Welcome, everybody, to the JPMorgan Healthcare Conference. My name's Anupam Rama. I'm one of the senior biotech analysts here at JPMorgan. I'm joined by my squad, Priyanka Grover, Malcolm Kuno, and Rocky Paull. Our next presenting company is Erasca, and presenting on behalf of the company we have CEO Jonathan Lim.
Thank you, Anupam. Great to be here at JPMorgan. Happy 2025 to everyone. I will be making forward-looking statements, so I encourage you to visit our website at www.erasca.com, where you'll find our latest filings. So at Erasca, our mission is to erase cancer. We have a very experienced management team and SAB that is fighting cancer daily. We also have a pipeline that's entirely focused on shutting down the RAS/MAP kinase pathway. In fact, that is our namesake to erase cancer and to eradicate RAS-driven cancer. Today, I'm going to focus on three main programs. One is our clinical stage, phase three program called naporafenib. I'll also talk about our RAS-targeting franchise exemplified by ERAS-0015, a pan-RAS molecular glue, and ERAS-4001, a pan-KRAS inhibitor. We're very well capitalized. We reported $463 million in cash at our last SEC filing in Q3.
And also, I'm excited to say that we've revised cash runway guidance from previously H1 of 2027 to H2 of 2027. We're very capital-efficient as a company. So this is our SAB. Kevan Shokat was a co-founder of Erasca, and Mike Varney is our chair of R&D. Also excited to announce today that Piro Lito from Memorial Sloan Kettering Cancer Center has just recently joined our SAB, so he's a thought leader across RAS, especially when it comes to resistance mechanisms. We also recently added René Bernards from the Netherlands. Our focus is to shut down RAS/MAPK with three complementary approaches. The first is to target upstream and downstream nodes. The second is to target RAS directly, both in the active GTP and inactive GDP state. And then the third is to target key escape routes.
And with this singular focus, we've assembled a deep modality-agnostic, pathway-focused pipeline with naporafenib in phase three. I mentioned this is a phase three program that is enrolling SEACRAFT-2. I'll talk more about this trial. We also recently reported data at the AACR meeting last fall from SEACRAFT-1, so I'll show you some of the really exciting melanoma data from that announcement. And then ERAS15 and 4001 will be going into the clinic this year, so that's going to be a big milestone for us in AURORAS-1 and BOREALIS-1, respectively. Cool thing is when you put those together, you get the northern lights. And then we also, in the discovery pipeline, have ERAS-12, which is a bispecific antibody. So we truly are modality-agnostic. We just want to find any way to shut down this wily pathway. So let's start with the RAS franchise.
We're getting a lot of focus and attention on this. I think investors are really excited about us making some noise in this space. So there's three main attributes when you look for programs that are targeting RAS. It's a very crowded space, as you know. So you really want to have very good preclinical potency, high oral bioavailability, and a strong proprietary position. And both molecules have all three at the nexus of these attributes. In particular, for ERAS-0015, we think that due to its high potency, if you can enable lower clinically active doses than what has been historically done with pan-RAS, you can lower the risk of solubility-limited absorption and exposure plateau that has been observed. You can also have lower drug load, which could result in better GI tolerability. And perhaps, especially when it comes to the pan-KRAS molecule, an improved therapeutic window.
I think in the G12C space, history has shown us the importance of potency and how that matters and translates into some efficacy advantages. Here's a summary slide that really highlights the key features of both molecules. Starting with 15, this is a potential best-in-class pan-RAS molecular glue. It basically binds cyclophilin A or CYPA and forms a tripartite moiety, very similar to Revolution Medicines' RMC-6236. RevMed's done a great job blazing the trail, but we're very excited about ERAS-0015 and its preclinical properties to date. In vitro, you can see it has low sub-nanomolar to low nanomolar IC50 across multiple mutations, including wild type. In vivo, what's particularly impressive is the incredibly low doses by which tumor regression is observed, so doses as low as 0.3-5 milligrams per kilogram PO on a daily basis.
The oral bioavailability is very high across small and large animal species. IP is strong with exclusivity expected through 2043 and no patentability roadblocks identified to date. ERAS-4001 in vitro has sub-nanomolar to low nanomolar IC50. In vivo, you're seeing tumor regression at doses between 30 to 300 MPK BID. The oral bioavailability for this class of compounds is non-trivial, but we're really excited to see very good bioavailability across small and large animal species. And similar to ERAS-0015, very long runway out to 2043 for the IP with no roadblocks. Now, what's interesting is, as a company, we can be the only company in the world to combine these two types of modalities: a pan-RAS molecular glue and a K-selective inhibitor. So the K-selectivity refers to the fact that it spares H and NRAS for ERAS-4001, resulting in a potentially wider therapeutic window.
And so when you put those together, we could literally clamp RAS and just not allow it to resist, at least at the mutation level. So I'm going to do a deep dive on 15. I mentioned that this is a CYPA binder. So relative to 6236, the binding affinity for this molecule is 8 to 21-fold higher, which may enable more potent RAS inhibition. In fact, there's a lot of key benefits that arise from this higher binding affinity, including potency, tumor residence time, and other attributes that I'll show you. That, in addition to the PK benefits, really the totality of the data are quite exciting. So when it comes to the cell growth inhibition assays, you could see in the middle column the cell growth inhibition with 15. In gray, you could see CGI with 6236.
Then the rightmost column highlighted in blue, you could see the fold potency of 15 versus 6236. And you can see as high as 13-fold higher potency in certain cell lines for 15 versus 6236. Even in areas where there is equipotency observed, for instance, in the PK59 cell line, in vivo, it tells a different story. So you saw equipotency. I'm going to show you a series of slides that show you the TGI growth curves on the left, the body weight change in the middle. And then these are for you to digest on your own. The TGI summary on the right really will draw your attention to the key points on these slides. And so, for instance, in this PK59 model where equipotent CGI was observed, RMC required 3 MPK to achieve tumor regression.
But if you look at 15, an impressively low 0.3 MPK was required to achieve the same degree of TGI, translating into a 10-fold higher potency advantage. So that's the message from this cell line. And then I'm just going to show you a number of different models where this translates. H727 is what Ryan Corcoran on our SAB would refer to as a bellwether model. So this is where good pan-RAS and pan-KRAS molecules go to die because it's just so intractable. It's a G12V non-small cell lung cancer model called NCI-H727. And RMC-6236 shows very good tumor regression at 10 MPK, which is impressive, but ERAS-0015 requires only 1 MPK, which is just amazing in this bellwether model. Similarly, in a CRC G12V model, it requires 25 MPK of RMC versus just 3 MPK for about an eight-fold higher potency advantage.
And then G12R is a mutation that is more often found in the GTP active state. So it actually is really tough for the pan-KRAS class of molecules to inhibit this type of mutation. But with ERAS-0015, you could see very good tumor growth regression at just 5 MPK. And in the same model, this was a different experiment, but also in PSN1, you could see side by side lower doses of ERAS-0015 translating into a 10-fold higher potency advantage. So achieving more than 95% tumor growth inhibition in this G12R PDAC model vis-à-vis RMC-6236. So now, as you think about the big three cancer types like lung cancer, colon, pancreatic, it'll be important to understand what combinations do. And so in this case, with non-small cell lung, you'll probably need combos with a checkpoint inhibitor. So what was done in this experiment is combining 15 with PD-1.
You could see in purple in the graphs on the left relative to the controls, just a flatlining, a complete eradication of the tumor in seven out of seven mice, and this is even upon cessation of the mono and combination treatments. On tumor cell re-challenge, you stop and then you inject tumor cells in the opposite side of the animal, and you could see the tumor growth just coming back, roaring back in the control arm, but the immune system has been primed with the combination, and so you could see continued eradication of the tumor with no body weight change, so very impressive and hopefully bodes well for combinations in the clinic with checkpoint inhibitors. I mentioned there's another advantage to the very high binding affinity of CYPA, so what you're looking at here is a tumor distribution kinetic study.
I'll focus you on the PK59 model on the left, so in blue is ERAS15. In gray is RMC6236. The light blue bar graphs show basically the dose-normalized concentration in the blood, so you could see for both molecules, from 4 hours to 24 hours, the blood levels decrease pretty rapidly, but in the darker blue graphs, you could see with RMC, there's a similar decrease of the tumor concentration of the drug, whereas in ERAS15, from 4 to 24 hours, you maintain persistently high tumor tissue concentrations, showing that there's a longer residence time. There's almost a depot effect of the CYPA where there's more in the tumor tissue than in the blood, and so 15 is just lingering.
So, as far as target coverage, we're really excited about that profile that we're, first of all, hopefully with the PK properties that I show you, delivering more drug to where it needs to go. And then once it goes there, it just lingers. And then this is not a model-specific phenomenon. You could see the same thing happening in another model on the right-hand side. So talking about PK, this is a pretty busy slide. So I want to draw your attention to three key parameters. You're looking at 15 in blue and RMC in gray. And across the board, when it comes to clearance, we're seeing lower clearance across multiple species translating into longer half-life.
There's also higher oral bioavailability as manifested by the high F%, including really a very high F% in monkey, which most small molecules that we've evaluated have less than 5% oral BA in monkeys. This is showing very high BA of 17%. Really across the board, showing very nice differentiation profile. Now, on the ADME and tox properties, the kinetic solubility in both the fasted and the fed state is very steady. It doesn't drop off in the fasted state, leading to solubility-limited absorption. The CYP profile looks good. The combinability should be pretty straightforward for this molecule. Some new data-generated GLP Ames was conducted. That's a measure of mutagenicity that was negative. GLP hERG, when I talk about 4001, I have an update there. You want to be well above one micromolar. This shows a very good safety margin.
And we also completed the GLP 28-day repeat dose studies in rat and dog. And so we're just waiting for the reports. So that's 15. I'm going to talk now about 4001, which is a potential first-in-class and best-in-class KRAS small molecule inhibitor. Now, this has the ideal profile in terms of if you want to hit certain mutations really hard, you probably want to hit D the hardest and then B. And then if you have C on board, that's great. There's a number of G12C inhibitors, but this really is sort of a DVC inhibitor. But also with KRAS, a lot of acquired resistance can happen with wild-type amplification. And so you do want to have some KRAS wild-type activity, which this does. And I think that differentiates it from some of the, say, mutant selectives like the D-selective small molecule inhibitors.
It's one reason why we prefer the pan-KRAS class of molecules, and also emblematic of the differentiation profile, you want to spare H and NRAS, so you really want to hit KRAS hard and spare H and N to have that wider therapeutic window, which this really doesn't touch H and N, so it's got a very nice profile. The other thing is you want to have activity against both the active GTP state and the inactive GDP state, which this molecule also has, so it's got single-digit nanomolar against both the GTP and GDP states. Now, like 15, this in the cell growth inhibition assays has very low nanomolar potency across G12D, V, C. It even hits A and 13D, and then I mentioned that it's a KRAS wild-type inhibitor as well, so I'll show you some TGI curves. On the left, you could see G12D, PANC model 0403.
So at doses at or above 100 MPK BID, you're seeing very good tumor growth regression. And that correlates with the PK and PD profile as well. These are different G12D and V models where doses as low as 30 MPK BID in the more sensitive models is showing very good tumor regression with no body weight change. And then this is that infamous bellwether model, the H727. Fortunately, this pan-KRAS molecule is able to show tumor regression, albeit we had to crank up doses to 300 BID. But we did see regression. And so that's good that even in the most intractable in vivo model, you're still seeing regression. And then this is a G12D PDAC model called HPAC, where you're seeing very good tumor regression at fairly low doses of 50 MPK. Now, I showed you that experiment with 15 where you combine with a checkpoint inhibitor.
The same experiment was conducted with this molecule. So we actually are really excited about the combinability for pan-KRAS, especially as you think about other agents that might induce rash, such as the EGFR antibodies, for instance. That'll be really important to be able to combine with those for colorectal cancer. In this case, for lung cancer, you're seeing similar story, just complete eradication of the tumor in seven out of seven mice tested. And then on KPC re-challenge, you're seeing persistently eradicated tumor. So you've primed the immune system with the doublet. In terms of PK and oral bioavailability, it looks promising for this molecule with double-digit % F across multiple species tested. And then on the ADME toxicology properties, the CYP profile looks very good. GLP Ames was negative. One yellow flag that we had when we first brought in this program was the hERG properties.
It was just hovering at around one micromolar, so that necessitated an in vivo study, so you run what's called a GLP dog cardiovascular study. Really pleased to say that we ran that, and there was no observed QTc prolongation, and so that tends to be a better litmus test of what you might expect in the clinic. Doesn't mean we're not going to do cardiac monitoring. We do cardiac monitoring for all of our molecules in the clinic, but it means that the risk of a cardiovascular liability is substantially lower based on the dog CV study. And then we also did repeat those studies in rat and dog, and both of those are complete in terms of the in-life portion. And we're just filing reports now.
So, KRAS alterations, here's the frequency, which is very high in the big three tumor types of colorectal, non-small cell lung cancer, and pancreatic. There's a long tail, but still very high frequency of patients across other tumor types. We'll be focused really on the big three, but also we won't be ignoring the long tail. So, we'll be looking at where can patients benefit the most and focus on tumor types with the largest number of patients and highest unmet need. We will be moving towards early assessment of combinations. We'll be very data-driven. And then we also have a portfolio of RAS-MAPK inhibitors. So, the fact that we could combine them not just with standard of care, but also with each other is very exciting. So, I'm now going to end with our clinical stage program, naporafenib.
This is a pan-RAF inhibitor that potently inhibits BRAF and CRAF with sub-nanomolar IC50 potency, single-digit nanomolar potency against ARAF, and is very selective, as you can see from the KINOMEscan on the right. Now, what's really cool about this compound is it's been dosed in over 600 patients to date. And so we in-licensed this from Novartis. They have run a number of different studies. I think what's emerged from there, in addition to our phase 1 SEACRAFT-1 study that we just reported on, as I mentioned, there's a very strong proof of concept demonstrated in NRAS mutant melanoma, as I'll talk about shortly. And so on that note, I'll draw your attention to the graph on the right. And so the standard of care for NRAS melanoma, which is a very aggressive disease, high unmet need, no targeted therapies that are approved for this indication.
It's really the only approved SOC is chemotherapy, which shows about one and a half months of response rates and a PFS, sorry, one and a half months of PFS and about 7% objective response rate. So really just a high unmet need. With binimetinib, you see a rough doubling of both of those metrics. But that was never approved because of the shifting standard of care. And so binimetinib is on the NCCN guidelines, but is not an approved regimen for this indication. Now, for naporafenib and trametinib, we do have FDA Fast Track designation. And we have really a potential to be first to market in this high unmet need area, which is very exciting for patients. And so these are data that we reported at the Triple Meeting last year.
So, you're looking at the melanoma cohort with Q61X mutations, which there's a high concordance of Q61X with NRAS. So, all of these, actually all of the patients on this chart had NRAS Q61X. And you're seeing 4 out of 10 responses for about a 40% response rate. Three of them had confirmed PRs at the time of this data disclosure with an 80% disease control rate. And then, when you look at the swimmer plot, very good durability. So, as of the data cutoff, all the patients, 70% of the patients remained on treatment, including all of the patients that had responses. So, very promising.
When you roll up all the data between standard of care, chemo in the middle, MEK inhibitors on the left, and then naporafenib and trametinib data on the right, you could see very impressive improvements of objective response rates, including the recently observed 40% response rate observed in SEACRAFT-1. And then importantly, from a regulatory perspective, we're going to be looking at co-primary endpoints with PFS or progression-free survival and overall survival. And so you're seeing a meaningful difference of the PFS from the pooled phase 1 and phase 2 data of about five months versus one and a half months for standard of care chemo and just about 2.8 months for single-agent MEK. And then almost a doubling of the overall survival. So we feel really good about the potential to win on either or both of the SEACRAFT-2 primary endpoints.
From a safety perspective, in the early phase 1 and phase 2 trials, rash was really a key area of concern, especially for the dose regimen that was tested of 200 milligrams naporafenib plus 1 milligram trametinib. When we took over this program with SEACRAFT-1 and 2, we're instituting mandatory primary prophylaxis. What that means is treating with oral antibiotics and topical steroids before you even receive study drugs so that you can get ahead of the rash. This resulted in really a dramatic improvement across grade 3 derm toxicities. You could see the dramatic reduction for about 37% down to 11.5%. The TEAEs leading to permanent discontinuation went down substantially, with none of them resulting from a derm tox. Then relative dose intensity, which accounts for dose reductions, you could see that the RDI improved substantially from about 50%-66% up to 98.5%.
We think that that safety profile, that improved safety profile, might have resulted in the improved response rates that we're seeing, albeit in small data set. This is the study design stage one dose optimization for SEACRAFT-2. We'll be testing two other doses because we could take the 201 dose from SEACRAFT-1 and evaluate that vis-à-vis other doses like 400.5 and 100 plus 1 against a single arm of trametinib alone. We'll take that stage one data from both SEACRAFT-2 as well as SEACRAFT-1. We'll evaluate that. We'll speak to FDA. And then that will initiate stage two, which is the randomized, basically randomization of the preferred dose that's selected versus physician's choice of trametinib versus chemo, with the study endpoints being PFS and OS. I'll end with the key milestones in clinical trial readouts.
So ERAS-0015, I'm really pleased to say that an IND was filed with the China CDE in Q4 of last year. What's cool about that is implicit in that is that it met the sort of GLP tox requirements of the China FDA authorities. And that's a similar bar to what FDA does. Now, our US IND, we're guiding to mid Q2 of this year because the CMC requirements are more extensive in the US than they are abroad. But we are parallel tracking both INDs for ERAS-0015 and ERAS-4001 with prioritization on ERAS-0015. So we're guiding to an IND filing for ERAS-4001 in Q2. And then we'll have phase one monotherapy data readouts for both programs next year, which will be very, I think, exciting for everyone. And then for naporafenib, SEACRAFT-2, I mentioned that we're in the stage one randomized dose optimization data. Enrollment's going well.
We will report data from that, as well as an update from the SEACRAFT-1 melanoma patients in the second half of this year. So I'll just end by saying that we have a very compelling investment thesis with a strong team, world-class SAB portfolio that's entirely focused on shutting down cancer via RAS/MAPK pathway. We are a phase three company, which is really an exciting place to be, but with an exciting future focused on shutting down RAS and multiple value drivers this year and beyond. So thank you very much, and look forward to the Q&A.
Yeah. Thanks, Jonathan. I just want to remind folks that there's three ways to ask a question. You can raise your hand and I'll call on you. You can email me or you can put it in the portal and I'll ask it from there. Jonathan, I just maybe competitively wanted you to comment a little bit about ERAS-0015 and what you've kind of learned from the evolving competitive landscape from Revolution Medicines and others. And you highlighted the preclinical differentiation, but just wondering what you've learned competitively.
Yeah. So I think what's interesting is that the scientific bar for working on a pan-RAS molecular glue is very high. So it's really hard, much harder to, I think, discover and develop a pan-RAS molecular glue than it is. There's many more pan-KRAS small molecules out there. So I think there's sort of a scarcity issue within this class. So RevMed clearly is the trailblazer in this space. I think we've learned a lot from their clinical data in terms of PDAC, as well as more recently non-small cell lung cancer. I think other tumor types will be interesting to see. But being second in an area of high unmet need across PDAC, lung, and other tumor types is a good place to be. There's one other company in China that has a pan-RAS molecular glue. They are further behind.
And also, based on the preclinical data we've seen, it looks to have a profile not too dissimilar from 6236 at this point, but it's still early days. I think, yeah, so I think we've learned mostly from RMC-6236 in terms of the clinical data for PDAC and lung. The PDAC data in particular are very exciting in terms of the ability to help patients. And we'll be following them as well as once we start generating our own data later this year, I think we'll be right there.
And then just for ERAS-4001 as well as ERAS-0015, just thinking about how you're thinking about the geographies where you'll be running the initial phase 1 study as well as number of sites. Have you kind of mapped the sites?
Yeah, so I would say that we want to, first of all, the enthusiasm among investigators for not just one program, but both programs has been very high. Unfortunately, it's a reflection of just the high unmet need of how many patients have these RAS a nd KRAS mutations. And so most of the sites, if not all, have actually expressed interest in running both trials and informing us that there'll be no sort of rate limitation in terms of the number of patients. In fact, they'll still have to turn patients away even if they have both studies, so I'd say we don't need a double-digit number of sites. I think single digits, it'll be predominantly U.S.-based. And we anticipate that enrollment will be quite brisk based on, unfortunately, the high unmet need.
Questions from the audience?
Wanted to squeeze a couple in on naporafenib. So maybe based on, can you provide some enrollment color and when we should be thinking about data? You have a pretty broad guidance there of 2025. And then what's going to be the size and scope of the phase one cohort and your assumptions around what dose you're going to move forward with, 400 mg or 100 mg?
Sure. Yes. So naporafenib, the pan-RAF inhibitor, we initiated that study in Q2 of last year in terms of the pivotal study. What we've guided to is that the stage one data readout will be in the second half of this year. The site enrollment has been going very well or initiation. Patient enrollment has also been good. The investigator enthusiasm has also been high on that program, again, just given the unmet need here, lack of a targeted therapy that's approved in this indication. We are evaluating two different doses in stage one. Plus, as Jonathan mentioned, we have that additional dose from SEACRAFT-1. So we can actually choose from different doses of naporafenib of 100, 200, or 400 milligrams.
And so based on the data that come out of SEACRAFT-1 that we showed at the triple meeting last year, as well as the two doses that are being evaluated in SEACRAFT-2, we'll be able to make a determination of which dose we want to move forward with. Now, the presumptive dose going forward at this point, just based on the data that Jonathan shared here and that we shared at the triple meeting last year, is that that 200 dose looks pretty good from a safety standpoint as well as from that efficacy standpoint. And so unless we see something dramatically different in stage one of SEACRAFT-2, that 201 dose looks really good. Now, the other thing I'll mention is that the study is an open label study. So that's why that stage one is ranged between 60 and 120 patients.
Because with as few as 60 patients or as many as 120, we may be able to make that dose selection decision to then move into stage two, which will be the randomized portion there as well. But from stage one, we'll get that contribution of component information versus trametinib as well as that randomized data versus trametinib.
And then just thinking about that update, just what are you looking for to just confirm the dose, right, on both efficacy and safety? Like you flip the chart, well, it's open label, but you look at the data and you say, "Wait, all right, we're good to go because we hit a certain efficacy and safety threshold.
Yeah. I think what's encouraging here is that if you look at the historical data that Jonathan went through today, the approved standard of care of chemotherapy or even the off-label use of single-agent MEK, unfortunately, it's abysmal, right? You have one and a half months of PFS for chemo. You have 2.8 months for single-agent MEK. Overall survival isn't that great. It's about seven months for both agents. The Novartis-generated data from the phase 1 and 2 shows about a 2-3x improvement on PFS and about a doubling on OS. So already, if we can just show that, it looks really good. Now, if you look at the SEACRAFT-1 data on top of that, with the introduction of the mandatory primary rash prophylaxis, the rate of AEs has improved dramatically. The relative dose intensity has improved dramatically. So that's helped a lot from the safety standpoint.
And then that also may be translating into an efficacy advantage where you're seeing now a 40% response rate, albeit small numbers in the SEACRAFT-1 study. So it's going to be a combination, Anupam, of both the safety tolerability side, but also the efficacy side in terms of which dose to move forward with. But we're already doing very well compared to where chemo and single-agent MEK are performing.
Questions from the audience?