Welcome, and thank you for standing by. I would like to inform all participants that this conference call, as well as any Q&A, may be recorded and made available to clients of JPMorgan. Where companies presenting any recording may also be posted on their website. Views and opinions expressed by any external speakers on this call are those of the speakers and not of JPMorgan. Parts of this conference call may also be reproduced in the JPMorgan research. Participants are prohibited from posting, sharing, or distributing any parts of this call or its contents on social media or any public forums without prior written consent from JPMorgan. If you have any objections, you may disconnect at this time. This call is intended for JPMorgan clients only. Press participants are not permitted on this call and should disconnect now.
Members of JPMorgan Global Banking Department may be present on this call. I would now like to turn the call over to Brian Cheng. You may proceed.
Good afternoon, everyone, and thanks, operator. Thanks for joining us for another installment of the biotech call series here. I'm Brian Cheng. I'm one of the Senior Research Biotech Analysts here at JPMorgan. Joining us today is Eikon Therapeutics. On the call with us is Eikon CEO, Roger Perlmutter. Roger, thank you so much for hopping on the call here. You and I just met days ago, lots to talk about just on the back of the data drops at ESMO. For those of us who are in the audience today who are not super familiar with the story at Eikon, can you give us a quick overview of the story, what you're working on, and maybe give us a quick overview of what we should expect now in the next 6-12 months? Thank you.
Super, Brian. Thanks so much for giving me this opportunity, and welcome everyone who's joined. Eikon Therapeutics is a late-stage clinical development company. We have multiple programs in the clinic. All of them are in oncology at this point, and all of them at the moment are generating data at the very high end of expectation. The company was founded, though, on a platform which was applying super-resolution microscopy to problems in biochemistry and molecular biology. Our platform permits us to visualize individual proteins in living cells in a nondestructive way with roughly 10 nanometer or so spatial resolution and sub-millisecond time resolution when we want it. It turns out that that's an extremely informative way to visualize processes in cells that ultimately affect physiologic processes in humans. Over time, the platform will be the greatest contributor to value.
We are developing drugs that have come directly from the platform, which permits us to identify targets that elude others and to prosecute targets that others simply can't pursue. The principal focus in the last year or so has been the clinical data that we're generating, and we're quite a large clinical program. We have roughly 1,000 patients under study in multiple clinical trials in oncology, as I say. You can expect that because of the scale of those trials and the fact that they are ongoing and delivering data, that over the next 6-12 months, there will be additional data from each one of those, which include trials in advanced melanoma, non-small cell lung cancer, breast, ovarian, prostate cancer, and in patients in particular who have defective DNA damage repair or mismatch repair.
All of those are areas where we have substantial expertise, have prosecuted targets previously, and where I think we're making real progress.
That's a great start. Let me just looking at TLR7/8 first. As we think about what you presented at ESMO, and I think a lot the prior data cut also came from ESMO, can you talk about just the breadth of the data that you've seen so far? What disruption that you see TLR7/8 could bring to the non-small cell lung cancer space?
Sure. I think it's important to put the TLR7/8 program, which is EIK1001, to put that in context, and the context is a long one historically. It began many years ago when I first had the opportunity to look at data from the molecule that became Keytruda. When I arrived at Merck, it was interesting to see that an antibody directed against PD-1, and this is, of course, now goes back a long ways in history, an antibody directed against PD-1 seemed to have activity not just in what we thought of as immunologically active tumors like malignant melanoma, but also in non-small cell lung cancer, where immune responses against the tumor were not felt to play a terribly big role. And ultimately, we pursued that quite aggressively.
In fact, at the time when I was leading research and development at Merck, my feeling was we were viewing something that was unprecedented, and it was our responsibility to attempt to enumerate the full breadth of response that could be triggered in patients with malignant disease. We began what became an enormous program, the largest program ever in clinical oncology. Keytruda, of course, has become an extremely successful drug and something that has made a big difference for an awful lot of people around the world. It's a fundamental tool in the therapeutic armamentarium, but the fact remains that the minority of patients respond in most cases. The question from the beginning had been, what's wrong with those people who don't respond? Why is it that there are such good responses in many cases, yet other patients don't respond at all?
That led us to do a variety of combination studies, and those combination studies were well-informed by the data that we were obtaining from the clinical analyses. What I can tell you is that pro-inflammatory signals in the tumor seem to be very important, and we sought to increase those pro-inflammatory tumors through the application of chemotherapy, which is a powerful pro-inflammatory stimulus through cell death, and that became the standard of care in non-small cell lung cancer. Those were the KEYNOTE-189 and KEYNOTE-407 studies in non-squamous and squamous cell carcinoma in particular. We did a lot of other studies as well, but my feeling was that there was still more opportunity to trigger, to enhance inflammatory responses that could make Keytruda an even more powerful drug and permit more patients to respond.
One of the things that we were interested in was the question of whether innate immunity triggered through toll-like receptors, in particular TLR7/8 , could be harnessed in this setting. The problem with these toll-like receptors is that they're very potent agonists for inflammatory response. If you trigger TLR7/8 agonism systemically, there's a high risk of what's called cytokine release syndrome, which is a very significant adverse experience and can, in severe cases, be fatal. We began studies of systemic administration of TLR7/8, working with a small company partner, which was called Seven and Eight Biopharmaceuticals, back at Merck. Those studies made plain to us that it was possible to administer a TLR agonist systemically if you paid attention to good pharmaceutical practice.
Since that time, now, we at Eikon, standing on the shoulders of the work that was done at Merck, we, having licensed the TLR7/8 dual agonist, we have in total more than 600 patients whom we've treated with that molecule, and we are very comfortable with the safety profile, and so are regulatory agencies around the world. We've taken that into malignant melanoma, of course, a responsive tumor generally, and we are in a seamless phase II/III registration-enabling study. We did the same sort of thing in non-small cell lung cancer, but to do that study, we had to first demonstrate that it was safe to give a systemically administered TLR7/8 agonist in combination with chemotherapy. We had not done that before, and so we began a phase II study, which is called TeLuRide-005, but the name isn't important.
The important thing is what we are doing is just combining the standard of care regimen, which is Keytruda plus appropriate chemotherapy in non-small cell lung cancer, either squamous or non-squamous, in combination with EIK1001, the TLR7/8 agonist. The data for that phase II study have been. Small parts of it have been presented in various different forums, but at ESMO, we presented the mature data, particularly for the squamous cohort. Those mature data show an improvement in overall response rate beyond what we had expected, and in addition, a quite long duration of response. In fairness, I have to say this is a single-arm study. It cannot be directly compared with any other study. It is unfair to do these cross-study comparisons anyway.
The important thing to recognize is that these data showed that we could use the systemically administered TLR7/8 agonist EIK1001 in combination with chemotherapy as well as Pembrolizumab, Keytruda. We also know we can use this molecule in combination with other checkpoint inhibitors, in particular PD-L1 binders like atezolizumab. With those were the basic findings that it was a safety study and we found it was safe, and that study was what permitted us to begin our registration-enabling studies. I would just say that when we had our conversations with the FDA some time ago now about doing a registration study, we brought the phase II data to them and we said, "We believe these data justify pursuing a registration-enabling trial." The FDA said to us, "Yes, we agree.
You should do a registration-enabling trial." But they found the duration of response and the response number interesting enough. They said, "Look, we need to remind you that if you see data like this in your randomized controlled study, then it is imperative that you, as sponsor, ask that we evaluate this for accelerated approval." They actually put that in the minutes of our record of contact, which is very unusual in my experience, and it tells you that as far as the FDA was concerned, the kind of response rate data and duration that we were seeing looked pretty significant.
That is what we hope for in our non-small cell lung cancer study, that by adding this pro-inflammatory stimulus, we will be able to take some of those patients who inexplicably do not respond to Keytruda plus chemotherapy and make them responders, and responders for a long period of time.
When you think about the phase, specifically TLR7/8, we have seen intratumoral injection of TLR7/8.
Yeah.
Those seem to have a lot of issues, specifically on the side effects perspective, where we see a lot of cytokine storm. Can you talk about, I mean, we're not seeing it here, but can you talk about what differentiates EIK1001 in terms of the design, your approach, that helps you to alleviate this issue that others have seen historically?
Yeah. Yeah, it's a great question, Brian. I think that we and others had studied pro-inflammatory stimuli given intratumorally, not just TLR agonists, but many other different molecules, STING agonists. In general, what we found is, while we occasionally saw responses in the tumor that was injected, we didn't see responses in other non-injected tumor sites, so-called abscopal effects were not seen. Part of the reason I believe that we didn't see that is that, if what we're trying to do is to expand the T-cell repertoire, in essence, and to recruit additional T lymphocytes to the tumor environment, they're not already there, and intratumoral injection is not going to help you very much. You need to get to secondary lymphoid tissue, lymph nodes, and spleen, and that means systemic administration is important. We know that EIK1001 is an active drug.
It's been studied before, and it's been studied in topical administration, and it has been shown to be active in squamous cell carcinoma of the skin, in basal cell carcinoma, in cutaneous T-cell lymphoma, a much harder tumor to treat. So we know it's an active drug. The question is, could it be safely administered systemically? Which was a risk. What we found was that there is a therapeutic window that one can exploit to safely administer EIK1001, a potent toll-like receptor TLR7/8 agonist, and the way that one does that is to administer the drug according to body surface area. There's some other tricks, if you will, related to formulation and other aspects, but administer it according to body surface area, which caps the top dose. If you administer on weight-based dosing, of course, weight increases as the cube, it's three-dimensional.
Body surface area, on the other hand, is two-dimensional. We are giving quite low doses because it is so potent. For melanoma, for example, where we have collapsed the dose after an interim analysis, this is entirely being under the review of an independent data monitoring committee, we collapsed the dose to 0.6 mg/m². That means that for most individuals, they are getting around a milligram of drug total, which is very little, and yet it is still potent enough to clearly have a treatment effect, which you can see easily by measuring certain cytokines that are elevated in patients who are treated. The adverse experience profile, though, is quite manageable. We see what we recognize as grade 1, 2 cytokine release syndrome in single digits, so very infrequent.
Grade 1 or 2 cytokine release syndrome is simply fever and chills, which, of course, you could get from chemotherapy or from Keytruda alone. So, it is very well tolerated, and it has a manageable safety profile in the context of patients who have advanced malignancy, and there is good reason to believe that it is having the appropriate treatment effect. What we also believe, based on looking at the duration of response, particularly in squamous cell carcinoma, is that we likely are having a good clinical effect. That depends, of course, on our acquisition of randomized controlled data.
As we think about the phase III that are ongoing for non-small cell and also melanoma, where are you in terms of the latest progress regarding to those two studies? We recall that there are interim analyses built in for those respective phase IIIs. Could you give us conceptually, what triggers the interim analyses, and then also, what will be the trigger for you to proceed after the next interim analysis?
Yeah, absolutely. So, exactly. We have two studies that are registration-enabling, one in patients with advanced malignant melanoma and the other in patients with non-small cell lung cancer. Both squamous and non-squamous are being investigated. Those studies are very simple in design. That is, we simply add either placebo or standard of care, or, I am sorry, either placebo or EIK1001 to standard of care. That is it. So everyone gets standard of care. They are getting what is best available therapy. Either they are randomized to placebo plus standard of care, or they are randomized to active drug plus standard of care. That makes the studies easier to enroll, of course, because every patient is going to get best available care anyway. The studies have enrolled at the very high end of expectation.
The melanoma study is ahead of the non-small cell lung cancer study because for non-small cell lung cancer, we needed to do a safety study first, the phase II study that I just referred to. The melanoma study has enrolled really quite well, and we were able to conduct our first interim analysis, which was to select between two doses. FDA, through something called Project Optimus guidance, prefers to have an element of dose optimization in the registration-enabling study. It is not a statistically powered dose optimization. That would be too difficult. But we compared two doses, 0.6 mg/m² and 0.75 mg/m² , and our independent data monitoring committee recommended 0.6 mg/m² . Not an enormous amount of data. At that time, they had roughly 30 patients per group that had been enrolled.
Importantly, for both studies, the future interim analyses are event-driven, right? This is typical for these kinds of oncology studies. You have to have enough events in order to be able to see whether or not there is something worth studying. For the melanoma study, the next interim analysis is the analysis that says, "Gee, it looks as if it is possible you could be having a treatment effect. We do not know, but you could be. It is not powered to see that difference. Since you could be having a treatment effect, it is appropriate for you to proceed into the large phase III portion." By the time we have enough events to evaluate that, we will have enrolled close to half of the study anyway, something like that. We do not know exactly when we will have enough events.
We, of course, the company, we are blinded completely, so all we can see is total events. We do not know how many are occurring in each treatment arm. There are now just two treatment arms that are relevant, and we have to look at the events in those two different treatment arms, either 0.6 or placebo. That is chugging along. We do expect that we will have the results of that interim analysis in the first half of 2027, and we cannot speculate on what they will be. As I say, it is event-driven. Similarly, for non-small cell lung cancer, which is a bit behind, we will first go through the dose adjustment, the dose collapse to a single dose, and then we will go through the event-driven analysis of response to expand the entire program.
Non-small cell lung cancer is a much more common tumor, so that will catch up with the melanoma program almost certainly fairly soon, because there are just a lot more patients, unfortunately, to be enrolled. It is also unfortunately the case that response rates are lower in non-small cell lung cancer, and hence events, progression events, are more common. As a result, the two studies will probably get to the point where we have definitive data at near the same time. What we have said, and this is really just an estimate because we do not know anything about event rates right now, but we have said that we expect that the data will be available sometime in the middle of 2029.
Maybe just on the melanoma side, in non-small cell lung cancer side, regardless of squamous or non-squamous, we do see the additive to synergistic effect, right, when you add it on top of IO, when you add it on top of pembro. It is very clear of that. In melanoma, maybe just can you talk about how confident you are today related to TLR7/8 additive effect in terms of on top of standard of care in melanoma? And then, how do you think about the durability response in general?
Yeah, it is a little bit hard to respond to that because the important thing is to look at randomized controlled data, right? I mean, you have to compare melanoma treated with the standard of care, which is Keytruda alone around the world. Although there are combination regimens that have been used, Keytruda alone is the standard of care around the world, and we are enrolling patients to that arm. You compare that with Keytruda plus EIK1001. All evidence that we have comes from single-arm studies that look promising, but we cannot say for sure that there is going to be an improved response, because cross-study comparisons can be very misleading in this regard.
When we started out, our feeling was, if we are going to be able to recruit new T-cell repertoire and to have an improved effect size, we should look first of all at a tumor that is already very immunoreactive, right? That would be melanoma. And then at a tumor that is much less immunoreactive, that is non-small cell lung cancer. We had data from our phase I program that told us that in patients who had failed all prior therapies, including checkpoint inhibitors, that we could see responses, both tumor shrinkage and stable disease. And that was true, for example, in renal cell carcinoma and non-small cell lung cancer. We had patients who responded with tumor shrinkage. In our melanoma patients, we had a large. Essentially, all of those patients who received drugs had at least stable disease. That does not mean, though, that the drug is especially active.
We really have to do the randomized study, and so that is what we are doing. And the study will enroll on the order of 750 patients or so, and we will have a very good idea at that point as to whether systemic administration of a TLR agonist does have a meaningful effect in melanoma. At this point, we would say, gosh, all the underlying mechanistic information that we have is very supportive, but without the clinical data, you just do not know.
I think we touched on this a little bit last week, and I'm curious, just on the back of what we see with the bispecific, the PD-1/VEGF, with ivonescimab. What do you think about just the prospect of that bispecific combining with 1001? Do you see a need to test out that combination? How do you think about the ability to combine and also the prospect of combining?
Well, I mean, yeah. I think, first of all, we are eager generally, of course, to see improvements in therapy for patients with advanced malignant disease. It goes without saying. I'm very proud of the fact that I was associated with the team that advanced Keytruda combination therapy and transformed the treatment of, for example, non-small cell lung cancer. But it doesn't make me feel especially good to know that the standard of care hasn't changed in more than a decade, and we're still using exactly the same treatment protocol that we began with KEYNOTE-189 and KEYNOTE-407. Of course, we'd like to do better. The bispecific data with various different combinations of PD-1, VEGF receptor molecules are interesting, and there are many of them. Obviously, ivonescimab is well along and approved in China and under review in the United States in a special circumstance.
It would make perfect sense to say, "Well, if we can really improve cancer care that way, can we improve still further by providing an orthogonal mechanism that we know is active?" And that TLR fits that right down to the ground. It's an orthogonal mechanism, and we know it's active, and the late Emmett Schmidt was the person who pointed out years ago that the single most important determinant of efficacy in combination studies with Keytruda, in that particular case, was single-agent activity. And we have single-agent activity with 1001. We know that. That makes it very attractive. This isn't the right time to do those studies with ivonescimab because ivonescimab isn't approved in the West. It's not approved in the United States.
Novel studies where you compare a novel agent, 1001, not approved, with ivonescimab, also not approved, don't provide a basis for subsequent approval and use. But as ivonescimab and other bispecifics are approved, it's perfectly reasonable for us to do combination studies because we're adding an active drug that is well-tolerated. We simply have to demonstrate that it can be used safely in combination. There are ways to approach that, and we, of course, would be interested in doing that.
Let's switch gear to the PARP1 selective program. You have two PARP1 selective, EIK1003, and then a CNS penetrant, EIK1004. Maybe just on a high level, when you think about what you're working on here versus the available PARP, what do you want to deliver in terms of the differentiation? I think you just presented the data at ESMO, the data also with abiraterone, paclitaxel. What helps you to build further confidence that PARP1 selective is the way to go?
Mm-hmm. Well, as you say, we have two highly selective PARP1 inhibitors. EIK1003 is about 650-fold selective at the level of enzymatic activity for PARP1 over PARP2. EIK1004 is about 800-fold selective. These compounds and the series that they're related to are structurally very different from any of the existing dual PARP inhibitors, the four dual PARP inhibitors that inhibit both PARP1 and PARP2 that have been registered for some time. We specifically sought high PARP1 selectivity at the enzymatic level because there's good preclinical data that says that PARP2 is much more involved in the hematopoietic toxicity that you see with PARP inhibitors. If you inhibit PARP2, then you suppress production of red blood cells, white blood cells, and platelets in the bone marrow, whereas PARP1 is not associated with those kinds of things.
Now, we don't know that that's true in humans, but we selected two very specific molecules. We chose these. We chose EIK1003 and EIK1004. The compounds were actually invented in a small company in China, Impact Therapeutics, and we scoured the earth using our tools to find the most selective molecules that we could. We found them in Shanghai. Then we selected the particular molecules to advance out of the series and file the INDs and move forward with those things. Simply on the basis of our choice for selectivity, we've ended up with compounds that have very little hematologic toxicity, and that's promising. EIK1004, for example, we presented the first data on EIK1004 at ESMO just 10 days ago, and that's 68 patients, not an enormous amount of data. But in those 68 patients, there was only a single patient, one, that had grade 3 anemia. One patient.
Whereas with olaparib, which is the dual PARP inhibitor that we helped to develop along with our colleagues at AstraZeneca, the grade 3 anemia rate is around 30%. With EIK1003, our more advanced but non-brain penetrant PARP1 inhibitor, we see single-digit grade 3 anemia. So, it seems to me that we clearly have shown that you can obtain PARP inhibitors that have meaningful activity. That activity is enriched in the patient populations that have DNA homologous recombination defects, those who have BRCA1, BRCA2 mutations, RAD51 mutations, PALB2 mutations. As you would expect for a PARP inhibitor, they seem to be quite active. That's a cross-study comparison, but they seem to be quite active. But their hematologic profile, toxicity profile is very different. The reason that we did all this was because hematologic toxicity limits very much the way in which you can use PARP inhibitors in clinical practice.
PARP inhibitors are used to maintain the treatment effect of chemotherapy. That's what they're indicated for. That's how they're used. Those are the labels that we obtained with olaparib, as I say, together with AstraZeneca, who invented that compound. So, maintaining a treatment effect is important. In patients with ovarian cancer, for example, who customarily are initially treated with platinum-based chemotherapy, platinum plus taxane often, if they are responsive, then they are eligible to receive a drug like olaparib to maintain the treatment effect. If they are not responsive, olaparib is not indicated because the hematologic toxicity associated with olaparib is not tolerable in combination with chemotherapy. We wanted to ask the question, was it possible to move upstream in the treatment paradigm? If we had a compound that's a PARP inhibitor, could we use it in patients in combination with chemotherapy?
Was it possible to go into second-line treatment or first-line treatment with a selective PARP1 inhibitor that had little hematologic toxicity? We began those studies by doing something that was extremely challenging, and that is we said, "Could we use our selective PARP1 inhibitor, EIK1003, in combination with paclitaxel?" Which is standard of care for second-line treatment of, for example, ovarian cancer patients who fail platinum, or breast cancer patients who are HER2 negative, and who fail, if they are ER receptor positive, fail hormone manipulation or chemotherapy. That study, a mature data set from that study, was presented at ESMO 10 days ago. In many ways, the most remarkable thing about our presentation was the fact that we could give it at all. That is, that we were able to combine EIK1003 with full-dose paclitaxel at 80 mg once weekly, which has substantial hematologic toxicity.
We could combine those two successfully, and patients stayed on for a long period of time, and there were really quite a lot of The majority of patients had either responses or stable disease. These are people who had failed platinum chemotherapy, so this was true second-line therapy. The average number of regimens that people had received was three. The median number was three. Many had received more than that. We look at those data, and we see the dose intensity is really quite high, especially for EIK1003, and we're using the two together. Well, that was an experiment that we had tried to do with olaparib many times, and we were unable to do it because the combined toxicity was too great. I don't think there's really much question that we have molecules that are different.
These drug candidates are having a treatment effect, and they have much less hematologic toxicity than other PARP inhibitors that have been introduced into clinical practice. The question is now, can we move into a registration setting where we could gain approval for treatment of malignancy with a combination of chemotherapy plus our PARP inhibitors, either EIK1003 or EIK1004? It's especially important, I think, when you look at the combination with novel hormonal agents in prostate cancer, because there's no doubt that the dual PARP inhibitors are active, that if you antagonize the androgen receptor axis, either because you block hormone production, testosterone production, or because you block the androgen receptor itself, those things in combination with PARP inhibitors are more effective than without the PARP inhibitors.
The use of PARP inhibitors, despite the fact that these drugs are registered for that indication, they are really not used very much. The reason is because elderly men with prostate cancer who have gone through a variety of different treatments are not very tolerant of the adverse effect profile of PARP inhibitors, particularly the anemia that they experience. As a result, they are fatigued enough anyway, they just don't stay on therapy. If we have something that has a much, much lower rate of anemia and transfusion dependence, then I think we could really expand the treatment paradigm, change the treatment paradigm in prostate cancer. Our data with the combination of EIK1003 with abiraterone demonstrates that the transfusion rate was pretty low.
It was about 15%, and that was mandated because in our study, anyone who had grade 3 anemia had to receive a transfusion, and our patients were heavily pretreated. Most of those who had anemia had anemia at the time that they entered the study. One of our patients was transfusion dependent. Even so, the rate of grade 3 anemia was pretty low, 15%. So, we are feeling quite good about those studies, and our goal is now to advance these programs into a registration-enabling setting in treatment. I think that will be, frankly, a paradigm shift for the use of PARP inhibitors.
Okay. When you think about the potential of tapping the treatment setting, which is much more mind-blowing than where they are today, which is primarily in the maintenance setting, how do you think about Well, there are two things, right? So one is, how do you think about where to tap first? If you were to run the first pivotal study, does it make sense to do a combo for the maintenance portion first and then expand into the treatment, or does it make sense to leap right into the treatment setting right away? Then second is, you have the CNS penetrant version that's actually evolving really quickly as well. We just saw the data at ESMO. Would you want to see that CNS penetrant version to become the lead for PARP1 selective? How do we think about the setting itself and also which asset to move forward with?
Yeah. Frankly, we are in a situation where we have many more studies that we would like to do than it makes sense for us to do. We are Eikon Therapeutics. We are not Merck & Company. We are not a large company that can pursue many, many studies at once. When we sit down and look at the range of opportunities for a highly selective PARP1 inhibitor that has favorable myelotoxicity properties, there are many.
There are many settings in which we would like to pursue this combination. The brain penetrant molecule EIK1004 is of special interest in patients who are prone to develop brain metastases, right? So, that includes, in particular, breast cancer patients because in the current treatment environment, many breast cancer patients fail therapy because tumor emerges in the central nervous system where existing therapies don't reach them. So, that's an attractive place to go with EIK1004. It's early days.
We just presented the first 68 patients treated with EIK1004, so there will be other things to think about there. But we certainly find the profile very appealing. For EIK1003, again, we have data in combination with taxane. We're eager to get additional data and have begun a study in combination with platinum and taxane, which opens the possibility of first-line therapy. As we look at the matrix of potential registration opportunities, we have to consider, well, where do we have data that gives us confidence, number one? That's always the most important thing, because you prefer to go to a place where you have confidence already based on a phase II dataset, number one. Number two, we have to ask, what makes sense from the standpoint of mechanism? Number three, we have to ask, what's practical to do?
That in terms of speed of enrollment, time required to get a treatment outcome, et cetera, et cetera. When you look at all of those things, of course, ultimately, if we could improve maintenance therapy so people stay on maintenance and we had activity that was similar to what you see with olaparib, obviously, we'd like to do that. But that's a long study, because those patients who are being maintained on therapy, they stay on for quite a while, and it takes a while to get the evidence. So, we're unlikely to move that into the pole position. I'm not saying that we've made that decision. I'm just saying it's unlikely. Whereas in many of the treatment combination settings where we already have data, combination with taxane, just to give you a feel for it, second-line therapy for women with ovarian cancer who have failed platinum is paclitaxel.
Paclitaxel, and these days, paclitaxel in combination with some other things, particularly Keytruda, but paclitaxel therapy in that setting, second-line therapy, gives responses, yes, but they're not very durable. Most patients fail therapy within three months. That's a place where, boy, if we could improve treatment outcomes, it would be really significant. So, that's obviously a place where we're interested. We already have phase II data in that setting. That's a place that's very interesting for us. Over time, we will evolve this, and in consultation with regulatory agencies, we'll evolve this into a set of studies that gives us the opportunity to improve treatment first, I think, for patients who have PARP-responsive tumors, those with the DNA damage repair mutations, either inherited or acquired.
Think about just as the Windhill case, still early. It's early days in terms of what you're seeing in clinic. But how does this fit in to the big picture? I think the ultimate question is, what do you want to see more in terms of the indications? It clearly has its way in terms of going for the DNA mismatch deficient repair type of setting. Highly attractive, but how do you make sure that you're on the right path in terms of flipping the registrational trial forward in the relatively near term?
Yeah. I love this approach. I think it's well-justified in terms of the fundamental research. Many, many years ago, we began these studies with Keytruda in the mismatch repair setting, and those patients who had evidence of microsatellite expansion and contraction in their DNA that indicated that for one reason or another, there was difficulty resolving secondary structures of mispairing because of repetitive sequences. The Werner helicase is an important molecule to help resolve that, to permit cells to replicate. In patients who already have a defect in mismatch repair, Werner helicase inhibition should be meaningful, should be significant. We can see that in cell lines very easily, pre-clinically, and we've done that. We show that the degree of mismatch repair deficiency sensitizes to Werner helicase inhibition. We undertook identifying a Werner helicase inhibitor. It was one of the very first programs that we undertook.
We actually had a pilot study that we did that was funded by Gilead. They were looking at some compounds, and we began looking at the Werner helicase. We had to first demonstrate that our platform could visualize the Werner helicase and that we could see what happened when things that bound to the Werner helicase interacted with the Werner helicase. It turned out to be a very good area for us to pursue and enabled us to identify new compounds. Structure activity relationships were easy to obtain. What we have is a, let's call it a second-generation Werner helicase inhibitor. The first generation began with Novartis, in my view, who assembled a Werner helicase inhibitor from small fragments that they linked together. It was a tour de force, and molecules that were related came from Roche in combination with Bayer and from GlaxoSmithKline in combination with IDEAYA.
Those molecules were brought forward into the clinic as Werner helicase inhibitors, clearly had some activity, but there were quite significant adverse experience profiles there, particularly gastrointestinal toxicity. The responses were not uniform. Our molecule is completely different. It comes from a completely different part of chemical shape space, and it comes directly from our platform. It would not have been easily discoverable any other way. We optimized the molecule, brought it forward. It was so clean in our preclinical tox that we took the molecule into healthy volunteers. It's an orally administered molecule, so it's a pill, but it has a nine-day half-life, so you give it once weekly, or it's compatible with once-weekly administration. You could give it with different frequencies. But we give it once weekly.
We fully occupy the target, and at ESMO, we presented just the first 10 patients that have come through our clinical study now, coming out of healthy volunteers into patients with either colorectal or gastric or endometrial malignancies. These patients could have microsatellite instability, which would make them candidates, or could be patients who had microsatellite stable tumors. What we found was in the microsatellite unstable patients, there were six of the 10 patients, and the unstable patients, all of those patients had stable disease receiving our drug, and they remain on stable disease. The unstable ones didn't respond. All of those patients progressed. That doesn't say for sure that our drug is active. It doesn't say that, but it certainly is intriguing.
We also found that in a single patient to whom we administered the combination of Keytruda, which is, of course, indicated in patients who have MSI-high tumors, Keytruda plus our drug, we got a good partial response, and that partial response appears to be durable. That's also intriguing. It's very early days. This is the earliest time point. It's important to keep in mind that a first-in-human study in cancer patients, we're mostly interested in the safety. That's what we want to see. Of course, we'd love to see activity, but we're mostly interested in seeing that it's safe to administer it. We don't know what dose is appropriate, for example. We've been doing dose escalation, and thus far, as we presented at ESMO, it looks actually very safe.
There are very few treatment-emergent or treatment-related adverse effects in this cancer population, where those patients can be quite brittle. We feel quite good about this molecule. It's very different in its profile from the first-generation molecules. It's chemically different. We're optimistic that the early evidence of a treatment effect will extend as we continue to enroll. I should say that the clinical trial is novel in its design and is enrolling exceptionally well. I think we're going to have quite a lot more data in the not-too-distant future to talk about, and I'm encouraged by what we've seen thus far.
Can I ask you a question about just in terms of still in combination with Werner helicase and PARP. Is that something that you may want to pursue? I mean, you have a lot of potential to combine with your own assets, right?
Yeah.
Is that I mean, you have done preclinical work. I think you have presented preclinical work in the past in terms of your combinability to further either extend durability or even show a deeper tumor response. But how do you think about what would be the right strategy, just given that essentially all three targets are now well on its way in terms of giving you the clinical data that you need to get to the next phase?
Yeah. First, let's just be clear. Combination therapy for cancer is entrenched for obvious reasons, and that is that we expect that because cancer is difficult to treat, that we have to approach tumor eradication with multiple different mechanisms that are orthogonal in terms of their mechanism of action. We will be using combination always. When we think about that, we're looking for new drugs that are especially combinable. That is where we can, in one way or another, engineer out the adverse effects that are associated with these things by selectivity, for example, or by novel chemical structures that have different characteristics. If we can do that, combinability is improved, and combinability is going to be very important.
The third thing I would say is that combinable regimens that can be administered from the beginning are very important because you have your biggest leverage in terms of treatment of malignancy early on before a lot of variants have emerged. They will emerge inevitably because of selection of the drug, whatever it is, the selective pressure of a drug that kills cancer cells. It will select out those that are resistant. Nevertheless, if you can go in at a time when patients have a lower tumor burden and fewer variants are present, you're going to do better. That guides our thinking in every way. Whether our drugs are used in combination with other drugs that we ourselves have produced or with other drugs that others have produced, the goal is still maximum treatment effect.
We have to do it in such a way that we're combining with drugs that already have a registration platform, because otherwise, it's very difficult to work out contribution of components and to obtain a label that explains how to use the drug and basically explains to physicians and to patients how the drug is best used. That's at a very high altitude how we think about it. We're fortunate to have an abundance of new molecules that thus far are performing extremely well. They're well tolerated. They have adverse effect profiles that are much improved over what is currently available in the case of PARP inhibitors, or that are extremely well tolerated despite concerns, as in the case of our toll-like receptor agonist program or the Werner helicase program.
They potentially can be used in combination with lots of different active drugs, and in some cases, it's possible they could be used together with our own drugs. We also have more new drugs coming out, new drug candidates. For example, we were able to develop a non-lutamide androgen receptor antagonist that will be a first, that is not only active against wild-type androgen receptor, which is the target of enzalutamide and darolutamide and apalutamide, but it's also active against all nine of the single point mutations that are selected for with enzalutamide therapy. We filed the IND for that molecule, which is called EIK1006, ahead of expectation. We had planned to do it next year. We filed it some weeks ago, and we expect very soon that we'll have the opportunity to bring that into the clinic and begin enrolling patients.
That's another, we hope, very combinable molecule based on its pharmaceutical properties. That's, in general, our goal is to make combination chemotherapy more and more successful.
As we think about JPMorgan Healthcare Conference in January, just actually just weeks away, how do we think about where investor attention should be? Where should we focus on? Where should investors spend more time on heading into the conference? And where do you think JPMorgan Healthcare Conference focus will be for you?
Well, first of all, I'm not going to presage what we will say at the JPMorgan meeting in January. We will have a lot to talk about. I think that we had seven abstracts accepted for ESMO, which is a lot for any company, but especially for a small biotech company. That's extremely unusual, maybe unprecedented. That's a lot of clinical data. And we had data across all of our clinical programs. That's four different molecules in multiple clinical programs. There's a lot to assimilate there, and I think investors are still working through the data that we presented. It's hundreds of patients' worth of data that were presented at ESMO. The data, as I say, are at the very high end of expectation and augur well for the future.
We'll have a chance to put those into a somewhat different context when we speak at JPMorgan, because, of course, that's not a scientific meeting, that's an investor meeting. We'll be able to put it into a different context and give people a sense of how much progress has been made at this small little company in Millbrae, California. And I'm looking forward to the opportunity.
Likewise. Well, Roger, we are at the top of our hour together. Thank you so much for coming on our fireside chat, and we really look forward to seeing you in just a couple of weeks in January in SF. Thank you so much for your time.
Thank you, Brian.
Thank you.
Bye-bye.
See you soon.
Ladies and gentlemen, thank you for attending today's call. This now concludes the conference. Please enjoy the rest of your day.