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

Good morning, and welcome to the Verastem Oncology Investor Conference Call on Monday, April 27, 2020. [Operator Instructions] Please be advised that this call is being recorded at the company's request and will be available on the company's website for a period of 90 days from today. At this time, I would like to introduce Mr. John Doyle, Vice President of Investor Relations and Finance at Verastem Oncology. Please go ahead.

Welcome, everyone, and thank you for joining us this morning. With me today to discuss the preliminary clinical data from the ongoing Phase I combination study investigating VS-6766 and defactinib in KRAS mutant solid tumors are: Brian Stuglik, Chief Executive Officer; Dr. Jonathan Pachter, Chief Scientific Officer; Dr. Udai Banerji, Professor of Molecular Cancer Pharmacology at the Institute of Cancer Research, Honorary Consultant in Medical Oncology at the Royal Marsden NHS Foundation Trust and Lead Investigator of the clinical study; and Dan Paterson, President and Chief Operating Officer. In addition, Rob Gagnon, our Chief Business and Financial Officer, is also in the room and will be available following the prepared remarks for Q&A. During today's call, Brian will provide some introductory remarks, Jon will provide a summary of the scientific rationale underlying our new RAF/MEK and FAK inhibition combination program and provide an overview of both VS-6766 and defactinib. Dr. Udai Banerji will then discuss the initial results from the ongoing Phase I study evaluating the combination of VS-6766 and defactinib in patients with KRAS mutant solid tumors, including data from the dose escalation portion and expansion cohorts in low-grade serous ovarian cancer and non-small cell lung cancer, all of which are being presented today at the AACR 2020 Virtual Annual Meeting I. Dan will discuss the market opportunity and next steps. After that, Brian will provide some concluding remarks, and we'll open the call up for your questions. Earlier today, we issued a press release reporting the clinical data from the ongoing Phase I study being conducted by Dr. Banerji in the United Kingdom. That press release is available on our website at verastem.com. Following the conclusion of today's call, a PDF copy of the slides being presented by the team today will also be available on our website. Before we begin our formal comments, I'll remind you that we will be making forward-looking assertions during today's call that represent the company's intentions, expectations or beliefs concerning future events, which constitute forward-looking statements for the purpose of the safe harbor provisions under the Private Securities Litigation Reform Act of 1995. All forward-looking statements are subject to factors, risks and uncertainties, such as those detailed in today's press releases and in our filings with the SEC, which may cause actual results to differ materially from the results expressed or implied by such statements. In addition, any forward-looking statements represent our views only as of the date of this recording and should not be relied upon as representing our views as of any subsequent date. We specifically disclaim any obligations to update any such statement. We refer you to the Disclosure Notice section in the press releases we issued today and the Risk Factors section of the annual report on Form 10-K for a discussion of important factors that could cause actual results to differ materially from these forward-looking statements. With that, I would now like to turn the call over to Brian Stuglik. Brian?

Thank you, John. Good morning, everyone, and thank you for joining us on today's call. We are really excited to share these data with you this morning as well as our thinking about the future development of VS-6766 in combination with defactinib, including future potential indications as well as the next steps in our anticipated registration strategy for this combination, which we believe represents 2 potential best and first-in-class compounds that could be made better through combination therapy and could be beneficial for a wide range of cancer types. Before we jump into today's program, I'd like to take a quick moment to provide some background about the strategy behind this new program and how it fits into Verastem Oncology. As we reported in our press release earlier today, the combination of VS-6766, our new RAF/MEK inhibitor, and defactinib, our FAK inhibitor, have demonstrated encouraging clinical activity against KRAS mutant variants, especially G12V and G12D. We believe with today's data, we have achieved clinical proof-of-concept in KRAS mutant low-grade serous ovarian cancer and that we have a strong foundation of evidence about the combination's activity in the KRAS variant G12D, which we intend to explore further. In addition, we believe that by pursuing low-grade serous ovarian cancer as our initial indication, we may be able to achieve a more rapid pathway to market for this novel combination. Beyond LGSOC, we believe G12V and other variants have blockbuster potential for this combination since 30% of all human cancers are driven by mutations in the RAS family of genes. In addition to this exciting new program, we're also well funded with a cash runway into the fourth quarter of 2021 and a strong supportive investment base. Finally, our marketed product, COPIKTRA, is actively generating revenues for the company and achieved a $5 million in net product sales during the first quarter of 2020, representing a 39% increase compared to the fourth quarter of 2019 and 194% increase compared to the first quarter of 2019. Next, we have a snapshot of our key pipeline programs. The VS-6766 plus defactinib combination study, at the top of this list, is the data we are going to be discussing here today. Study #3 in the list under defactinib is also a study I would like to highlight because data from that study will be presented today in a virtual poster session at AACR by Dr. Andrea Wang-Gillam. This study is important because it demonstrates the beneficial impacts of defactinib on 2 key areas: the underlying tumor microenvironment and a dense pancreatic stroma, one of the more challenging aspects of pancreatic cancer and why so few therapies have succeeded in this setting. This is also important as you will learn more later about our plans for the combination of VS-6766 plus defactinib in pancreatic cancer. The bottom half of this slide in green depicts the ongoing duvelisib development programs. As I mentioned earlier, duvelisib is currently marketed as COPIKTRA in patients with certain relapsed or refractory hematologic malignancies. And we are actively working to differentiate duvelisib's unique PI3K-gamma inhibition properties through development in peripheral T-cell lymphoma and other key combinations. In fact, all 3 of these molecules had abstracts selected for presentation at the AACR Virtual Meeting I. Here at Verastem Oncology, the impacts from the COVID-19 pandemic have been somewhat modest, but there have been a few key ways that it has impacted us, in particular Dr. Banerji's ability to present this Phase I data in a typical form at this year's AACR Annual Meeting. As many of you are aware, the in-person program was canceled, and instead the AACR is now hosting a series of virtual meetings. Another key area where the pandemic impacted us was that Dr. Banerji had to use a November 2019 data cutoff rather than something more recent due to physicians and participating institutions' time and attention being diverted to attend to COVID-19 patients in the U.K. And finally, while there wasn't an impact on patients being treated in the ongoing Phase I study, there was an impact on available personnel to clean and create the data reports and other administrative elements surrounding this data readout. Overall, we feel very fortunate to be reviewing these data with you today, and for Dr. Banerji to be presenting these data in a virtual format today as part of the -- this year's AACR meeting. With that, I will now turn the call over to Jon to discuss the importance of more complete RAS pathway blockade and the role of resistance for patients with RAS mutant tumors. Jon will also cover the scientific rationale for why we believe that VS-766 plus defactinib combination is well positioned to overcome the known limitations of current available therapies and potentially improve therapy in many tumors with high unmet need. Jon?

Thanks, Brian. There is a critical need for drugs that effectively block the RAS pathway. RAS is the most commonly mutated oncogene in human cancer. And as Brian just mentioned, 30% of all human cancers are driven by mutations in the RAS family of genes. The slide on the top shows examples of tumor types in which mutations in KRAS, NRAS and BRAF are prevalent. And patients with mutations in the RAS pathway often present with an overall worst prognosis. Accordingly, there have been intense efforts across the oncology research community, for essentially 3 decades, to effectively target this pathway. And although there has been some recent success, challenges remain. Blocking a single target generally leads to resistance and -- rather than a durable response. And additionally, tolerable combinations, for example, with MEK inhibitors have been challenging. Next slide, please. So with the objective of durable responses for patients, it is critical to understand mechanisms of resistance to current agents and address them. So I'll go into those now. On the left side of this slide, you see the RAS/RAF/MEK/ERK pathway that cancer relies on. Just to orient you, receptor tyrosine kinases, such as the EGF receptor, activates RAS to the RAS PTP state; RAS than activate RAF; and then RAF then phosphorylates and activates MEK; and then MEK phosphorylates and activates ERK. And because of these activating mutations, cancer becomes addicted to this pathway and doesn't give it up easily. So Channing Der and others have shown that blocking a single node in this pathway typically doesn't lead to a durable response and cancer finds a way around it. So for example, it is believed that blocking KRAS G12C or blocking RAF or blocking MEK or blocking ERK commonly leads to resistance. So why is this? Next slide, please. So this slide shows that one mechanism of resistance that Dr. Banerji and others have been looking at is that if you block RAF, you get a feedback activation as shown in green of FAK, and that's shown -- and you see that is activation of phosphorylated FAK at tyrosine 397. When FAK is activated, it can drive downstream pathways. Well, what Dr. Banerji has found is both in cancer cells and now in patients, if you block MEK, you'll also find that, that activates phosphorylation and activation of FAK as a resistance pathway. And once activated, it's been shown that FAK can then drive the ERK pathway itself as well as the RhoA and YAP pathways. And all 3 of these pathways are capable of driving tumor growth and bypassing the RAF/MEK blockade. So that's one mechanism to deal with. A second mechanism is shown with the green arrow on the left. And this is really work out of Neal Rosen, who has been a pioneer in this field. And what's been found is that if you block MEK, ERK no longer is getting the signal that it needs and the cancer reacts to that; ERK feeds back and activates RAF to restimulate and drive through the blockade in this pathway. Next slide, please. So our RAF/MEK inhibitor, VS-6766, and our FAK inhibitor, defactinib, we believe, play important roles here. And I'll describe their roles in blocking this critical RAS pathway and also in blocking the resistance mechanisms that cancer employs. So now let's focus on VS-6766, which is a unique small molecule, RAF/MEK inhibitor. As you know, there are numerous MEK inhibitors that block MEK only, and VS-6766 effectively blocks MEK as shown on the upper right of this slide. But what's unusual about this drug is that it binds MEK in a way that it blocks the ability of RAF to phosphorylate MEK, and that's shown in the far upper right corner, so that a VS-6766 potently blocks a BRAF or a mutant BRAF or CRAF's ability to phosphorylate MEK. And the importance of this comes back to this feedback reactivation pathway, as shown in the green arrow again. So what's been shown by Neal Rosen at Memorial Sloan Kettering is that, for example, looking at the PD compound on the bottom right, if you block MEK, ERK feeds back and activates RAF, and RAF then drives MEK activation, which you can see as phospho-MEK or pMEK on the lower right corner. So paradoxically, this MEK inhibitor is actually driving MEK phosphorylation, and the result is that you're not getting complete blockade of ERK, which you need. In contrast, VS-6766, because it blocks MEK and RAF activity, when you block MEK the feedback occurs, but now you don't see that increase in phospho-MEK and you're able to get to complete blockade of ERK, as shown by the phospho-ERK decrease. One thing that I'd really like to highlight, and you'll see the importance later in the presentation, is coming back to the -- I mentioned that VS-6766 blocks RAF kinases, so that's BRAF and V600E mutant BRAF, but I really like to focus on CRAF for a moment. So it's quite important and unique that VS-6766 inhIbits CRAF activity, and the relevance of that is shown in this slide. So this is work from Mariano Barbacid primarily, and he has found that if you look at models of KRAS G12V mutated lung cancer, as you see in the middle panel, if you knock out CRAF, it's essentially a mouse clinical trial, you get a very good objective response rate of 62%, you get a lot of tumor regression; whereas if you have wild-type CRAF, you don't get any regression. So what that's showing is that for KRAS G12V mutant-driven lung cancer, CRAF is critical. And that's further shown on the right panel. So you can see that with CRAF knockout, you're greatly extending the lifespan of these mice with G12V-driven lung cancer. And in sharp contrast, if you knocked out BRAF, for example, on the bottom right, you see that you don't get that increase in survival. So Dan Paterson will come back to this later in the presentation as we look at the activity of VS-6766 in cancers with this KRAS G12V mutation. This slide also taken from Neal Rosen's paper shows that unlike some drugs that are limited to a single KRAS mutation, VS-6766 is effective across multiple RAS and RAF mutations. So for example, you see that regardless of whether you have various KRAS mutations or an HRAS mutation or BRAF mutations, VS-6766 is capable of effectively blocking phospho-MEK and phospho-ERK. So coming back to these bypass mechanisms for feedback reactivation, let's now look at how Verastem's drugs impact these feedback pathways. Next slide. So as just mentioned, VS-6766 effectively blocks both RAF and MEK. Next slide. And defactinib effectively blocks FAK and the effect of this is that you'll block the downstream signaling to ERK, you'll block the downstream activation of RhoA and the downstream activation of YAP, and ultimately you'll block proliferation survival and migration of the cancer cells and -- with the result of blocking tumor growth and hopefully getting tumor regression. Next slide. So with that, I'd like to come to a few brief slides on the single-agent activity of VS-6766. Dr. Banerji is here and can comment later, but this is from an ASCO 2017 presentation, Dr. Banerji was the lead author. And what they looked at here was various schedules for VS-6766, looking for an intermittent schedule in which VS-6766 would be both well tolerated, but maintain its efficacy. Because like other effective MEK inhibitors, adverse events do occur with this agent. So you can see that the conclusion of this slide was that a twice-weekly schedule, essentially Monday, Thursday was found to be the best tolerated schedule, and that became the recommended Phase II dose of 4 milligrams to go forward. And you can see here in the middle column, the safety advantage of this twice-weekly schedule. So you can see for -- looking at grade 3 or greater adverse events, for example, rash and CK elevation, you see 19% and 8%, respectively, with the twice-weekly schedule, and that compares quite favorably to the higher rates of these adverse events, given in the far right column with typical once-daily administration of VS-6766. So a better tolerated regimen. Now let's look at the efficacy of this twice-weekly schedule. So here, we're looking at 10 patients with KRAS-mutated lung cancer. And you can see that VS-6766 was quite effective with 3 partial responses, in the waterfall plot on the left. And probably more importantly, going beyond 12 weeks, progression-free survival is meaningful for these patients. A number of these patients, as shown by the dotted line, went beyond 24 weeks of PFS, which is quite extraordinary. The study also looked at this twice-weekly schedule of VS-6766 monotherapy in a handful of patients with gynecological cancers. And you can see partial responses, for example, in BRAF-mutant ovarian patients and a KRAS-mutant ovarian patient and a KRAS-mutant endometrial cancer patient, and these became signals that Dr. Banerji then followed up along as he moved to the FAK inhibitor combination studies. So to conclude -- I mean, sorry, to move on to defactinib now. First talking about the target on the left side of the slide, focal adhesion kinase, or FAK, is a non-receptor tyrosine kinase that plays key roles in drug resistance, not the subject of today's presentation or the fact that FAK is also important in immuno-oncology and the tumor microenvironment. And Andrea Wang-Gillam will give a presentation at AACR later today on defactinib in combination with pembrolizumab and gemcitabine in pancreatic cancer that takes advantage of these observations. But it's been shown that FAK inhibition reduces stromal density, which allows entry of cytotoxic T-Cells into tumors, and FAK inhibition also reduces immunosuppressive cells in the tumor microenvironment. So defactinib is a selective inhibitor of FAK, and the other member of the FAK family, PYK2. It has good phospho-FAK target inhibition, it has been shown in the tumors of patients who have been treated with oral defactinib. We've seen early signs of clinical efficacy, and I'll allude to that in a moment on the next slide. And it's been studied now in more than 500 patients with a good safety profile observed to date. So for example, looking at grade 3 or greater toxicities, over 2.5% of patients, only hyperbilirubinemia was shown there. And I'd like to mention that hyperbilirubinemia with defactinib is an effect on glucuronidation of bilirubin and is not associated with other liver sequelae. And preliminary results show that defactinib is well tolerated in combination with various agents, as mentioned on the slide. Next slide. So the next slide gets into how we got interested in defactinib in the context of KRAS mutant cancer. On the left slide (sic) [ side ] is one figure from the paper from UT Southwestern in 2013, and they looked at a series of lung cancer cell lines and found that interestingly, the cell lines with the mutation in KRAS were more responsive to FAK inhibition than those with wild-type KRAS. They also went on in the paper to look at various in vivo models that showed the same, and they also delineated the signaling pathway, which I won't go into today. But based on these preclinical data, David Gerber, also UT Southwestern, conducted a clinical trial with defactinib monotherapy in patients with KRAS mutant lung cancer. And as I mentioned, 12 weeks PFS is meaningful in these patients, a 12-week PFS rate was set as the primary endpoint. And you can see in this slide from David Gerber's presentation at World Lung in 2015 that even though the patients treated with defactinib had more prior lines, 3 median prior lines in this case, compared to some of the drugs that were compared by Dr. Gerber, the fact that he had compared favorably with a 36% 12-week PFS rate as a single agent in KRAS mutant lung cancer, so Dr. Gerber concluded with 2 statements: one is that defactinib is showing activity in KRAS mutant lung cancer; the second conclusion is that it's important to think about what to combine defactinib with going forward for even better response. And one conclusion in that World Lung presentation was that based on existing data combining defactinib with a PD-1 inhibitor would make sense, and we have trials ongoing, as Brian mentioned earlier, in that regard. But on the next slide, we also set out to look for what would be the best agents to combine with defactinib. So we did a screen of approximately 30 different agents representing different mechanisms of action to find the best synergy with defactinib, and we found that MEK inhibitors, including VS-6766, a RAF/MEK inhibitor, were the top hit. So the way the screen works, just briefly, looking for synergy, which is effectively 1 plus 1 equal 4. So the green line -- I'm sorry, it's hard to see, the blue line is a Loewe additivity calculation. So that's representing what would be expected in each case if you had additivity; if the red line is below the blue line, that indicates synergy. So as shown here, if you have, in the upper left, a KRAS mutant ovarian cancer cell line, or on the upper right, KRAS G12V mutant lung cancer cell line, or on the bottom, a BRAF mutant cell line as well as other cell lines that have ERK pathway, RAF pathway, driven dependence, even without mutations in RAS or RAF, you still see the synergy between a MEK inhibitor and defactinib. And this was seen on the top panel using trametinib and on the bottom panels using VS-6766 with defactinib. So next slide, we also went on to look at several in vivo models, and these are just a few examples. And what you can see on the left is that combining a MEK inhibitor with a FAK inhibitor gives better efficacy. So for example, in a KRAS driven -- I mean, a KRAS mutant ovarian cancer model, trametinib is a MEK inhibitor or a FAK inhibitor, or VS-6766 is a RAF/MEK inhibitor, all give essentially tumor stasis, but combining VS-6766 with the FAK inhibitor gives tumor regression. Similarly, in a uveal melanoma model, in the next panel, the FAK inhibitor or the MEK inhibitor, trametinib, each give tumor stasis again, though combining a MEK inhibitor with a FAK inhibitor gives tumor regression. Moving on to the right side of the slide. If you look at a -- these are other data from the literature. If you look at a BRAF inhibitor or a FAK inhibitor, they both show some activity in this BRAF mutant melanoma model. But if you combine a BRAF inhibitor with a FAK inhibitor, you get a much better tumor growth inhibition. And interestingly, on the far right panel with the BRAF mutant colorectal model, you see that BRAF inhibitor plus FAK inhibitor is effective, but inhibiting BRAF plus MEK plus FAK is the most effective. So with that, I'd like to turn the presentation over to Dr. Banerji, who will discuss his results of the Phase I combination of VS-6766 and defactinib. And I'd also like to take a moment to say that we've been working with Dr. Banerji for several years on this. We've had shared excitement for quite a while. And it's really exciting for us to share this story with you today. Dr. Banerji?

Thanks, Jon. If you could move the slides on -- so as Jon said, we've had -- I've been working with both compounds actually of 766 and defactinib on different trials for a period of time. What I found exciting was the mechanism of action of the 766 compound was actually quite different from MEK inhibitors. I have worked with MEK inhibitors for about 10 years now, both the selumetinibs and the MEK162s and trametinibs. And because of its unique mechanism of action of not just being a MEK inhibitor, but also a functional RAF inhibitor, I was very keen to actually take this drug forward. So as with any MEK inhibitor, you will get toxicity if you give it continuously. So we did devise a intermittent dosing schedule, which Jon alluded to, which is twice a week. So it is very intermittent, Monday, Thursday or Tuesday, Friday. The reason for this is that drug has a half-life of 50 hours plus. But like any MEK inhibitor with a long half life, you can't give a loading dose because you'd probably run into different toxicities. So we decided to go with a twice a week. And with defactinib, we have been working with the drug for a while on other trials. And we were very keen to give it as the schedule found in other trials, which was twice a day. So we did go through 3 different dose levels. And once we found the dose levels, and the details of that are there in the presentation today, we wanted to have expansion to test the hypothesis about the drug being active in KRAS mutant cancers. So these are the schedules we've got, and we've got our recommended Phase II dose, which we will come to in a later slide. And the expansions we've got ongoing are low-grade serous ovarian cancer, non-small cell lung cancer and colorectal cancer, and we're in discussions with Verastem for further expansions. And if you have the next slide. So this is something, again, in the actual presentation, I've got some further data with proteomics, which showed phospho-FAK increasing -- levels of phospho-FAK increasing upon pertubation of the signal transduction system with a MEK inhibitor, particularly in RAS mutant cell lines. And I think we wanted to make sure that, that was one of our hypotheses for actually starting the study. The other hypotheses, as Jon alluded to, is also based on immunotherapy aspects of it, that phosphorylation of FAK -- reduction of phosphorylation of FAK, or FAK inhibition can reduce regulatory T-cells. And we've got preclinical evidence to -- published preclinical evidence about this. However, in the trial itself, we did 3 biopsies. So we did one biopsy before treatment, one biopsy after a single dose of the MEK inhibitor or RAF/MEK inhibitor, and then one biopsy about 15 days into treatment. And as you can see, individual patients, a vast majority of them showed a increase in phosphorylation of FAK somewhere between 18 to 24 hours after the first dose. And they all -- the phosphorylation did come down when the combination was instituted. So it was very pleasing for me to see that actually what we had hypothesized in our preclinical experiments actually played out in patients. If you have the next slide. So this is the details of the toxicity and side effect profile seen. So the actual dose that we have recommended taking forward to a randomized Phase II study is the dose of 766 at 3.2 milligram twice a week and defactinib 200 milligram twice a day. It's given in the schedule 3 weeks on, 1 week off. And what you see is the side effect profile is -- some of it's very similar to what has been seen with MEK inhibitors. But because it's highly intermittent, it is less so than before. So the rash is seen with all MEK inhibitors, it's acneiform. It's -- to put this in perspective, because it's given twice a week in this instance, it does -- it's not that evident. It's certainly less than what you would see with, say, cetuximab treatment in colorectal cancer. We've had ladies with low-grade serous ovarian cancer, been on it for more than now 2 years, actually go about their lives normally. So they don't feel it's -- or with a little bit of makeup, you can actually not see it. We've got some biochemical abnormalities with CK elevation. Again, that's quite normal with MEK inhibitors, hasn't been limiting. And as Jon pointed out, there was 1 or 2 cases of asymptomatic rise in bilirubin, but that's due to the glucuronidation rather than hepatic injury; again, visual disturbances, you can see a few flashing lights of blue halos after the first dose. And after that, it is very tolerable. Again, that's very common with MEK inhibitors. And if you move on to the next slide, and that's the pharmacokinetics. So we did -- we weren't there to look for pharmacokinetic interactions in detail. We did the PK of both drugs around day 15 into the study. The results are not very different, AUC, Cmax half lives are very not -- very different from what's published for a single agent in both these studies. So there doesn't seem to be any interaction. And next slide. So this is the area we are very excited, and we have been working with MEK inhibitor for a long time. And this is -- and in RAS mutant cancers for a long time. This is almost a validation that we can inhibit RAF signaling and show responses, which are both reproducible and durable. So the waterfall plot on the left of your screens are the patients with low-grade serous ovarian cancer, which have been treated until the November cutoff. As you can see, a majority of the ones, the patients who came to us, actually had RAS mutations, but 4 out of -- depending on if it's KRAS mutant or not, it was 4 out of 8 patients of all KRAS mutations, all 4 out of 6 patients, if you're looking at only KRAS mutant, so that's about 67% of patients with RAS mutant low-grade serous ovarian cancer actually responded and that was very exciting to us. And what was probably even more exciting for us is, if you see the histograms with dash lines across them, so they are patients who had actually received previous MEK inhibitors. So these are MEK pre-treated low-grade ovarian -- serous ovarian cancers responding, and for a swimmer plot on the right, as you can see, the -- it's number of weeks on the X-axis. So we have patients well over 1 year, and definitely so a few patients now over 2 years on treatment, which is also very exciting. So the -- both reproducible in small numbers, you'd have to do larger numbers, but in expansions, in expansion of a Phase I study, this is very exciting, and it's also durable. And to put this in perspective, if you look at chemotherapy, the options open to a patient response rate is only 10%; letrozole for a long time was the only other treatment, 13%; and trametinib more recently, which was a randomized Phase III, was presented with 26% response. And here, you're seeing -- and if you're looking at RAS mutant, it's 67%, or 50% if it's all low-grade serous ovarian cancer. Small numbers, which will need to be validated in larger sets for all, but it's still very exciting. And if you have one more slide, so this is small -- non-small cell lung cancer, which are RAS mutants. We purposely chose cohort with RAS mutant lung cancers. So what Jon suggested from our monotherapy study, we had seen a signal. What -- a couple of interesting aspects of this, so one of them was that -- G12C was not as common as we thought. So RAS mutant G12C seems to be in the literature 60% or 50% of all non-small cell lung cancers. It wasn't so in our hands, but that could be just the core subset of cohort of patients we have here in the U.K., which is -- and the numbers are small. However, we did see proof-of-concept responses. And again, depending on what you've sort of put as your cutoff, 12 or 24 weeks, we have seen patients -- I think, 24 is definitely something that lung cancer docs think are important, but there are people who will be very excited about 12 weeks as well. The other important thing is that the excitement that's been there in RAS mutant lung cancer all seems to be due to the Kevan Shokat's work of -- on G12C KRAS mutant lung cancers. And actually, there's nothing full, at least in our hands, a vast majority of the RAS mutant lung cancer cohort. So this is for the first time I think we've actually seen a glimmer in non-G12C KRAS mutant lung cancer, we've seen responses. And for the last slide, I think, from my point of view, VS-6766 and defactinib have shown both single agent activities in small subsets and different clinical trials. So it's not just preclinical data. There is clinical data as single agents, it's modest, but it's definitely there, so which also makes me have more confidence when we see responses in the combination. The RAS pathway blockage activates FAK, as we've shown preclinically and in post-treatment biopsies. The FAK and MEK inhibition are synergistic in preclinical models, and it's not just one particular G12C model. For example, there are multiple RAS models, G12V and G12D. And for the -- in the FRAME study, the most adverse events were grade 1 and 2 in intermittent dosing schedule. There doesn't seem to be any clear PK interactions. We've shown proof-of-concept biomarker inhibition and feedback loops and biopsies. And for us, what's most exciting clinically was the promise of responses in cancers with unmet need, particularly low-grade ovarian cancer, low-grade serous ovarian cancer, and even more exciting was the fact that previously MEK pretreated patients responded to this combination, and we've seen proof-of-concept responses in lung cancer. But I think there are other cancers, which we should look at in a larger subgroup. But as Jon said, you have to have to go in with the sort of indication with the easiest path to registration. And I think low-grade serous ovarian cancer is definitely one to go for, but it obviously needs specialized discussions with gynecological oncologists. So I think that's my set of slides, I'll hand back to Dan Paterson.

Thank you, Dr. Banerji. Clearly, we're very excited about all the data being reported today, but especially in low-grade serous ovarian cancer. And as I'll go through in a moment, KRAS mutated G12V mutations. In this early data in LGSOC, we observed an encouraging response rate, tumor shrinkage in every patient and long duration on therapy. Collectively, these elements speak to the potential for meaningful efficacy and tolerability. Next slide. Given the impressive activity in this study in LGSOC, I'd like to take a few moments to review the clinical unmet need and potential market opportunity. LGSOC represents 4% to 10% of epithelial ovarian cancers with an incidence of approximately 1,000 patients diagnosed annually in the U.S. and probably more importantly, given the high unmet need and the fact that these patients go through many therapies, a 10-year prevalence of approximately 6,000 patients living with the disease, who would be the addressable market. Overall, there's a high rate of RAS pathway mutation somewhere in the range of 50%, and there are no approved therapies in this disease. Importantly, as Dr. Banerji mentioned earlier, chemo has a response rate of approximately 10% in this setting with MEK inhibitors showing a response rate between 15% and 26%. Taken together, all of these make for a very attractive initial regulatory path. The potential for long duration of therapy and high unmet need with available therapies make an attractive and attainable commercial opportunity. And finally, within LGSOC, G12V and D are the dominant KRAS mutations with G12V conferring a more aggressive phenotype. Next slide. We then examined the data across various KRAS mutant variants, which unveiled an exciting finding. These are small numbers so far, I want to have that caveat, but the signals we're seeing in the G12V and G12D variants were quite unexpected and surprising. As you can see from this slide, we're seeing signals across response rate, disease control rate and/or durability for G12V and D. There may be potential and additional variance as well, but those areas are very early. Given the CRAF G12V association that Jon presented earlier, it's gratifying to see the results playing out in patients in the clinic. As we think about where to go next, we're awaiting the additional patients from Dr. Banerji's study to read out, and at the same time, we're doing a deep analysis of the RAS mutations and looking at other co-mutations. We can then look mechanistically at other potential combinations and areas where we can combine with targeted agents. Next slide. Based on the observations of a higher response rate seen in patients with KRAS G12V mutations in the investigator-initiated Phase I combination study, we then conducted a combined analysis with data from the combination study and the prior single-agent study that Jon referred to earlier that used the twice-weekly dosing schedule of VS-6766 to get a more complete picture of activity in KRAS G12V mutations. As you can see here on this slide, the response rate in these 7 patients was an impressive 57% of 4 out of 7. Importantly, these are all confirmed responses. From the single-agent group, 2 of 5 patients responded for a 40% response rate. In the combination group with defactinib, 2 out of 2 patients responded for a response rate of 100%. The consistency across the mono and combo responders is also promising. On the right, the time on treatment for the subset of patients was promising as well, especially for these heavily pretreated patients. I do want to point out that following the November 2019 data cutoff for this report, we learned from Dr. Banerji there was one additional G12V patient who had been enrolled and that patient is a responder who was included in this analysis. We conducted these initial analyses to understand the impact that various KRAS variants may have on response to identify potential signals to pursue in future prospective studies. It's important to note here that we are not aware of other products in development for cancer patients with the G12V mutation, we're, of course, aware of the assets being developed for G12C. Interestingly, the KRAS G12V variant represents a very significant opportunity, both in non-small cell lung cancer as well as across other tumor types. As you can see here, the incidence of non-small cell lung cancer is estimated to be approximately 194,000 patients per year in the U.S., of which G12V is 7%, second only to G12C, which is around 13%. And as you can see on the right, the G12V variant accounts for about 21% of patients with KRAS mutations across cancers. We looked at all of the clinical KRAS G12V patients in the 2 studies, including gynecological and non-small cell lung cancer. We were surprised to see a similar response rate of 58% or 7 out of 12 with all of these being confirmed responses as well. From the single-agent group, 3 of 7 patients responded for a 43% response rate. And in the combination group with defactinib, 4 out of 5 patients responded for a response rate of 80%. The time on treatment data here on the right also remained favorable. As we look beyond lung cancer at the broader opportunity for G12V and D variants, we find a very intriguing picture. First, the G12V and G12D variants together comprise approximately 50% of all KRAS mutations. Given the evidence we've seen from the Phase I Wang-Gillam study investigating defactinib in pancreatic cancer, combined with a high rate of G12V and G12D mutations in this tumor, this is a key area of interest for us. And our intention is to add a pancreatic cohort to Dr. Banerji's ongoing Phase I study. After that, we may also look to expand into uterine cancer where G12V and D variants also appear to be quite prevalent. Longer term and building further on the pipeline within a product concept, we believe that the mechanism, along with the preclinical and clinical evidence, allows us to pursue many different possible directions in order to maximize the asset's blockbuster potential. On the upper left area of the picture here, where we have both clinical and preclinical data, these will be our high-priority areas to explore, including low-grade serous ovarian cancer, KRAS G12V non-small cell lung cancer and pancreatic cancer. In the upper right, we see several other expansion opportunities, including additional G12V and D cancers, uveal melanoma and certain BRAF-mutant cancers. We expect to also explore VS-6766, plus or minus defactinib, in combination with other agents, including G12C inhibitors, EGFR inhibitors and the MTOR inhibitor everolimus as well as anti-PD-1. Given the mechanistic underpinnings, we could also explore other related mutations like GNAQ, NF1 and MAP pathways. On the intellectual property front, we've carefully designed our IP strategy to protect the VS-6766 defactinib combination and to maximize both the length and strength of the commercial opportunity. We currently hold rights to composition-of-matter patents that run through 2027 and 2028 for VS-6766 and defactinib, respectively. This is before any Hatch Waxman patent term extensions we might be eligible for, which we believe would extend coverage through 2032. We filed an application on the VS-6766 intermittent dosing regimen, which would run through 2038, if granted. We also hold newer issued patents that cover the FAK/MEK inhibitor combination and these extend to 2035. Beyond that, we've applied for additional coverage that will protect the specific combination until 2040, if granted, and the method of manufacture through 2032, if granted. And of course, we continue to explore further patent opportunities to maximize and protect our exclusivity. And finally, for the key takeaways from today's presentation. From a mechanistic perspective, we believe that VS-6766 and defactinib has the potential to be best-in-class RAF/MEK and FAK inhibitor. They provide more complete RAS pathway shutdown addressing key resistance mechanisms. The combination uniquely targets CRAF to shut down KRAS G12V. From a clinical perspective, we've identified a novel, all-oral regimen that has no overlapping safety issues. The initial data on the combination are encouraging, including both objective response rate and durability. And KRAS G12V mutant cancers appear to be particularly responsive to the VS-6766 defactinib combination, and we'll be prioritizing and planning our next steps here. From a market opportunity perspective, we have a first-in-class approach to KRAS G12V and G12D mutations. There are no approved therapies in low-grade serous ovarian cancer, and we believe that's an interesting opportunity. And we see blockbuster downstream potential in other human cancers, driven by the RAS family of genes. Looking ahead, our goal is to initiate a registration-directed study in low-grade serous ovarian cancer during 2020. We want to complete the expansion cohorts in the ongoing investigator-initiated Phase I combination study. We will be adding new KRAS G12V and G12D expansion cohorts in both non-small cell lung cancer and pancreatic cancer, and we also intend to explore BRAF-driven indications as well as combinations with KRAS G12C and anti-PD-1. With that, I'll now turn the call back over to Brian for some closing remarks. Brian?

Thank you, Dr. Banerji, Jon and Dan. In closing, I would just like to say that opportunities like you have seen here today, don't come along very often. During my 30-year career at Lilly, I was fortunate to be part of the team that rescued ALIMTA from excessive toxicity by adding a folate supplement and created lung cancer history, targeting ALIMTA in non-squamous non-small cell lung cancer on the basis of thymidylate synthase expression levels in various subtypes of lung cancer. Both of these findings were underpinned by a strong mechanistic and preclinical understanding, with ALIMTA eventually going on to exceed annual sales of $2 billion. I believe the data presented today on VS-6766 plus defactinib are equally compelling. The combination is underpinned by a strong biologic understanding of the pathway and has been validated in multiple in vitro and in vivo models and shows early promise in translating into the clinic across multiple studies and tumor types. We are extremely excited about the new VS-6766 and defactinib combination program, and we look forward to commencing a company-sponsored, registration-directed clinical trial in LGSOC and further elucidating its blockbuster potential downstream. With that, we'll open up the call for your questions. Operator?

[Operator Instructions] Our first question comes from Alethia Young with Cantor Fitzgerald.

Congrats on the early promising data in these particular subgroups. So I guess a couple for me. One, can you talk a little bit more about the CRAF? And why you think you saw efficacy in 12D and 12B but not 12C? And then I'll ask a couple more.

Jon Pachter, would you like to start with Alethia's question and Dr. Banerji, feel free to add any additional comments.

Yes. I'd be delighted to answer that. So we've been very interested in CRAF, and work from Neal Rosen has shown that VS-6766 is unique in putting -- in inducing a complex of CRAF with MEK, or with RAF with MEK that inhibits all CRAF activities. And putting that together with Mariano Barbacid's work CRAF is critical for downstream signaling of KRAS G12V. We don't yet know whether that CRAF is also critical for G12V in other variants. We suspect it may be, but the data still need to emerge. So I think that it's based on that CRAF inhibition, which is quite rare among other agents, we believe, that, that's what's leading to the G12V activity that we're seeing.

Great. And then another question. So with the ovarian cancer, you think that you'll be able to take this and move it in? Or are you kind of going to do a bigger cohort and then move into a registration-directed study? Just trying to clarify.

Yes, I'd like to ask Dan Paterson to comment on the ovarian plans?

Sure. Alethia, thanks for the question. We believe with the data we have in hand, that's enough to have a dialogue with the FDA about the ultimate study design for a registration-directed study. So our intent is to seek a meeting with the FDA, reach out first half of the year and discuss the design of that study.

Okay. And then when you started this, did you think that the promise was more in non-small cell versus ovarian? Or did you have a hypothesis there? Maybe I would ask Dr. Banerji to give his perspective on that as well?

Yes. Let me start Alethia and then ask Dr. Banerji to add. When we started this combination based on the preclinical activity that Jon Pachter showed, we thought there was going to be activity across all the different variants. We did not know to what level and what the durability would be in each of those variants. And that's obviously the data that's starting to emerge here today. So Dr. Banerji, given your experience with MEK inhibitors, it'd be great to hear your perspective as well.

So I think there's a sort of level of complexity -- 2 levels of complexity. So one is what does -- how do K -- RAS mutations drive cancer? So as you've seen with BRAF, obviously, BRAF inhibitors work in melanoma, but they don't work in colorectal cancer unless there's a combination. So I think the initial feedback loops of FAK was seen, and we were quite excited and we're seeing some single-agent activity in lung and ovarian, to be fair, and uterine cancers. But they're all RAS mutant. And that's what sparked it off. Now whether the combination will be equally effective in low-grade serous ovarian and colorectal and lung? It's hard to say because there's this 2 levels as I said. One is whether the RAS mutation signals the same way, and we've published on this before. And the second is the level of sort of the stromal cell, the effects of the stromal cells. So that might be very different in pancreatic cancer, driven by KRAS from, say, colorectal or ovarian. So the only real way is actually to do these expansions and get a rule of thumb. So if you see something like you see in low-grade serous ovarian cancer, that's very obvious to me. I mean, there's no second doubt if you're getting more than half the patients respond, we did olaparib in BRCA-mutant cancer. It's very obvious those signals are seen. And then some more subtle signals are seen where you're not 100% sure you want a larger cohort. And I would say, there are things like pancreatic cancer, which we haven't yet explored. So I think the -- to answer your question, preclinically, it's a very good place to start with, but you actually have to do expansions to see them. I'm a great believer and see, if you do 10 patients and see no responses, I think, then it's probably not going to work. And if you see more than 5 responses, it's definitely going to work. And then there's a gray area in between. So I think where we are now is -- we are in low-grade serous ovarian, it's very clear. It's very likely to work. And in lung, we need larger cohorts and to see whether there are any more and then there are these unchartered territory like pancreatic cancer, which is the Holy Grail for RAS mutant cancers. But they are -- the fact that defactinib acts on stroma is that definitely is important and is different from what we've done before. So I think let's see how it plays out.

And so maybe my last one is just kind of -- I'm trying to get the numbers straight in my head. So I think the abstract had about 42 patients or so. You guys added another 7 with non-small cell V. How many more patients have been dosed from November to April since then? Can you kind of frame that for us as well? Maybe as, particularly, Dr. Banerji's cohort and/or, I guess, at Verastem you guys have dosed more people than where you are today?

Yes, why don't I ask Dr. Banerji to provide an update in terms of how many patients have been enrolled into each cohort? And then I'll wrap up, Alethia, with the other question that you had.

So yes. So our intention is to treat 20 patients with low-grade serous ovarian cancer and 20 patients with RAS-mutant non-small cell lung cancer. So we've got to -- in the low-grade ovarian, we've got about 9 patients with low-grade ovarian cancer treated in the expansions. We've got about 11 patients left. And in the non-small cell lung cancer, we've been -- we've got further, there's about 12 to 14 of patients treated. But these are patients actually on treatment now. So that's why we haven't commented on them at this point. And as people pointed out, a lot of our data monitoring actually is done remotely, but you actually have to -- can't verify because they're not letting people into hospitals, but thankfully, we haven't lost any patients to COVID and all are well and shielded. And we're managing to telephonically see them and for them to come in, in a less intensive schedule, but all as well.

Yes. And then, Alethia, on the second question then, the data that we covered today was from the November 2019 cutoff. There has been, as Dan mentioned in his presentation, one additional G12V responder that occurred after that 2019 cutoff that we wanted to see what the activity was, given our finding across tumors. That represents in the slide what's referred to as combo patients. Next, what we did is go back to the Gerber paper -- I'm sorry, not the Gerber paper, it was the 2017 ASCO presentation that Jon Pachter covered where we had cohorts of non-small cell lung cancer patients and gynecologic patients. We took the mutation status from those patients and incorporated those patients into the combined analysis, and they are referred to as model.

And then this non-small cell cancer patients, Dr. Banerji, are they all -- are they G12V or they all [indiscernible] on KRAS?

Just on KRAS. So there are multiple, there's V,C, D. But as I -- as that slide showed, I was quite surprised of how few C we saw -- I mean, at least what's been published out there is that -- it's more than 50% of G12C, but that certainly wasn't our experience. But these things will pan out in larger studies, but we're treating all patients with RAS-mutant non-small cell cancer.

Our next question comes from Bert Hazlett with BTIG.

Yes. Just a quick follow-on and a clarification. So the additional patient -- the additional response on the combo in the later slides from Verastem, it includes an additional response since the November data, is that correct?

Yes, that's correct, Bert. We specifically, given -- given our observation around G12V, we asked Dr. Banerji, has there been any additional patients enrolled with G12V since the November cutoff? And what was the response rate? And that one patient was pulled in to the combined analysis that we referred to.

Okay. Great. And you were kind enough to mention the patient -- numbers of patients that have been dosed in lung cancer and in other settings. How many you made it to the first scan? Do you have that number potentially? I know we're down in the weeds?

Yes. Let me -- as a backdrop as Dr. Banerji mentioned, it's been challenging to get all the data cleaned, given the COVID crisis. However, I would ask Dr. Banerji to comment on what his observation has been within the trial.

So what I can say, I guess, I don't have the exact numbers in front of me. But the swimmer plot for the first 10 patients that were shown here and is going to be presented today at AACR is not very different from that. So there isn't a big cliff of people at basically 8 weeks falling off. It's roughly the same. And I can only say roughly because I don't have the numbers in front of me, but certainly not people falling off after 2 cycles. There are an occasional patient who does that. But it's mainly around 3 out of 10 went on to 24 weeks, and 7 out of 10 went on to 12 weeks. It's roughly this -- I would say it's in that pattern. And the only way really you can get a handle of this is in a randomized study to show whether your treatment is better than chemotherapy. I got a feeling it is. As I say, it's -- the signal is not as strong as low-grade serous ovarian cancer, but it's definitely not a 0 out of 10. So -- and these are durable. In the monotherapy cohort trial, which I think Jon mentioned on the twice a week schedule of the drug itself, we've had 1 patient who went on to 3 years on treatment. So -- and it's durable. It's just -- we have to figure out the exact numbers and the nuances between, as was discussed today, G12V or whether it's something more broader than that.

That's a nice segue to another question, which is the robust activity you're seeing in low-grade serous ovarian cancer. What is your thought with regard to the MEK pretreatment or, let's call it, pretreatment or the MEK experience? Did that in some way set these patients up for a more robust response with the combination? Or were there any other similarities in this patient group that might be important?

Bert, as I -- on that particular question, I'll refer to Dr. Banerji. And -- as he's mentioned in the slides, several of the patients here were pretreated with MEK inhibitors, and I'm sure he can share what you saw within each of those patients.

So it's an interesting question. So to me, what was exciting was that patients who were exposed to MEK inhibitor before have responded again which makes me feel that maybe there is -- the [indiscernible] is adding something to it, only way you can really figure that out is in a randomized trial. But I think -- I don't think the fact that, that MEK before would, in any way, make them respond better. In fact, we would think that there will be resistant clones that have been generated or have been enriched for by giving MEK before. So to me, that's very exciting news. The other aspect, I guess, is trametinib on its own. There was -- last ESMO, there was a really interesting randomized Phase III, which came out, that showed overall survival benefit, but again, the response rates there were 26%. And what -- the early -- very early glimpse we're seeing here seems to be close to 50% or above. And I think that might be because of the combination. Again, early days, I don't know the full answer to this, but it's looking encouraging.

I have just one last one. Could you comment on any dropouts due to AEs with regard to the combination? And then just any more general comments you might have with regard to management of the adverse events in the combination?

Yes. Dr. Banerji?

Yes. So AEs, what I would say is I've worked with 766 when -- at one stage, Roche was running the trial, and they were doing 7 days on, 7 days off and 4 days on, 3 days off. And certainly, I did learn a lot from that. And it's also been my sort of strong laboratory experience that actually -- you've got to hit the target hard and inhibit it rather than to sort of suboptimally inhibit it for a long period of time. I know it goes against the grain of a lot of targeted treatment. But I believe the cancer cell is going to die from your addicted signal transduction, you got to hit it hard. So that's why I came up with this twice a week, but you hit it to the maximum and as intermittently as you can. So with the side effect profile, that's definitely compared to any MEK inhibitor or for that matter, it's acneiform rash, which is more BRAF inhibitor. So we don't see -- at the 3.2 milligram, which is the recommended Phase II dose, I've got at least 2 patients are more than 2 years on it, and they are fine. So that gives you an idea of how tolerable it is. It's true if you go more frequency, if you give it 3 days a week or if you get 4-milligram 3 days a week, you'll definitely get to lots of drop-offs. But it took quite a lot of PK/PD modeling before this -- before we came up with a twice a week schedule. So the MEK rash is the sort of driving toxicity, which we have to dial down. And I think with twice a week, 3.2, we've definitely done that. So as I said, we've done little bit of makeup, patients are managing completely to lead normal lives. Now I'm not saying -- I mean, if you go expansive 50 or 100 patients, you might have 1 or 2 patients who you might have to take off and that's happened with the trametinib trial as well. So it's a numbers game as well. So we're talking smaller numbers, but it's based on a lot of experience with MEK rash, and we seem to be getting away with it. The visual toxicities are Cmax related, they occur 10 or 12 hours after treat, and that's quite normal. So we haven't gone -- you can give up to 4, but we've gone for 3.2 for long-term tolerability. So far, so good, and I'm quite confident that we can use it in a larger population.

Yes. And I would summarize, Bert, that with both agents, we've seen a low incidence of Grade III, IV toxicities. The agents have nonoverlapping toxicities. And importantly, as Dr. Banerji showed, we don't see any PK interactions to date with the regimen. So again, thanks for your questions, Bert.

Our next question comes from Matthew Cross with Jones Trading.

Wanted to start up first by asking about if there had been any trend that you'd seen across dose levels in responses, I guess I'm curious, in particular, if you've seen anything related to -- within your recommended Phase II dose, that cohort whether response was driving to some degree, the decision to have that be carried forward into that -- the expansion dose or if that was a safety or PK consideration and then -- but really trying to see if there was any kind of dose response across your cohorts or where results were most exciting for either low-grade ovarian or non small cell.

Yes. Thanks, Matt. Dr. Banerji, maybe if you can comment on how you arrived at the recommended Phase II dose because I think that will cover responses that you have seen at various cohorts.

No, that's a really good question, and it's a question quite very close to my heart of range of pharmacodynamic effects, at least academically. So I think because we started off in a combination, like most combination studies, we went up and it started at quite a high doses and then worked our way back. But one thing with MEK inhibitors, as you get the Cmax related eye toxicity. So in the monotherapy study, when Roche was running it, I think they went up to 5 milligram, and they saw sort of eye toxicity, which was -- we decided that dose was intolerable, so we went out to 4 milligrams. So you don't have a very much higher -- by going intermittent, you still can't actually go up a sort of very high dose, like EGFR, you can go up to a very high dose, if you give it 1 day a month or something. We couldn't do that because of the Cmax-related eye toxicity talk that inhibits the upper limit of a MEK inhibitor. Now the lower limit is -- again, there's quite a lot which is published since clinical cancer research. It's mainly on peripheral blood mononuclear cell ERK. So they did quite a lot in the first study. And you can do a PK/PD modeling and actually look and maybe go down to even 2.4. And I don't know that. But we went at 4 and 3.2, and both are effective. I don't know whether if you go down even further, it will be effective, but I've just done it because I've designed the trial the way we did because obviously, we wanted to get an answer as quickly as we could. But if there are patients on it, our patients are now 2 years, so I don't know if they go on for very much longer, whether they will be okay with a lower dose and I don't know the answer to that. That's a very good question.

Would I -- Dr. Banerji, it may be also important to share what you're seeing the expansion -- in the expansion cohorts. Because as referenced in the press release and in your presentation, you did indicate additional patients have been enrolled into the expansion cohorts, which would be the recommended Phase II dose. And that you've seen responses across all 3 cohorts. So maybe a little update there would help with what we're seeing?

So we have seen, as you say, responses in -- at this dose level across multiple tumor types, as we've discussed there's one extra patient certainly in lung cancer. We've even got a colorectal cancer with KRAS mutation responding at this dose. Again, I don't know whether they'll respond to a lower those, but I think the only way to figure out this out is to do larger cohorts. What I can say confidently is that actually patients will tolerate this dose at this efficacy for as long as they need to because again, lung cancer patients often have COPD and lots of other issues or just poor respiratory reserves. So none of that seems to be a factor of dose reduction. But the only way I'm finding out is doing larger cohorts. But going down further, I don't know the answer. But I guess if it's a tolerable dose, then this is where registration studies should go to and there might be a 25%, 30%. But coming back to my sort of conviction in signal transduction, you probably can't inhibit low levels for long periods of time, that usually doesn't help. You have to have significant inhibition of the pathway. And 3.2 and 4 certainly do -- I'm not sure of lower down, but it might be possible that it could go to 1 dose further down below, but you need to do biopsies and prove that.

Got it. Okay. That's very helpful color from both of you guys. And I'm glad to hear you're seeing responses already in the expansion cohorts, including in colorectal. I guess the other thing I wanted to address was, I guess, some feedback from you all on is kind of the meaningfulness of these pooled results from the monotherapy and combo in non-small cell and ovarian. I guess, looking at these analyses, does this kind of -- are you suggesting that you may plan to take 6766 forward as a monotherapy in one or more indications or to focus particularly on KRAS G12V or some of these other variants where it's been most effective? I guess, if we're comparing monotherapy in these combo results, do you feel there's a real benefit to adding in defactinib? Or are these kind of more modest combo improvements, some level of just variation in 6766's response outcomes?

Thanks, Matt. I'll kind of open and ask Dan to comment on what our plans are going forward and then maybe Dr. Banerji, your thoughts on the importance of defactinib here. So as we mentioned in the presentation today, these results are early but extremely encouraging. And maybe, Dan, if you can expand on what our current thinking is in terms of where we would go next.

Yes. I mean, obviously, we have strong interest in further exploring the G12V signal. And it does appear that the combination adds something. It's small numbers, but we also believe mechanistically, that the combo adds something. I think, ultimately, we'll have to sort out the contribution of each agent. But I would say we're going to pursue low-grade serous ovarian and follow the signal in G12V and in G12D.

Yes. And then Dr. Banerji, maybe your thoughts on the role defactinib is playing here.

So -- actually, it's a really good -- clinically very relevant question. I would say I wasn't very surprised that the combination was better or at least looks better. I think because there was quite a lot of preclinical evidence of, I do a lot of proteomics sort of feedback loop, work in the lab. And certainly phosphorylation of FAK was something that we've seen with MEK inhibitors. And I know GSK had done work. And as Jon has pointed out, Verastem has done work on the combinations showing that you actually see regressions in xenograft models when you use the combination, but you get stasis when you use each individual. So there is activity individually. There's no doubt about that. Because you get stasis, but how often a stasis in a xenograft model actually got you to a registered drug, it's -- there is quite a lot of them have actually shown stasis and not got you to a registered drug. So I think regressions are much higher bar, and that we definitely seem to be seeing when -- in the slides that were shown previously. So that's one thing. So preclinically, there is evidence of that. Now clinically, I would say, 2 things, I guess, which we make me think defactinib does something. One is the fact that you're seeing response rates very small numbers again. But if you -- all the sort of early selumetinib studies, the 13% response rate, 10% response rate, and the MEK162 had a very small response rate. And the strongest signal seems to come from trametinib at 26% response rate. So this is early days, but it's clearly much more than that, which makes me feel defactinib probably is contributing something. And the second thing is, as we saw -- there was a question previously that there are people who had MEK inhibitors before and have progressed after that and then gone on to this combination. So that's, I guess, again, both indirect evidence that there's probably something from -- coming from defactinib, but the only real way of finding out is a randomized trial. But these are the sort of 3 areas, the preclinical areas, the fact that pre-MEK treated people have responded. And the response rate seems in the smaller numbers that we've seen but seems to be higher than what's been seen with MEK inhibitors alone.

Matt, maybe we have time for one more burning question, and then I'd like to ask the operator that I'll make some wrap up comments if there's another burning question you have.

I'm not showing any further questions at this time.

All right. Thank you very much. With that, I'd like to thank everyone again for taking the time to dial into today's call. We sincerely wish for the safety and well-being of everyone, especially the patients we serve, living with cancer. Have a great day.

Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.