Okay. Welcome, everyone, to our next fireside chat here. I have the Verastem team, Ting Wu and Jonathan Pachter. Thank you both. Thank you to both of you for joining us today. To kick it off, maybe Jonathan or Ting, we have seen a lot of updates in this space lately. You guys had your own comprehensive update for the G12D. Ultimately, the number one question I get is, maybe there is too many fish in one pond here. Where do you guys see yourselves differentiating in the RAS landscape? What do you think investors are still missing about your molecule following the comprehensive update that you just had?
Yeah, I think the differentiation, we feel that for a patient with a KRAS G12D-driven cancer in 2026 or 2027, they deserve an agent that is selective for that variant, and we think that ours is best in class among both G12D and pan-RAS for fulfilling that role. There are several properties of our drug that really stand out. One is being dual on/off really potently, and I think others claim to be that, but I think a problem with some pan-KRAS, for example, is they tend to be off-selective and therefore do not have activity. So I think the dual on/off, we have a long residence time or slow off rate so that, for example, if you see the grade 1, grade 2 GI adverse events, you can stop dosing, but you are still covering target because of that long residence time, which is valuable.
Ours is extremely G12D selective, and KRAS G12D does not mediate any normal function in the body, so that really gives you great tolerability. We do not see rash or stomatitis for that reason. But I think the thing that is probably underappreciated is really simple, is in the RAS space, I think most agents fail to maximally hit the target, and I think that that is for one of two reasons. For some agents, for pharmacological reasons, you dose up and you hit a ceiling, and you cannot keep getting higher exposure. I think zoldonrasib might be an example of that when you look at their PK. Then for the pan-RAS inhibitors, it is more toxicity limited that you just cannot dose to fully hitting the target, and that has led some to combine a couple of different RAS inhibitors of various flavors to try to deal with that.
I think that we can really maximally hit the target, and when you look, for example, at knocking down KRAS G12D variant allele frequency, we clearly do that much better than others, and it is reflected in the clinical data as well that we shared that Ting will talk about more. So I think there are many properties that are really best in class about our agent.
What is misunderstood or not picked up? I think a couple points. One is I think the world is moving toward that appreciation that even though pan-RAS or pan-KRAS, if there was ever one good enough, could be a larger market, I think that the world is really moving toward variant selective. I think if there is a variant driving your cancer, you want to be something really selective for that. I see that shift in the last six months. People are appreciating that.
People are thinking, "What's the best G12D selective?" as opposed to, "How would you ever compete with the pan-RAS space?" The second thing that I think is maybe misunderstood or not thought about as well as it should be is the value of single-agent activity. I think that many of the RAS inhibitors, if they have insufficient efficacy on their own, they might obfuscate that by combining very quickly and showing you what the combination does with EGFR or with chemotherapy, and I think it's really about your drug. I think that for durable response, for example, in pancreatic cancer, you can't only rely on the combinations.
I think that, for example, when you treat with chemotherapy, you might, for the first two or three cycles, you might give full dose chemotherapy, but then you're dropping out day eight of day one, eight, 15, you're dropping the doses, and then by four or six cycles, you're not giving chemotherapy anymore because it's not tolerated. So, it's your drug underlying that you really have to depend on, and we're very combinable, so we are doing those combinations, but I think that as you're comparing between drugs, it's really important to understand what the single agent does as a starting point.
Got it. Okay. One of the more common questions we've received for you guys is the 600 milligram dose versus the 900 milligram dose, the choice of going towards the 900, whereas your partner GenFleet has not done that, or I don't think is planning to do that. Is there any evidence, I guess we have some empirical evidence so far, but anything that suggests that the 900 is going to be materially better than the 600?
Yeah. Thanks for the question, Kalpit, and I think we can look at it in three ways. The first one is, of course, the clinical data, and the disclosure of the data we have disclosed is that there's a clear dose relationship between the 600 and 900 milligram. The dose response is pretty clear. I think that's the most, I would say, direct evidence. The second thing is about the PK, because the 600 and 900 PK is dose proportional. It's a meaningful increase from 600 to 900 milligram, give you more confidence that there's a different exposure that leads to the dose response. The third point is about the safety and tolerability, is that the tolerability profile for 900 and 600 milligrams are identical.
Mostly the doses are limited by safety or tolerability, but we don't see a difference. If you see a dose response, you should pick the higher dose. Those are, I would say, evidence at this point that give us confidence that 900 is a go for a dose. Regarding GenFleet, I think it's more about they haven't fully characterized 900 mg. As you probably noticed, they only dosed three patients at 900 mg and did not fully explore that dose and decided to go for a speed for 600 mg.
This is not the first time where in different countries there are different approval doses. Just an example, lenvatinib has a lower dose in Japan, as an example. I don't think that's kind of unique, but it's more like, okay, did you fully explore the dose? If there's any kind of racial differences there that we need to continue to explore further.
It's also intriguing when you look at the waterfall plots with our RAS inhibitor and different RAS inhibitors. If, again, to my point of wanting to maximally hit the target, if you are submaximal, you look at the left side of the waterfall, you see tumor growth is the best response for some patients. I think at 600, we see that sometimes. At 900, across indications, we never see that. Our waterfall's only full. That's also showing the value of 900 versus 600 as being the right dose.
What would be a good outcome for you guys for PFS and maybe ORR and PFS difference between the 600 and 900 doses?
Yeah. Firstly, I believe that we have sufficient data to support 900 milligram as a go-forward dose. I frame it as less of a comparison, but what 900 milligram would need to achieve. I think the clearest benchmark, probably in PDAC, because there are RAS inhibitors out there for standard of care in G12D. I would say something that we demonstrated, like 40% north of a single-agent activity in a previously treated, and a PFS probably north of seven to nine months is what we are aiming for. I think one of the things is that we continue to observe is that our datasets, that as data mature, I think we are moving towards that kind of goal that we want to achieve. For the 22-patient dataset at PDAC 900 milligram that we disclosed, we will have a follow-up on PFS in future medical conferences.
Regarding 900 milligrams, it is important to emphasize that 900 will be our go-forward across indications, across cancer indications. We are probably getting to the point where we are hoping across combinations as well. We have DLT cleared 900 with anti-EGFR. We have DLT cleared it with gem nab paclitaxel. As to the combinations with chemo or chemo-IO for frontline colorectal and lung, we are in the process of clearing those. Ultimately, our expectation at this point is that 900 would be well tolerated with those as well.
Okay. When looking at safety, one of the other common questions we have gotten is the China dataset safety seemed a little better than the Western dataset safety, if you look at the publication from GenFleet. Any comments?
The other way around, right? The U.S. seems better than China?
Oh, sorry. U.S. seems better than China. Yes, correct. Any explanations for that? Anything you want to add to what you have already previously communicated for that?
Yeah. Certainly, I feel like comparing across different standard of care and geographic regions is always a bit challenging. However, I think a few things that we have done slightly differently on the GI is that we ask the patient to take antiemetic prophylaxis. We manage these AE more proactively. Most importantly, we have learned a lot from our investigators and patients on how to manage these AEs. That will help with the GI AEs that you have seen. The other thing is about, I think the two points is the hematologic AE and also the liver function test abnormality. Starting with hematologic AE, in the U.S. population or I would say ex-China population, we consistently did not see any meaningful hematologic AE.
This is not just, I would say, just looking at AE table, but looking at the raw data, we haven't seen that trend. The most evidence that we can demonstrate is that we are able to combine with full dose of chemotherapy across indication that include non-small cell lung cancer, PDAC, and also we're starting for CRC. I would say neutropenia or anemia or thrombocytopenia is not an issue for this combination. I think the combinability speaks themselves of lack of hematologic AE. The other thing is the liver function test abnormalities. Again, we look at the raw data very closely. We also have enrolled, I would say, a sizable number of non-small cell lung cancer patients where immune checkpoint inhibitors are standard of care in the prior lines.
We didn't see that trend that it kind of stimulates these liver function test abnormalities, reactivation via irAEs in theory. We really don't see those. Again, we are able to combine with either single-agent pembrolizumab or the frontline like quadruplet, what was it? Pembrolizumab, pemetrexed, and carboplatin. We don't see those creeping up in terms of liver function test abnormalities. So, I believe that these are probably of reasons that we don't see on paper, probably co-meds, probably like background higher, I would say, HBV prevalence, hepatitis prevalence. Those are possible explanations. But I think focusing on our U.S. data, we have evolved a lot. We have collected a lot of data across indications and also across time frame. So, we have confidence that these are not an issue in the U.S. population.
I think Chinese traditional medicines are also a huge factor that's not really documented very well. I think there can be drug-drug interactions with them. When we've spoken to Chinese companies, they've said that as well. I actually, at ESMO, I made a point when I talked to clinical development leaders from other companies, I asked, that have global studies, they often, even outside of the context of RAS inhibitors with other agents, they see poor tolerability in China in the same study relative to others, and the Chinese traditional medicines may be part of that.
Yeah. Makes sense. Maybe moving on to some of your combination-based regimens that you guys are targeting for TARGET-D 301. You're including both a chemo-based regimen and then a chemo-free regimen. What's the rationale for doing both? I think one of the more looked at regimen, at least from our inbounds, is the cetuximab combo, right? Why do that combo? Why not just go straight to a pan-RAS combo or something else? If you can give additional rationale there, that would be useful.
Yeah. It's interesting that for colorectal, I think everyone accepts that anti-EGFR is the right combination. I think it's from BREAKWATER with BRAF, that clearly played out, and with G12C inhibitors in colorectal, that played out well. I think that in pancreatic, that's newer. I think others are doing that, but I think we've really been very excited about that. I think pre-clinically, there was a nice paper in February from Channing Der's lab where they knocked out each gene in the genome one at a time in pancreatic cancer cell lines, CRISPR-Cas9.
And they said, "Which target, if you knock it out, best facilitates the activity of RAS inhibitors in pancreatic cancer?" One of the number one hits was EGFR, and they found that in all cases when they treated pancreatic cancer cells with a RAS inhibitor, you see an upregulation of phospho-EGFR. So there's good pre-clinical rationale. We've also got nice data where we look at cetuximab with our G12D VS-7375. We see regression when we combine the two in pre-clinical models. And Ting could speak to what we're seeing clinically, but it's also bearing that out.
Yeah, certainly. I think aside from the preclinical models that demonstrate strong synergy, in the clinic, we have seen responses at suboptimal doses or subtherapeutic doses in PDAC. The other thing is that even at slightly higher doses, we have seen faster and more deeper responses compared to monotherapy. We are able to show some preliminary comparison in future medical conferences, even before our 201, the PDAC main cohort reads out. I think we will be able to share some evidence to give everyone some confidence that cetuximab is adding to either the response rate and also some of the durability.
Then on October 14, after our update, a few hours later, GenFleet, our partner in China, gave an update as well. They showed that in the frontline pancreatic cancer with a combination of cetuximab and the same G12D inhibitor, they saw a 71% response rate. I think it is really, in frontline, everyone is going for 60%-80% response rate. I think you also want tolerability to make that sustainable. I think we have made a decision not to combine with FOLFIRINOX because it is so toxic, and with RAS inhibitors, especially ours being as active as they are, we do not see the need for that.
So, to your other question about why gem nab and the anti-EGFR, I think the world is ready for a non-chemo regimen. I think that if the combination with anti-EGFR can deliver the kind of efficacy that you can have with chemo, then let us leave chemo out. But I think we have two shots on goal there. I think that gem nab is well-tolerated. We are seeing that, and we will see which one wins, and maybe both win.
And probably last point to John's, I would say aspiration or the field's aspiration of going to a non-chemo regimen. It was also, you mentioned about the pan-RAS combination. I think we are very open to novel combinations in the future. It is more for this phase III study, we believe providing a chemo-free option is good for patients, and we will continue to explore those novel combinations in the future as the next waves of therapeutics.
With pan-RAS, we and Erasca are each doing experiments combining the two. Their pan-RAS, which we think could be best in class, and our G12D, then next year, I think we expect to start the clinical trial combining them. We are also in advanced discussions around PRMT5. I think pre-clinically, that's really exciting, and certainly the data that Tango have shared are exciting. So, I think that's another combination that we think is near-term for us to combine with a PRMT5 inhibitor, especially in MTAP-deleted G12D pancreatic cancer, which we think could give a really nice all-oral frontline therapy that might give a really extended response for patients.
Would you ever consider a pan-KRAS?
If one ever emerges that has both good bioavailability, which I think is one reason why they've fallen out, and is not so off-selective that it lacks any single-agent activity, then we would certainly consider it. That would be quite interesting. I think we're all still waiting to see that, if anyone can succeed with that.
Just to be clear, for TARGET-D 301, either of those regimens can pass and the trial could be successful. You don't need both.
That's correct. We don't need both.
Okay.
Of course, we haven't showed our details of the statistical plan, but the plan is to either or can be positive.
Talk to us about the potential accelerated approval path, now that RevMed's in the market, at least in second line, part of first line. Curious to hear your thoughts on whether you see this as still being a viable path or first line all the way through and hitting PFS and getting there?
Yes, that's a very important question, and I'll start with probably the previous treated, the 2:1 study. First is that once FDA announced the multidisciplinary review document that daraxonrasib has a lower response in G12D population compared to their overall 30%, they have 24%. I think that really opens the door and also highlight the unmet need in the KRAS G12D population. I think if our data consistent to hold on what we announced around 40+% response rate, I think we have a fair chance for targeting AA approval in the previous treated setting. Because the rule for AA approval, I think there's several elements. One is that your response rate lower confidence interval needs to be better than standard of care to demonstrate that superiority. Second is that it comes with substantial DOR as well with that response rate.
And thirdly, there's also another term that if you can demonstrate favorable benefit risk, then it also will take into consideration. I think we were able to hit all three of these, especially we can demonstrate substantial benefit superior after OR, DOR, and with that lack of rash and stomatitis, I think we have a fair chance in the previous treated.
From a timing point of view, before we go to the front line, I think we've said that this will be fully enrolled, the phase II, in second line by the end of the year. That means that we could have data as early as the middle of next year to submit for accelerated approval. I think we believe that we might have the first G12D inhibitor that's approved.
And that's despite RevMed's drug being on the market, you're going to-
We've had no trouble enrolling.
I'm just saying, you're going to differentiate on efficacy or safety or both.
Probably both.
That would be the basis for the accelerated approval.
Yes.
Okay. Can we anticipate data next year, second half, or have you publicly commented on that?
Yeah, I think a few things, right? If you think about our development plan, we have these phase II studies, single arm in the previous treated population aiming for AA approval. I think for next year, the big milestones will be if we can do multiple SR approvals on these three studies. Along with that, we will disclose data once we align with approval plan, which means that for all three studies, there is a chance that we may be able to present at medical meetings.
To be clear, the three phase IIs that are ongoing are one in PDAC, one in colorectal, and one in lung.
On top of that, of course, people want to see combinability, want to see how your regimen in the front line will look like. So we plan to disclose frontline combination either in PDAC, CRC, and also non-small cell lung cancer in future medical meetings as well, to demonstrate we are up into the ORR even higher than the competitors. That is what we wish to achieve. Of course, back to the previous point that John made, I think these frontline combination, really high ORR, where they can translate into PFS, that is still a question mark, because when you lose the chemotherapy and you have a long PFS, then the rest of the PFS will need the single-agent activity to sustain.
So I would say a very data-rich 2027, and not just the main cohorts for the phase II studies, but also some frontline combination data we wish to share. There may be along the line some translation work as well as smaller cohorts, such as CNS cohort in lung. If we have sufficient data to support, that might be another consideration.
And continuing on your accelerated approval question, we should speak to the phase III as well in frontline and the role of Project FrontRunner in that regard.
Yeah. We haven't got to that. I think for the Project FrontRunner, this is not the first case in GI oncology. I think the recent precedents are the BREAKWATER BRAF V600E with a HER2 mutant gastric cancer. We're not doing something really novel. I would say just doing probably more novel in pancreatic cancer with this ORR endpoint. We believe it's meaningful that we can hit a meaningful higher ORR together with the DOR. It will open the door for a potential accelerated approval. Again, FDA stance is always data-driven, that they agree that we incorporate this endpoint, but whether we will get this accelerated approval in the phase III study will really depend on the data they see.
The fact that they've agreed to ORR and PFS as primary endpoints and not overall survival means that we can make it a much smaller study. So it's less than 360 patients, which we can enroll extremely quickly, the phase III.
Makes sense. I want to switch over to colorectal cancer for a second. You've had a 43%, sorry, not 43. You've noted that you've had some proof of concept data with VS-7375 plus the EGFR inhibitor. How do we know what the contribution of each might be in the combo, and what is ultimately your bogey that you're looking for to advance this?
In KRAS mutant colorectal, anti-EGFR is actually contraindicated. There's not efficacy. We don't need to show that. I think we expect that the RAS inhibitors alone all give about 10%. I mean, if you look at sotorasib in colorectal cancer, if you look at daraxonrasib in colorectal cancer, 9% or 10%. I think you need the EGFR. I think we are comparing, though, in the current trial, the single agent as well as the combination with anti-EGFR, so we'll see that directly for the contribution of parts. What were the other elements of the question?
When's the next update for that, and how many patients' worth of data?
We have disclosed six patient worth of data in our October disclosure, and those are in the 101 or the phase I study. We have moved our CRC patient into phase II because that's the original directed study. We do have more patients in the phase I, but not that many. In the total, we will disclose the data set around 10 patients around Q1 next year. We need some time for the data to mature, as these patients are dosed pretty recently. However, I want to mention that the TARGET-D 203, which is the second, I would say, phase II CRC study, has been enrolling extremely well. We have already 50 plus patients dosed in that regimen. Ultimately, I think that cohort will be most important, and we plan to disclose that cohort data probably around Q3 to Q4 in 2027.
Got it. Then switching over to lung, you showed a 43% confirmed response rate at 600 milligram dose, and now I think we're waiting for the 900 milligram dose there as well. What delta do you want for efficacy differences between the 600 and 900 lung?
Yeah, I guess, like, speaking with our KOLs, and this has been quite consistent, they probably want to see around, like, 15% of delta. I think it's not just the response rate, but also the PFS needs to be meaningful longer. Of course, we understand that our bar is quite high.
But to your point, it's not about the delta between 600 and 900. It's about 900 being best in class.
Of course.
Right? Just to be clear about that.
Yeah. I think that 15% means that not just with our agent, but also with, like, competitors. The 15% differences in OR, they will see as clinical meaningful, corresponding with a longer PFS. That is what we want to see. The other differentiation, not about the efficacy itself, but the CNS activity is also important. Because in KRAS G12D, there is 30% of patients with CNS mets. If we can address those patients, we will be fulfilling an unmet need. Especially for some of the comparator drugs, they are larger molecules in nature. We are a small molecule, which give us a better chance for CNS penetration. We are running that cohort in our phase II study, and hopefully should generate some data next year as well for CNS.
We have been told by KOLs in the lung space that if we hit brain mets, that everybody would choose our RAS inhibitor over others. I mean, that is really important. If you have brain mets, certainly you want to treat them, but you also want to prevent them from occurring. The other thing to highlight in the lung space is that some of our competitors have just made a point of not showing patients with prior docetaxel treatment, and for reasons we can only guess. I think we see that our efficacy, as we have shown, it does not matter whether you have prior docetaxel or not. Our agent is very effective in lung cancer.
Maybe one last one from me. You recently completed a financing. Walk through what that financing gets you through, what milestones, and do you have enough cash on hand to get through the phase III in PDAC?
Yeah, certainly. We raised $250 million, and we are able to have the pro forma cash around more than $400 million. That will carry us to first half 2028, and that allows us to complete all the phase II studies and get all three phase III studies up and running. That will carry us over, like, important inflection points for 2027 and give us more optionalities. I think most importantly, I want to emphasize that we are not dropping any indications. We are going full steam ahead for three large indications. We will be also looking at smaller indication and additional opportunities along the way.
Our speed to the phase IIIs is not at all dictated by the status of the phase IIs. We are moving as quickly as we can to get the frontline phase IIIs set up and started.
I think I have the answer to this question, but if you had to pick a favorite indication of all the programs you are running, which one might that be?
I think all, because, like, patient cannot wait, and there are a lot of KRAS G12D patients out there.
Yeah, we feel really bad for the patient that has a cancer that we did not name, so I agree with that. I do have a purple ribbon on, and you might know the color code, but that is just me.
Okay, wonderful. Anyone in the audience have any questions for Verastem?
Nate? Very good, Kalpit. Thank you.
Thank you. Thank you for joining us.
Thank you.