Thank you, Audrey, and welcome to today's Acrivon R&D and Investor webcast. My name is Adam Levy, and I'm Head of Investor Relations at Acrivon. As a reminder, during this call, we will make certain statements that are considered forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995, including expressed or implied statements regarding our strategy, business plans, and objectives, the expected therapeutic and clinical benefits of our product candidates, and the potential of our platform and our product candidates, and progress in timing and execution of our clinical trials. Such forward-looking statements are not guarantees of future performance, and therefore you should not put undue reliance upon them. These statements are subject to numerous risks and uncertainties that could cause actual results to differ materially from what we expect.
I refer you to the SEC filings on our website for a discussion of the risk factors that could cause our actual results to differ materially from those discussed today. Joining on the webcast today from the company will be Peter Blume-Jensen, our CEO, president, chairperson, and who is also the inventor of our AP3 platform and co-founder. Kristina Masson, our Co-founder and EVP Business Operations and President of the company's Drug Discovery and Phosphoproteomics Research subsidiary Acrivon AB, located in Medicon Village, Lund, Sweden. Dr. William R. Bradley, MD, attending physician and board-certified radiologist, Department of Radiology, Massachusetts General Hospital, clinical instructor, Harvard Medical School, Senior Clinical Advisor, Acrivon. Eric Devereaux, our chief operating officer, and Jean-Marie Cuillerot , our Chief Medical Officer. We are also pleased that the following KOLs will be joining us today. Dr.
Mansoor Raza Mirza, MD, Chief Oncologist at Copenhagen University Hospital, Medical Director of the Nordic Society of Gynecologic Oncology Clinical Trial Unit, Vice President of the European Society of Gynaecological Oncology, board of directors, Gynecologic Cancer Intergroup. Dr. Mirza is both a medical and radiation oncologist with a primary focus in non-surgical treatment of gynecologic cancers and is a recognized expert in endometrial cancer. He has broad experience in clinical protocol development, trial conduct, and clinical trial regulations. He is a senior author of national guidelines for the management of endometrial cancer and other tumor types and has been lead investigator of multiple trials leading to marketed approvals, including PARP inhibitors and the most recent frontline therapy introduced for endometrial cancer, combining platinum-based chemo with anti-PD-1. Dr. Robert L. Coleman, MD, Co-Director of the GOG Partners Foundation, Chief Medical Officer at Vaniam Group, gynecologic oncologist, Texas Oncology, U.S. Oncology Network.
Dr. Coleman is one of the country's preeminent gynecologic oncologists over the past three decades. He has authored or co-authored more than 700 scientific publications, including 410 peer-reviewed articles, numerous book chapters, monographs, invited articles, and textbooks. In addition to being Co-Director for GOG Partners, he serves on the board of directors for GOG Foundation, Inc. Yes for Olsen, PhD, professor at the University of Copenhagen, executive director of the Novo Nordisk Foundation Center for Protein Research, academic co-founder of Acrivon. Dr. Olsen is the undisputed pioneer of phosphoproteomics and opened the entire field with a breakthrough Cell paper from 2006 that is a citation classic in the field. He is one of the most cited scientists in the field of protein sciences, and among his many achievements, he is now also a recognized leader and pioneer in the field of single-cell phosphoproteomics.
At the end of the webcast, we will have a Q&A session. To submit questions at any time, simply use the text bar at the bottom of the webcast screen. With that, I am pleased to hand the call over to Peter.
Thank you so much, Adam. Welcome, everyone. Super excited to share with you our R&D update today. We are going to talk about our AP3 Acrivon Predictive Precision Proteomics platform, and for the first time, I'm going to introduce our generative AI capabilities overview, which includes generative phosphoproteomics. We haven't talked about that before, but have actually relied on that for the last couple of years to understand the proteome-wide effects of drugs, and we use that, leverage that to optimize drug design, enable desirable pathway effects of the drugs we generate. That's how we generated our dual WEE1, PKMYT1 inhibitor that we are now advancing rapidly in the clinic. This will be an overview of that. We are planning in the second half of this year to have a more detailed AP3 R&D day, but today we'll give an overview of these new capabilities.
A main part of our program today will also be an update focused on endometrial cancer of our phase two program, where we are and our esteemed clinical KOLs will help comment on the data we have there to date. We'll then end up on the clinical data we already have emerging for ACR-2316. I encourage you to stay on for that. We have exciting program updates there already. We'll end with a brief update on our preclinical cell cycle regulatory program and also share our extended runway guidance. As you all know that have gotten to know us over the last couple of years, our approach is really centered around Acrivon Predictive Precision Proteomics, or the AP3 platform.
Instead of inferring from genetic alterations shown here at the top, we are how drugs will act on disease cells, we are directly measuring the activity states of the dysregulated, of the disease-driving pathways within a cell and are able to accurately match that with the drug mechanism of action. Acrivon stands for that exact or accurate match between how the drug acts and how it modulates the upregulated or perturbed signaling pathways that are causing disease beyond cancer. We are currently focused on cancer, but later this year, we are planning to announce an autoimmune inflammatory program. As outside oncology, there is even more need for a method based on that is independent of genetic alteration based on measuring pathway activity states.
We are doing this with cutting-edge mass spectrometry and generative phosphoproteomics, AI-driven analysis of the networks, and this is what enables us to optimize drugs based on intact cells and the pathway effects. We do use co-crystallography for structure-guided drug design as well, so it's rational chemistry, and that enables us to generate compounds with very, very high selectivity against the targets that we are hitting. At the same time, we can see inside the cell basically how the drug acts and optimize the drug for desirable pathway effects, and that's how we generated our superior single-agent active molecule, ACR-2316, as we'll talk about later. You can appreciate that when we're able to see everything a drug does, we can see upregulated resistance mechanisms, we can identify predictive and pharmacodynamic biomarkers, and we can also do what we call indication finding.
We can match drug pathway effects with disease-driving pathways involved in different, for example, cancers. We have shown this particular slides also before in our corporate deck, but basically what we have is structured analysis, which are enabled by an AP3 data portal where all our multimodal data are filtered through and resulting in structured machine learning, AI machine learning, amenable data that is the basis for everything we do. The focus for us is understanding here lower right how the pathway effects are, and a key thing to what we do is understanding what is upregulated, indicated here in red as an example, and what is downregulated within the protein signaling network in a cell by a particular drug. To do that, we are working on so-called consensus substrates, recognition motifs. We are inferring the kinases and their activity states based on these substrates and their phosphorylation level.
This is core to what Acrivon is doing, measuring how a drug acts and regulates activity states of signaling pathways inside a cell. One cannot do that with genetics or transcriptomics or even pan-proteomics. It is based on the post-translational modifications and phosphorylation of proteins, and we have focused entirely on this for many, many years. The turnover is very rapid. We generate about 150 million quantitative data points every 12 days, and all this is sitting in the cloud, which is now amassing to terabytes of terabytes of proprietary data that we apply our proprietary machine learning-enabled analysis to. Here is an overview of our platform. As you can see, we have on the left side, we have our internal mass spectrometry-based high-resolution data that we generate week out. We have all kinds of additional wet lab data that are of different types.
It can be crystal structure data. It can be Western blots. It can be tissue imaging with our quantitative multiplex biomarker platform. Of course, we can also tap into publicly available data. All this multimodal data set is going through our AP3 data portal. It's a computational pipeline that results in generation of machine learning-friendly structured standardized data that are amenable for our analysis. As part of our whole platform, we have built the AP3 Interactome, which we have previously introduced. It's an integrated interactive data portal accessible for all scientists at Acrivon that allows us to, by push of a button, pull out any kind of data on substrates and pathway regulation and phosphorylation sites by push of a button. We are going to today share just a high level of our so-called AP3 KSR predictor. This is the basic tool by which we have learned.
We have applied generative phosphoproteomics to now cover a full understanding of the entire kinome beyond what is publicly available, understanding protein kinase substrate interactions, and from that, we deduce how the activity states are after drug treatment. Based on all this, we get insights and can validate that. Very importantly, our proteomic data sets are used also for validation of our generative phosphoproteomic zero-shot predictions. We also have a beta version of an AP3 chatbot. Out of this, we are applying all these things for drug design, target identification, disease-driving targets, resistance mechanisms, mechanism of action of a drug. We have about one-third of all drug regulated, of all quantified MAP phosphorylation sites are statistically significantly up or down regulated. We have in one experiment about 50,000-60,000 pharmacodynamic biomarkers in every single experiment.
We're going to show you how we also are applying that to our clinical program. The final slide from what is in our corporate deck is shown here. This is an example of the AP3 Interactome. You have these so-called consensus substrates, which are the basis for understanding how drugs act on pathways. We can look at how these substrates are up and down regulated in terms of phosphorylation and from that deduce the activity states of the signaling networks. We have these sunbursts, which are all by push of a button, all kinds of conditions and time points and concentrations of all the different drugs we have tested. We always synthesize benchmark competitors when they're available for drugs like our dual WEE1/PKMYT1 inhibitor, or for novel drugs, we try to use so-called tool compounds.
We use co-crystallography and have a very high success rate of that and are using structure-guided drug design for our medicinal chemistry in general. I want to go into just a brief overview of our learning-based generative phosphoproteomics. We're working, or the computational science team here in Boston is working on different types of language models. We have protein language models, we have large language models, and we have chemical language models. These are applied for three particular areas. One is to understand, again, how a drug acts and regulates the activity state of signaling pathways inside the intact cell. We also are applying it to both understand in detail at the pathway level the mechanism of action of drugs we are developing or existing markets of drugs or drugs we're in license. It's applicable regardless of modality and developmental stage.
We also use it to identify both predictive and pharmacodynamic biomarkers. Finally, when we're developing drugs, we also use it for in silico screening. I'm going to show you some examples of the importance of the KSR predictor, which we have almost completed now to cover the entire kinome. The publicly available databases, ICIP and others, the PSP databases and so forth, cover about 50%-75% of the kinome, and we are having pretty much full coverage of that now. As an example, I'll go back to our starting point, which was our first fully internally developed molecule, which is our dual WEE1/PKMYT1 inhibitor, developed specifically to overcome the limitations of the existing WEE1 and PKMYT1 inhibitors. They were characterized by a very narrow therapeutic index. We know there has been clinical activity, but the therapeutic index has been challenging.
The only thing known pretty much to date about these two kinases, WEE1 and PKMYT1, is that WEE1 has a direct phosphorylation site in cyclin-dependent kinases, tyrosine 15, and PKMYT1 phosphorylates 314. These are adjacent amino acids. What we have done is we have applied our KSR predictor to now completely expand the understanding of WEE1 mechanism of action without going into any details, and please don't do so, but take my word for it. These are so-called consensus sequences for how this WEE1 kinase operates inside cells. It phosphorylates all proteins inside the cell that have these amino acid sequences listed here. The importance of that is that based on that, we understand how a drug, a WEE1 inhibitor, for example, regulates the cellular signaling networks inside the cell.
This here is fundamental to our understanding of how a drug acts and how to optimize it based on its pathway effects. We build this out. We can do pathway cluster analysis and understand which interconnected proteins are operating and how they're regulated. Even with the most specific drug, whether it's an antibody that only is identifying one protein or the most selective small molecule inhibitor or an oligo-based oligonucleotide modifying agent, et cetera, it doesn't matter how specific it is. Due to the interconnectivity in the signaling network, you have the pathway effects leading to massive regulation of thousands of proteins inside the cell. That's what we can understand with our methods, and that allows us to optimize drugs. We can see which pathway mechanisms are regulated. We've done the same for PKMYT1.
By just this KSR application, we've built out our understanding of how WEE1 and PKMYT1 operate inside the cell, how they're regulated by, in this case, our lead molecules as we were developing them. This was fundamental to generate our ACR-2316, which is a highly differentiated molecule. It's the only WEE1 inhibitor we have had in our hands to date, and PKMYT1 that we have had to date compared with all the benchmarks that causes complete tumor regression across models. It was developed for superior single-agent activity and with exquisite selectivity achieved through co-crystallography and medicinal chemistry to only hit WEE1 and then have a balanced inhibition of PKMYT1 based on our generative phosphoproteomic analysis of the pathway networks. We were able to deduce PKMYT1 as the dominant resistance mechanism to WEE1 inhibition.
It's a method called reciprocal quenching, and it's in our corporate deck. We talked about it for quite a while, but basically what we could see was that there was an upregulation of certain mitotic proteins and DDR proteins by WEE1 inhibition. And we could see from the primary amino acid sequence, the kind of things I've showed you a few slides back, we could deduce that this was originating based on PKMYT1 activation. And we could confirm that by inhibiting those upregulated resistance mechanisms, PKMYT1. This insight led to the balanced inhibition of PKMYT1 combined with a potent inhibition of WEE1 to achieve superior single-agent activity. We wanted to make sure that this was reflected in the pathway activation. We know that WEE1 inhibitors, and to some extent, PKMYT1 inhibitors cause massive CDK upregulation.
We saw that even more potent than with any of the benchmarks. We also ensured, and we had many choices, some did and some did not. We wanted to ensure that these ultra-selective inhibitors also led to activation of PLK, especially PLK1. PLK1 is essential for what is called the mitotic execution, and that is how we achieved a very, very robust mitotic catastrophe. This molecule hits the tumor cells and causes pro-apoptotic tumor cell death. We have massive tumor shrinkage, as we will show you from compounds like this already during dose escalation, and that will be shared with you at the end. We are very excited about this molecule. Finally, we use our AP3 platform to predict highly sensitive tumor types like we have done for endometrial cancer for ACR-368. Our dose escalation phase one study is actually conducted in selected solid tumor types.
I mentioned in the beginning, or didn't really, but we have built out spatial phosphoproteomic spatial technologies and have published that. We have clinical phosphoproteomics, which is supporting our first internally developed ACR-2316 clinical trial. We have PD readouts already, and we'll share that later. Also, we have developed early stages of single-cell phosphoproteomics. The undisputed pioneer, not only in phosphoproteomics, but also in single-cell phosphoproteomics, is actually our academic Co-founder, Jesper Olsen. Some of his publications are listed here. I'll give the word for just a few comments from Jesper on the importance of this field and where he sees it move and how that maybe links to what Acrivon is doing. Jesper. Yeah, thank you, Peter. Yeah, it's really an exceptional platform that everyone has built for phosphoproteomics. I think it's unique in the way that you take this signaling pathway approach.
This links in very well with all the developments and the trends in the field now, where we can see that in signaling, that phosphoproteomics is really the way to tease out drug effects. This will be even more important now when we can start to look at the signaling networks in individual cells. We can look at cellular heterogeneity. We can look at the tumor microenvironment. It is going to be very, very interesting to basically combine the single-cell phosphoproteomics with the AP3 platform, especially the computational parts of it.
Awesome. No, that's exciting. You know we're developing it, as I know you are, to be able to address, for example, tumor cell heterogeneity, but even more so also in autoimmune inflammatory disorders. Jesper has been instrumental for us to get this whole single-cell phosphoproteomics set up, established in our lab.
We are applying all these learnings to that now. We'll now move into the important phase two clinical update. We are focused on endometrial cancer. As you know, this is an exclusive in-license asset from Eli Lilly and Company. Endometrial cancer is a tumor type predicted by our AP3 platform to be sensitive to ACR-368. Lilly actually treated over 10 years all kinds of solid tumors with single agent and in combination and never identified endometrial cancer. Using our AP3 approach, we could preclinically identify and predict sensitive tumor types. We predicted endometrial cancer to be particularly sensitive. I hope the data today will demonstrate that convincingly to everyone on this call. Just as a reminder, we have conducted our trials in our three lead indications such that all biomarker-positive patients are treated with monotherapy.
The biomarker is called OncoSignature and is tailored for ACR-368. It predicts sensitivity to monotherapy. Using our AP3 platform and the types of pathway analysis I showed you earlier, we have identified some of the mechanisms behind resistance to ACR-368, which include pathways that are involved in nucleotide synthesis. Some of these can be overcome with low dose of gemcitabine. The way we are conducting our trial is we are having a registrational intent in about one-third of patients that are biomarker-positive of our tumor types, and the two-thirds are getting an exploratory combination of low-dose gem and ACR-368. Importantly, the OncoSignature or biomarker-negative patients are not predicted to be sensitive to the combination. They are simply explored based on the fact that we anticipate and have shown in all our preclinical data.
We now have some clinical data support for that, that low-dose gem truly does sensitize a subset of these patients. That is exploratory. The registrational intent trial is the biomarker-positive patients. Just a real quick comment with the experts on Dr. Mansoor Raza Mirza and Robert Coleman are part of our ad board, which also includes Dr. Ramez Eskander. Dr. Ramez Eskander and Dr. Raza Mirza were behind the recent frontline approvals of anti-PD-1 combined with platinum-based chemo backbone. After about 18 weeks of chemo plus anti-PD-1, one goes over to a very long maintenance phase of either pembrolizumab or dostarlimab. This has completely changed the treatment landscape over the last year and a half. These two New England Journal of Medicine articles that led to these approvals were published in June 2023.
This frontline therapy is so effective that we now have a whole different setting in second line. This is almost, if you add this up, one and a half years or so of frontline therapy. While we, until this new frontline therapy was introduced, had lenvatinib and pembrolizumab as the recognized second-line therapy after chemo backbone, we now have a situation where most, if not all, patients, both pMMR and dMMR stage three, four patients will have had prior immunotherapy for a very long time. In the pMMR patients, and in particular, it is questionable how much value you'll have a continued anti-PD-1 in the second line. There is definitely going to be an anticipation of reduced usage of that regimen with a new frontline therapy regimen.
This leaves us with a very, very big unmet need in second line if we exclude the patients that are very sensitive to trastuzumab deruxtecan. There is an FDA accelerated approval for HER2/3+, which is about 10% of these patients. For the majority here, we are left with a very, very high unmet need in second line. The control arm in Vicki McIlroy's study that led to the second line approval was about just under 15%, 14.7% and 3.8 months. These patients had never seen anti-PD-1. We know that long-term anti-PD-1 therapy is very, very causing a lot of tumor heterogeneity. The third line was about 9-10% from previous studies. It is anticipated that current standard of care is about 10-12% response rate and PFS of about three months here. I don't know if either Dr.
Mirza or Coleman will make a comment here, but otherwise, I'll give the word to Dr. Bill Bradley.
Just, just, Rob, what I would say just briefly, just for orientation purposes, that the landscape for endometrial cancer is now increasingly being defined by the attainment of biomarkers. That has brought the upfront testing and more comprehensive testing about markers to help define the therapy in the frontline setting. That is a new finding that has come from the expansion of molecular subtyping and characterization of the tumors that we see. That is what you're seeing here on this slide.
Excellent. With that, Bill, please.
Yeah, thanks so much, Peter. We will begin with the demographics and baseline characteristics of our endometrial adenocarcinoma patient population.
We'll start with biomarker-positive patients, of which we have 20, versus biomarker-negative, of which we have 38, maintaining the expected one to two ratio of one OncoSignature positive for every two OncoSignature negative patients. Looking at the left-hand side of the column, a few things to point out. We do have a healthy representation of Black African-American patients, which has been identified as a key mission in many academic centers to get better treatment to this historically oncologically underserved population. We look at the stage of the cancer here. We see a majority of our patients are stage four, with 80% in biomarker positive and 58% in biomarker negative. We also have overrepresentation of some of the more aggressive histologies, including serous and carcinous sarcoma, which together account for 85% of our biomarker-positive patients.
Coming to the right side of the column, we see these patients have had several prior lines of therapy with a median number of prior lines of two in biomarker positive, a mean of 2.6, and three in the biomarker negative groups, which is much higher than some of the number of prior lines in some competing ADC trials. Also, very importantly to really emphasize is that over half of the patients in our population were refractory to their last prior line of treatment, meaning that the best overall response to last prior line was progressive disease. At the first imaging time point of their last prior line, they had progressive disease, again, indicating a very aggressive underlying tumor biology. Vast majority, two-thirds are pMMR, meaning that these patients are less likely to have a good response to immune checkpoint therapy.
Also, the majority of patients are P53 mutant, again, emphasizing that these have an aggressive molecular phenotype. All of these patients have progressed on prior backbone platinum chemotherapy and PD-1 or PD-L1 therapy, 100%, essentially the two not being eligible in the biomarker negative group. Of those, two-thirds have also progressed on prior pembrolizumab . I think this demographics, again, and I've reviewed every scan on this trial, and these patients are very, you know, have large tumor burdens. They have multifocal disease, multi-organ involvement, and the tumor biology in these patients is often very aggressive with cases of tumors invading critical mediastinal structures, pulmonary artery, the stomach, tumor fistula. This is a very aggressive baseline tumor biology as well seen in some of these molecular and clinical characteristics.
Yeah, Bill, if I may add to that. Thanks.
I think the most important features are that they all are post-PD-L1 treated patients. This is the biggest unmet need today where we are seeing these results from this drug. When you look at the population, it's much more aggressive. 75% of the patients are either carcinosarcomas or serous. Exactly 60% refractory to the previous treatment. This is really tough population.
Yeah. No, thank you. Actually, I have no two of them we don't have the status for. It's actually, if you want, two-thirds of those where we know what the response was to the last prior line. We should probably keep the pace up, Bill. Yeah, excellent points. Thank you. This is just, we'll go over this quickly. We have 23 biomarker-positive patients who have received at least one dose, 44 biomarker-negative patients. Slightly smaller number for efficacy available.
Some patients have not received the first imaging yet. We have five patients still on treatment for biomarker-positive, seven for biomarker-negative. You can see the reasons for discontinuing treatment or discontinuing the study listed below, which are PD among death, not unexpected given the kind of the aggressive nature of these tumors. Overall, ACR-368 has a very encouraging safety profile. Most AEs are limited, transient, reversible mechanism-based hematologic AEs, as we would expect with DDR inhibitors. Most of the AEs typically occur during the first one to two cycles of therapy and often would decrease in intensity or not be seen in subsequent cycles of therapy. Notably, there is absence of GI toxicity and long-lasting myelosuppression or the more typical severe non-hematologic AEs we commonly see with ADCs and chemotherapy.
That's very important in this patient population, which is typically post-menopausal elderly females where quality of life is an important consideration for these patients, especially in terms of the GI toxicity. Overall, a very encouraging safety profile.
This is, again, I mean, no ILD, no ocular toxicity, no mucositis, no GI toxicity. It makes it really a great molecule to combine with novel drugs.
That's a good point. The AEs here, obviously, as you know, we have a confirmatory trial strategy to look for moving up more frontline. Obviously, the AEs are completely non-overlapping with what you typically see with the immune checkpoint inhibitors. That's a very or ADCs. I think that's where it was getting. Yeah, or the antibody drug conjugates, which are prolifically going forward here. Yeah. Okay.
Here's a waterfall plot for the best overall response in our endometrial biomarker-positive patients. We can see that several patients had a very good depth of response below listed in the squares. We'll see the best overall response on last prior line, histology, MMR status, prior pembrolizumab and P53 status. Just to emphasize again that this is a very aggressive tumor population when these patients have had several prior lines and have a lot of unfavorable characteristics. As Peter said, the primary goal of the OncoSignature assay is to predict response to monotherapy with ACR-368. We can see that the OncoSignature very accurately predicts the tumor drug sensitivity with 80% of the tumors shrinking and 80% disease control rate in predicted responders.
Furthermore, the confirmed response rate is two times higher than in the last prior line of therapy, which is very striking, not something we typically see. The specific numbers are 35% response rate with ACR-368 versus 15% response rate in the last prior line. It is very unusual to have a response rate increase in a subsequent line of therapy to that degree.
It is despite the fact that you have 60% of the population who had refractory disease on prior line of therapy, right? This is a median of 2.6 or mean of 2.6 prior therapies. This is really impressive. All anti-PD-1 also, right? Yeah, and all anti-PD-1s. Only four patients where I do not see tumor decrease, but otherwise tumor is shrinking all the way. I would like to hear, Rob, this. Yeah.
You know, we always say that, you know, in the clinic, we always love to see responses so we can tell the patient we treat to the lack of progression. Patients that do not show objective progression meet criteria for it, stay on treatment. Tolerance of treatment over multiple cycles leads to this disease control rate, which is actually pretty important at the patients, particularly at this stage in their treatment history, when we are basically just trying to continue to provide them active compounds to keep them alive. This is, you know, to see this kind of portfolio in a patient population that we just described is very exciting for the clinician. It looks very much like the ADC. Yeah, exactly. Waterfall plots, what we are seeing.
We saw a week ago the B7H4 looks very similar, 35% response rate, despite the fact that there only one prior line of treatment was. Exactly. Here you go. The toxicity portfolio. Toxicity portfolio. I think this is really impressive. Only 60% or so prior anti-PD-1. I think that's important too.
Yeah. One less line and only 60% prior anti-PD-1.
Yeah, because Peter, now it's the post-PD-1 is the unmet need that, and you are right on the spot here.
Yeah. Excellent discussion. Great points. Here's the spider plot in the biomarker-positive patients showing the patients that do respond often have a dramatic initial decrease in tumor that is sustained for very good durability for most of the patients.
A little bit underestimated here, but we can see the duration of treatment better on the upcoming swimmer plot, which we can take a look at. Then we have the swimmer plot again showing the biomarker-positive endometrial cancer patients. You'll see that the arrows at the top there indicate that those patients are still on treatment, you know, about a year to 14 months after beginning treatment, which is a very impressive duration of response. If we look at the median duration of treatment in confirmed responders, which is still not reached, it is greater than 7.9 months.
Looking again at the boxes on the left, we can see that the patients colored in blue that had some degree of response on prior line are doing extremely well in terms of duration of response in PR, whereas a lot of the kind of what we're calling refractory patients here had a slightly poorer performance, which we can also will be split out as well on the next slide. Here we go. Here's the ACR-368 activity compared to the last prior line. We have 18 patients for which we have data on the best overall response for the last prior line. In the best overall response for the last prior line, we have a 17% response rate and 33% disease control rate. Compare that to ACR-368, where we're having a 39% response rate and an 83% disease control rate.
If we again break that out into activity and patients relapse after the last prior line, that bumps up to a 50% response rate with a median duration of response of greater than 10 months, 100% disease control rate with a duration of treatment of 8.5 months with the end not yet reached. We see the four arrows there indicating that four of those patients are still on treatment. Patients, the ACR-368 activity patients refractory to the last prior line drops a little bit, but it's still better than standard of care single agent chemotherapy with a 33% response rate, duration of response of three months, 75% disease control rate and duration of treatment of about 3.9 months. The take home messages for this slide in the biomarker positive patients is, as Dr.
Raza Mirza has emphasized, all patients have progressed on prior platinum-based chemo and prior anti-PD-1 prior to beginning treatment with ACR-368, which really puts the responses we're seeing in a very exceptional light. Like we talked about, the median mean of prior lines was two and 2.6. This is much higher than some of the competing trials in ADCs. 66% of our patients were refractory to the last prior line, meaning the best overall response was PD at the first imaging time point, and 33% of patients had relapse on last prior line. Again, to emphasize, the response rate is two times higher with the ACR-368 in the current line than the response rate in the previous line, which is a very striking change in response.
ACR-368 is active even in the refractory patients with a 33% response rate and a 75% disease control rate. We have a long durability with ACR-368 in relapse patients, four to six still on treatment. The conclusion is ACR-368 may be optimally positioned in patients who are not refractory to the last prior line of therapy.
Bill, I think the upper set of patients which says relapse, these come closer to what we see in the trials of ADCs. Although in ADC trials, we have, for example, in B7H4 again, only one prior line of therapy, and you are seeing 50% overall response rate, which is very impressive. Even in the refractory, which is which you do not see in clinical trials in ADCs, you see 33% response rate. This is really impressive.
Yeah. Awesome.
I'm looking a little at the clock here, but this is a great discussion. I would like to bring you in also on the way forward. Real quick here, and I'll go very quick. We uncovered the main resistance mechanism to ACR-368 through phosphoproteomic analysis, where we found that the OncoSignature three biomarkers were upregulated post gemcitabine in resistant cells. We showed in a number of pre-clinical studies that we could sensitize resistant cells with very low doses of gemcitabine. We followed up with rodent models and showed that we had an upregulation of the OncoSignature in post low doses of gem. Here we have increasing doses for the three biomarkers, and you can see more and more biomarker positivity in the nuclei as we increase the doses. All this led to, which is consistent with the sensitization.
This is the basis for which we think that it was worth exploring whether we could help patients that are predicted completely insensitive to monotherapy with ACR-368 by having low-dose gem as added and then explore that combination. Just from independently at Dr. Christine Chung, Chairman at the Moffitt, is running an IIT in head and neck cancer. Independently, we are there collecting pre- and post-LDG treatment tumor biopsy samples. There is an induction dose of LDG. Right enough, that is the first human clinical data we have there. We are not able to take serial biopsies in our ongoing ACR-368 registration intent trial. We see indeed an upregulation of the biomarker signature completely consistent with that. We do believe that LDG does sensitize.
You see here that we have in the biomarker-negative patients, we have five confirmed responders out of 36 patients here with imaging, which results in a confirmed response rate that is comparable again to the prior line of therapy. We do believe, based also on the tumor shrinkage that Bill is going to show you in a couple of images now, that this is real sensitization with ACR-368. We believe that we potentially can explore this in a low-dose gemcitabine combination, all-comer, since we also believe that low-dose gem is upregulating the OncoSignature and also enhancing sensitivity in OncoSignature-positive patients so that we could increase durability and response rate there. That will be subject potentially for a future trial in an all-comer. Again, heavily pretreated patients, many of them refractory and so forth.
Yet we have a respectable response rate that compares to the last prior line. Bill, will you go relatively quickly through this to make sure we can cover the whole?
Yeah, absolutely. I'll just go through these quickly. These are just some case stories to illustrate some of the points from the larger data set we just shared. This first case story is a biomarker positive patient who had a confirmed PR. They were stage four serous pMMR P53 mutant. They had a biopsy of hepatic metastasis, which was OncoSignature positive, tolerated treatment well. As you see on the right, the prior treatment, the patient was refractory to the last prior line, which included anti-PD-1, dostarlimab. This patient had primarily hepatic disease and peritoneal disease.
The hepatic disease was chosen as target lesions by our independent radiology reviewers, and those completely disappeared at week eight and week 12. If you look at the scan in more detail, this patient actually had more than 10 additional hepatic metastases spanning both lobes and all hepatic segments. All of those hepatic metastases disappeared at week eight and week 12. To see this burden of disease disappear after progression on the immune checkpoint inhibitor is something we really don't see in the clinic very often in MGH or Dana-Farber. It's a very impressive response and really demonstrates ACR-368's ability to fully eradicate multifocal disease, even in initial high disease burden cases, which is very striking. Second case story is also a biomarker-positive patient, endometrial adenocarcinoma pMMR P53 mutant. They had a biopsy of a vaginal mass, which was OncoSignature positive.
You can see that this patient was refractory to all prior lines, including anti-PD-1 as well in the setting of pembrolizumab. This case shows one of the challenges we have with endometrial cancer in RISCs. These tumors are often very infiltrative, grow across multiple tissue planes in the pelvis. Sometimes RISCs is a little bit primitive in that the long axis may not represent the true degree of tumor volume. Doing a more sophisticated volumetric analysis including all of the tumor in the scan, we can see that even though the long axis diameter by RISCs only decreased by 4%, there was actually a true volumetric reduction here of 44%. As Dr. Coleman said, sometimes we have these patients that have SD by RISCs, but we're actually seeing some degree of tumor shrinkage doing more sophisticated analysis of the actual tumor volume.
This case story three is a biomarker-negative patient. This patient had a confirmed PR, dMMR p53 mutant, stage four serous. They had a biopsy of hepatic metastasis, which was OncoSignature negative. Best overall response in prior line was PD on pembrolizumab. This patient really had a truly massive disease burden, and they had over 50 hepatic metastases occupying 40% of the liver, which is kind of where we clinically start to worry about some compromise in liver function. After week 16, there is a dramatic reduction in the size and number of the liver lesions, and a lot of them are not even enhancing anymore. The remaining areas may consist of just some fibrosis and necrosis, indicating a very impressive treatment response. They additionally had equally strong regression in necrotic lymphadenopathy in the retroperitoneum and peritoneal disease.
Again, this degree of tumor reduction is striking and really supports low dose gem's role in sensitizing tumors to ACR-368, driving a robust treatment effect as we see here. The last story is also a biomarker-negative patient with the best overall response of SD. This is a stage four endometrial adenocarcinoma. We do not have the molecular subtype. They had a biopsy of a peritoneal lesion, which was OncoSignature negative. You can see that their prior treatment, they were essentially refractory to all prior lines, including immune checkpoint and carboplatin and doxil, bevacizumab. These are one of the, this is a patient that nicely shows kind of the massive kind of disease burden that some of these patients have. If we, Peter, just play that video just to get a sense, you can see that there is really a tremendous amount of peritoneal disease here.
Doing a more sophisticated analysis, this is a volume of disease of over 1,020 centimeters cubed. After treatment with ACR-368, we see a profound volumetric reduction in peritoneal disease from 1,000 to 168, about 85% volumetrically. That would be equivalent to like a liter of soda or three full glasses of beer down to a large egg, which is a tremendously tremendous volumetric reduction. The striking thing is RECIST. Please do not kill beer. Yeah, exactly. Yeah. RECIST, radiologists are always the worst with our comparisons. RECIST, this is another one where RECIST kind of falls short because this was categorized as SD. I think really failing to capture the true extent of tumor shrinkage, we can clearly see here from an imaging perspective.
Again, this profound volumetric decrease in tumor really strongly supports low-dose gem's role in sensitizing tumors to ACR-368, enabling a significant treatment response. Absolutely.
Thank you so much, Bill. Yeah, these are very established tumors. Obviously, the goal for us is to learn from our data here and then maybe zoom in on where we want to go. Here is an important update. Based on our very encouraging and emerging data, the competitive positioning here, based on limited number of emerging therapies still and limited activity of the ADCs, for example, compared to ovarian and an attractive commercial opportunity, it is and remains. We've said that for the last six months since the ESMO update last year. This is our prioritized and first potential registration opportunity. We've identified this tumor type, predicted sensitive by AP3 and verified that in the clinic.
We got a BDD in January this year. We have a huge unmet need in second line, debatable. I guess the exact numbers will emanate, but it is in that range also, I think, based on Dr. Eskander and Coleman on this call and Dr. Raza Mirza. The incidence is estimated based on our blinded carrier market research that we conducted August, September last year, 25,000-30,000 in the second line, stage three, four recurrent or locally advanced across all histopathologies. Obviously, the prevalence is significantly higher and a much bigger number is probably there to tap into. We are talking about real significant numbers here. We think it is very clear based on these drastic tumor shrinkages we see in patients that are predicted absolutely not sensitive to monotherapy that we have refraction there that are sensitized by LDG.
We are continuing to pursue our registrational intent in second or larger line biomarker-positive patients. We want to try to really position us closer to post the new frontline therapy where our data combined clearly show that we have, although the numbers are small, clear signs of very, very encouraging activity. Also, our confirmatory trial strategy is what we pursue as well, aiming for switch maintenance, label expansion, frontline, potentially in an all-comer based on sensitization with anti-PD-1. I'm going to show you a data set later. Peter,
I think this is a great opportunity, especially also looking at the toxicity profile. Right. With PD-1s, with ADCs, amazing way to combine.
Yeah. No, thank you. I do want to say that we have alluded to that for a long time, but we now are completely deprioritizing ovarian and bladder cancer.
We have talked about the increased competition and very, very successful emerging therapies there, smaller, decreasing market opportunity based on the 10-12 emerging therapies there, and also a very high clinical bar that is not met based on our preliminary data after the 23 biomarker-positive patients. This is very preliminary data. Based on this and combined overall focusing our resources on where we think there's the biggest return of investment, we are deprioritizing ovarian. Likewise, bladder, where we had a lower than predicted biomarker-positive fraction based on our preclinical analysis, simply resulting in a challenging enrollment. Combined, endometrial is what we're focusing on with all our resources, as well as ACR-2316, which is showing very early signs of clinical activity.
I'll just show you our excitement about the switch maintenance combination strategy with anti-PD-1 in the new frontline setting as a confirmatory opportunity. Not only is there strong preclinical data, but we have also in-house generated our own immunocompetent data. You can see here we are almost four months treating mice here. These are immunocompetent mice that were treated with either single agent 368, different doses, or with anti-PD-1. None of these two agents were able to completely regress tumors. Most of the mice, seven out of eight, got complete regression. We stopped treating. The tumors did not grow back. We can re-challenge with these tumors and in control mice that have not previously been subjected to the anti-PD-1 ACR combo, the tumors grow as expected while they do not grow, there is immune memory. We come in a second time and re-challenge the same mice.
We now can see that there's contribution when we deplete the CD8 positive cells. They're the ones responsible for the immune suppression here, while those that are re-challenged and just basically not changed in terms of treatment, they are completely regressed here. This is very, very exciting data showing complementary activity of 368. We believe we can tap into that delta here where both anti-PD-1 is not fully expected, especially in the 80% pMMR, and where ACR-368 is not fully active. We think this year is a real exciting opportunity. Quickly, we are running short on time, so I'll now accelerate a little bit here. ACR-2316 that we introduced briefly in the beginning is moving very, very rapidly, advancing rapidly in the clinic. We are about to complete our third dose level.
We have selected tumor types identified with our AP3 platform, as we've done for ACR-368. Long story short, as of April 1, we are anticipating to be ready for dose level four. We believe that RP2D will be around DL5, but as I'll show you, we are excited about already seeing signs of clinical activity here during the early part of the dose escalation, as well as early drug target engagement. The dose levels one and two were clear without safety concerns. The preliminary PK data on these two lower dose levels indicate a nice approximate dose proportionality. Very importantly and very excitedly, I can report that we saw significant drug target engagement already at the very first dose level using our in-house clinical mass spectrometry method that Kristina Masson has set up in Lund, Sweden, using mass spectrometry and in human PBMCs.
We are focusing on the selected tumor types, and we have just now also seen the first signs of clinical activity already at dose level three, two dose levels before we actually believe we'll have a very high threshold drug target engagement. This patient had received prior chemo and anti-PD-1, three prior lines of therapy with metastatic lesions that were pretty established. There was significant decrease of these throughout chest, abdomen, and pelvis. We had almost a PR here on the first imaging, but a very significant tumor shrinkage, and the patient remains on therapy. This is obviously very exciting data. Kristina, please.
Yes, thank you, Peter. We have established our in-house clinical proteomics capability here to support the 2316 trial.
Using AP3, we set up a clinical mass spectrometry-based PD PBMC assay for high resolution and very fast turnaround of quantitative measurements of drug PD effects. Schematic shown here. If you go to the next slide, thank you. This allows for a completely unbiased global measurement of the most up and down regulated PD markers. Here showing you data from the lowest dose in our phase one study, the lowest dose. In this example, from over 30,000 identified phosphorylation sites, we pull out all significantly up and down regulated drug-regulated sites and key kinase substrates. What is really remarkable here is that the direct evidence of PLK1, CDK1, and CDK2 activation in red here, as well as AKT1 and 2 substrates in blue, are downregulated. All of these are crucial for the proapoptotic effects of this molecule.
This is indeed the mechanism of action which we specifically designed for. It's important to note here that the 2316 molecule is a highly selective molecule. What is shown here are the downstream effects of targeting those kinases specifically.
This is completely consistent with our corporate deck where we have from our AP3 Interactome, the top activated and the top inhibited kinases. It's the exact same ones that are pulled out here now for the first time from our human study. We are really, really excited about that.
Yes, we definitely are. Here, I just want to share a very brief overview of our very early discovery program. This is an undisclosed and very attractive cell cycle target, which has been shown to be essential for cancer cell viability and is well suited for the AP3 platform.
A highly selective two compound shows strong anti-tumor efficacy in rodent models, and AP3 profiling of benchmarked compounds, as well as our own internal compound, is ongoing. This will enable mechanism of action-based SAR. Below here, you see some representative co-crystals in the middle of two candidate series together with target X. The curves to the right are examples just highlighting the impact of compounds on viability in wild type versus targeted mutated cancer cells. On the next slide, I'm not going to go into details here, but just want to point out that we do apply our generative phosphoproteomic approach that Peter talked about earlier to all of our targets and programs at Acrivon that we are working on.
This really allows us to uncover the network of proteins and ultimately decide which pathways to activate or inhibit for a desired mechanism of action for our molecules. With this, I'm going to give the word back to Peter.
I'll just end on our updated pipeline. This will come out with our AK filing, and the focus is on endometrial cancer. Not shown here is our further confirmatory trial strategies and thoughts about LDG all comer, but you can anticipate that these are things that will be forthcoming. We're going to update on the registration of 10 trial and confirmatory trial design at a later time point. We maintain our initial clinical data readout second half of this year for ACR-2316. We are planning to announce a development candidate for our cell cycle regulatory program that Kristina just covered now.
We also are going to declare our first autoimmune inflammatory program later this year. With that, Adam, please. Yes, thanks, Peter.
This is a flash cash, $185 million on the balance sheet as of end of Q4, and importantly, additional runway guidance into 2027 now compared with previous guidance.
Thank you. With that, Adam?
Yeah. Let me read the questions. Our first questions, Peter, are from Lee Watzek of Cantor. She has three questions. The first one is, with the 35% confirmed overall response rate, are you still confident that you can get accelerated approval given the low benchmark in the post-PD-1, PD-L1 setting?
I would say if the data hold up, it's still relatively limited numbers. I would say that given the patient profile we have, I think it looks extremely promising.
We are probably going to try to really make sure that we also get better representation of more typical patients after the new frontline, meaning lower percentage of refractory patients going forward and patients that have relapsed or recurred on the prior line and get more towards what the current ADC emerging therapies are coming in at. They come in at one prior line. They come in with a much smaller percentage of serous and carcinosarcomas, these aggressive histopathologies. They come in with a subset of them having received anti-PD-1. I would say, conversely, given the type of patient population we have here and the massive tumors that we're seeing, this is a single agent targeted agent that we're seeing this kind of tumor shrinkage with. I'm very, very confident that we will meet the bar here.
I also think we are going to flesh out, and I can ask Dr. Coleman and Russell Mercer to comment more on that, but I think we're going to get a better understanding of the true standard of care. I think we are going to see it's going to be down in the very low double-digit percentage of response rate and very, very short duration of response. All this combined, I would say, yes. I also think we have an agent that can do particularly well as we move up towards after the new frontline. Meaning second line, we expect based on the data to see maybe even better activity. There we saw 50% confirmed response rate and duration of response not yet reached at over 10 months in the small. This is serious numbers there.
I think we should really try to get the patient full mood, move more up towards that. Dr. Coleman or Russell Mercer, please comment.
Yeah, if I can jump in here. I think that what you said was right on, Peter. We do need to narrow the confidence limits here a bit on the sample, which would come from an increasing sample size. This is definitely in the ballpark for what would meet criteria for accelerated approval when we are comparing a population for which single agent chemotherapy is the historical control. This is our target. We try to get around 35% of an injector response rate. We'd like to be able to reject 20% of the lower limit of the confidence, and we're already close to that.
A little bit more sample size with this kind of activity, I think, will be in good shape.
Completely agree. Yeah.
And her other question, Peter, do you need to wait for duration of response to mature further before going to the FDA to discuss path forward?
We haven't disclosed when we are talking to them, but you can imagine where we are now with this data that we are going to move towards that aggressively and get a formal alignment. With our advisors here also, we are definitely in a position, I think, to be able to both share our strategy, share our data. Of course, meanwhile, the data will mature. It is maturing data, and we keep enrolling and dosing aggressively. All this will be something we look forward to updating on in the hopefully not too distant future.
Our next question, Peter, comes from Roger Song of Jefferies. Roger asks, with this promising data, please elaborate on the registrational path for both biomarker positive and negative patients.
Our registration intent path for the biomarker positive is monotherapy. The current trials, I think it is remarkable to have 100% prior anti-PD-1 progression, as well as 100% prior platinum-based chemo backbone. For comparison, AstraZeneca in their Blue Star trial has just gone into a phase three trial with a 35% confirmed response rate with their ADC with two dose levels. That is in patients that have only one prior line of therapy that has 60% of them only have had prior anti-PD-1, and where they have about 40%-45% of serous, or maybe it is actually 30% of serous histologies, which are very difficult to treat together with carcinosarcomas, where there is a huge amount of need.
If we compare that line more or prior lines, much more aggressive histopathologies, prior than Lenvatinib and Pembro, which is actually very toxic as well. I think our registrational intent here is to, as I said, aim for a second line and higher and come up in patients that 100% of them have received prior anti-PD-1. The confirmatory trial strategy will be to use our data and leverage the synergy we have with a preclinical rationale for that published by others, including on ACR-368. We have our own in-house data. And there's also actually a clinical study that was conducted by Geoff Shapiro in ovarian cancer patients, small number of patients, but that showed consistent with what Dr.
Russell Mercer said about non-overlapping toxicities that RP2D of ACR-368, estimates we 105 mg per square every 14 days combined with standard doses of anti-PD-1 were well tolerated in that study. It was, I think, 17 or 19 patients up to some years ago. All this combined, I think, is our confirmatory trial strategy. For the biomarker-negative patients, that is an exploratory phase one B2. It's never been a registration intent trial. The biomarker, when it's negative, does not predict that the combination of drugs will work. It simply is a way for us to say, when you're negative because of our AP3 platform uncovering sensitization, let's give these patients a chance to benefit. You can see that already there we have promising signs of activity that actually compare with a prior line of therapy very favorably, about the same or even a little higher.
All this combined means that the biomarker negative, I think, could be an interesting way to boost response. One thing we are discussing, contemplating to pursue is an all comer where we look at low-dose gemcitabine plus ACR-368 in both biomarker positive and biomarker negative patients, as we also expect the biomarker positive patients to be sensitized by low-dose gemcitabine.
Roger, as a follow-up question, can you comment further on what would be the clinically meaningful ORR, DOR, and PFS in second line versus first line endometrial cancer for a novel therapy?
I think our esteemed KOLs here said it, and we have the data from the control arm of Vicki Mcilroy. We have also the data from Ray Coccar way back that are kind of also important here, I think. We have the physicians' choice chemotherapy control arms in various studies now.
I would say it is the bar right now. The standard carrier is about that, maybe plus minus 12% response rate, and chemo probably three plus minus months. As Dr. Coleman said, if you aim for a 35% response rate and maybe five months of duration, five and a half months, maybe more duration of response, I think you're starting for the accelerated intent to have something meaningful. You saw that in our relapsed patients that they are few, but the value of that data is that it's completely differentiated from the patients that were refractory to the prior lines. All the ongoing patients ongoing on therapy are those from the relapsed. We have duration of response there of over 10 months, and it's not yet reached. We have a response rate of 50% confirmed responders, few patients.
But I do think that's a patient population where if we move up one line there, we'll see even better responses there. Not eliminating the responses we're seeing now or ignoring them, but beefing up everything there. I do think that we have a nice margin to that. The frontline therapy, I think we will need to kind of discuss further. I think Dr. Coleman and Russell Mercer probably have insights on after the New England Journal of Medicine papers came out, there was a lot of censoring in those. I think the PFS for pMMR patients has dropped a little bit from the original 13 or so months to maybe 10, 11 months. I think we need to improve that by probably three months, four months maybe in pMMR.
The response rate is already very high, so it's probably more overall survival that will be a long-term endpoint. I'll let Dr. Robert L. Coleman comment on that.
Right. I mean, the bar is getting higher and higher. If you're going to have switch maintenance, a good thing would be that you have a little bit lower standard of care PFS because you are not taking the chemotherapy part in it. That would be a good strategy to enter combination with IO, so it's not against IO. I think all this will help, but you would need a substantial increase in progression-free survival. I'm not sure are we soon going to see, while we are doing the trials, that the survival past progression is getting longer and may become longer and longer.
That may put the bar of getting OS benefit quite difficult. We have to do the same strategy that we have to show a clear trend of OS, probably even lotting alpha. I think I would like to hear Rob on that as well. I believe that we are quite rapidly improving the outcome of our patients, also past progression in second line. We would be entering with ADCs and these drugs with ACR-368 and so on and so on.
Yeah. What I'd say is that to answer the question maybe a little more even specifically, our targets for what are meaningful differences in the clinic are essentially two assessment cycles. We generally will assess patients on an every other cycle basis. If it's a monthly infusion schedule, then we're looking for 16 weeks difference, which translates to a hazard ratio around 0.6.
The reason that's important is that the upper limit of detection based on sample size gets us in the low 0.7 to mid 0.7 range, which is felt by the agency in many different settings in GYN to be considered relevant. That's kind of the targets. That applies both in the frontline and in the recurrent. The question related to accelerated approval, the targets, the benchmarks for there will definitely, most definitely include duration of response and a totality of safety data. The issue about overall survival has become more debatable of recent because the post-progression survivorship, as Sir just mentioned, has increased, which challenges the ability and the fidelity of an OS endpoint. Generally, what we're looking for is the lack of decrement. The question always comes up is, what does that mean?
What we've started to do is we've started to power these trials on the basis of a hazard ratio for OS around 0.85, which provides us reasonable confidence that we're not actually hurting patients. That's really how we're going about this. We can get much more specific to the questions, but that's the kind of the general benchmarks that we're looking at for accelerated approval, objective response, duration response, safety, PFS, and OS endpoints for the frontline and recurrent settings.
Just one question for you, Rob and Mansoor. If you subtract the 18 weeks chemo and anti-PD-1 from the, I think it was 13.1 months, I'm going by memory here on the New England Journal of Medicine paper with Dr. Ramez Eskander. It was similar, I think you had Dr. Robert Coleman.
I think we have about nine months or so for the switch maintenance phase. Maybe 13 months or so there would be a meaningful delta. I do not know where the data have matured. I think it might have decreased a little bit from the original 13 months to maybe 11, 12 or so. Yeah, that is to Roger the best, I think. I think you heard Dr. Coleman said maybe four months would change the hazard ratios here meaningfully. Yeah. I think we have your patients are mostly in pMMR and that is right. Yeah. The pMMR is also common and that brings them even shorter PFS. That is right. Expectation. What is the pMMR patient population we are focusing on there, the 80%? It is not grade one, grade two endometrioid, which have a longer. No, no. It is the stage three, four, and the higher grade.
Yes, Adam, we have time for one. Yeah, I think we have time for one more. Our question is from Mark Fram of TD Cowen. Yep. He says there are six biomarker-positive endometrial cancer patients in the safety database that are not in the efficacy evaluable population. Can you elucidate those?
Yeah. We have excluded all patients from, if you remember, we did an OncoSignature lock now quite a while ago, and we've included all patients that we have for the safety evaluation. It's that simple. Three of the six are on the locked OncoSignature now, and they were included in the disposition table that Dr. Bradley looked over. There were three patients prior to the threshold lock, but they're still part of the safety evaluable patient population.
Unfortunately, that is all the time we have today.
I'll hand it back to you, Peter, for any closing comments and thank everyone for participating in the webcast.
Yeah. I want to thank everyone participating today here late afternoon and also provide a special thank you to Dr. Robert Coleman, Dr. Russell Mercer, and Jesper Olsen, our Co-founder from academic Co-founder from Copenhagen, the Novo Nordisk Foundation Center for Protein Research there. I'm very excited about our data and especially how we can potentially position this drug, this drug to serve to come out and help patients. It has amazing anti-tumor activity and noticeably so across histopathologies and especially in the very difficult to treat histopathologies and is already differentiating itself from where emerging ADCs are currently placing them with similar response rates in early lines, less heavily pretreated. I think we have reasons to be excited about this.
We are looking at MDS and MPN as a potential additional potential low-hanging fruit with sensitization to CHK1/2 inhibition and their human transplant data on usage of ACR-368 in that population. That is something that's still being looked into. We have rapid advancement of ACR-2316 and additional pipeline programs following shortly thereafter. We are excited and all this is possible by the platform. Thank you for joining today and wishing everyone a great evening.