Erasca, Inc. (ERAS) September 2026 Conference Transcript

Good afternoon, everyone. I'm Sean Lyman, head of U.S.

Mid-Cap Biotech Equity Research at Morgan Stanley, and welcome to the Morgan Stanley Global Healthcare Conference.

Before we commence, to make you aware of some certain disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com forward slash research disclosures. And if you have any questions, please reach out to your Morgan Stanley Research Sales Representative. For this session, we have the pleasure of hosting a RASC with Chairman and CEO Jonathan Lim and CFO and CBO David Chacko. Thank you for your time, gentlemen. We really appreciate it. Thanks, John.

Great to be here.

So we'll jump right in. So let's start with 0015. You've now shown the Aurorus 1 data across dose escalation with first responses at doses as low as 8 milligrams and a 50% UORR eight-week-in-12-plus KRAS G12X pancreatic cancer at the 32-milligram recommended dose per your July update, as well as with other data. Can you give us a view on what's most differentiating in the data set that you've shown so far and how you think about durability in data in the future?

Yeah, I think taking a step back, we're really excited about URAS 15. This is a potential best-in-class pan-ras molecular glue. It's one that is in a Phase I dose escalation study called Aurorus I. And going into that program, preclinical activity looks really interesting because this is a molecule that's been optimized on both potency as well as PK. So from a potency perspective, the binding affinity is about 10 to 20 times higher to cyclophylline A than the predecessor compound, and that translates into about a 5 to 10x potency advantage, and then also much stronger PK characteristics in terms of longer half-life, shorter clearance, and then much higher absorption within the local tumor environment because of CYP-A overexpression. So going into the clinic, we weren't sure if we would need to make tradeoffs between higher efficacy or stronger safety and tolerability or, in that unique case, whether we could actually see both. And I think from the early clinical data in the Phase I setting with Aurora S1, we've presented data twice this year, our April R&D Day and then a July update. And what's interesting is that you're seeing very promising single-agent efficacy within the second-line-plus pancreatic cancer setting. To your point, at pharmacologically active dose ranges, which is between 16 to 32 milligrams, we're seeing response rates in the 40 to 57 percent range, which really is very compelling. that if that persists within that range, that truly is best in class for that indication. And then, you know, as a company, we'll talk about this later, but we're really focused on first-line pancreatic cancer, where the unmet need is about, unfortunately, four times the size of what it is in the second-line pancreatic cancer space. Within non-small cell lung cancer that's RAS-driven, about a third of all patients with RAS lung have a G12X mutation or a RAS, mostly KRAS mutation of some sort. And what we've seen there is response rates in the 62% to 75% range, which really is extraordinary for a targeted therapy. That really puts it in a select group of molecules where you have some of the best targeted therapies that are seeing response rates that high. and that's at that pharmacologically active dose of 16 to 32. When you limit the data set to just second, third line only, post-checkpoint and post-platinum, that's where we see response rates in the mid-70s. So that's on the efficacy side where we're, I think one of the reasons we're seeing that higher efficacy is that the PK and exposure is dose proportional. So when you move up from 16 to 24 to 32 milligrams, you're seeing very good dose proportional increases in exposure, which then is really grabbing that additional efficacy. But you'd worry if that's coming at a cost of safety and tolerability, and we're just not seeing that. We're actually seeing very reasonable frequency and severity of rash, very low grade, mostly grade one, some grade two. and then very minimal GI in terms of diarrhea and vomiting, very low vomiting and nausea, if at all. And then in terms of stomatitis, those rates are quite low. I think what's interesting is a measure of how well-tolerated a drug is is what's called the relative dose intensity, and our median RDI stays at 100% through that dose range. So we think we're in that sweet spot of having both encouraging safety tolerability as well as encouraging efficacy.

Sure. Thank you. Thank you, Jonathan. Looking ahead to the 1H27 expansion data for 2015 and in combination data, what are you hoping or what are you focusing on with respect to release of that data set?

Yeah, so the key combinations that we're looking at. So all of that data that I just mentioned in lung and pancreatic is monotherapy data. Now, as we think about combinations, we sort of have a hierarchy of priorities. So the top priority is really what does this molecule do with standard of care therapy within certainly the big three tumor types of lung, pancreatic, and colon are key areas of focus for us. So with lung, a key priority is to see what does this do with Pembrolizumab, and then what does it do with Pembro plus chemotherapy if you need that, for instance, especially for patients with low PD-L1 expression, then whether a chemo-free or a chemo-containing regimen is needed, that's going to be a data-driven decision. I think when it comes to pancreatic cancer, you certainly want to be able to explore both chemo-containing as well as chemo-free regimens, and in that regard, we're already starting to explore combinations with gemabraxane because that really is a standard of care in that frontline setting. We also have disclosed a partnership with Tango for a combination with Volpi, which is a PRMT5 inhibitor, And so the public guidance from Tango is initiating a dose escalation of the combo of ERAS-15 with BOPI in the second half of this year. And then finally, with CRC, you really need to have an answer for EGFR activation, which can be a mechanism of resistance, especially when you put RAS map kinase under pressure with a pan-RAS inhibitor. So in that regard, we're excited to have disclosed in July the successful completion of the lowest dose of 16 milligrams where that passed the DLT window, and so we were able to dose up to 24. So 16 is a viable go-forward dose with panitumumab full strength, and that's something that at the time of the disclosure we were expanding. And so in the first half of next year, we're going to be disclosing various monotherapy and combination arms. At minimum, I think panatumumab is something that we're, because we started that first this year in the first quarter, we are calling that out as part of the first half 27 disclosure.

Great. Thank you, Jonathan. How confident are you that the therapeutic index holds into expansions and combinations? and is any rash prophylaxis built into the program?

Yeah, we're confident based on the early clinical data that we have a meaningful therapeutic window because we are seeing meaningful efficacy differentiation but not at a cost, as I mentioned. And so, in fact, as we in July showed the data of safety and tolerability by dose, We were worried that if you see a certain rash rate at 16 and 24 milligrams, do you suddenly see a spike at 32 milligrams? We didn't see that. We actually saw numerically lower frequency and severity of rash at 32 milligrams. Now, that could come up over time, but it's really in that same ballpark. So I think the therapeutic window seems very reasonable from 16 to 32. From a durability standpoint, time will tell, but we're really excited enough and have enough conviction based on the early clinical data that we've disclosed for both non-small cell lung as well as pancreatic to initiate three registration-enabling trials next year. So the first one is a potentially registration-enabling trial with monotherapy ERAS-15 for second-line-plus non-small cell lung cancer. This is something that we plan to initiate in the first half of next year. And then we are going to be conducting a confirmatory combination trial of ERAS-15 with pembrolizumab plus or minus chemotherapy. And that could start sometime between the second half of next year to first half of 28. And then in first-line pancreatic cancer, based on the compelling activity that we've seen in second-line plus pancreatic, we think that first-line could be even more interesting, and we've guided to starting a phase three for first-line pancreatic in calendar year 27.

Wonderful. Thank you, Jonathan. How are you thinking about competitive positioning with those registrational studies?

Yeah, so, you know, we're moving fast. So far, we haven't seen any abatement or slowdown in the enrollment of our AuroraS1 study. So, unfortunately, that speaks to the high unmet need, but also speaks to the enthusiasm amongst patients and investigators about the promise of ERAS-15 from a safety and tolerability and efficacy perspective. So, we're really excited to make that available globally. So, our phase threes will be global trials, and, you know, the unmet need outside the U.S. is even higher because there are no registered drugs outside the U.S. But it's going to become a more crowded space, I would just say, as a second entrant in the space with respect to pancreatic cancer. We're feeling very good about our competitive positioning.

Wonderful. And I guess on combinations, you cleared the first escalation cohort at the full commercial dose in CRC. With no DLTs. How central is the backbone plus EGFR strategy in CRC to the thesis? and how you think about combination versus monotherapy approaches.

Yeah, I think in CRC, the name of the game is most likely combination because, to your point, EGFR is a key escape mechanism for reactivation of the MAP kinase pathway. So if you put it under duress with a PAN-RAS agent, the pathway can circumvent and reactivate MAP kinase through EGFR reactivation. So you really want to take a combination approach with whatever RAS targeting strategy you have by having an anti-EGFR antibody on board. And so in our case, we chose panetumumab because of the fact that it's slightly better tolerated than cetoximab, and you don't have to have certain carve-outs in southern United States the way you do with cetoximab.

Sure. Wonderful. Thank you. I guess now that there is an approved RAS on benchmark with defined efficacy, safety, and pricing profile, how does that shape how you're thinking about your own development and positioning?

Yeah, I would say it certainly has dissuaded us from going in a second-line pancreatic, but outside a second-line pancreatic, I think it's still wide open. I mean, in the history of oncology, with all of these high unmet needs, whether you look at the checkpoint inhibitor space, whether you look at the CDK4-6 space, whether you look at a number of different areas of targeted therapy, there's never sort of, you know, one first mover that just takes over everything. And so I think patients and KOLs and, you know, the oncology community would prefer multiple options for patients. And so we're right in the mix. but I think we're in an exciting time for offering different choices to patients, and we really want to make sure that ERAS-15 is one of those choices.

Thank you. I've got a few questions now on the PRMT5 collaboration and the strategy debate around that, so maybe just take us through the scientific rationale of the combination of the pan-ras inhibitor with PRMT5 and maybe talk about how you see the commercial opportunity in an MTAP-deleted setting.

Yeah, it's an exciting area. I think, you know, the whole idea of targeting PRMT5 as a synthetic lethal strategy to go after MTAP-deleted tumors in pancreatic cancer, that frequency is anywhere from 25% to 40%, so pretty meaningful. And then I think it's about a third or so in lung, thereabouts. And so it's pretty meaningful in both PDAC and lung. And there's been really compelling early data presented by our partner Tango in a dozen patients where they showed very high response rates in the, you know, 92% range with a combination of Diracxon Rassib with Volpe. So we think that scientifically and clinically that's an area to look at. and so we're excited to be working with Tango to be on the leading edge certainly for pancreatic cancer and then I think for non-small cell lung the key differentiation for PRMT5 inhibitors is whether they have CNS penetration or not and so that's something that we're going to be paying attention to as well.

Sure, thank you. And on the deal, it's capital light and non-exclusive and you keep your molecule and they keep their molecule but how do you think about these collaborations as a way to explore combinations without diluting focus or spend, and should we expect more of them?

Yeah, I mean, I think we're convinced that we have a potential best-in-class pan-ras molecule, and we're getting a lot of inbounds from companies of various sizes to really get access to our pan-ras molecule. So to your point, without diluting focus and resources, is if we can make our drug available as a pan-ras molecule of choice, we'll do that. But we do have to sort of prioritize because we're just getting inundated based on the early data that we've presented. People are really excited to work with us on that. And so, you know, if there are these sort of capital light, low bandwidth requiring, you know, collaborations, then we're certainly open to those.

Sure. Thank you. And still on collaboration, so you may have signed a collaboration with Merck for 15, with PEMBRO and Aurorus 1. How central is the checkpoint combination to positioning, especially in lung? And when do first combination data come? And how do you weigh it against the other pathways, like with PAMI and monotherapy?

Yeah, I think in lung, you have to have an answer for checkpoint. So can you combine appropriately with a checkpoint inhibitor like PEMBRO? You know, if we can, then that opens up a front-line strategy for us, and so that's something that Merck and us are excited to see what happens. So we've disclosed that we're in dose escalation for that combination, and then the question is if the doublet has an attractive go-forward dose, then will you need to layer on chemotherapy on top of that? So will you need sort of a CIS or CARBO-PEM-PEM strategy on top of ERAS-15? So that's where we really need to complete the dose escalation with PEMBRO alone, see where the dose falls out, and then consider layering on chemo on top of that.

Got you, got you. And stepping back on the broader strategy debate, the field now spans, you know, on specific approaches, PAN-KRAS, PAN-RAS-GLU. Like, how do you frame where PAN-RAS wins versus a more selective approach? And as your thesis said, the broadest possible RAS shutdown is the durable answer or that different tumors and lines will call for different tools?

I think all things being equal, taking a holistic strategy to shut down RAS because there's so many different mutations and isoforms and ways in which the pathway can get reactivated. If you can, just shut it down with a pan-ras approach, that would be preferred. And then the question is just, what is your therapeutic window within different indications? And then what is your combination strategy? So all things being equal, if you can have a combinable pan-ras agent, which we think we have, then that would be the preferred approach, especially to your point, Sean, about the importance of durability. I think if you have a combo, let's just take pancreatic cancer, for instance. In the first-line setting, if you have a chemo or chemo-free combo, if you have lower dose reduction and modification on both sides of the equation with chemo, that should translate into really compelling efficacy and durability. And so all things being equal, you'd much rather take a pan-ras approach than a mutation-specific approach. Now, if you can't achieve that therapeutic window with a pan-ras, then you might have to pivot to a pan-K or a mutant-selective strategy. But we're really taking the position that if you can thread that needle with pan-ras with a wide enough therapeutic window to not only have really compelling monotherapy activity, which we've seen, but also a safe and tolerated combination. and the fact that we took the highest bar of panetumumab, which has stacking skin tox with a pan-RAS, and cleared the first active dose, that is promising. So we're encouraged by that early data.

Great. Thank you. Thank you, Jonathan. And moving over to 4001, which is a pan-K-RAS. So I believe we're going to see the Borealis data, monotherapy data, sometime this year. What would you consider a differentiated profile versus other PAN-K-RAS efforts?

Yeah, so 4001 is our PAN-K-RAS molecule, and as you mentioned, that is in the Borealis I study, which will have its Phase I monotherapy dose escalation data in the second half of this year. How to think about that data disclosure is that, you know, we are guiding to it being dozens of patients. In terms of a Phase I dose escalation, you traditionally see safety, tolerability, PK, initial signs of activity at relevant doses. In terms of how that benchmarks against others in the space, as you know, the pan-KRAS space has really only had minimal data disclosures to date, a couple of companies disclosing a single-digit number of patients, and so there really isn't a good benchmark out there for us to say what to point to. I think we as a field are still learning the role of pan-KRAS, And for that matter, some of these other approaches that you were mentioning, Sean, as well. And so as we disclose data, as others disclose data, I think we'll know more about what that looks like.

Sure. On the mechanism of 4001, it's designed to hit both GT-bound active states and the GD-bound inactive state of KRAS. Can you explain why targeting both states matters?

Yeah, so as a reminder, our molecule hits both the GTP and the GDP state, more so against the GDP state, so about 1.6 nanomolar against GDP, about 6.8 against GTP, so about a 4x difference there. There's a couple of reasons, Sean, that's important to be able to hit both. So the various KRAS mutations sit on a spectrum of those that are more shifted towards being in the GDP state, like KRAS G12C, or more in the GTP state, like the D and the V and other mutations as well. And so having a molecule that's able to hit across the spectrum, I think is really important, number one. Number two, you know, with our molecule having activity against both GTP and GDP, preferentially against the GDP state, that could be an advantage, especially when you think about combinations with our pan-RAS 0015 molecule. 0015 works through two mechanisms, the first of which is to block downstream effector proteins, but the second of which, which is important for this discussion, is that it causes hydrolysis of the RAS protein from the GTP state to the GDP state. And so if you pair those two, and we are one of only a few companies that have both molecules in-house, if you pair those two approaches where one causes hydrolysis to the GDP state and the other one mops up that GDP state, that could be a unique proprietary combination that we have.

Sure. You just answered my next three questions as well. Zooming out on sequencing, if PAN-RAS and PAN-K-RAS and mutation selective agents all end up with a role, where do you think broad agents versus codon selective agents get used?

Yeah, I think it comes back to what Jonathan was mentioning as well in terms of, you know, if you do have the ability to go after a broad swath with a pan-rasp with the right therapeutic window, that would be preferable. Both because what we've shown to date has been very good efficacy and very good safety, but also because of the potential for there to be, you know, resistance that can be through other mutations or other isoforms of the protein or whatnot. And so having that ability to target more broadly with a pan-ras would be advantageous. Now, I think we'll know more of the field as more and more data gets disclosed. There was a hypothesis, for instance, about whether pan-ras is too toxic and do you need to spare H&N-ras, especially when you're combining with an anti-EGFR that causes its own rash. Now, to Jonathan's earlier point, the fact that we actually did combine with panatumumab and we've cleared that 16-milligram cohort, which is a pharmacologically active dose, that actually bodes well for that combination. And so we'll know more, Sean, as there's more data, not only from us and from others, but I would say that if you can take that widest lens possible, that would be ideal.

Awesome. I've got a question here on China, Joyo, and the global strategy. So both RAS assets came from Joyo, and you exercised your option in March to take 15 worldwide, you know, added China, Hong Kong and Macau. A meaningful portion of the data you're citing also comes from the Joyo-sponsored China trial. How do the U.S. Aurora S1 and China data sets integrate? And can the China data support U.S. filings? And how are you thinking about China as a development engine?

Yeah, so I think China is a very important region in terms of the strength of the science that comes out of there is really quite amazing. And that's why we and others have been spending time over there to really find the right collaborators. And so it's really a great ecosystem. And I think it's a win-win in terms of leveraging what's the best of the U.S. and the best of China. I think in our case, a lot more of our data now is coming from the U.S. So in our July data set, that was U.S. only. We were accepted for an oral plenary session at the triple meeting this November in Spain, and so that will be all U.S. data. But we are taking a global approach, so I think it's important to keep in mind that both Phase 3s that we're going to be launching in lung as well as pancreatic will be global trials under one sponsor, that is Araska.

Sure, sure. I guess what we're seeing, you know, it's heralded in, you know, all the different approaches, you know, targeting RAS, you know, multiple assets, there's more MOA, there's, you know, sort of RAS targeting ADCs coming. Like, how do you maintain, you know, surveillance, you know, over the competitive dynamic and ensure you have a viable position going forward?

Yeah. You know, I think we're very active. We're aware of all the different developments that are going on, but also it's really important to have a high degree of focus and be able to execute what it is that you have before you. And so the fact that we have really, you know, we are one of the leaders in the PAN-RAS space, and then we also are amongst the leaders on the PAN-K-RAS space. I think the fact that we have two shots on goal positions us well. And then with that combination potential, that's where we think that could lead to some interesting explorations that we can do as early as next year.

Wonderful. Moving on to financials and the strategy. On the balance sheet, you raised in January, upsized again in July Q2. There was about $384 million before the raise. You know, given you've got three registration enabling trials and two phase one programs, How far does the current cash take you, and through which catalysts?

Yeah, we are very grateful to the investors who helped support us in these various financings and helped us put together a very strong balance sheet. You know, we are now in a position to be able to initiate, as we said, the three potential registration-enabling studies, two of those in Lung, one in Pank, and the capital that you just alluded to, Sean, does fund our corporate milestones, including what I just mentioned.

Wonderful. Thank you. I got through most of my questions. I have a couple which I've got here pending time, and we do have time. So we belong central to the story in brain metastases common. How are you thinking about CNS penetration and intracranial activity for your programs, and is that an axis of differentiation?

Yeah, I think it's hard to say right now. Now, we haven't done preclinical benchmarking of our compound from a CNS penetration perspective, but in non-small cell lung, there are patients where their blood-brain barrier is already compromised, and so in that case, it shouldn't matter. And then given that combination approaches with pembrolizumab plus or minus chemo are important, then I think especially where the BBB is compromised that should be no issue in the event that patients do progress in the brain beyond other parts of the body then that's something that we'll have to just track from a durability standpoint with a PRMT5 space we do think that having CNS penetration will be good for lung cancer Wonderful.

Thank you. With that, I'll finish all my questions other than one final one. So is there anything I didn't ask that I should have or a message you'd like to leave investors with?

We've covered a lot in this fireside, so thank you, Sean. I mean, I think, you know, as we've tried to outline here today, we're very excited about our pipeline. We've got two really exciting assets between ERAS-0015, the PAN-RAS, where we've already disclosed data. That data looked really strong in lung with that 62% to 75% response rate, in pancreatic with the 40% to 57% response rate, and then in CRC, it's an early start, but the fact that we were able to clear that first cohort with panitubumab is very exciting. And then on top of that, as Jonathan mentioned, we are able to have our cake, Anita 2, in terms of having what appears to be a really good safety profile across the major areas of rash, GI, and then especially in terms of interruptions and reductions where our molecule is looking quite strong. So in totality, we're very excited about what our pan-ras can do based on the data that we've shown. Pan-K-ras will also have a data update later this year, and we think that these two molecules really do address our core mission about erasing cancer and eradicating RAS-driven cancers.

Wonderful. Well, it might be a nice place to leave the conversation, but thank you for your time today, gentlemen. We really appreciate it. Thank you.