Good evening, everybody. Those of you here in Madrid, it's been a wonderful ESMO for us at Incyte, and we're really excited about telling you a lot of the progress we're making in solid tumors. Thank you very much for joining us here today. We will be making forward-looking statements that you can see here. We have a great set of speakers today. You can see them here. Our own Steven Stein, Chief Medical Officer and Head of Late Development at Incyte, Dr. Elena Garralda, who's Co-Director of the Vall d'Hebron Clinical Research Program and Group Leader for the Vall d'Hebron Early Clinical Drug Development Group, among many other titles, and Dr. Manuel Hidalgo, Professor of Medicine, Director of Gastrointestinal Cancer Center at the Perlmutter Cancer Center at NYU. We're really pleased, and we're very grateful that they agreed to join us here today.
We have a great agenda. I'm going to make a couple of very quick comments to get us started, and then we're going to shift to reviewing the data for two key programs, '890 and '734. Our speakers will review the data for you, and will give you an overview of how we're developing these programs going forward. I'm going to make some closing remarks, and then we'll have plenty of time for Q&A. Please prepare your questions. All our speakers will be available for the Q&A. The focus today is the solid tumor development strategy at Incyte. I think over the last couple of years, we've been busy at work to build what I would consider a world-class solid tumor franchise. The focus today will be more on GI oncology, but we're building a broader solid tumor franchise within the organization.
We are focused on three basic principles to build that strategy, to build that franchise. One, we're trying to go out to novel targeted therapies, whether we are to novel biology, that in the case of TGF-β by PD-1, or we approach a particular target at G12D with a different mechanism from what was there before. Instead of an on inhibitor, in this case, we go with an on-off inhibitor, for example. We are focused on cancers of high incidence. We're trying to address large medical needs in patients that have limited treatment options. One thing that we've done pretty consistently in both programs, as well as in our CDK2 program, is go as quickly as possible to frontline. We believe that that is where the largest patient benefit resides, and obviously the largest commercial opportunity as well.
That's something that has been a consistent pattern in these programs, and you'll see it as we review the data. This is our portfolio today. We have one approved medicine, ZYNYZ, we shouldn't forget about that. It's approved for patients with anal cancer. The first PD-1 approved in combination with chemotherapy for frontline anal cancer, as well as Merkel cell carcinoma. '890 and '734 will be the bulk of the presentation today. Of course, on the right side of the slide, '667, a CDK2 inhibitor, which is in pivotal trials for patients with ovarian cancer, and will continue to progress in the clinic. The three programs address approximately 250,000 new patients per year. You can see the numbers here, taking into account only the indications that we've listed in the slides, which are the initial approvals for these programs.
For '734, pancreatic cancer with G12D mutations, for '890, MSS colorectal cancer, and for '667, CCNE1 overexpressing ovarian cancer. Almost 250,000 patients with very poor outcomes today. Not only is this a very large opportunity for us as a business, but is a very important medical need that needs to be addressed with better medicines for these patients. The programs have grown and will continue to expand, and this slide gives you an idea of what we're doing. Steven, after Dr. Garralda and Dr. Hidalgo show the data, Steven will give you an idea how we're deploying our resources in this program. For '890, our first indication in pivotal trials is colorectal cancer. We'll talk more about that.
Let's not forget that we've shown evidence of efficacy in head and neck cancer, ovarian cancer, and non-small cell lung cancer, all in patients that most of which had poor likelihood of response to PD-1 inhibitors. We moved initially in first-line colorectal. We are already implementing an adjuvant proof of concept study in colorectal cancer, very important, and we've generated data that was seen in this meeting as well in late lines. Very important for both programs is to expand into novel combinations. With '890, not only we've done chemotherapy FOLFOX, that is the bulk of the data you're going to see today, but recently we announced a collaboration with the next generation anti-CTLA-4 antibody.
We think that's an important approach to be tested with what we believe is a better PD-1, and we are in conversations to collaborate with an EGFR antibody, as well as a novel ADC, and you'll hear more about those in the next few weeks. For '734, our lead indication is pancreatic cancer. However, as you saw today, over the weekend, we presented data in colorectal cancer as well in late line in combination with cetuximab. That is going to be implemented as a pivotal trial. In first-line pancreatic, we're well underway. The enrollment is going on globally and accelerating as we speak, and we're moving into the adjuvant setting in pancreatic cancer as well. In addition to the combinations with chemotherapy and cetuximab, we announced today and on Friday, three very important collaborations to continue to expand the scope of this program.
Those are with navlimetostat, PRMT5 inhibitor, with SNV1521, a PARP1 inhibitor. On Friday we announced a collaboration with petosemtamab with a focus on colorectal cancer. These novel combinations will allow us to continue to expand and improve, hopefully, the results that we're seeing with combinations with chemotherapy and existing EGFR inhibitors. Highlighting one more time the importance of the breadth of these programs that we're trying to address, a number of patients across two core indications to begin with, colorectal cancer and pancreatic cancer. A very strong focus for us initially on gastrointestinal oncology. As I mentioned, we have interest in continuing to expand and grow within GI oncology as well as other therapeutic areas. With that, I'm going to hand it over to Steven Stein. He's going to walk you through elements of the development strategy for these two programs.
Dr. Garralda and Dr. Hidalgo will then walk us through the data, and we will come back with more details. Thank you.
Thanks, Pablo. As Pablo said, we are going to focus on the solid tumor side of our development at Incyte. Just to start you off on our brilliant logic on our numbering systems, INCA, and the A is the antibody, 33890 is our TGFβR2 antibody. It gives you some sense of the number of compounds that are screened before these things get into the clinic. I will refer to it as '890 going forward. Then INCB, in this case, a small molecule, 161734. I will refer to as '734 going forward, our G12D mutant-specific small molecule. I will take you through some of the development highlights related to this. We had four rapid oral presentations at ESMO, and I hope most of you saw them, but in case you did not, we will go through the data in them tonight.
Firstly, '890 in first-line Microsatellite colorectal cancer in combination with standard of care chemotherapies. Very importantly, the ability to show you data with FOLFOX-bev in this setting, both in terms of response rates and some evidence of durability. In terms of '734 presentations in pancreatic ductal adenocarcinoma, and again, very importantly, with standard of care chemotherapies in both [gem/ABRAXANE] and modified FOLFIRINOX, which is somewhat of a differentiator for this compound. Then '734 in second-line colorectal cancer in combination with an EGFR inhibitor, cetuximab, was also an oral presentation. Lastly, not the focus of tonight, but as Pablo mentioned, '667, our CDK2 inhibitor in CCNE1, overexpressing in this setting platinum-resistant ovarian cancer with bevacizumab, but is a very good setup for going forward our maintenance study in platinum-sensitive ovarian cancer, but not the focus of tonight's presentation. Let us talk about TGFβR2 by PD-1.
We think, and as Pablo alluded to, this has the potential to be a next-generation immunotherapy backbone. Go beyond PD-1 alone and addressing probably the two most important tumor mechanisms in terms of evading the immune system in PD-1 and TGFβR2. Broadly speaking, the takeaways we would like you to get from what we are going to show you tonight is the ongoing development of this bispecific antibody in Microsatellite colorectal cancer being the lead indication, and as Pablo said, well underway and enrolling very well, but with expansion opportunities in additional solid tumors. The first-line activity that we will take you through tonight we think is enabling in a large way for the ongoing first-line study.
We'll show you the FOLFOX-bev response rates and way above historical benchmarks for standard of care chemotherapies, and then the durability thereof as a read-through to potential down the line, both progression-free survival and an overall survival signal. In keeping with the rationale of the bispecific, the really encouraging and notable activity in liver metastasis, which themselves are TGFβR2 high expressing and supportive of the rationale in its entirety for developing this clever construct in terms of the bispecific antibody. Expanding, again with a phase II study, but an elegantly designed proof of concept in patients with Microsatellite colorectal cancer who have had curative resection, attempted adjuvant chemotherapy, but are then ctDNA positive and then are treated with '890 for up to 12 months, and I'll show you that schema. High TGF-β tumors are PD-1 inhibitor resistant.
The pathway is hyperactivated in immune-excluded tumors that are resistant to these anti-PD-1 inhibitors. Binding to the TGFβR2 receptor leads to suppression of CD8-positive T cells and then, as a consequence, immune exclusion of those cells. Additionally, high TGFβR2 expression is shown to be significantly associated with non-response to checkpoint inhibitors and reduced survival overall. In terms of the biology behind this schematic shows on the y-axis T-cell TGF-β gene expression and then uses a Pearson correlation coefficient on the x-axis to look at TGFβR2 protein expression in solid tumors. You can see circled for you both Microsatellite colorectal cancer itself and then the liver Mets are high scoring in terms of regard and are enriched for the biology that's addressed by the bispecific antibody. In terms of the construct here, it's, again, a very cleverly designed bispecific.
Given the history with TGF-β biology, where toxicity was largely limiting over the last couple of decades with different mechanisms trying to address the biology, this is a high-affinity PD-1 arm. It's more than tenfold higher affinity if you compare it to the TGFβR2 arm. The mechanism is sort of docking in terms of the PD-1 and then blocking on the immune cells, the TGFβR2 signal, and thus sparing normal tissues which enables normal function to continue and then avoids the known toxicity of the broad pathway blockade. So a smart construct that we've taken forward. Pablo alluded to it, highlighted in dark blue are the ongoing efforts, firstly in the middle, the first-line phase III NEXERA-303 study is well underway. It uses the care standards of FOLFOX-bev plus minus the '890 antibody.
On the left of that, the adjuvant study is also starting as we speak, in patients who've been resected, have had adjuvant chemotherapy and are then in a window of up to a year ctDNA positive and then treated with '890 with the proof of concept being ctDNA clearance and documenting that. There's ongoing work with the muzastotug CTLA-4 antibody there on the right, a bevacizumab combo alone, some of the data shown at this meeting. In ovarian cancer, combinations with chemotherapy and bevacizumab, and then an ADC that'll be announced shortly. A novel EGFR antibody in head and neck cancer that'll also be announced in chemotherapy combinations in lung cancer, but not addressed tonight. With that, I'm going to hand it over to Dr. Garralda to present the data further.
As Pablo said, she is the Co-Director of the Vall d'Hebron Clinical Research Program, the group leader of the Early Clinical Drug Development group, and the Director for the Research Unit for Molecular Therapy of Cancer. Dr. Garralda.
Thank you very much. It's really a pleasure to be here and walk you through the data that has been presented at [INCA33890] and during ESMO. Let me see if I manage to. There you go. I think you've seen already the need that there is for therapies in colorectal, but I'll just walk a bit through perhaps a concept that has not been stressed. The 85% of cases of colorectal are going to be MSS. For these cases, still immune therapy has not really managed to provide any substantial benefit to these patients. It's only the 15% that are going to be MSI-high that really have obtained benefit from immunotherapy. Just to put the things in context, it's also important to understand what is still the current first-line treatment, and this is still the chemotherapy with FOLFOX and bev.
What would be the benchmark would be a response rate of around 45%, 50%, medium PFS of around 10 months and 25, 30 months for median OS in phase III trials. These would be the numbers that we would be benchmarking the results today. INCA33890 is really a big phase I study. We did the dose exploration that explored different doses, and really we explored and reached that 900 mg IV every two weeks was the recommended phase II dose, and this led to several expansions. We have done expansions with different chemotherapies based in the different settings. You see that it's all patients with advanced colorectal, and then depending on the chemotherapy, there were different lines of treatment. I'm going to focus on the combination with FOLFOX-bev for frontline, but you can see in general the patient situation for all treatment indications.
The median age was 56. It was a good prognosis that most of the patients, all of the patients were ECOG zero or one. What it would be important to highlight is in the frontline, first line, 67% had liver metastases. This is, I guess, the normal pattern of colorectal. As you've heard in the previous presentation, liver metastasis is a worse prognostic factor for colorectal cancer and also has been proven to be a predictive factor for immunotherapy with those patients that have liver metastasis responding worse to immune checkpoint inhibition. Also regarding PD-L1, you can see that 61% of patients are PD-L1 above 1%. As you know in colorectal, there's also importance about the sidedness. Just to say that 65% were left-sided versus 30% right-sided. This is the normal prevalence of left and right. 63% were KRAS mutants.
These have slightly worse prognosis, but in general, a representative population of what we see in clinical trials, in global phase III clinical trials. It is a phase I study. What did we see in the combination? When we are combining drugs, of course, one of the first things is to see if we can actually give this combination. The combination was safe. We did not observe any DLTs. It is important to stress that we did manage to maintain the relative dose intensity of the standard of care. That would be one of the concerns when you are combining in frontline with standard of care. There were some adverse events leading to those delays or discontinuations, but you can see the numbers are low. 9% for '890. What would be expected from the combination of around 67%.
You need to remember that FOLFOX when it is given in a long time, there is normally reductions of the oxaliplatin because of the toxicities. Important to mention that we did see some immune-related adverse events. These are similar to the proportion that would be seen with regular checkpoint inhibition, with only 14% having more than a grade three event. As I have mentioned, we did not observe any fatal events or grade five events. If we dive a bit deeper in which were the most common treatment-emergent adverse events, you can see that they are really well-established adverse events due to chemotherapy. You are going to be seeing the neuropathy of oxaliplatin that was mainly low grade. We did see 2% only of grade three neuropathy. Then the GI toxicities that would be expected due to the FOLFOX toxicity.
We did see some infusion-related reactions, but they were mainly due to the oxaliplatin, so also the standard of care, very few infusion-related reactions due to '890. In general, adverse events that would be consistent with what is known of FOLFOX chemotherapy and also of immune-related adverse events due to the checkpoint inhibition. Also perhaps important to mention the hypertension. Only 28% of cases had hypertension, also similar to what we would expect with treatment with bev. Here you can see the waterfall plot. We did have 42 patients were evaluable for efficacy in this first-line MSS colorectal cohort with a median follow-up of nearly six months. We did see 71% response rate, and this was confirmed in 67% of the cases with a six-month PFS of 81% and a disease control rate of 100%.
You can see with the arrows that most of the patients are still on treatment, so it is still early times to actually understand the durability of the response. But as mentioned, it is promising in the stage that we are. If we go to subgroup analysis, it is also important to highlight different patient characteristics. As I have mentioned, liver mets is still a big area of need in colorectal. You can see in patients with liver metastases that are actually drawn in blue in the presentation, the response rate is 83%. Peritoneal invasion also a hard place to reach metastatic site, 62%. Then also important to mention that in those patients that were PD-L1 high that we already had seen in the monotherapy that were enriching for response, actually the response rate reaches 81% versus 50% in those patients that were PD-L1 less than 1%.
The response in the right side is 92%, and the left side it is 64%, but these are small numbers. To conclude, I think that the data has shown that there is a tolerable to combine FOLFOX-bev with '890. We haven't seen any DLTs, still early times, but for now, no safety concern. There is notable preliminary efficacy in this frontline MSS with '890 FOLFOX and bev. The response rate has reached 71%, disease control rate of 100%, and six month PFS of 81%. We have seen responses in patients that are hard to get responses, such as patients with liver metastases with a response rate of 83% and patients with peritoneal metastases 62%. Also, this is enriched in patients with PD-L1 positive reaching out to responses of 81%. This data support the phase III NEXERA study that you have heard in the previous presentation.
Thank you very much for listening.
Thank you, Dr. Garralda. To reiterate a few points in terms of where we are on the efforts in microsatellite colorectal cancer and then expansion beyond that. It is a high incidence tumor, as you saw from Pablo's data, and currently to date has not been sensitive to checkpoint inhibition in any way. The phase III study NEXERA-303 is ongoing, and as Pablo said, is enrolling very well. Beyond that, we will be starting a proof of concept study or is starting as I speak in colorectal cancer resected post-adjuvant chemotherapy that is ctDNA positive, and I will show you that study schema in a moment. The novel combinations that were mentioned to you as well are underway. There is a few important disclosures on the slide. I will highlight them. One is that the top-line data is expected in 2029, right at the top there.
This is histologically confirmed metastatic colorectal cancer, no prior treatment for the metastatic disease. Patients, if they have received either neo or adjuvant therapy, were allowed to enroll if they didn't have a recurrence within 12 months of the end of that therapy. They could not be MSI-high, obviously no BRAF mutation, and no immune checkpoint inhibitor within the prior three years, and an ECOG of [zero to one]. The goal N is 700. It is a standard 1:1 randomization of '890 at a 900 mg Q 2-week IV dose with standard FOLFOX-bev chemotherapy versus placebo.
The primary endpoint is progression-free survival, and then an important disclosure just after that based on the data Dr. Garralda showed you that the [inaudible] PD-L1 positive group first, which is based on the data we have has a much higher chance of success in terms of at least the response rate data. A key secondary, obviously, in this disease is overall survival, and then other secondary endpoints listed there for you. In terms of the adjuvant setting, there is obviously significant unmet need for patients with colorectal cancer who have documented MRD by ctDNA post-surgery. About 15% of stage two patients and double that, 30% of stage three will experience recurrence. The curves show you, and for stage two and stage three, median disease-free survival if you're ctDNA positive.
For stage twos, if you're positive in this particular data set, which is large, 1,100 + patients, a median DFS of 12.7 months if you're positive. For stage three, 16.2 months. Overall, that's about a one-year median DFS if you're ctDNA positive post-adjuvant chemotherapy. There is also evidence from other work that if you're able to clear the ctDNA, and that's the POC for this particular effort, that would alone be a biological surrogate for disease-free survival. Hence, the design of the proof of concept study. Here it is for you. You have to have microsatellite colorectal cancer, a curative resection, prior adjuvant chemotherapy, be stage two or three, ctDNA positive with a much larger window that's typically used in the setting up to 12 months, and have no evaluable disease on PET-CT screening. It actually needs quite a low number to get there, 40 patients overall.
You can see the schema above. We treat for up to 12 months, and the primary endpoint is ctDNA clearance at three months with other endpoints below, including disease-free survival and some exploratory biomarker work. Just to reiterate, a potential immunotherapy backbone that goes beyond CPI alone by addressing TGFβR2 biology. Topline data for the registration study in 700 patients in 2029, based on enrollment currently. The efficacy measures that were shown tonight, we think point to, as we said, notable clinical activity and overall response of 71%, disease control rate of 100%, and six-month landmark PFS of 81%. Hopefully over next year in 2027, we'll able to update the PFS data as data matures.
Including, as we've shown to you, activity in difficult-to-treat subgroups, liver mets being a clinical marker of checkpoint resistance, an 83% response rate there, peritoneal mets 62%, and the response rate in the PD-L1 positive by TAP scoring of 81%. Then the move into the adjuvant setting, which we think is really important for the program in that ctDNA positive population. Then ongoing work, which will be the subject of later updates, ovarian, head and neck, and lung cancer. Turning to G12D, the idea is to build a franchise across G12D mutated tumors. That's where we are with '734, and we're very proud of the data at this meeting, but also where we're going with the compound beyond that. Again, the key takeaways that we want to show you here tonight, that this itself has the potential to be a backbone for G12D mutated gastrointestinal cancers.
The phase III in PDAC is ongoing and enrolling very well. A planned registration study, which I will show you in a minute, in second and third-line colorectal cancer in the first half of 2027, and then the adjuvant PDAC study. The first-line PDAC activity in phase I supports ongoing phase III development. We were able to show you activity with both standard of care chemotherapies, modified FOLFIRINOX or [gem/ABRAXANE]. Again, what we think are historical benchmarks above standard of care. The promising colorectal cancer activity that supports expansion across lines of therapy with the cetuximab response rate above an historical benchmark for second and third-line standard of care chemotherapies. Combinations that Pablo alluded to with a PRMT5, a PARP1 inhibitor, and an EGFR LGR5 bispecific that will all begin in 2027. '734 itself is a highly selective G12D on and off inhibitor.
It is a scaffold that binds at the switch-II pocket with picomolar affinity. That translates to a GDP on binding affinity of [0.38 nM] and a GTP off binding affinity of [0.32 nM]. The selectivity of a wild type is greater than [80-fold], and we think that the construct of an ON/OFF inhibitor may translate to more durable efficacy as seen with either only on or only off inhibitors, and we can talk about that more in Q&A. It is, to our knowledge, currently the only KRAS G12D ON/OFF in phase III development. Again, the entirety of the program is listed here for you in the middle. The phase III DAWN-303 study ongoing with both standard of care chemotherapies and adjuvant study on a chemotherapy backbone. So chemotherapy plus minus '734 that will begin in 2027 with curative intent.
The novel combination work on the far right there in PDAC. In colorectal cancer, the most established data we have is with cetuximab, which we will show you at this meeting, but then ongoing work with an EGFR LGR5 bispecific, a PARP1 inhibitor, and that phase III, that data with cetuximab would enable a registration study to start in the first half of 2027. We are obviously interested in other G12D mutated tumors as listed below, including non-small cell lung cancer. With that, I will hand the podium over to Professor Hidalgo, who, as Pablo said, is the Director of the GI Cancer Center at the Perlmutter Cancer Center at NYU. Professor Hidalgo,
Good evening, and thank you very much for inviting me to discuss this exciting data. I think most of what I will be presenting today was already discussed or presented at ESMO, so it will be a nice opportunity to summarize some of the data. Pancreatic cancer, very serious lethal disease. It is very commonly RAS mutated. The majority of patients have a mutation in the KRAS oncogene, and G12D allele is the most commonly mutated, up to 40%. Management of patients with advanced disease, with metastatic disease, which is basically half of the patients at presentation, and another 60% of those of the patients who present with earlier disease will progress a few years after resection and adjuvant chemotherapy. It is with the systemic multi-agent chemotherapy, not very effective in phase III trials, response rates in the 20%-30% range, median survival around one year in the best case.
For that reason, the overall survival for the entire pancreatic cancer population is about five years. 10% for stage four disease is very dismal, only 3%. A very significant medical need. We will never get this point right. It is the other way around. The data that will be presented today is the part of the phase I trial, the combination of the novel agent '734 with both of the standard of care chemotherapy in patients with untreated advanced adenocarcinoma of the pancreas, locally advanced or metastatic disease. Of course, they needed to have a confirmed G12D mutation, good performance status, no prior treatment with any of those RAS inhibitors. The study proceeded in two steps. First, dose escalation from 600 mg to 1,200 mg of the '734 with either gemcitabine ABRAXANE or modified FOLFIRINOX at standard of care dosages used in this disease.
The selected recommended dose was expanded. In aggregate, if you look at the patients who were treated at the recommended dose, we have 26 patients for GA and 28 patients for modified FOLFIRINOX. The primary endpoints were safety and tolerability and then ORR, disease control rate, and PFS. These are the patient disposition. Let us look at the tables. Very common population for this type of first-line advanced disease pancreatic cancer clinical trial. The only noticeable difference is that in the gemcitabine ABRAXANE group, if you can see, there are significant number of patients, 38%, who had received prior chemotherapy. Those patients likely have received prior modified FOLFIRINOX in the adjuvant setting and then they progressed and were entered into the study. The majority of patients, 85%-86% with liver metastasis, which is the most important prognostic factor.
This is a very standard population for patients with advanced pancreatic cancer. These are the safety data for the gemcitabine and nab-paclitaxel combination. In general, as expected with chemotherapy, for this particular chemotherapy, very low rate of grade three events. Perhaps a little bit more neutropenia, but as I said, a large proportion of these patients had been likely pre-treated with modified FOLFIRINOX. Then the GI toxicity, nausea, and vomiting that we see with '734, which is grade one, two and is easily manageable. If you look at the relative dose intensity, it is about 85% for the '734 and then about 70% for the first few cycles, four cycles for chemotherapy, which is very similar to what you would be expecting in a standard of care. No really major concerns or concerns at all from the safety perspective. This is the data for safety with modified FOLFIRINOX. Similar.
Actually, the incidence of neutropenia is lower here, which is rare compared to the gemcitabine ABRAXANE, probably because very few patients have been pre-treated, I think only 11%. Probably most of these patients with modified FOLFIRINOX, at least in the U.S., will get growth factor support. Expected safety profile of the modified FOLFIRINOX backbone with, again, higher incidence of mild, easily manageable GI toxicity, nausea, and vomiting. Again, the dose intensity, very high in the 70% range for chemotherapy, which is what you would be essentially expecting. This is the combined response rate for the two groups. The ORR for gemcitabine and nab-paclitaxel is 69%, and for modified FOLFIRINOX, 89%. The confirmed and pending is 58% and 81% with close to 100% disease control rate. A few patients with complete responses, which is totally unheard of in this particular disease. These data are maturing.
We don't have yet a fully analyzed PFS and, of course, OS. But when you look at the swimmer plot for PFS, the six-month landmark analysis for PFS, which with chemotherapy is around 50% for a standard of care chemotherapy with this regimen, is 80%, 90% for the two regimens. Compared in this non-comparative trial, I think very well with what is expected based on standard of care chemotherapy. Responses seems to be durable, which is very important, of course. We need more the data to mature to analyze this in more detail.
The conclusions for the pancreatic cancer component of the presentation is that there is a favorable safety and tolerability profile in combination with both standard of care [inaudible] and modified FOLFIRINOX, very consistent with what is expected for these standard of care chemotherapy regimens, with only perhaps a higher incidence of easily manageable GI toxicity, nausea, and vomiting, which was observed in the single-agent phase I trial as well. Very deep and appears to be durable responses with both regimens. Responses in the 70%- 80% range for this, 70%-90% range for these two regimens with six-month progression-free survival of around 80% and 90%. Expected, it's about 50% in those chemotherapy regimens. This supports the ongoing phase III trial comparing '734 with the gemcitabine ABRAXANE and/or modified FOLFIRINOX in combination in this patient population.
We go now to colorectal cancer, which there is less data, but also very interesting. Colorectal cancer is, of course, more common than pancreatic cancer, but the frequency of RAS mutation is lower, is about only 50% of the patients. If you look at the RAS 12, the allele is only 15%. As discussed before, patients receive combination chemotherapy, cetuximab, bevacizumab. Patients with the mutant KRAS do not benefit from cetuximab, so basically, they're managed with FOLFIRI, FOLFOX, bevacizumab. Then the standard of care second or third-line regimen, depending if they got triple chemotherapy or doublet chemotherapy, is with LONSURF bevacizumab or with regorafenib. Very low response rate, less than 5%, and median PFS of 5.6 months with median OS less of a year. So very significant medical need as well in patients with later stage colorectal cancer metastatic disease with a G12D mutant allele.
As said before, many years ago, it was established that the EGFR inhibitors don't work in this particular situation. For that reason, they are not used, but there is preclinical and early clinical data that the combination of the RAS G12D inhibitor with cetuximab is synergistic, and that has been explored in this trial. This synergy, which probably is also applicable to other diseases, pancreatic cancer, less data, lung cancer, some data as well, but where we think is more relevant is in colorectal cancer. This is the clinical trial in combination with cetuximab in patients with metastatic colorectal cancer, locally advanced or metastatic disease with G12D mutation, good performance status, who have progressed to standard of care chemotherapy. There was a single agent, dose expansion, dose evaluation from 200 mg to 1,600 mg.
The drug is active per se, a single agent, but most of the activity, as you will see, is in combination with cetuximab and is very well tolerated at both 600 mg and 1,200 mg. The recommended dose for the clinical trial moving forward is the 1,200 mg, but it is very well tolerated. These are the characteristics of the patient population. Really nothing very special. Majority of patients with liver metastases. As said before, all of them have received a significant number of prior chemotherapy regimens and bevacizumab or similar VEGF inhibitors, which is the standard of care for this population. Nothing really too special in these two populations. Very classic second, third-line patients for clinical trial. This is the safety data. On the left in blue you have the single agent safety data.
That is, as mentioned before, basically the only important side effect is GI toxicity, nausea, and vomiting.
But as you can see, the majority of them is grade one. On the right in green, you have the safety for the combination with cetuximab, which the major difference is in the acne, the rash. Cetuximab causes rash. This is well known. It is manageable. There is a tachyphylaxis mechanism, and with time it goes away. Patient in general can take cetuximab without problem. Because this RAS inhibitor alone has very little rash, as you can see in the slide, can be combined with cetuximab, which other agents, as you know, have more skin toxicity, and these combinations are simply just not feasible. Here, I think this is the only real opportunity to test this combination in clinical trials. Again, most of the patients were able to stay on trial with very little or very few dose modifications or interruptions.
This is the data of the combination. It is a small cohort, but the ORR is 36%. Remember, with the standard of care, it is less than 5%, actually it is close to 0%. There is basically no responses. This is control rate of about basically 100%, no patient have disease progression in this cohort. The single agent monotherapy, there is some activity. Actually, it is more than with the standard of care, is 8% and 79%. For this reason, because of the tolerability and the higher efficacy of the combination, it makes sense to move this one forward into more definitive clinical trial. As I said, these responses are durable. Median PFS is already 7.2 months. It is about five months with the standard of care. So it is a very interesting data set and in a population for whom there is no very effective treatments.
This is biological data, cell-free DNA, tumor biopsies collected in the context of the trial. Two pieces of information. The first one is the changes in the G12D VAF, which as you can see is very, very profound. We are seeing this across different diseases, with the majority of patients in this particular trial having more than 90% control or elimination of the mutant allele. It is becoming clear that this correlates with outcome. You will see data down the road, yet unpublished, that patients with a stable disease who have this clearance of the mutant alleles also derive substantial benefit. So I think this surrogate or intermediate or biological endpoint, whatever you want to call it, will be progressively incorporated into clinical trials.
And on the right, you have the baseline genomic characteristics, a more detailed mutational analysis beyond the G12D, as expected for colorectal cancer, but nothing that really correlates on the left, the responders, on the right, the stable disease, nothing that really correlates at baseline with outcome. For this reason, the only selection criteria should be G12D mutation for this trial. Now, it will be very interesting in these trials as we move forward to repeat these post-treatment biopsies and see where the escape mechanisms of patients when they progress. Similarly to the profound correlation between allele, the VAF decrement for response when patients start to progress, in some patients, we see that going up, not necessarily the RAS allele, but some other mutations start to emerge. So that's a very interesting translational field that we will navigate as we go. For colorectal cancer, these are the conclusions.
Very manageable toxicity profile as monotherapy and in combination with cetuximab with only the rash, but because the single agent doesn't have a skin toxicity or has very little skin toxicity, this makes this combination feasible. No DLTs in the escalation and low rate of side effects leading to discontinuation. Very interesting preliminary activity in a small cohort, but 36% response rate, universal disease control, and PFS 7.2 months, and this supports clearly the continuation of this clinical trial in patients with KRAS G12D mutant colorectal cancer. And I think this is my last slide. Thank you very much.
Three visits to the podium. It's testing my ability with flashing lights and not experiencing a seizure. We'll see how we go here. So '734, I know, our mutant-specific small molecule in pancreatic ductal adenocarcinoma is ongoing in the phase III DAWN-303 study, enrolling really well, and the adjuvant study to begin early 2027. The colorectal data just shown to you, leveraging the synergy with EGFR blockade, a phase III second/third line colorectal study planned in the first half of 2027. And enabling work in first line setting is ongoing as well as a potential going forward. And then the novel combination work as shown to you as well. Another important takeaway here, the top-line data for the phase III study, given the enrollment characteristics to date, is expected in 2028.
Study eligibility is histologically confirmed metastatic PDAC with a KRAS G12D mutation, no prior therapy for metastatic disease, ECOG performance status of [zero to one], and very importantly, measurable disease by RECIST is needed with an N of 590. I'll come back to the why behind the RECIST criteria in a second. Randomization is 1 :1 . Chemotherapy plus '734 versus chemotherapy plus placebo. Stratification factors are there for you, the chemotherapy regimen, ECOG status, the presence of liver mets, and geographic regions. In terms of the endpoint, the primary are three overall survival, progression-free survival by a blinded independent review committee, and an overall response rate, which you noted in the ClinicalTrials.gov listing. T hat's why the measurable disease by RECIST is required, and gives a potential opportunity to take that forward at a regulatory level should it deliver the data that's encouraging in that regard. Standard secondary endpoints.
The adjuvant setting beginning in 2027. In PDAC, we think the best way forward is to hit it early and hard with the chemotherapy plus minus '734 combination study being the likely study that we do there, enabled by the data shown tonight. The rationale is obviously to use this in a curative-intent setting to eliminate micrometastatic disease and extend disease-free survival. Study details, broadly speaking, are post-resection PDAC chemotherapy backbone with a planned phase III start in 2027. In colorectal cancer, as shown by Professor Hidalgo, again, a phase III start first half of 2027, enabled by data you've seen here tonight. Study is cetuximab plus '734 in second third line colorectal cancer. Again, a high incidence tumor with significant G12D mutated disease aiming to take advantage of the synergy with EGFR antibodies.
Study details, broadly speaking, second third line colorectal cancer that bevacizumab experienced and that helps give you the control arm to beat likely PFS. The regimen again is cetuximab plus '734, a start first half of next year. I hope this illustrates for you why we think this will be a multi-combination franchise. The pancreatic combinations are obviously both chemotherapy regimens, [gem/ABRAXANE], modified FOLFIRINOX, then a PRMT5 with a BMS navlimetostat, the PARP1 from Synnovation. Colorectal combinations, you just saw cetuximab data. Ongoing enabling work in earlier settings with FOLFIRI and FOLFOX with cetuximab, and then FOLFIRI, FOLFOX with petosemtamab as well will be ongoing. The next steps, establishing it in first pancreatic ductal carcinoma topline data 2028 into adjuvant 2027. New tumor type with colorectal cancer pivotal initiation first half of next year. Ongoing work in the first line setting, and that's up and going.
The novel combinations as elucidated for you earlier. With that, I'll hand the podium back to Pablo.
Thank you, Steven, Dr. Garralda, Dr. Hidalgo. Let me close emphasizing a few key points here. You heard data and the development plans for these two programs, '734 and '890. I should add here that we're very proud that both programs were entirely discovered and developed at Incyte, in fact, by the team led by our CSO, Patrick Mayes, who's sitting right here. So we're very proud of that fact that we are really delivering on our promise of discovering and developing novel therapeutics. Both are in phase III, as we discussed and Steven reviewed in great detail. Based on the data that has been presented at this meeting, we remain convinced that not only they're very competitive in this space, but that we have shown data that substantially de-risks the ongoing pivotal trials. I think that, to our eyes, is becoming increasingly clear.
Both programs have relatively near-term milestones. As Steven Stein described, the '734 pancreatic cancer study will deliver top-line data in 2028, which is just around the corner, and the TGF-β/ PD-1 in colorectal cancer in 2029 will provide more details on the other pivotal trials as they get initiated. As I mentioned in my introduction, we think where the greatest patient benefit is the earlier we can intervene in these diseases, hence the important push in the direction of adjuvant therapy. Proof of concept with '890 in the adjuvant setting in colorectal, and we are initiating adjuvant in pancreatic cancer next year. As we described at the beginning, not only these are very large medical needs, but they provide substantial commercial opportunities for Incyte in what is a growing franchise around GI oncology, the focus that will continue to expand as we described over the past hour.
With that, concludes our remarks. Thank you everybody for listening. We can now open for questions here or online. If I can ask, are we going to bring them to the stage or they can. Do we have a microphone? Oh, here we go. We have chairs for you if you do not mind joining me. We can get started, Alexis. Oh, there is a lot of questions. Oh, we want to give a minute. Eric, Michael, Evan, one sec. I will stay here so we can stay charged on it. Okay, that is fine. All right. Okay. Eric? Oh, Michael grabs the mic. Go ahead. [audio distortion]
[audio distortion] AEs. Just curious how you think about that in a competitive environment. The second question I had was on the DAWN-303 study. Just to clarify if overall survival is necessary to hit to declare success and if you think if that is possible, just given possible post-progression treatment with other KRAS inhibitors now in PDAC?
I am going to ask Dr. Hidalgo to comment. One point I would make is the data that has been reported for zoldonrasib in combination with the same two chemotherapy regimens and ours, our duration of exposure is longer. It is between 1.5x and 1.7x longer duration of exposure. You treat patients for a longer period of time, you have more time to develop adverse events, therefore, those rates of adverse events, you need to keep that into account. We are not presenting exposure-adjusted event rate, which we could do, but that is not the way we presented that. We presented all the data so you have it all, but the exposure is different, therefore, the numbers are different. Let me ask Dr. Hidalgo to comment on how he sees the profile here, and then we will go to Steven with the question on DAWN-303.
The question was zoldonrasib plus daraxonrasib or zoldonrasib plus chemo?
Chemo combo.
The chemo combo.
Do you see any differences between that and the '734 plus chemo?
I don't know. I think it's very early to say. I think they appear to be very effective, which is good. But I think we need to look at the full data set, more follow-up, the duration of the responses, the timing of the side effects, the molecular analysis of the cell-free DNA. At this moment, I think they are both very interesting combinations, but time will tell.
The median dose intensity for zoldonrasib is 72% and 59% for the two regimens. For us, it's 67% and 69%. I don't find those numbers to be different. The rate of events, as I discussed, you have to keep the duration of exposure into account. Steven, on DAWN-303, please.
Michael, thanks. The primary endpoints are three. Progression-free survival, overall response rate, and overall survival. The response rate one was disclosed in the clinicaltrials.gov listing. We're able to do that analysis at an earlier time point, but likely when the study's fully accrued, given how well it's accrued. It has the potential to be an accelerated approval strategy, but that'll be determined at the time based on the data plus the durability. The good thing is the confirmatory part of it's inbuilt in the study itself. Given that the PFS and OS is in that study, that could serve as confirmatory and accrual is very likely to be incompleted. So we won't be dealing with that issue in that regard. The meat of your question is OS actually required? Given the short duration historically, we're likely to have all three endpoints.
Probably, PFS with the overall response rate and the right direction on OS would be adequate, but we are likely to achieve all three based on the data we have to date. Thanks.
[audio distortion] , BMO. Thanks for taking my questions and congrats on the data. So two from me. One, as you think about the benefit of targeting on and off with '734 versus the competition, when do you think that will start to play out in the data that we actually see? My second question is, I want to just discuss a little more about your partnership with Genmab, looking at '734 in combination with petosemtamab. What did you see in their data set or in the biology that had you want to partner with them?
Why don't you address the ON/OFF question, I'll come back to the petosemtamab question, Steven.
Our Chief Scientific Officer is here as well, Patrick Mayes, if he wants to add anything to what I'm about to say. So obviously we have an ON/OFF inhibitor, with picomolar affinity on the switch-II pocket for both the GTP and GDP states. That was intentional in what we do. We think there's potentially signaling through both the on and off state, and that may translate to a benefit. It may also be the reason, by the way, in colorectal cancer, that they didn't achieve ultimate PFS benefits, but that'll be determined obviously by the final efficacy data. I'll just see if Patrick wants to add anything in terms of the construct.
Yeah, thanks. The nuance here is that the mutant KRAS sits in two different states in the cell, right? The off state GDP-bound and the on-state GTP-bound. The off state is more pr`edominant. So 60%-70% of mutant KRAS in a mutant cell is off at any given time. So if you think about a once-daily drug, which these are, you have a peak to trough profile of that drug over time. For ON/OFF, you're maximally inhibiting all mutant KRAS in the cell across the dose interval. For an ON only, you're inhibiting a small fraction of the total KRAS pool, and then that pool increases over time as you reach trough. So the theoretical advantage is you're getting more maximal inhibition, which you know is very important for these mutant proteins. We're getting target coverage well above IC95.
We see that as important to drive these deep VAF responses. That's the nuance on the ON/OFF.
On the petosemtamab question. The discussion said very eloquently, in KRAS mutant colorectal cancer, EGFR is always a driver. Always. That's why very early on, we did the combination with cetuximab, and you saw the data, which we think it's actually quite compelling considering the background response rate you expect with cetuximab in this population, which is single digits at best. So we accelerated that combination. We're entering a pivotal trial early 2027. In parallel with that, I think Steven mentioned it, but there were so many new pieces of data in the presentation, just to emphasize that we have done work in combination with FOLFOX cetuximab and FOLFIRI cetuximab, that those cohorts have been enrolled. They need longer follow-up, which is why we haven't presented them, but they will enable potentially front line.
As we started to think about the broader colorectal strategy for '734, it became clear to us that there's a next generation of EGFR inhibitors that deserve attention. Of those, we decided petosemtamab, based on the data summary, which was presented here at ESMO, was the most compelling data set, and that's why we put in place that combination. I can't tell you today what the front-line colorectal study will be with '734. I can tell you that we're going to conduct one, whether it's with the backbone of cetuximab or petosemtamab. I think the data will determine that. Eric. I don't know where the mic went. Eric has been waiting, sorry, for a minute. We'll go to the back there. Yeah.
That's all right.
And then we'll go into the back.
Maybe for Dr. Hidalgo or the company, what is the role of chemotherapy in the first-line setting going forward if you do have a chance or a choice of using a G12D with a pan-RAS or with an EGFR or even with a PRMT5 inhibitor? Where do you think the patients are going to want to go? And how does that depend on the data, of course. For the company, are you doing any GI prophylaxis on '734? Thanks.
The chemo question, all is going to be, of course, data-driven. I won't be surprised, and actually I hope that not too far from now, there are non-chemo combinations that in selected populations or in larger populations that appear in D5, results in substantial responses and duration of responses that can displace chemotherapy from the front line. But I think that chemotherapy still will remain part of the standard of care for later lines. No patient is being cured, unfortunately, with these combinations. We delay the progression, but eventually we're seeing that with that action, we're seeing that in all the studies, there will be a point we treat them with chemotherapy. So one question would be, are they more sensitive or differentially sensitive to chemotherapy? But as we speak today, I think the chemotherapy is the standard of care in the first-line setting.
The combos with chemo that are well-tolerated, I think, are going to get there sooner than the targeted combinations. Then we will see how the impact on OS is going to be also important. Probably, these are killing different cell subtypes. So do you want to kill them all at the beginning or kill a few and then kill the others? Or is the impact of that in survival will be important? So it's really interesting questions for the future, but I don't think chemotherapy is going to move from the front line in the next few years.
Steven, on the GI prophylaxis, please.
Eric, thanks for the question. In terms of nausea, vomiting, and diarrhea, learning as we went through the phase I program is to very proactively manage them, both with the antiemetics and the antidiarrheals, and then training in that regard. I think you saw it play out in the data sets in terms of very low discontinuation rates despite the relative amount of those different adverse events. Then certainly in terms of, and Professor Hidalgo alluded to this, patients getting used to on how to manage themselves in terms of when to take the therapies and how, because you see amelioration of those toxicities as the cycles progress. Second and third cycle and beyond rates actually get much, much lower.
I think the combination of training, appropriate use of treatments, and then patient adaption to the regimens is what's happening there.
Andrew Berens, Leerink Partners. Congrats on the data and the very extensive programs. Maybe to follow up on Eric's question. When I look at '734, it's obviously very extensive, a lot of different combination partners, but there's one that's conspicuously missing, and that is a pan-RAS inhibitor. Why would you not want to do that trial now just to protect yourself in case that's the direction that the clinical paradigm evolves to? We're hearing a lot of enthusiasm amongst clinicians. Also for '734, is it brain penetrant? If it is, what kind of side effects do you see from that?
On the second, is brain penetrant, it's not. That's pretty clear from the preclinical data. On the first part of your question, look, we are not done establishing combinations. As we make clear in the slides, both for '734 and '890, we have ongoing discussions for additional combinations. I agree with you on the potential merit of a combination with pan-RAS inhibitor. We haven't selected one at this point, we haven't put that in place, but we're not ruling out doing that as well. We thought other combinations were more urgent, because we have, particularly on PRMT5, I think we need to accelerate that collaboration, which is one of the reasons why we selected the compound that is in the lead. It's in pivotal trials already.
But stay tuned because more collaborations will be announced in the future, and one could very well be with a pan-RAS inhibitor.
Combinations, if I. It's that every day, there is a paper, a preclinical paper that shows synergy, right? There are, I don't know, 20, 30 very cool combinations that everyone is kind of. I think it's going to be quite a crowded space. What are the right combinations and which one of those we need to advance to the clinic? But there are many, many others that in preclinical studies look very, very interesting. Which ones we select, which one do we pick is going to be a challenge, yeah.
Matt?
Thanks, Matt Phipps, William Blair. Thanks for all the details. First of all, as you think about some of these combinations, any thoughts on an EGFR combo in [pink], maybe even petosemtamab? Is that a possibility under the current agreement? Then, as you look longer term, and this might be also a question for Dr. Garralda. If EGFR bispecifics become left-sided KRAS wild type standard of care, and then some KRAS combo moves into the KRAS mutant across these different variants, do you need to try to fit '890 into those? You're running this ['890/bev] combo. Is that patient population going to shrink as these others come up? Thank you.
Why Dr. Garralda addresses the interplay between future EGFR inhibitors and '890 on left-sided colorectal, then I will come back to your question about pancreatic.
No, I think it is a very relevant question, and I think in the front line for colorectal is. Right now, there are a lot of pivotal studies, which is a first thing for a long time, because it has been without changing for long. It is true that the impact of RAS inhibitors in colorectal right now is a bit more complex because of the combination that require, because of the feedback loops. The durability of responses that have been seen mainly in second line and third line are not very good. We will be seeing, of course, all of these combinations in the front line, but I think they are a bit behind from what is happening in pancreas.
It is true that the right-sided, left-sided is something that is there, but the reality is, most of, or I would tell you, the four big phase III programs that are being run right now in frontline colorectal are enrolling globally, so right side and left side, although they are stratifying for right versus left. I think there are some challenges and there is, I would say, some differences on how this is treated in Europe versus in U.S. regarding the use of cetuximab or not. But I think having a global approach still remains a very defensible position for both.
On EGFR and pancreatic, we did generate some data. It is active. We honestly decided to accelerate the EGFR combination colorectal because, based on what has been disclosed, we think we can probably be in the lead in colorectal cancer as opposed to pancreatic, and we thought there were other combinations in pancreatic that have more merit. Now, the combination with petosemtamab could change that, but we will decide that in the future. But right now, the focus in pancreatic is where we describe this number novel combinations, even though there is activity with EGFR inhibition. The role of EGFR as a driver in pancreatic is less clear than in colorectal, as you know, so we thought it was a little bit of a less urgent question to answer. Alexis? No, go ahead, please.
You also need to think that the frontline, the median PFS is 10 months already with the chemotherapy. So, you need to have a good tolerability to combine in colon frontline, because this will be important because of the number of adverse events that are going to be happening for a long time. So I think this is also a bit different from pancreas frontline, from colorectal frontline, where most of the patients, or a lot of the patients are going to have medium PFS that are longer than in pancreas.
Salveen?
Sure. Salveen Richter, Goldman Sachs. With respect to the combinability and duration of treatment, can you provide more context around the chemotherapy relative dose intensity for both cohorts and benchmark it against what we've seen historically, but also with competitors? Secondly, just from a high-level standpoint, just given the multiple KRAS G12D programs now showing promising data here in first-line PDAC, what do you believe will ultimately differentiate your program?
A couple of points. First, on the dose intensity. If you looked at the historical number with FOLFOX and [Gem/Nab], it's about 75%-78%. Those regimens are rarely delivered at full dose intensity. Particularly, the second dose, the week 8 dose of Gem tends to be dropped, and FOLFIRINOX requires adjustments. Our competitor, as I mentioned a minute ago, delivers 72% to 59%, and we deliver 67% to 69%. I find all those numbers to be roughly similar. When it comes with chemo alone, not with another G12D inhibitor, I think the important thing to remember is maybe there is a little bit of a reduction in dose-intensive chemo, but the gain in efficacy is, to our eyes, dramatic. We're almost doubling or basically doubling the response rate. The PFS continues to run. We have very few progressors, as Dr. Hidalgo showed.
If there is a little bit of a reduction in dose intensity, I think the benefit that you're getting from adding a G12D inhibitor, '734 in this case, this well compensates for that with much, much better efficacy. In terms of differentiation, we talked about the ON/OFF versus on mechanism. Whether that plays out in any meaningful way or in efficacy, time will tell. Right now, it doesn't look that way. To our eyes, the two programs in the lead, ours and our main competitor, are very comparable. I wouldn't say they're identical, but they're very comparable, both in efficacy and safety. We are both in the front. I think that our studies enrolling globally is enrolling extraordinarily fast. It will finish very rapidly.
I don't see right now, and maybe you disagree, is anyone behind us with a better data set than what we put out. I would simply say we're in the lead, co-lead if you may, and I haven't seen a better program coming behind us. We really like to be in that position today. Yes.
It's Darout, Barclays. Thanks for taking the question. Congrats on all the data. My first question is for Dr. Garralda, and second question for the company. First one is just thinking about the evolution of the frontline CRC space with multiple bispecifics moving forward, how you sort of see these different bispecifics being positioned in the space? Then the second question, we've seen in other tumor types, the synergy between TGF-β and EGFR. Just curious for your molecule, TGF-β bispecific, what you've done to maybe look at that combination and maybe be able to capture some of the synergy that we've seen in other tumor types, companies like Bicara moving their EGFR TGF-β bispecific forward in head and neck?
Okay. Dr. Garralda, you want to comment about the evolution of the landscape in colorectal cancer?
Yeah. Thank you very much. As I mentioned before, it's exciting times for the frontline of colorectal, and you have three PD-1, PD-L1, VEGF bispecifics running frontline clinical trials that are running very parallel to this phase III clinical study. The reality is that I would say PD-1, PD-L1, VEGF bispecifics have started a race. The data that they have presented in colorectal is small. I think the reality is that the combinations previously of [atezo-bev] were not very successful. Here they are adding the extra component of the bispecific. But how much that is going to add in colorectal, for me, it's difficult to know. I don't think there has been data presented, and I think there's a big race. Of course, an effect that is being shown in other tumor types, and probably this leads the fast programs that have been initiated in colorectal.
But the biology explanation, I think it's clear that colorectal is TGF-β driven. So that's why I think there's perhaps more a plausible biological explanation of why this could be a mechanism of tackling the immune system in colorectal more than the VEGF PD-L1. But it's just a hypothesis. There's three clinical trials that are running. We are seeing the effect of those drugs in other tumor types. We've seen the effect in biliary, and they're clearly effective drugs. What will happen in colorectal, for me, it's really difficult to know right now. But as I mentioned, the tackling TGF-β has been an important need, mainly also because of the liver metastasis and the immune suppression in the liver. So that's why it makes more sense, but later you need to see the data.
I will second that, and then I'll have Patrick comment on your second question. Colorectal cancer is one of the tumor types where Avastin is the most important, right? With colorectal cancer, ovarian cancer, GBM, basically. And we're able to deliver full dose Avastin, which is an important thing. You cannot do that with a PD-1 VEGF. Now, obviously, you have a VEGF arm, and that's important. And the second is what Dr. Garralda said, and we showed you in the presentation. We've done the work to understand the biology of why we should go after this tumor type. The TGF-β profile of colorectal cancer, metastatic colorectal in general, and particular liver mets is striking. So the data that we showed you today is striking with that biology. We have 83% response from patients, 31% patients with liver mets, which is, I think, an extraordinary number in this disease.
T he best part is that that is explained by the biology, and I think makes for a wonderful story. You want to comment on the combination with EGFR inhibition?
EGFR.
Yeah
Combination question. The concept's been proven in head and neck, I think. And then the logic would be in CRC should play out as well. We've dosed the combination, now established the safety of the combination. It's on our list of things to explore. So I agree. I think there's logic there in CRC as well.
Can't see you with the column. Who has the mic?
How are you? Reni Benjamin from Citizens JMP. Can you provide maybe a little bit more detail on the NEXERA and DAWN studies in terms of the improvement that you need to see in order for it to be statistically significant? I can claw this, but I'd rather hear it from you as to-
Yeah.
-what it is. [crosstalk]
Both NEXERA and DAWN, what is the hazard ratio we need to see for positive trial? We tend to keep those details a little bit quiet, but maybe you're lucky, and Steven will talk about it.
I can't see you, Reni, but I'll address the question. We try to put the historical benchmarks on the slides, and I hope I reflect them correctly. If you look at either the [gem/ABRAXANE] regimen or modified FOLFIRINOX, broadly speaking, you're dealing with 40%-50% response rates, PFS of around five months, and an overall survival that tends towards 10+ months a year. We design, in terms of hazard ratios, to beat that by amounts that are both statistically and clinically meaningful. In terms of the response rate, just to double down there a little bit, we have the ability to analyze that at an interim at some point next year when we will have completed accrual. If the data tracks where it is currently, we could be dealing with a doubling of response rate, and we'll see the durability at the time.
In terms of FOLFOX/bev in metastatic microsatellite stable colorectal cancer, you're looking at, and I probably should let Dr. Garralda address this, but she alluded to it, medium PFS of around 10 months and overall survival of 25-30 months in terms of the benchmarks you have to beat. You saw our six-month landmark data. Obviously, the data will mature over next year, and we'll get somewhat of a better sense where we're heading. Again, we set up hazards to beat that are both clinically and statistically significant. I think we like the setup currently. It's enabled the phase III, and it should continue to. It's unlikely that patients are going to immediately progress now. We like where that data set is currently.
Based on the final data that you get, can these phase IIIs be modified, less patients, potentially expanded based on how that's coming? I guess just to follow up on that, with '890, you have additional tumor types that you're looking at. What's the basis for that? Is it TGF-β? Kind of expression or preclinical data you have? Thanks.
Two questions there. On the first one, you can always amend a clinical trial. You can amend the statistical analysis plan. I do not think it is going to be necessary in this case, based on the data we have today and based on the benchmark that we have been discussing for the past hour. The studies are well powered. Remember, you would have an expectation of benefit in a study like this, and then you have a minimally detectable difference based on the number of events. The studies are well powered to really deliver with a high probability of success if the data continues to hold up. Yes, you can always modify if you need to. I think that is going to be very unlikely in this case.
In the case of the TGF-β/PD-1 program, we had a very broad phase I/II program that enrolled almost 400 patients, 385 patients, I think, to be exact. We threw a number of tumor types, and we saw a signal in three other tumor types that we presented not that long ago. Not large number of patients, but we saw efficacy in head and neck cancer, non-small cell lung cancer, and ovarian cancer. Particularly intriguing is ovarian cancer, since the response to PD-1 single agent in that disease is negligible. We saw double digit in the 20% response rate range. We are discussing internally and with collaborators how to pursue one or more of those three tumor types, and you will hear more about that in the next few months or early next year.
Hey. We have a number of questions on the line.
Okay.
I am just going to ask one on behalf of Brian Abrahams at RBC. It is a little bit of a two-part question. The first is regarding '734. Can you talk about your level of diligence in looking for an optimal PRMT5 partner, and what made you partner with BMS, and why did navlimetostat stand out? What kind of signal are you looking to see from the combination, and how quickly can you move forward to pivotal? The second question is regarding '890. Can you comment a bit more on the myocarditis and whether this is potentially drug or TGF-β related?
So two parts there. We looked at the landscape of PRMT5. To be direct, the decision to go with navlimetostat was because it is in the lead. It is in pivotal trials already. We thought we would have a great collaborator in BMS, and that is how the decision was made. The profile of what I would call this generation of PRMT5 inhibitors, it is hard to argue that there are substantial differences between the different compounds. Basically, the fact that it was in the lead made the decision easier for us. On '890, myocarditis, there have been four cases of myocarditis in the entire program. Almost 400 patients, I mentioned, 385. So in instance, it is about 1% or less. All four have been asymptomatic. One was in combination with FOLFOX/bev, again, asymptomatic patient.
The other three, one of them was at 1,500 mg, which is a dose we are no longer exploring.
All of them resolved. That is basically the background. Now, if you look in the literature about the incidence of myocarditis with PD-1 inhibitors, there is a background incidence, just myocarditis, about 0.5%, 0.7%. So pretty much indistinguishable from what we have seen. Then if you add a number of other cardiac complications associated with PD-1 inhibitors, you end up with an incidence of cardiac complications with PD-1 inhibitors much higher. So we do not think this is a signal that should worry us. Obviously, we keep patient safety top of mind. This seems consistent with what has been seen with PD-1 inhibitors more broadly. I do not know if there is one here. There is one here and one here. So we are going to need a mic over here, Alexis, please. Go ahead, Tara.
Hi. Tara Bancroft from TD Cowen. So I am wondering if you can all talk about, and particularly the doctors here, how important a first-mover advantage is in both PDAC and CRC. Let us say that the G12Ds all have relatively similar efficacy safety profiles. How much would that actually dictate your utilization, and what timeline would actually indicate a first-mover advantage to you? Like few months, a year? How long?
Well, clearly it has an advantage. When we get drugs approved, The one that gets approved first, then we learn how to use it. In this particular class of agents, as alluded to before, there is a little bit of an art on how you manage the GI toxicity. You need to educate the patient, you need to educate the nurses, you need to educate yourself. Every patient is a little bit different, and then you become familiar with a drug, you learn how to handle it. Second one will need to show something very different for you to change. I think that for that type of side effects, first comer will have an advantage. But I think that's common across the entire. There was another part of the question?
No.
That was it?
That was it.
Yeah. Okay.
Hey, thank you, guys. Faisal Khurshid from Jefferies. Just a quick one from me. Just wanted to ask, how should the street be thinking about the risk profile on '890? Because I look at the waterfall plot, it looks pretty amazing, but it's frontline CRC. We've kind of been burned by ORR in this disease as well as this mechanism. Just curious how you're thinking about risk and how you think we should be thinking about risk?
Yeah. Let me start from the back of that question about the failures in this mechanism, because I think that's a central point of this program, which is this is an entirely different idea. People trying to drug TGF-β have done basically two things. One is give a systemic agent that blocks TGF-β, like a kinase inhibitor against TGF-β. That is a systemic inhibition that led to substantial toxicity clearly related to TGF-β inhibition in normal tissues. A different approach was to use a trap, a ligand trap. Those tended to be not very active because it's really hard for a ligand trap to work across entire tumor spectrum in a patient.
What we did, and Steven alluded to briefly, was by combining the PD-1 arm with very high affinity for the target and the TGFβR2 arm that only binds to the receptor when PD-1 is fully docked, we really are blocking TGF-β only in the tumor microenvironment. We are convinced that that has made a huge difference in our ability to deliver full coverage of the target without the toxicity usually associated with TGF-β inhibition. So that's point number one. I think, as I mentioned, when you look at the entire safety profile, we've reported all the single-agent data in colorectal cancer a few months back. I think you look at the data set, this medicine, considering the mechanism, is very well tolerated. The second part of your question is response rates that have not delivered.
I think the question is always by how much you increase the response rate and obviously durability. At the end, this is a durability game. We need to show that these responses continue to last. If you look at the waterfall plot, as you pointed out, a lot of the patients, almost all of them, are still on drug. The response rate could continue to get better. There are several patients with stable disease, close to a response, that are still on drug. Unless the responder starts to crash in the next three months, I think we're going to have a very strong PFS that we'll be happy to report next year. I think at this stage of the program, we have high degree of conviction that this intervention is going to deliver a positive trial for frontline colorectal cancer as designed.
We will provide updates on the data with longer follow-up as we obtain them over the next few months, which I think those of you that still have questions, hopefully they will reassure you that we're going in the right direction.
Running up on time, but going to squeeze in one last question from the online folks. This is from Steve Willey at Stifel, who said, just to clarify, the 2028 top-line data disclosure for DAWN-303 is presumably for PFS. If you do choose to move forward with an ORR analysis to potentially support an accelerated approval, does that potentially come sooner? When do you expect to complete enrollment in this study?
Yeah. The top line is on the PFS, firstly, to answer that directly. We haven't given the specifics on the ORR analysis because it may take place extraneous to us and the study team just to protect the integrity of the study. So that'll be conducted on a certain number of patients, probably approximately in the middle of next year, but it'll be post complete accrual and about a year before the PFS analysis. But should it not suffice to meet the accelerated approval criteria, it wouldn't be something that's made public to protect the integrity of the study on the other endpoints. I think that addresses those questions.
All right. With that, we're going to close. Thank you everyone for attending. Thank you those online for listening. Have a good trip home