if it's somehow – I think that was really the point about not wanting to leave activity on the table. I just realized the metaphor is you want to leave it all out on the field that you don't want to leave anything on the table to make those two metaphors. So one can imagine that there may be some benefits in some settings too if you can achieve a higher dose intensity for particular tumor types or there's a particular combination in which that's particularly logical to do. But I mean we're out really of our [indiscernible] for third metaphor to say anything like that at the moment because it should not what we're experiencing. And by the way, pre-clinically, as you know, the vast majority of what we showed was daily dosing, admittedly in a mouse with different pharmacokinetics, but nonetheless. We had target coverage for 24 hours a day, and we were able to do it. So this – what's happening in human seems to reflect that, at least within the limits of what we can detect so far.

Interesting. Okay. Thank you.

And I think it does contrast. Maybe I'll add one other bit of color here since we're amongst the groups that I spent a year setting SHP2 inhibitor. It does contract significantly with the SHP2 inhibitor experience, where in humans really to get the concept overall dose intensity that we felt would be needed, it was mandatory, in our view, intermittent dosing. And pre-clinically, the anti-tumor activity even at the best optimized SHP2 inhibitor schedule and dose just didn't – wasn't in the same league as for RMC-6236. There just seems to be no question that inhibiting RAS is far different from inhibiting something upstream from the downstream.

I see. Okay. So it was like the guiding light here is that just – is it really like the PK? And are you achieving the concentration thresholds that are needed? Or is it really about efficacy?

I think the guiding light here is how much efficacy can we get at a tolerated dose. And so far, we're encouraged by the efficacy signal we're seeing so far at the tolerated doses. I mean the PK is not going to make a decision for us. It's going to be the clinical outcomes.

Yes. Okay. Thanks so much. I appreciate it.

Thank you.

[Operator Instructions] Our next question comes from Alec Stranahan with Bank of America. Your line is now open.

Hey, guys. Thanks for taking our questions. Congrats from me as well on your initial responses. Interesting that the two PRs are G12D. Could you give us a sense of the G12X makeup for the other 10 patients in the efficacy group? Was it entirely G12D and G12C or something else?

Well, there's no G12C patients in this study. But right now, those patients are all being channeled into KRASG12C inhibitors, so the RMC-6291. The patients in the group overall really cover the full range of KRASG12 mutations, and the ones on the waterfall are indicated. So you can see that there are really only three types: G12V, G12D and G12A. If I'm counting correctly here, there are 1 2, 3, 4, 5, 6, 8 of the 12 are G12D mutations. So that just, I think, is completely consistent with the epidemiology of the KRAS mutations in these particular histotypes.

Got it. That's helpful. And just a follow-up, thinking through the balance of safety, efficacy, you said your goal is to maximize responses. But would this be through the lens of keeping this forward as a monotherapy? Or would you take a hit on efficacy at a lower dose if safety is good with an eye towards combinations, since I know you've shown in the past some preclinical data combining 6236 with anti-PD-1.

Yes. That's a great question. But I don't think we're there yet. I just don't think that we know enough about the tolerability profile or what's going to be dose limiting to know whether – either to know at what level of toxicity we would be prepared to tolerate as a single agent or in due in combination with anything. So it's really too early. It's a very reasonable question, but it's just one which we don't have an answer to right now. Because as you know, toxicity is not all created equal. I mean some toxicities are very well managed and easily tolerated, and some are absolutely horrible and unpleasant and are going to be clearly less limiting. We're just going to have to keep dose escalating and following these patients and see what happens. With a view to maximizing the dose for response, that's not actually the primary – it's not the primary driver here. The main driver is durability. We're looking for durability because there are only two approvable endpoints that are going to make people prescribe this drug. One of them is progression-free survival and the other is overall survival. A response rate per se without either of those two things is not particularly helpful for patients. And so we're really looking not just consumer shrinkage but the durability of that.

And if I could just add just one small point. On the combinations, we've shown data that 6236 can combine with, for example, 6291. Effectively, it goes lower than just that we typically use for monotherapy studies in the mouse. So we have a lot of flexibility in the combination realm. And so determining the recommended Phase II dose for monotherapy doesn't necessarily dictate the limits of what we do in combinations where we may lower the dose impact that ensure the combination study we will start with a lower dose.

Okay. That makes sense. And just on the duration versus ORR point, it sounds like the mid-year update, the median follow-up is just not going to be long enough to get to that duration question, and it's going to be more about just the gross responses, correct?

I can't answer that because we don't really know. But again, all 12 of these patients are still on study. And in pancreatic second-line cancer, one doesn't have to wait 18 months to find out whether or not you've got a drug that’s active compared to standard of care. So I think one has to really talk about individual indications in order to know what the reference, the benchmark is for that particular indication.

Okay. One last quick one, if I may. Just on the differences between 6236 and 9805 as it relates to the binding to G12D, whether it's covalent inhibition or another aspect that you think could result in different clinical profiles beyond just the therapeutic window that was mentioned before.

Right. Well, it has the differentiated feature that is selective for G12D it forms a covalent plus selective bond with the G12V of RAS and hence, it's basically irreversible. RMC-6236 is non-covalent and so – and it's reversible. So it's binding, will go up and down with wherever the plasma levels are, whether it's intracellular concentrations are at that minute in time. But the trade-off is you capture activity against a wider range of RAS proteins that might be part of the overall oncogene addiction for RAS tumor cell. So there are trade-offs between those. And I think it's just really yet to be determined which of those is more favorable. Obviously, overlaid on top of that is tolerability, how much dose can you give, et cetera. But I just think it's too early to be making projections about which is better. I tend to think that RMC-6236 maybe the mother of all RAS inhibitors, but for us to be effective in treating all RAS cancers, we're probably going to need a whole family of RAS(ON) inhibitors. And that's why we designed our portfolio the way we did.

Okay. Thanks, and congrats again on the progress.

Thank you.

[Operator Instructions] Our next question comes from Ami Fadia with Needham. Your line is now open.

Thank you for taking the question and the data. My first question is just around how to interpret the evolution of response with duration of treatment. If I'm reading the chart on Slide 12 correctly, some of the patients who are on the chart are cycle seven, day one on stable disease. But then there are – the last two patients in that chart which are cycle five, and they seem to have shown a response rate. I'm just trying to understand, so those clearly were on treatment longer. Is that a fair way to look at it? And overall, is it just too short of a duration for us to really interpret responses over time?

Well, we definitely need more data, and we need more time on study to collect those data. But you're right, so that lung cancer patient who shows cycle seven, day one. That means that patient has been on long enough to have likely a third scan. That's what cycle seven would be – cycle seven day one would be a third scan. So that patient, that 40 milligrams, if I'm interpreting this correctly, if we're talking about the same page, has stable disease with some tumor volume reduction, but not PR. We don't know if they will go on to shape of PR. We're not trying to imply that they will or won't. We don't know. We'll have to follow them. So I'll ask Steve comment on this.

No. I mean – I think that you – directionally, you are correct and inferring that whether a patient has a response by RECIST is a function of both dose and time on treatment. And clearly, the – what this chart is beginning to perhaps indicate is that responses at 40 milligram might be uncommon, which is why we dose escalated to three times that dose level where you are beginning to see RECIST responses if we wait long enough. And as Mark said earlier, we don't know whether there is a dose at which majority of responses we'll be seeing the first response evaluation or not. It's just not as – it's not clear to us whether increasing the dose changes with the assets of response. Again, I want to really – I want to reemphasize here that these two diseases are diseases where chemotherapy induces responses, but it doesn't last very long. And that's what we're trying to solve for. We're trying to solve for the very short duration of response or benefit the patients get with cytotoxic chemotherapy. The median time to progression for patients with pancreatic cancer that received prior therapy is 3.5 months, and we're trying to solve for that. So it's really about what – how long it takes the tumor to get to 30% or 31% reduction. It isn't really the point. The point is how long do they stay on set before they progress again, and we would like to push that beyond 3.5 months.

Yes. Okay. That's helpful. And just clarify for me, after how many cycles is it a patient to receive a scan? Is it every two cycles?

Every two cycles.

For the first six cycles and then it goes – I think it goes to every three months.

Yes. So cycle three day one means they've completed two cycles is the day after they've completed two cycles.

Got it. Okay. Just with regards to safety, you mentioned that you dosed up to 40 milligrams per kg in sort of the preclinical models before you started to see some safety. What type of safety emerged at that point? And what was that – and then maybe what duration of treatment?

Okay. So this is very species-specific. So mice, we just see weight loss. The tox species were mice and [indiscernible] monkeys. And obviously, the monkey is a little bit more representative of what happens in humans. And there, we saw very clearly wild-type RAS-mediated toxicity, predominantly gastrointestinal toxicity and some suppression of myeloid hemopoiesis. Now interestingly, in the human study so far, we have seen some GI toxicity. It's not – it's a little bit diverse in terms of the exactly which part of the GI track is affected and how that manifests itself. So sometimes, we've seen nausea, sometimes we've seen diarrhea. Sometimes we've seen loss of appetite, for instance. It's a constellation of gastrointestinal effects that almost – well, all of them have either been grade 1 or grade 2 and very easily manageable. But the predominant toxicity, the one that's coming through with the increased frequency is skin rash, which we were not able to detect in the preclinical toxicities largely because that we're limited by what we can detect in those species. But it is interesting, the one organ in which we showed – in our JPMorgan presentation, we show that it is the one organ where the clearance of the drug is slightly delayed compared with the other organs, nowhere near as much as tumor, by the way, but it is the clearance of the drug from skin is slower than from, for instance, colonic epithelia or other organs in the body. So it maybe the tissue kinetics of 6236 are contributing to the skin rash. And that makes it the pattern of toxicity a little different in humans compared to the preclinical species.

Understood. Okay. Please remind me – I don't know if you mentioned exactly what dose the patient was on, the patient that had to pause the dose and then restart at a lower dose?

I don't remember. I'll have to take that off-line. I'm sorry, I don't remember.

Okay.

I think we have time for one more question. Is that right?

Tremendous. Another speaker. I think there's one other person.

[Operator Instructions] Our last question comes from Jay Olson with Oppenheimer. Your line is now open.

Oh, hey, thank you for the update and for taking the questions. We had a couple on 6236. What are you expecting to be the minimally effective dose? Is there any saturation of target occupancy that could lead to a plateau of efficacy? And what dose would you see that plateau? And then since the AE table excluded events that occurred in fewer than four patients, can you just comment on what those AEs were and if any of them were grade 3 or 4? Thank you.

Well, it's easy to answer the second question, we see no grade 3 or grade 4 events other than that one bowel perforation that we described, which was a grade 4 AE and also a serious adverse event. But as we said, the bowel perforation was most likely due to the efficacy of the drug not due to the toxicity of the drug on the gastrointestinal epithelium.

Sort of question about MAD...

There is no plateau in our dose response. We continue to see increases in efficacy up to the point where the drug becomes intolerable due to inhibition of multi-RAS. So there is no expectation of a plateau as we continue to dose-escalate. And this makes – this is really important because this compound so far, everything we know about RMC-6236 does not follow the paradigm for Project Optimus. We continue to see increase in efficacy as we see increase in toxicity. They are absolutely related to each other, but we are seeing efficacy at doses that are tolerable. And that is incredibly important both for the platform and for the further development of the compound.

Okay, great. Thanks for taking the questions.

Thank you.

I would now like to turn the conference back to Mark Goldsmith, Chief Executive Officer, for closing remarks.

Thank you, operator. 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.