All right, let's go ahead and get started. Welcome, everyone, to the 44th Annual JPMorgan Healthcare Conference. My name's Anupam Rama. I am one of the SMI-Cap biotech analysts here at JPMorgan. I'm joined by my squad: Ratih Pinhe, Priyanka Grover, and Joyce Zhao. Our next presenting company is Erasca, and presenting on behalf of the company, we have CEO Jonathan Lim.
Thank you, Anupam. Good afternoon, everyone. It's great to see all of you here. 2026 is off to a very exciting start, and I look forward to telling you more about what's in store for Erasca. Before I do so, I will be making forward-looking statements, so I encourage you to visit our website at erasca.com to find our latest SEC filings. At Erasca, our name is basically aligned around our mission to erase cancer and also to eradicate RAS-driven cancer, so it's a portmanteau with a double meaning. I started this company with Kevan Shokat, luminary from UCSF, who's really the father of G12C RAS targeting, and he and I co-founded this company in 2018 around a disruptive idea to target RAS. We've since then assembled a world-class scientific advisory board of leading pioneers. We're entirely focused on the RAS-MAPK kinase pathway.
We have a team with deep expertise in planning and execution of global development, and I'm very excited to share some recent data that we have for our lead pan-RAS program. This is ERAS-15, which is a pan-RAS molecular glue with best-in-class potential for various RAS solid tumors, and we also have ERAS-4001, a pan-KRAS program with first-in-class potential for KRAS solid tumors. So this is really our one-two punch for addressing RAS-driven tumors, and we're nearly entirely focused on those two programs. We're well-capitalized with $362 million in cash reported in our Q3 earnings and cash runway into the second half of 2028. I mentioned that Kevan is a pioneer in the industry. We also have Mike Varney as our chair of R&D, and we're really surrounding these individuals with amazing luminaries across the R&D spectrum in academia.
This is our pipeline, so I mentioned the bulk of our activities and focus are on ERAS-15 and ERAS-4001. We also have a program called ERAS-12. There's a bispecific antibody against EGFR that's in the discovery stage. So I'll focus this presentation on our RAS molecules. We've been in this space for quite a while, and I'll say that the ideal RAS-targeting molecules integrate three key attributes. So you want to have strong preclinical activity. You want to have strong PK performance in the form of oral bioavailability and clearance, and then you also want to have a strong proprietary space. At the center of that Venn diagram is the ideal RAS-targeting molecule, and we think we have two ideal RAS-targeting molecules.
Now, on the preclinical and PK front, you ideally would want to have a lower clinically active dose that could translate into lower risk of solubility, limited absorption, and the exposure plateau that's been observed with the most advanced pan-RAS molecular glue in development. You also could potentially have some upside in the form of better GI tolerability because of lower drug load and the effect on local tissues in the GI tract, and then ideally an improved therapeutic window. So ERAS-15 and ERAS-4001 exhibit very good profiles that exceed our target product profile along those three dimensions. So starting with ERAS-15, this is a highly potent molecule with sub-nanomolar to 13.8 nanomolar IC50 across multiple RAS mutations. In vivo, we've been able to observe very strong tumor regression at doses as low as 0.3 milligrams per kilogram, up to about 5 MPK.
The oral bioavailability, as measured by F% or the free fraction across multiple species, is very high, and the IP is strong with exclusivity expected until 2043 and beyond with patent term adjustment. We're pleased to announce U.S. patent covering composition of matter that was issued in the fourth quarter of last year. ERAS-4001 is also a very potent molecule with sub-nanomolar to single-digit nanomolar IC50. It does take slightly higher doses to see tumor regression, so between 30-300 MPK twice daily. This too has good oral bioavailability, which has been difficult for the class of pan-KRAS molecules, and this also coincidentally has strong IP exclusivity out to 2043 and beyond. The thesis behind a pan-KRAS molecule is potentially a wider therapeutic window because you are HRAS and NRAS wild-type sparing.
So by hitting KRAS really hard and sparing H and NRAS, you could potentially have better tolerability. Now, the question is, with ERAS-15, if we have a pan-RAS molecular glue that also has potentially a wide therapeutic window, then can you achieve with a pan-RAS molecular glue what you're trying to achieve with a pan-KRAS molecule? That thesis also is something that we're testing, and we think ultimately, if you could take these compounds and combine them, well, that would be a source of differentiation as well. So I'm going to start with a deep dive on ERAS-15. This is the potential best-in-class pan-RAS molecular glue, and I'll also update you on the clinical update that we shared earlier this week.
ERAS-15's potential differentiation really stems from the high binding affinity to cyclophilin A or CyPA, and so this works by a similar molecular glue mechanism where this has about 8- to 21-fold higher binding affinity to CyPA versus RMC-6236. And as a result of that, you just get many more of these bipartite moieties. So if you have 0015 with CyPA, that bipartite moiety, we call it a Pac-Man molecule, goes hunting for RAS to form a tripartite or ternary complex that then takes RAS out of circulation. And so by having the higher binding affinity, you just have a lot more, this is a nod to all of you who grew up in the '80s, but pan-RAS Pac-Man molecules that just take RAS out of circulation. Now, what does that mean? Well, it results in better potency across multiple cell lines.
You can see various degrees of fold potency versus RMC-6236 against different mutations of interest across the spectrum of KRAS and other drivers, and you can see sparing of RTK-altered cell lines, and importantly, in vivo, we're seeing ERAS-15 demonstrating comparable anti-tumor activity to RMC-6236 at just one-tenth of the dose, and so I'm going to show you on this slide a sensitive KRAS G12D pancreatic model where you could see it's called PK-59. RMC-6236 was able to achieve very good tumor regression with 3 MPK, and what's really impressive is that ERAS-15 was able to achieve the same degree of tumor regression with 0.3 MPK or one-tenth of the dose, so that just gives you a sense of how potent both of these molecules are, but RMC-6236 in this comparison requires 10 times the dose.
On this insensitive model, KRAS G12V non-small cell lung cancer, this is a model called NCI-H727. In the industry, among academia especially, this is widely viewed as a bellwether model of sorts because it's where you have, it's very difficult to see tumor regression with pan-KRAS molecules in this type of model, but you could see that RMC-6236 was able to achieve good tumor regression with 10 MPK, and also impressively, ERAS-15 was able to achieve that with just 1 MPK. So again, the one-tenth of the dose in an insensitive model. So we have a lot of other models that we've shared. They are in the public domain, so I encourage you to view our prior presentation. So we won't walk you through the half dozen or so other models, but importantly, from a PK perspective, the kinetics of ERAS-15 also look very promising.
So if you look at the tumor distribution and residence time, 15 does look like it has preferential tumor distribution and longer residence time. So if you focus on the graphs on the left, this is the same PK-59 model, that sensitive PDAC model that I just mentioned a couple slides prior. So if you look at the light blue bars for ERAS-15 versus RMC-6236, you could see that the levels of both drugs go down pretty substantially from 4 - 24 hours. But if you look at the medium to dark blue bars, you could see that 6236 diminishes or decreases over that 24-hour period, but ERAS-15 levels are sustainably high.
And so what that tells us is that, well, the reason for that is that there is CyPA that's overexpressed in multiple solid tumor types, and because of the higher binding affinity of ERAS-15 to CyPA, it's just lingering in the tumor and surrounding tissues a little longer. And so that tumor distribution and residence time, if that translates, well, certainly the PK kinetics look preferential in that case, but if that translates into a safety or efficacy advantage, then that'll be really cool to see. And then that's not just one model, but it's also seen in another model called PSN-1 on the right. From a PK perspective, ERAS-15 showed promising PK across multiple small and large animal species. So I'll draw your attention to the boxed areas.
Clearance-wise, ERAS-15 had lower clearance across the board relative to RMC-6236, longer half-life, and a higher oral bioavailability as expressed by F%. So I'm now going to switch gears to talk about AURORAS-1. This is the name of the first-in-human clinical trial that started dosing last year. The key eligibility criteria are patients with advanced RAS mutations where RAS is a driver in different solid tumors. They received prior therapy and had no previous treatment with a RAS inhibitor. So ERAS-15 was given as a single agent, and it was oral once-daily dosing or QD dosing.
So the goals of this single agent monotherapy trial are to show signs of clinical efficacy, ideally to show well-behaved PK characteristics, reasonable safety and tolerability, and to also inform combination therapy where, as you know, combination therapies really are standard of care, especially for the big three tumor types of colon cancer, pancreatic cancer, and non-small cell lung cancer. AURORAS-1 was designed to evaluate the potential for single agent differentiation and also whether ERAS-15 can be a potential best-in-class RAS-targeting molecule for combo development across multiple solid tumors. Now, I would say that the pace of enrollment, we were hoping that AURORAS-1 would enroll briskly, and we were pleasantly surprised by the pace of enrollment that has advanced even faster than we anticipated, which unfortunately underscores the significant unmet need.
There's still a lot of work to be done, but it does highlight that there's been very high investigator and patient enthusiasm, so just to summarize the preclinical differentiation, the thesis is really around the in vivo activity, so seeing comparable efficacy at one-tenth to one-fifth of the RMC-6236 dose in multiple mouse models, and then the PK properties in terms of higher BA, lower clearance, and longer half-life, and the preferential PK kinetics around the tumor distribution, and all of those stem from the cellular potency as well as the CyPA binding affinity. Now, we were very pleased to see that with early clinical data, the preclinical thesis does seem to be reinforced by what we're seeing with the efficacy, safety, tolerability, and PK data seen to date.
So we have seen ongoing confirmed and unconfirmed responses in multiple patients with differing tumor types and differing RAS mutations, with ongoing responses in the form of two confirmed and one unconfirmed partial response in patients, also with different tumor types and different RAS mutations at a low dose of eight milligrams QD, and additional ongoing responses in multiple patients at greater than eight milligrams QD. So I will say in terms of the CPRs versus UPRs, the reason why some responses are unconfirmed versus confirmed is that those patients have not been on study long enough to have confirmatory scans, so it's not a function of them falling off of treatment. In fact, I will say that all responding patients, as of the data cutoff, were still on treatment, and so that's a promising sign.
I can also say that we have been cleared since this press release. We have been cleared to start the 40 milligram dose cohort. So that just puts it in perspective that at a low dose of eight, there's still a lot of headroom when it comes to safety, PK, tolerability, and efficacy. So we're very pleased with that. The safety and tolerability profile has been favorable with no dose-limiting toxicity and predominantly low-grade AEs observed at all doses evaluated, and then well-behaved linear PK seen across all doses evaluated. In fact, there's been no evidence of the exposure plateau that was seen with the other molecule. So I know that Anupam's going to ask me about all the different doses that were evaluated in this trial. We will disclose all of that in the first half of this year.
Top-line safety, tolerability, PK, and initial efficacy data for dozens of patients are planned for this first half. But in the meantime, based on the early signs of activity as well as safety and tolerability and PK, we think 15 has the potential to become a preferred RAS-targeting backbone for combinations. This is a case study. One of the confirmed PRs was in a 70-year-old heavily pretreated patient with KRAS G12V non-small cell lung cancer treated at eight milligrams daily. You can see multiple lines of treatment, both with chemotherapy as well as checkpoint inhibitors. At the first restaging CT scan at cycle three, this patient had UPR, and that was confirmed with the first confirmatory scan at cycle five, and that was at eight milligrams, after which the patient was dose-escalated to 16. That patient continues on treatment as of the data cutoff.
You're looking at CT scans where the patients pretend the patient's feet are coming out towards you, so the left lung is on the right and the right lung is on the left, but everything in white means that the lung has either collapsed or is full of fluid, and so you could see in green that there's almost complete whiteout. There's basically complete whiteout of the right lung, and the left lung has heavily collapsed as well. After cycle three, you could see significant expansion and aeration of both lungs, and the patient, in fact, clinically was able to come off supplemental oxygen after just the first week of eight milligrams of treatment, so the patient is currently asymptomatic from disease and from treatment, and this is really why we're doing what we're doing, and it was just really exciting to see this patient doing this well.
So switching gears to ERAS-4001, this is the potential first-in-class pan-KRAS molecule. This has very high affinities and long residence times to KRAS G12D, G12V, and even G12C. It does hit KRAS wild-type, which is important because there is a propensity for KRAS wild-type-driven amplifications to be potential oncogenic drivers in different solid tumors. It is HRAS and NRAS wild-type sparing, which leads to the potential wide therapeutic window. So this is both a GTP and GDP-bound KRAS inhibitor, so it has single-digit nanomolar IC50 against both the GDP and the GTP states. And then we're seeing potent and selective inhibition of cells across G12X and G13D, as well as wild-type cell lines, which is nice to see.
In vivo, you could see a pancreatic model where at doses of 100 MPK BID, you get good tumor regression, and that's also seen across other tumor types and basically in vivo models at doses as low as 30 MPK BID. And in terms of that Bellwether model, I mentioned to you that pan-KRAS inhibitors have a more difficult time to see tumor regression in the H727 model. Well, it was really nice to see regression in this model with 4001, but it was at a dose of 300 MPK. And in terms of the PK, ERAS-4001 shows promising PK across multiple species, including mouse, rat, and dog. And I think of note, if you look at the bioavailability, it's really quite high across multiple species. So we have a number of key milestones this year and next.
ERAS-15 for AURORAS-1, I mentioned that in the first half of this year, we'll have a phase one monotherapy data set. And then in the second half of this year, we plan to initiate monotherapy expansion cohorts, as well as to initiate combination dose-escalation cohorts with readouts from those activities in 2027. And then for ERAS-4001, we're enrolling the BOREALIS-1 trial. We mentioned in our press release that this trial is enrolling as anticipated, and we are guiding to phase one monotherapy data in the second half of this year and to initiate monotherapy expansion cohorts as well as combination dose-escalation cohorts in 2027. So I just want to end by summarizing that we have an experienced team with a track record of serial success, world-class SAB, a promising pipeline that targets large underserved markets, really focused entirely on RAS-MAPK, and then clinical advancement of an industry-leading RAS-targeting franchise.
I think the fact that we sort of straddle both worlds of pan-RAS as well as pan-KRAS makes us very unique in the industry with multiple potential near and long-term value drivers and a strong capital base to be able to go after that. So thank you very much, and I think we will morph into the Q&A.
Thank you, Jonathan. I'll ask the first couple of questions, but I'll open it up to the audience, so feel free to raise your hand when I prompt you. I got to ask, man, what do you think about all these headlines that we've been seeing in the space related to you and your competitor?
Well.
Go.
I mean, I'll say that we respect Revolution Medicines as a pioneer in the space, and maybe I'll just echo what my colleagues did, which is, as a company policy, we don't comment on rumors, but I think it just speaks to the excitement around this space and just the fact that there's scarcity value for programs that can really move the needle in such a high unmet need.
Questions from the audience on this point?
Really? That's your only question? Come on.
So actually, now digging into ERAS-0015, you said in the first half, I think you said we'll get safety, PK, and maybe some efficacy. So maybe you could walk us through two things: dozens of patients, how much follow-up, the types of tumors you're going to be focused on, and then ultimately, when we do think about the early Revolution data, what are you looking for in that data on the key metrics?
Yeah, so we'll definitely have efficacy data, so it's not maybe, it's definitely, and we'll have safety, PK, and efficacy. I think what we look for on the safety tolerability front is for this class of compounds, and really we just have N- of- 1 right now in terms of what's publicly available, now N- of- 2 with our molecule. But I think you want to see good safety and tolerability, and we'll probably pay particular attention to sort of the rash as well as the GI AE profile. So I think that's going to be key areas of focus. I think the fact that we haven't seen DLTs and the fact that we've seen predominantly low-grade AEs across the board, that's promising.
The fact that we're able to open a 40 milligram dose cohort, if you just take this one-tenth framework to the natural conclusion, that would be the equivalent of 400 milligrams of RMC-6236 just in terms of sort of that potency PK argument from in vivo. So I think we're hopefully going to enter uncharted territory. Now, from a PK standpoint, the fact that we were able to say that we have linear PK observed at all doses evaluated, and this is not including the 40 because 40 is now open. The fact that we're not seeing the exposure plateau also is entering uncharted territory for the pan-RAS molecular glue class.
I'd say the fact that we saw first clinical responses, and it wasn't just one response, which might have been a fluke, but the fact that there were multiple responses across multiple tumor types and multiple RAS drivers. I think the fact that we saw that with eight milligrams, which is a really low dose. I mean, I haven't worked on a drug where single-digit milligrams has seen clinical responses like that. The first clinical response with RMC-6236 was seen with 80 milligrams, and so we were pleasantly surprised that at one-tenth of the dose, we were able to see first clinical response. So I would say that if you just look at the recommended dose across panc and lung for 6236, that range is 200milligrams - 300 milligrams.
At doses above 8, once we get into biologically relevant doses, I think we certainly can look at what the response rate looks like in the different tumor types. To answer your question, it'll be predominantly the big three tumor types. Technically, it's all comers, so you may see some patients that fall outside the big three, but it's really concentrated on the big three. Durability is tough because it really depends on which dose levels. Of course, the earlier dose levels, you'll just have more follow-up for them. If we're opening 40 now, I'm not going to give you a year of follow-up on 40 milligrams, but the doses in between will have various lengths of time of follow-up, and we'll just report the data what they are. And then you'll work your magic and interpret it.
Is there an ORR threshold you're looking for?
Nothing numerical because I think when we open up the expansion cohorts, then that's probably a more fair look at it. Now, we do have in our protocol the ability to backfill cohorts, but still, you just never know what you get during the dose-escalation phase. So I mean, I'm sure everyone will look at the numbers, but there's not sort of a threshold in our heads as to what we're expecting or guiding to. I think it'll be apples and oranges, and it really is the totality of the safety, tolerability, PK, and efficacy that we'll be looking at.
Questions from the audience?
So tell me if I'm getting too literal with the math, right? But if the RevMed dose is 200-300 and you're going with this 10x potency, you said you're starting a 40 milligram expansion cohort. So do you think that's like a testing ground, like a thesis tester for kind of higher doses relative to what they saw? And when would we have a decent enough mature data set for 40?
Yeah, that's a good question, so first of all, just one clarification. It's not an expansion cohort. It's actually an escalation cohort, so 40 milligrams is part of the ordinary course dose escalation, which follows a modified Fibonacci. And so I would say that, yeah, I mean, if you believe the one-tenth framework, then 20-30 would be sort of the imputed dose that would be equivalent to 200-300. If you believe the one-fifth, then it's more 40, so maybe reality lies somewhere in those bookends. I'd say if we're just starting 40 now, then it probably wouldn't be fair to evaluate durability in the first half, but later this year, that might be something that's more meaningful.
Additional questions from the audience?
Then I guess maybe a similar-ish question for ERAS-4001. Kind of what are you thinking about size and scope of the update in the second half of the year? And kind of in your mind, what's an encouraging update?
Yeah, so ERAS-4001, what we're guiding to there is that we'll have that date in the second half of this year. That will be similar in terms of what Jonathan described in terms of dozens of patients. We're going to be looking there at the safety, PK, as well as initial signs of activity. This is a very interesting class in the sense that for a while, there actually hadn't been much in the way of clinical data disclosures. Obviously, that's no longer the case, but we will have our data update in the second half of this year.
Question.
Hi there. Congrats on the progress. I'm just wondering if you could share some thoughts on how you envision the commercial space evolving. I know it's early days, but if you have a market leader that's going after specific mutations and then you're coming on as the follow-on, I'm just curious how you're looking at that and how that's in some ways coming back to your development strategy and how you're looking to take things forward.
Yeah, thank you for the question. I'd say in a world where, let's say that ERAS-15 just turns out to be a me-too compound in terms of no real clear differentiation, but it's just sort of second in class, still a pretty good place to be because you have such a massive unmet need. I'd say maybe certain lines of treatment within pancreatic would be a tough nut to crack, but everything outside of that is still pretty wide open for a second-in-class molecule. I mean, if you just look at history of the checkpoint inhibitor space and other areas where there's certainly room for more than one molecule, if not half a dozen to a dozen. And so with this, if we're comfortably in second in that first wave, then that's a good place to be.
Now, in a world where we show best-in-class or me-better potential, then everything's sort of wide open for us, and that would be a really exciting place to be, and it'll be great for patients, of course.
Questions?
Yeah, up here.
Thanks, Jonathan, for sharing. So it seems like the RMC-6236 is going to have early proofs next year. So is there a chance that FDA would require a head-to-head comparison for your molecule with the Revolution stuff?
It's possible in the U.S., yeah, for second-line pancreatic. I wouldn't rule that out, but I'd say that if you wanted to enroll a trial outside the U.S. and seek sort of alternative strategies, that's a possibility. Now, in a world where you have true differentiation, then a head-to-head wouldn't be as daunting as if it was just a me-too molecule.
Go ahead.
Very nice, encouraging early data for your pan-RAS. My question is around you have two portfolios, right? One is pan-RAS and pan-KRAS. And the pan-RAS, you show relatively good safety and tolerability, at least in early clinical data. So where do you see how you position these two assets if the RAS turned out to be also tolerable?
Right. Yeah, I think there's sort of swim lanes for each, and it really depends on what the data show for both molecules. I think if you believe the thesis around HRAS and NRAS wild-type sparing, then it's great to have a pan-KRAS molecule to do that. Now, if our pan-RAS molecule is able to be very well tolerated, then is there a world where you just kind of double down on that? Possibly, but it's still nice to have a pan-KRAS molecule. I think we're going to continue to be data-driven, and so we won't know until we see what 4001 is capable of, and really, that timing difference is because ERAS-15, AURORAS-1 started before BOREALIS-1, and enrollment has exceeded our expectations for that trial, whereas BOREALIS-1 has been performing. It's been enrolling like any typical phase 1 trial has. It just hasn't been enrolling as quickly as AURORAS-1.
That's really the timing delta.
Any final questions from the audience? I have one more. All right, go ahead.
Great progress. Just on ERAS-4001, it's all maxed out at 300 mgs per kilogram. What do you think is happening pharmacologically there? And can you say a little bit about what drug combinations you think RAS inhibitor will look most effective with?
Yeah, so in terms of combinations, I mean, it depends ultimately, going back to the previous question, what kind of combinations are we thinking about for the pan-RAS versus the pan-KRAS? And so you would think, for instance, about combining with standard of care agents by the tumor type. You can also think about potentially investigational agents as another wave of combinations for us to consider. So we're considering all of that, and we'll be data-driven in terms of how exactly we want to pursue that based on what we're seeing from the initial data that we generate as well as other data that are generated externally. Can you repeat the first part of your question? I don't think I followed that part.
With 4001, you used 300 milligrams per kilogram in the animal model, whereas you're using eight milligrams or starting eight milligrams in man. So can you just sort of talk about what you think is happening pharmacologically with that drug?
Oh, yeah. Okay. So I believe the one you're referring to is in the NCI-H727 model. That's the one that Jonathan mentioned is typically an insensitive model, particularly the pan-KRAS molecules. And so the fact that that molecule was actually even able to show regression, albeit at a high dose, but the fact that it was able to show regression is actually really encouraging for that molecule. So I think it just speaks to more of that model being a very tough model for a pan-KRAS.
And just broadly speaking, for both molecules, how do you think about unlocking sort of the combination potential? Because that's where you're going to ultimately go, right?
Yeah. I mean, I think if you need a wider therapeutic window for combinability, then a pan-KRAS molecule for certain indications might be the way to go. And then for certain cases, especially pancreatic and lung, where pan-RAS has been able to demonstrate combinability with both chemo as well as checkpoint inhibitors, so that could be the swim lane for pan-RAS.
Thank you, David and Jonathan.
Thank you.
Thank you.