Verastem, Inc. (VSTM) Earnings Call Transcript & Summary

As you know, I see a lot of familiar faces in the room. You're familiar with our story. We're really focused on RAS/MAPK-driven cancers. Within our own portfolio, we've got the RAF/MEK inhibitor, avutometinib, defactinib, which is our FAK inhibitor and then the G12D inhibitor. In addition to that, we do have 2 other targets as part of our GenFleet collaboration that we haven't disclosed. And so both within our own portfolio and then combining with things outside our portfolio, we think we have a lot of different options to really directions to go in. I won't go through the guiding principles here, but really, what we're trying to do is develop treatments for cancers with high unmet need. AVMAPKI FAKZYNJA CO-PACK, which is the brand name of the drug that just got approved. It was approved on May 8, almost 2 months early, very excited. The data in low-grade serous ovarian cancer, 44% overall response rate with durability is really unprecedented and the fact that we can offer really an oral therapy for patients is just wonderful. So what we're going to do today is I've got Dr. Picozzi here really to talk about the pancreatic study, so the RAMP 205. Jon Pachter will talk a little bit about some of the preclinical information. Dr. Hong is here to talk about the G12D inhibitor, and then John Hayslip will talk us through really our development program going forward for those. So Dr. Picozzi, we're happy to have him here. He's the Director of the Pancreaticobiliary Program at Virginia Mason and really a real expert in this disease. So we're just honored to have him here to give some background. Thanks.

Thank you. Well, I appreciate the kind introduction, Dan, and it's a pleasure to be before all you today. My task is to update you on the pancreas cancer field in general and in specific, how the RAMP 205 study fits into that. So just by means of introduction, pancreatic cancer is actually a continuously increasing disease both in the United States and worldwide. Here you see the incidents in several major countries. The most current numbers are actually somewhat higher than this. The expected incidents of pancreas cancer in the U.S. this year is about 66,000 and over 0.5 million worldwide. You see the difficulty with survival. The 5-year survival of all pancreas cancer is 13.3%. But if one takes adenocarcinoma, which is what we're focusing on today, it's actually less than 8% even now. In its most fundamental sense, the disease has a biological abnormal substrate and foremost among the abnormalities seen in pancreas cancer are mutations in KRAS. I think you can think of KRAS and mutation -- you can think of KRAS is like the on-off switch for pancreatic cancer growth and when it's mutated, it's permanently on. The vast majority of people with pancreas cancer will have KRAS mutations. And here you see their list in order of frequency with KRAS, G12D mutation the most frequent, followed by V, then R, and a variety of others. About 10% of people are what we call so-called KRAS wild type, they don't have a readily demonstrable mutation. So currently, this is how we treat people with pancreas cancer, and they divide into 2 groups. About half of people have disease that is confined to the pancreas visibly and about half of people have disease that's visibly spread, local and metastatic, although it turns out that literally every -- almost everybody has metastatic spread in a way that you can't -- either can see or can't see so that people operated on if they receive no additional treatment will recur 90% of the time. So the way I think of it, everybody has metastatic disease is just you can't see it in some people. This is how we treat people with the more advanced forms of pancreatic cancer. There are multiple standard regimens that are used, which really, there's no consensus on them in terms of which is best. And I think the truthful answer is different regimens are best for different people. And generally, if one of those regimens fails, you try as a second therapy, something else. So -- and FOLFOX maybe has been substituted for by NALIRIFOX, which is actually the only approved drug regimen in second-line pancreas cancer. And below that, you see the survival of -- the rates of response and survival still far from what we'd like to see, and more importantly what the patients would like to see. So in that environment comes drugs that intend to block the RAS/RAF pathway, and the first drug in the combination we're going to discuss today is a drug called avutometinib. Avutometinib, as you see from the diagram blocks the RAS/RAF pathway and thus inhibits the effect of KRAS. But you also see there's a second pathway that's very essential to tumor growth and development, which is referred to sometimes as the FAK pathway. And so a FAK inhibitor drug, defactinib, okay, which is the second experimental drug in the combination as we discussed inhibits that pathway. And when the first pathway is switched -- is inhibited, more activity switches to the second. So an important theme of this discussion is the duality and synergy of both of these drugs together. It turns out the drugs also have other effects, in particular, the FAK inhibitor, which induces a decrease in the stroma, which is the cellular surround of -- which is the surround of the cancer, both in terms of collagen and fiber structure as well as cells. And this reduction in the stromal density may play a role in therapy, including drug penetration and penetration of immune cells. So the point is it's a broader concept of how drugs work and typically now in pancreas cancer, we think not just of the effect of the drugs on the cancer of the cells, but also the surrounding environment, what we refer to as the TME, the tumor microenvironment. And furthermore, if we explore the combination of avotometinib and defactinib in conjunction with chemotherapy, you can begin to see at least in a preclinical setting, the synergy I was talking about. The slide on the left relates to frequency or degree of tumor regression. And you can see that there's some tumor regression by using standard chemotherapies, much more by using the 2 investigational agents. But the biggest gain of all comes when all 4 agents are used together. On the right-hand side, you see the development of liver metastases. And again, you see some effect when you add either defactinib or avutometinib to chemotherapy, but it's the combination of the 4 drugs that almost completely abrogates the development of liver metastases, which is actually directly related to the cause of death of many people with pancreatic cancer. So this is an overview of the RAMP 205 study. It looked at patients who had untreated, pathologically proven metastatic pancreas cancer that could be readily assessed radiographically with what we call a performance status or a liver function that permits chemotherapy. And the first part of the study, Part A, looked at different doses of the 4 drugs in the experimental design or the experimental group to try to find what would be the best combination in doses of drugs to move forward into the second part of the study, which is to look in more detail at that most successful combination. So it turned out that there were 5 different dose combinations examined. And as you will see in a moment, the 1 out of the total of 60 patients, 12 in each group, that ended up being the most successful and you understand why, in a sec, was actually the very first level tested. And here you see the doses of the 4 drugs and their schedule over time. You can see that most of the other combinations look at different adjustments in doses and scheduling of the different drugs. And here's the response of dose level 1. Out of the 12-patient study, you can see that 10 of the 12 actually had evidence of what would be considered a response radiographically. And to put that in some context, a typical treatment for metastatic pancreas cancer might show a radiographic response of about 30%. And here, it was, 10 out of 12 or 83%. On the right of the slide, you see some of the more detailed data in terms of what's been confirmed and unconfirmed. I actually think almost as equally important is the absence of progressive disease. About 25% of people with metastatic pancreas cancer will progress despite therapy in the first 2 months. And in this early experience, there wasn't any and something you'll see even more as we examine all 5 dose levels. So a very high response and what we call the dose control rate, which means the absence of progression after 4 cycles of therapy was a very robust 92%, again, based on a literature comparator that's much lower than that. This is called a swimmers plot or a swim plot of the patients. And you can see that many of the patients remain on therapy. The red dotted line indicates the historical length of time that patients remain on just gemcitabine, nab-paclitaxel. And you can see that, first of all, many of the patients remain on study, but it will turn out that a vastly significantly greater percentage of patients will, in fact, have therapy extended beyond what would be the normal amount of time using gemcitabine, nab-paclitaxel. And with that, presumed better outcomes in terms of response. And in the corner, there, you see some comparator data from 2 very significant trials in pancreas cancer that's metastatic, the MPACT trial, which looked at gemcitabine nab-paclitaxel. And the NAPOLI 3 trial, which also use gemcitabine, nab-paclitaxel is one of its arms, and you can see the literature is very consistent in terms of the median progression-free survival. This is a CT scan from a patient who was treated with dose level 1. And I think even to the relatively untrained eye, and I include myself in that, my wife is a radiologist and she cautions me never to look at x-rays because I'm not a radiologist. But I can even because when my kids get sick, they tell my wife, we want to go see a real doctor like dad. You can see there's an obvious regression of the cancer in the liver, particularly in the bottom set of slides. But I think that the CT scan doesn't really tell the true story. And it turns out the very last patient I saw before I came to ASCO is a man in his 60s who is one of the patients at this dose level. And when he started this treatment, he was in a lot of pain, very limited in terms of his lifestyle because of that. And actually, even wondered if he wanted to take treatment at all and continue to live. And now here we are 8 months later. He just -- as I said, he was the very last person I saw before coming here. He had just gotten back from a trip to the Cancún. He was snorkeling and eating what he calls gourmet Mexican food every day. And he asked me if he could move to Cancún and get his treatment there. And I said, no, I don't I think we can do that, but you can have another vacation, if you want. The toxicities of the treatment are actually similar to what you see in general with other treatment regimens. There wasn't anything that was previously unknown that was detected. Most of the toxicities were relatively mild and as I've said, are characteristic of what's been known in the past. Probably the big issues relate to hematological toxicity as is characteristic of gemcitabine, Abraxane and GI toxicity. The avutometinib also has a peculiar ocular toxicity, so patients are managed by serial visits with an ophthalmologist. But overall, it's quite doable. Here you see actually the responses of all 5 cohorts. So you see that there were significant responses at every dose level, but there were many more at dose level 1, which was a big factor in the choice. And to me, again, all what is significant is the number line is pointing significantly downward and the absence of lines pointing upward. Again, in a typical pancreas cancer regimen in this setting, about 25% of these bars would be going up instead of down. And if you look at dose level 1, there aren't any bars going up, they're all going down. And to me, that speaks as much about the activity as the magnitude of the response. Many of the patients also in all 5 dose levels remain on treatment. But again, the dose level 1 seems to be the one that is the best. And so the conclusions that have been drawn to date by the study team are that the toxicity of this regimen is certainly manageable and has a very impressive response rate, particularly at dose level 1, and that was chosen as the recommended Phase II dose, the response rate that's to date, which is, again, early, but what further experimentation is going to be based on was 83%. And that optimizing dose intensity, particularly with respect to nab-paclitaxel may be important. The vast majority of the patients are expected to do better than what would be expected historically. And the additional enrollment of additional patients, additional 17 up to 29 is already underway. In fact, we have one patient as a part that now. So thanks very much for your attention. I'd be happy to answer any questions you got. Yes, sir.

So what is unique about dose level 1. Why does the similar [indiscernible]. Jeff from RBC Capital Markets.

Yes. So let me try that one. So I think that's not entirely clear because of the sample size. But if you look at dose level 1, it's the one that has the highest overall dose intensity of all the drugs, particularly the Abraxane.

And a related question. So why do you think the 2.5 milligram dose for avutometinib works better than 3.2 that's been used for LGSOC. I guess, do you think there's perhaps an additional dose optimization that could help identify potential differences in terms of efficacy between different tumor indications?

So I don't have an exact answer for that. But I think -- first of all, your questions are very sophisticated. Thank you. And the way I would think of it is that I would think of the drugs not individually, but collectively. And -- so it's like cooking and in a particular dish more of one spice as opposed to another, maybe better or not. So -- and there probably are differences between the cancers. So in this particular cancer, that particular set of concentrations may come to the best result.

So I think we'll do a Q&A at the end. So we'll see the rest of the questions for the end.

Okay. Okay. Sorry about that, Jon.

No, that's fine. Good questions. Thank you. Thanks, Vince. So I'm John Pachter. I'm Chief Scientific Officer at Verastem Oncology. If you can advance the slide. Oh, I guess I can advance the slide. So now we're going to change topics to our KRAS G12D (ON/OFF) Inhibitor, VS-7375, which we developed in collaboration with GenFleet Therapeutics. And I have about 5 slides to convince you why I think it's a best-in-class molecule. So let's see if you agree. One of the key things, as you can see on the lower right of this slide is that it binds both the GD -- sorry, the GTP ON-state and the GDP OFF-state with low single-digit nanomolar affinity. And I think being a dual on/off inhibitor is important. And to talk through that a little bit, if you look at an ON-only inhibitor such as RMC-6236, which is RevMed's RASMULTI. There's been some elegant work from Piro Lito, which shows that it can induce GTP hydrolysis and convert the ON-state to the OFF-state. But 6236, as an example, is not covalent. So by definition, it's going to come OFF, it can't bind to the OFF-state, whereas our molecule combined regardless of where the KRAS lives ON-state or OFF-state. And theoretically, at least, there should be a pool of the OFF state that could build up and be ready to come back and signal, and we think it's an advantage to be able to occupy no matter whether it's in the ON-state or the OFF-state and also use some preclinical data where we've compared directly to the ON-only inhibitors, and we feel that there's really exciting in vivo data that shows that difference. So just quickly, we've looked here at the selectivity of our G12D inhibitor versus a couple of others. And what we're doing on the right side of the slide, if you look at the table, we're looking at 3D proliferation of a number of tumor cell lines, those with G12D mutation, those with other KRAS mutations or KRAS wild-type cell lines. And you can see that 7375 is potent against an inhibiting proliferation of all of the G12D cell lines and very selective relative to the other KRAS mutant cell lines or the KRAS wild-type cell lines. And that's in contrast, for example, the 9805, which doesn't hit a couple of the G12D cell lines very well but does hit a number of the other KRAS mutations. It does seem to be selective relative to wild type and the Mirati molecule is less potent in these experiments. So I think this is probably the most important slide that convinced me other than the theoretical that our dual ON/OFF mechanism of our G12D inhibitor is really advantageous. So what we did here is we looked at a comparison of VS-7375 as G12D ON/OFF inhibitor to a couple of ON-only inhibitors, one RevMed's 9805, their G12D and the other RevMed's 6236, which is their RASMULTI. And we compared it in a couple of ways, but we used typically the highest concentration that RevMed has used in their in vivo models for each of the two agents to give it the best chance of full activity. On the left is KP4, which is a KRAS G12D pancreatic cancer model. And you can see that we treated with 7375 compared to 9805 or 6236. And for the first 9 or so days, the agents look similar. We continue dosing all of the agents and the RevMed agents let go. So less durable response, whereas with VS-7375, we're really -- it's really striking to me that it's able to maintain for quite a while. The antitumor efficacy, we're currently doing some RNA seq and other translational studies to better understand why the RevMed compounds are letting go as ours is holding on to the tumor growth inhibition. And similarly, on the right side of the slide, you see a G12D colorectal model. And in the bar graph on the right, you see individual animals. And you can see, for example, where it says maximal dose. We used 100 milligrams per kilogram of 7375, which is the highest we go with that agent. And we use maximal doses of 9805 or 6236 from RevMed. And again, you see really deep regressions with 7375 in this model, which are really impressive relative to what the ON-only inhibitors do. So I'm very excited to see as the clinical data play out, whether we can see that efficacy advantage, relative to the ON-only inhibitors and also the OFF-only inhibitors. And this slide, I'm showing mainly to set up how we should think about the clinical PK that Dr. Hong will talk about in a moment. So what we did here is we ran 4 different xenograft models, 2 pancreatic and 2 colorectal. And we're looking at different doses. And you can see in green at 30 milligrams per kilogram orally 7375 give strong tumor regressions in all the models we've tested and impressively in pink, at 100 milligram per kilogram, we see partial responses, if you will, 30% or more reduction in every mouse in every model we've run. So we had thought this is before we had seen any of the clinical data from the China trial that was presented today. We felt if we can get essentially the equivalent of green or the 30 milligram per kilogram dose we're doing well if we can get the equivalent exposure to the pink, we're doing extremely well. And what we found in the China trial was that when we correct for protein binding, we get the equivalent exposure of the 100 milligram per kilogram dose that's giving these very strong responses. So as you know, Mirati BMS' G12D fell out for lack of oral bioavailability. These are relatively large molecules, 650 molecular weight or higher. So it was really a challenge as we selected this particular compound with GenFleet to make sure we had really excellent oral availability across species. And I think this might be my final slide. This is a colorectal model, where we're looking at the combination of 7375 with anti-EGFR, cetuximab. And I've been running preclinical models my whole career. I get less and less impressed with them as time goes on, but this one really did impress me. You can see that all 8 mice had a complete response with the combination of cetuximab and VS-7375. We stopped dosing on day 28 and the complete responses persisted. So we will move in our clinical program very quickly for colorectal to the combination with cetuximab. So with that, it's my great pleasure to introduce Dr. David Hong, I can't think of anyone better to represent the RAS inhibitor space. Dr. Hong is Deputy Chair of the Department of Investigational Cancer Therapeutics, the Phase I program at MD Anderson Cancer Center. David?

All right. Thank you, guys. Thanks for coming. All right. Let's go through this. So KRAS space is pretty big. We enrolled 1,500 patients last year in Phase I trials. I'd say 35% of my patients have some kind of RAS. And somebody asked me, you have a lot of RAS trials up going. I don't have enough RAS studies going because I have so many patients. If you look at D itself, the most common mutant allele, you can see here in pancreatic, it's 37%, colorectal 12.5%, endometrial 8%, non-small lung 4.9%, but there are other rare tumor types that are not listed here, like jejunal, small bowel, cholangio, there are subsets of head and neck that are almost 100% D. And if you take all of those together, that's 61,000 patients a year. It's a lot of patients. So you all saw the GenFleet data. And I think when the abstract came out, it wasn't a whole lot of initial patients. I think it was 11 in non-small lung and 11 in PDAC, but you see the data now. And it's definitely impressive. I mean especially the PDAC data, 23 patients, response rate -- confirmed response rate of about 52%. And you can see here, based on the different dose levels, all the way from 400 -- sorry, from 300 BID, 600 daily and 400 daily. And I think they're moving forward with the 600 daily here. But it looks good. I mean -- and I'm sure I'll get questions about comparisons, et cetera, that's always hard to do. But you can see here, in the list here on the very bottom, the number of prior therapies, there are patients up to 4 prior lines of therapy, et cetera, all looking like they're having a relatively robust response. Too early, too early to figure out PFS yet, right? And that's an important metric, right? Too early to obviously a cut off for overall survival, and that's an important metric, but we'll see that this initial response rate looks very encouraging. The adverse events here. One, obviously, contrast to what's been out there, particularly RevMed's 6236 is you don't see rash, right? There's no real rash. But there are other what are probably expected side effects with these pan ON/OFF inhibitors, which are the GI side effects, some diarrhea, nausea, vomiting, et cetera. Some of you guys have asked me this question, how much of this is related to the actual class of drug versus and may -- some of it may be pharmacological, whether it's pill burden, whether it's certain types of coating on these pills. I'm sure we'll figure this out as these trials advance a little bit more. And what's a little bit more interesting here is the neutrophil, they did see some neutrophil, Grade 3 neutrophils, but neutropenia is not something that we don't know how to manage. We all know how to -- oncologists, we all know that's standard fare in oncology. Here's the PK. I think the one thing that we learned from Mirati's 1133 is PK matters, okay? PK matters because I will tell you I had many, many preclinical scientists tell me when they were running these trials of Mirati 1133, you're going to cure patients, David. You're going to cure all these pancreatic patients. This drug is amazing in preclinical models. And I put it in patients, we didn't cure anybody nor did we really see any significant responses. So the fact that they can actually get this in human -- this is human data, right? This is human data, I'm assuming, right? Not animal data, that's important, right? You see the steady state effect here, which is way above this phospho-ERK IC50. You see that in the Cmax, Cmax trough which we think is probably a very important metric and then you see the AUC. Likely compatible with once-a-day dosing, patients love once-a-day dosing. They hate taking their pills more than once a day. And so all this suggests that the 600 milli QD, which was recommended by -- well, but we'll see -- we're going to -- there may be some differences as in Asian and Western population, but we'll see how this goes. Okay. Here's the trial design, very simple, straightforward. Let's try to move this as fast as we can. Full disclosure, I'm actually involved in the trial, and we just activated today, but we're going to get patients on this month. So you can see here, dose level 1, 2, 3, move from 400, which they saw activity in, up to maybe 900, then move forward very quickly to second line and second line non-small cell lung patients and then move into the combination, which I think will be important because we don't know yet whether or not like the RevMed can actually combine it with cetuximab. I don't -- I'm not involved in those trials. We'll see what hopefully they can, but we don't know. I mean we will see where this all goes. But this is a good straightforward simple Ford design, and we will accrue rapidly. I know I will. So there is a high unmet need, okay? I just -- I want to reemphasize this again. 61,000 patients a year just in the United States, KRAS G12D, diagnosed, not just in pancreatic, lung, colorectal, other solid tumor types. Unmet need, huge market. The initial Phase I is very encouraging, the pharmacokinetics, the responses, et cetera, proof of concept. And what -- part of the reason I'm also involved with them is because I know they're thinking of combinations beyond just a single agent, right, beyond just the standard of care, right? And you saw the data from the ramp. I mean, could this -- there's clearly -- I mean, there's clear rationale to combine it with FAK inhibitors. There's clear rationale to maybe combine all, triple, we'll see, but that is important. There are some key questions we talked about. I think I've talked about to some of you all. This shows proof of concept that these ON/OFF inhibitors can work, right? In the world of kind of RAS development right now, there are these pure OFF inhibitors, ON/OFF inhibitors and these pure ON inhibitors. And the ON/OFF inhibitors are right now the most frequent type of RAS inhibitors in development, but they're all still in dose escalation. But if this GenFleet data is -- I'm assuming it's real, then it really shows that these ON/OFF inhibitors can work. One question that we don't know yet of any of these RAS inhibitors because they haven't done the experiments is how brain penetrant are they. And that's going to be an important question in lung because you talk with any lung doctor, after they see a response rate, they'll ask you, does this go to the brain because about 40% of those patients, 45% of those patients will get brain metastases. Okay. So hopefully, this thing will -- and I'm sure you guys will at some point, see what activity it is and then maybe have a cohort that allows for untreated brain metastases, but we'll see where that goes. Okay. That's me. Thank you.

Thank you, Dr. Hong and Dr. Picozzi for sharing those data with us this morning and importantly, for providing your insights and interpretation. My name is John Hayslip, I'm the Chief Medical Officer at Verastem Oncology, and I'll review with you now our clinical development plans and time lines. Acknowledging that we're earlier in our journey in pancreas cancer than we are in low-grade serous ovarian cancer, for example. I believe avutometinib and defactinib is showing the potential to reshape the expectations in metastatic pancreas cancer. Today, most patients with metastatic pancreas cancer, they do not expect to achieve deep and durable responses to their treatment. But with the data that Dr. Picozzi shared with us today, I see that there is potential to reshape expectations for patients with metastatic disease. Our clinical focus right now is moving forward rapidly in the RAMP 205 trial. Dr. Picozzi I think, reviewed with us that we've expanded at dose level 1. We're focused on enrolling additional patients to expand that cohort out to 29 patients in total. Our team is planning now for a randomized trial to initiate in '26. We're excited about the potential to evaluate at least two different indications in metastatic pancreas cancer. Of course, newly diagnosed patients with metastatic pancreas cancer and also importantly, the potential to see what we can bring to patients with borderline resectable pancreas cancer. There may be no indication more in need of a new novel therapy with high rates of deep response in advanced pancreas cancer. We're planning and working on further regulatory interactions to align these plans, and I look forward to sharing more about that in the future. And as also mentioned, we'll continue to evaluate additional novel combinations to bring forward in pancreas cancer in the future. Based on the results shared this morning by GenFleet Therapeutics and reviewed for us by Dr. Hong, I am incredibly excited by the high rates of objective response that was presented for the 7375, our KRAS G12D ON and OFF inhibitor, also impressed by the nonclinical or preclinical findings regarding depth of response and how that translated to durability of response in those models. I'm very excited to see updating clinical data throughout the year. Our focus here at Verastem Oncology is on activating the study in the Phase I, Phase IIa trial. Very happy to announce that we've actually initiated 3 sites here at the end of May, and those sites are actively screening patients right now. We're initiating the trial at the 400-milligram dose level on the 7375 study. The 400-milligram dose level is a dose level that was just shared this morning with a monotherapy activity for patients with heavily pretreated recurrent resistant metastatic cancer. As reviewed, we have planned expansion cohorts in pancreas cancer, non-small cell lung cancer, in the combination with cetuximab in colorectal cancer previously treated, which Dr. Pachter reviewed with us has shown complete responses in all of the mice tested to date. Additionally, our team is evaluating more cohorts for potential activation in the future. We're actively targeting newly diagnosed pancreas cancer, newly diagnosed lung cancer and additional indications such as KRAS G12D mutated such as endometrial cancer, building on our strong focus in gynecologic oncology. So with that, I thank you for your attention. I'll turn it back over to Dan Paterson, our CEO.

Thanks, John. So as you can see, some very exciting data today to build on our recent launch, exciting year for us, really, I'd say, a transformative year for Verastem in 2025. So with that, I will leave it and we can take a few questions, I think we get a little time, for any of the panelists.

We'll start here because I know you had your question before.

Michael Schmidt with Guggenheim. I had one for Dr. Hong actually. I know you've presented the 9805 data for RevMed last year. Yes, just curious how you think the 7375 data compares? And do you think the tolerability can be managed at this 400-milligram QD dose?

Of this drug?

Yes.

Well, I think both drugs look pretty good. It's hard to do -- the #1 thing that all clinical trials will tell you, it's hard to do cross-trial comparisons. And that's what I'm going to say. But both drugs look good. I'm not going to dismiss 9805 results because, obviously, you know the results. And it looks like the 400 seems tolerable. I mean we'll see what happens. There could be some differences between Asian and Western populations, but I don't expect necessarily any change. But both drugs look good. And Jon and I have talked about like how do we -- how do they -- and they have an opportunity, I think, to move very quickly. They have an opportunity to do combinations that RevMed doesn't have. They don't have a FAK inhibitor. They don't have these other things. So we'll see what happens.

And then yes, I mean the drug is obviously very selective, right, but is it more selective than other D inhibitors? And so I think because of that, some of us were a little bit surprised by the GI AEs, I know they were mostly grade 1 and 2. But is it just idiosyncratic as Dr. Hong mentioned perhaps or is there something...

I can -- maybe John can come on that. But I think GI toxicities can be very difficult to tease out exactly what are the causes, right? So they can be related to the actual mechanism, which I would argue -- I will probably say I'm just going to say all oral agents to have some extent, whether RAS or what other therapy have some GI toxicities, mostly diarrhea, mostly nausea, vomiting. Some of that, though, is like -- is complicated by, especially during the trial, things like pill burden. Like Mirati's adagrasib, for example, before they reformulated it, they had a lot of nausea, vomiting because the -- somebody who's taking 12 pills a day in one sitting, right? So I don't know what the exact pill burden on this is. There's also other aspects of the actual pill, not necessarily the mechanism such as you can enteric coat a pill, which makes it much more tolerable. So all these things kind of come into play. And so I guess John can comment on it, but I don't necessarily think it's just purely the RAS inhibition.

What I was going to say is when I think about side effects, but particularly GI side effects, I think, not just about the drug with the disease and the patient. So for example, some patients have had pancreaticoduodenectomies, some patients have other GI pathology. Some patients have other issues that -- or symptom -- or other important symptoms that require management. So it's a very complicated thing to associate a particular symptom with just one factor. It's often multifactorial.

And I was just going to add that the switch II binders have a common kind of structural component and so I think you're going to see some things. I don't think that KRAS G12D is expressed in normal tissues that's mediating what you're seeing. So it is structural base rather than target based. And I think the similarities to that adagrasib make sense from that point of view.

And we've already discussed some strategies in the Phase I to mitigate that. It might be different than how the patients were treated in China.

As Dr. Hong reviewed, in this trial, we'll be testing patients with food. So they'll take their medicine with food. The data presented, of course, this morning comes from the first-in-human study of this medicine. And so in the initial patients dosed, all patients were treated in the fasted state. They did not know when they initiated the trial, whether they would be really any nausea or diarrhea. And so, the results that we reviewed this morning come from patients who had not received any prophylaxis or prevention. And also remember, when you read an adverse event table, it's showing you the table as to whether patients had any adverse event even for 1 day. So a patient who had nausea for a day and it does not the next day, still is counted as a person with nausea. So is this -- this has not been a reason for discontinuation. The drug has been -- to our appearance, it looks to be tolerable in that study. And we're optimistic that by evaluating in the fed state, evaluating by patients who initiate, they start the treatment with some nausea prevention with the opportunity to come off of that if they're doing well, we may see an even improved adverse event profile in the future, and we're excited to see what we can see.

Yes. Most oral agents when you give it in the fed state, the nose usually gets better. That's my experience.

This is Clara Dong for Kelly Shi from Jefferies. So my question is on G12D. So from the ASCO presentation, we saw there were a few colorectal patients enrolled in China Phase I. So since I don't see the CRC activity data specifically broken down. So just wondering if you maybe can qualitatively comment on what you see to support you moving forward with the CRC EGFR combo study besides the preclinical combo data you presented.

Well, I'd say, first and foremost, that amazing preclinical data, but I don't know if John or Jon, you want to comment on.

Yes. I can say a bit. I mean, I think that we just had an approval of our FAK inhibitor, which doesn't show a lot of single-agent activity, but combined with avutometinib, it's clear that the two together really got over the finish line and the FDA accepted the contribution of parts. I mean that's an example of -- in colorectal, I think it's been a real challenge for KRAS inhibition alone to drive -- to be sufficient. But across this -- the MAP kinase pathway, BRAF inhibitors, et cetera, there are examples, and also G12C inhibitors previously and where combining with anti-EGFR greatly increases. So I don't think that you need a huge response rate in colorectal as a single agent to predict that you could have success when you combine with an anti-EGFR.

Yes. I mean, most of the -- even with -- these are the reported data from 6236 response rate where the pan-RAS inhibitor is probably around 9% in signal agent in colorectal. They didn't have a whole lot of patients either. I don't think any of us, especially GI docs really expect more than single-digit response rates in either the pan-RAS or even the D-specific cohorts with these molecules unless you combine it with cetuximab or with panitumumab. And the fact that they don't have any rash is key, right? Because you're going to have -- with the pan-RAS inhibitor like daraxonrasib, the big question is can you dose combination to the full dose of cetuximab and to daraxonrasib without really threading that therapeutic window, and we don't know yet.

And also just wondering if you could elaborate a little bit more on the 400-milligram daily dose to start in the U.S. Phase I trial because -- and also I just want to confirm, are you sticking to only 400-milligram dose because you presented both 400 and 600-milligram for efficacy and safety.

The 400 is the starting point for dose escalation. So the advantage of having the Chinese data is we didn't have to go back to 100 to start. We're starting in an effective dose range and escalating from there. Now one of the things we'll explore is there are probably differences in how the regulators look at your chosen dose given Project Optimus. And we're going to have to make sure we nail that before we move on.

Jeet Mukherjee, BTIG. Obviously, just still on the topic of the adverse event profile we're seeing with the various agents here. Let's just say, assuming equivalent efficacy, what adverse events from these agents becomes a nonstarter or a limiting factor for you in the clinic?

Of the current profile? Really nothing. I mean, like nausea, diarrhea -- I mean the neutropenia rate was very, very low. So I doubt and I don't recall exactly at what dose levels they had the higher levels of neutropenia. But these are standard toxicities that -- because you're not really talking to me. You're talking about the community oncologist who's going to prescribe this thing, right? These are standard side effects that community oncologists all know how to deal with.

Understood. Helpful. And at least among the lung cancer patients presented today, did any of them have brain mets or were they excluded from the trial?

I don't know.

John, do you know?

My guess is most Phase Is, we exclude -- we essentially -- the eligible criteria for most Phase Is is that they either cannot have brain mets or they have to have treated brain mets. That means they get XRT and then they have to be stable for a period of at least 4 weeks before they get enrolled. And by definition, that there's -- the tumor is dead, right? So most of the -- at least the trials that I've run, most of the Phase I RAS studies have excluded brain mets at this time. And they usually do that experiment at a later time because they don't want to -- if you have -- even if you have a small brain met, it can rapidly grow and so you don't want to have these patients fall off while in the middle of the dose-limiting toxicity window. So my guess is that at some point, we'll probably in the expansion, we'll probably try to see if that's the case, right?

Got it. Helpful. And just lastly, to the frontline panc data seems standard of care provides about 5.5 months MPFS. What improvements or delta on PFS would be considered clinically meaningful here?

Okay. I wanted to comment a little bit on the first question you asked, and that is to emphasize two things. One is that, as Dr. Hong mentioned, managing GI toxicity to a GI oncologist is one of the, if not the most, aspect of supportive care that they're most experienced with. But I also want to add that you have to remember this from the patient's point of view, these patients have desperate prognosis. And many times, they're going to do darn or anything if they can prolong their life. And so there's a greater tolerance for this sort of thing than there are in many other in oncology. Now with respect to the PFS question. Actually, oncologists look more at overall survival and progression-free survival. And the reason for that is that progression-free survival is sometimes hard to gauge radiographic such as CT indications of response are infrequent given the natures of disease and somewhat difficult to assess. And it's turned out there was a study done some years ago that was presented at the ASCO GI that looked at 20 years' worth of studies in pancreas cancer. And there was a very low correlation coefficient between PFS and OS. And a good example would be the PANOVA-3 trial I presented on Saturday, where there wasn't a clear benefit in PFS or local PFS seen, but there's a clear benefit in OS. So again, oncologists look more at OS than PFS.

Kalpit Patel, B. Riley Securities. Maybe one for the KOL. So there's been some ongoing discussion about patients who are diagnosed as locally advanced versus metastatic and what their prognosis is like over time. And we saw here that almost all of the patients were metastatic at diagnosis, whereas RevMed's data about half were locally advanced and half were metastatic. So I'm curious, does it matter after you have a median of two priors, your outcomes or your response to these therapies in the advanced setting.

Do you want to start?

Okay. I'll start. So as you go through different lines of therapies for curable pancreatic cancer, I think the difference between locally advanced and metastatic narrow. But still, patients with locally advanced disease would have some survival superiority over metastatic disease. And so therefore, when I read trials, the percentage of patients that are locally advanced versus metastatic does matter. That would be my answer. Dr. Hong?

I mean, locally advanced is pretty bad still. I mean as you mentioned -- and as you mentioned, these patients with locally advanced tumors. I mean you can't resect them with the Whipple, et cetera. Invariably, I mean CT scan, if you think about it, is a very blunt instrument, right? It's not like we can -- I can't -- on a CT scan, I cannot see usually a lesion that is bigger than a centimeter. And these patients -- a lot of these patients, locally advanced, will probably have studying on their peritoneum that we cannot see by CT scan. So locally advanced is -- he's is the -- Dr. Picozzi is a pancreatic expert, but locally advanced is pretty bad.

Yes. Actually, locally advanced patients, if they're not laparoscopically staged, I mean literally, you go in the abdomen and look, will have occult metastatic disease about 1/3 of the time. So there is overlap between the two.

I think we have one more over here.

Eric Schmidt at Cantor. Maybe 2 questions. I'm curious about your decision to dose escalate from 400. I know it's not your data, but when you look at the China data, it looks like the PK for 400 and 600 is overlapping and then you're way above phospho-ERK inhibition levels at Cmax and Cmin. So why not go lower? That's number one. I know Dr. Hong thinks PK matters. And then two, could you explain the two Grade 3 liver dysfunction abnormalities that were observed in Phase I. Again, knowing it's not your data.

Yes. So glad to take that. Thank you. So regarding the first question, the data shared today represent the data as of the data cut on May 16. So the data shared today will include some additional observations beyond the data, I believe GenFleet had in hand at the time they decided to expand at the 600-milligram dose. So I won't speak for them, but -- so the data have evolved, there's more data than what they have when they made that choice. I think as Dan mentioned, and it's important just to be clear about where we stand at Verastem, so we have -- we're developing 7375 outside of China. So we have United States, Europe and other locations. And so we will continue the development program and work to evaluate what is the optimized dose and schedule of 7375. Regarding the design of this trial, this is an oral agent. And so we are excited to boldly initiate the study, and we are treating patients at a dose that we have seen clear efficacy in the study coming out of China. In addition to, of course, the more blunt dose level evaluations of efficacy and safety, we'll be doing all of the standard exposure response, exposure response for safety, exposure response for efficacy, and if the data lead us to suggest that other dose levels would be appropriate, we'll be right there to do that. The Grade 3, so as you mentioned, that's not my data nor my study, but I would say I think what I can share is that both of those cases, we regard as highly confounded and don't impact our starting doses or how we'll be evaluating in our trial.

I know we had one question from online. Maybe I can give to the KOLs here. Can you put the RAMP 205 data in context with other data at ASCO for experimental agents looking at first-line PDAC.

Okay. I think that's an excellent question. And I would say the data is actually very competitive. For example, on the -- in part of the GI oral session, there was an examination of a GSK3B inhibitor in association with Gem/Abraxane and the data that is being displayed here with the avutometinib, defactinib combination is significantly -- going to be significantly superior to that, for example.

All right. I think we'll wrap it up. Thanks, everybody, for your attention. Thank you.

Thank you.

Thank you very much.