Curis, Inc. (CRIS) Earnings Call Transcript & Summary

Welcome, everyone. Thank you for joining us at the H.C. Wainwright 2021 Global Life Sciences Conference. My name is Stephen Bersey. I'm an equity research associate here at H.C. Wainwright. Now I'd like to welcome our next presenter, James Dentzer, President and CEO of Curis. Curis is focused on the development and commercialization of drug candidates for the treatment of human cancers. So now I'll hand it over to James.

Thank you, Stephen. Thank you for having us, really appreciate it. It's nice to be here today. I'm going to go through a general overview of Curis. I'll go through our corporate presentation deck, which I think you can see on-screen. I'm not going to belabor every single slide, but there are an awful lot of really exciting things going on at Curis. And I think in a brief period, we can go through the highlights. So if I skip right to Slide 4. Slide 4 gives you a good overview of what makes Curis different compared to other companies in biotech. What we're looking to do is invest exclusively in novel, first-in-class cancer drugs. In particular, we want to look for those opportunities that have significant potential and unmet need. What I really mean by that is something that has blockbuster potential, something where the revenue can exceed $1 billion per year. Something doesn't have that kind of pay off at the end, it really doesn't justify the investment that we're going to need to put in, in terms of time and money and effort in order to get that drug through to approval. There are 2 drugs in our pipeline that I'm going to focus on today. One is IRAK4 and one is VISTA. So IRAK4, the drug name is 4948 and our VISTA program is 8993. IRAK4 is a small molecule, it's a pill. It is a first-in-class inhibitor of IRAK4 in oncology and VISTA is a monoclonal antibody. That's an injectable, and it is the first-in-class drug addressing VISTA. I'm going to go right to the IRAK4 program because that's the one that's gotten all the press over the last 8 weeks. So as I said, we have the first-in-class drug in oncology, addressing IRAK4. We've been going at this drug in the clinic now for about 2 years. And in the early stages -- of course, we spent a lot of time talking to investors and talking to physicians about the scientific theory for why IRAK4 is a great drug in oncology, but we were the only ones going after it. We started to get a lot of momentum and attention over the last 3 or 4 months really, when a couple of things happened that were very favorable. First in November, we signed an agreement with the NCI. So the NCI has now said that they believe IRAK4 is a great target in oncology. And of course, as they look across the entire spectrum of IRAK4 inhibitors -- and there are other IRAK4 programs, they're just not in cancer. The one that they decided they wanted to focus on for clinical and nonclinical study was ours. So we were very excited by that. We signed a CRADA with them, a research agreement, with the NCI in November. That, of course, caught a lot of people's attention. And then, of course, a month later at ASH, we announced data that I'll go through in just a moment. What makes our IRAK program different from others is, I would say, we have the only IRAK4 program that's a dual targeting inhibitor. It hits both IRAK4 and it hits FLT3. Both of those targets, we believe, are very important in oncology. And then the second big advantage we have is, of course, we were the ones that first discovered IRAK4 as a target in oncology. So we are the first program in. We've got a 2- to 3-year head start on competitors. And of course, if the data keep panning out the way they have so far, we expect to be the first one to approval. And therefore, as you might imagine, gain the lion's share of the market first. If we walk through our mechanism of action of why we think IRAK4 is such a great drug, and I'd to start with AML/MDS. So this is a type of leukemia, very aggressive leukemia that, up until a year ago, the world did not associate with IRAK4. There was a paper that came out in 2019 that was published and its authors on it, Amit Verma and Dan Starczynowski, presented their findings at ASH last year. What they were looking at in their paper was they were analyzing some specific genetic mutations that seem to occur frequently in the AML/MDS patient population, but nobody really understood why. What Verma and Starczynowski showed is that the reason these mutations show up frequently in the patient population is because they are, in fact, driving disease in these patients. Further, they highlighted how these mutations are driving disease. And you can see on this slide -- I'm on Slide 9 in our deck. On the left-hand side, you see a normal cell. A normal spliceosome processing results in the short isoform, the normal or wild-type isoform of IRAK. If, however, you have one of these genetic mutations, these are mutations in the spliceosome, what happens out of that aberrant processing is you get a long-form of IRAK4. And critically, in this long isoform of IRAK4, the N-terminal is retained, and that's very important because it, of course, causes IRAK4 long to be oncogenic, and you see this on Slide 10. So on the left-hand side, what Verma and Starczynowski are showing is what happens when they created a knockdown model in the lab, and the only thing they're knocking down on IRAK 4. When they do this, they get a dramatic reduction of leukemia colonies. This is really exciting. They are now seeing in the lab, a new driver, the first new driver, genetic driver to be implicated in AML/MDS population in years. They were very excited by this. And of course, what they wanted to know is, is there a drug available that can do that same sort of thing, they knew that Curis had the IRAK4 program. So they asked us for a drug, which we gladly gave them, and you see the results of their work on the right. When they treat their mice in the lab with our drug, blocking IRAK4, they've got the same results that they got in the knockdown models, meaning a dramatic reduction leukemic blasts. So now they're even more excited. They found the first new target in years, and there is a drug available that appears to hit that target. So the next question that they want to answer is, of course, how important a target is this? Yes, it's new, but does it affect the sizable enough proportion of the population? What we found in our experiments was that the mice that appeared to benefit from blocking IRAK4 that, that would be sufficient to get this dramatic reduction in leukemic blasts, they had 1.25 times as much IRAK4 long as they did short, and everyone has a spectrum. You might have 50-50 long to short, might be 60-40, 70-30. What they found is that if you had 1.25 times as much IRAK4 long as short, that was enough for it to be driving your cancer. So based on that, they then modeled well what percentage of the human population, the actual patient population of patients with AML/MDS fit that profile. How many patients as a percentage have 1.25 times as much IRAK4 long as short? And the answer to that question is 50%, it's actually 52% to 53%. So this is groundbreaking. Not only have they identified the first new genetic target in AML/MDS in years, but it turns out that it is the largest genetic target, largest genetic population in AML/MDS yet discovered. And in fact, there's a drug that's currently available that you can go into the clinic to treat, our drug 4948. So we, of course, took their findings. We went right to the FDA. The FDA was as excited about it as we were. They encourage us to go very quickly into the clinic, which we did. And we went from presenting the findings of this work at ASH last year to 7 months later, getting through all of the processes and dosing our first patient. We had 6 patients treated on drug and got the results back in time for ASH. And when we were trying to figure out what results might define success, we looked at the only other program that we could find that went in that very same patient population. So these are patients that have failed prior therapies. They've relapsed, typically from azacitidine, but they've -- they relapsed/refractory patients, and in monotherapy. There was a drug from a company called Forty Seven. When they went into that patient population, they saw 15 patients. Out of the 15 patients, they saw blast account reductions in 6 of the 15, which was pretty good. They didn't get any responses, so they didn't get any blast recovery down into the normal range. But simply being able to move the blast counts down, which are, of course, clogging up the bone marrow in these patients and causing disease, getting 6 out of 15 was terrific. We thought, given that the mouse modeling suggested that half the patient population would benefit, that we would get blast count reductions in half our patients, 3 out of the 6. Assuming the 6 is a representatives sample, of course. What we found when we actually opened the envelope and looked at the data was something very different. What we found was we went 6 out 6. This was shocking. It could be that it's statistically possible that all 6 patients have 1.25 times as much IRAK4 long as short, or that all 6 patients have one of these splices on mutations. And of course, we don't know that at the time of the ASH presentation. We will be getting that answer and, of course, presenting that later this year. But when we first opened the envelope on those 6 patients, we were shocked. So it's statistically possible that these patients are all part of that 50%, but it's unlikely. What's more likely is that we can simply affect a larger percentage of the population than what we were originally anticipating. We will, of course, be looking to enroll more patients in the study and follow them for a longer period of time. We wanted to see can we get more than 50% of the patients to show blast count reductions? That would be fantastic. But also, as you can see in this chart, 3 of these patients finished in the normal range, 2 of them started well above the normal range, so those 2 are officially responses. You remember when we were setting our expectations for the study, that Forty Seven saw 6 of 15 blast count reductions, 0 of the 15 got a response. We went 6 of 6, and 2 of the patients got a response. So this is shockingly good data. Our view would be, we were originally anticipating with this drug we were going to go into combo therapy. Azacitidine is more commonly used with venetoclax increasingly used as well. And there are other therapies often used in the AML/MDS. We believe, because this is a small molecule, because it's oral, it's a pill, because it is not myelosuppressive, and because it's clearly disease-modifying as the single agent, that it is an ideal companion as part of a combination therapy of drugs for patients with AML/MDS. And we're clearly going to pursue that path. But we think, given these data, there may well also be a monotherapy alternative. We are going to look for opportunities where patients have relapsed or refractory patients who have already been on a prior therapy and have relapsed and now really, there's nothing for those patients. We may also be able to go earlier stage of disease, where really nothing gets used with chemo. We just announced a few weeks ago that Dr. Uwe Platzbecker in Germany, who chairs the MDS consortium in Europe, and he has 17 sites in his consortium, that he signed on to run a study in low-risk MDS. So we are now going after those early-stage patients. We are about to kick off a combination study in the U.S. in high-risk MDS/AML patients. And of course, we're continuing our monotherapy study as well. So we're doing all of these things simultaneously, and we look to present data in all of those studies at varying points throughout this calendar year. So that's the AML/MDS story. We're also planning to pursue NHL. This was the original area that we went after. As I mentioned in the beginning, there are lots of different reasons why IRAK4 might be a target for a drug in multiple different types of cancer. In NHL, the key is that we're trying to indirectly target NF-kappaB. What Curis discovered is that NF-kappaB overactivity, which has been implicated in lots of different types of lymphoma, you want a drug that can downregulate that. For the last 10 years, it has been known that the BCR pathway is a major driver of NF-kappaB, and drugs that -- especially ibrutinib, the drugs that target BTK in that pathway can, in fact, downregulate NF-kappaB and be an effective treatment for patients with non-Hodgkin's lymphoma. What Curis scientists discovered was that the Toll-like receptor pathway is the second driver of NF-kappaB activity, and there were, in fact, no drugs that were capable of blocking that pathway. The gene that's implicated in most overactivity of the Toll-like receptor pathway is the MYD88 gene. The MYD88 gene is a central component of the Myddosome and that gene, in particular, is very difficult to target directly. What we learned was you don't have to target that directly. You can target IRAK4. Myddosome will not function without IRAK4. So if you can block IRAK4, you block the Myddosome. And because the Myddosome is a choke point in the Toll-like receptor path, any oncogenic activity that happens upstream, that it has to go through the Myddosome, will be blocked as well. So the bottom line is, if you're looking to downregulate NF-kappaB activity, today, with current therapies, you have one choice, block the BCR path with a BTK inhibitor. Tomorrow, what we look to bring to market is a drug that can block the Toll-like receptor pathway. And that either one of these drugs on its own should be beneficial in down regulating in NF-kappaB. Of course, the ideal would be the [ novel ]. The data that we have studied to date has been monotherapy data. So we've literally been looking at patients who have been on ibrutinib or on another BTK and have relapsed. Some patients that are naive, and for whatever reason, haven't gone on BTK, and we looked at those patients as well. But we've been looking to prove the case that if you block the Toll-like receptor pathway, if you block it with an IRAK4, that, in fact, you can downregulate NF-kappaB. And we looked for 2 pharmacodynamic markers to determine whether or not we can do that actively. The first is NF-kappaB phospho-p50. So we use that as a proxy for NF-kappaB activity. If phospho-p50 stains positive, that means you do have active NF-kappaB which means that either blocking either of those 2 pathways off to work, whether you use a BTK inhibitor or an IRAK4 inhibitor, if you're positive for p50, one of those 2 ought to be effective. And then we also look MYD88. If we run the genomics, if we profile a patient's tumor, and we find that this patient does, in fact, have a MYD88 mutation, it is more likely that the activity driving NF-kappaB is coming from the Toll-like receptor side rather than the BCR side. So we look at both of these biomarkers in concert. We are still collecting data to verify them. But the data that we have so far seem to suggest that both of those things are very accurate predictors of the potential for our drug to work in these patients. On the left-hand side on this next slide, you get a hint of that in phospho-p50, and these are all -- it's half the patient population that we've tested so far, and it's usually a subtherapeutic level of drug. But even at subtherapeutic levels, you get the indicator. On the left-hand side, these are patients that tested negative for phospho-p50 baseline. So of course, if you don't have NF-kappaB activity, it stands to reason taking a drug that down-regulates NF-kappaB shouldn't be effective and in fact, it isn't. Every single one of those patients progressed in the study. On the right-hand side, what you find is these are patients who tested positive or stained positive for phospho-p50, meaning they did have NF-kappaB activity. And what you can see, even at the low levels of drug, even at subtherapeutic levels of drug, you did see stable disease in all of the patients, except for one. And oftentimes, actual tumor reduction even at subtherapeutic dosing. On the bottom right-hand side, you see a before and after. We do have 1 patient, unfortunately, only 1 but we do have 1 patient that has pre and post samples of the tumor immediately before and after drug. So -- or not immediately before it. Baseline in the before case and then afterwards taken. And what you see is that the patient tested positive for p50 at baseline, meaning this is a patient that had NF-kappaB activity. After taking drug, that patient now stains negative. So clearly, it's a small sample size. It is only 1 patient, but what it shows from our perspective is that the drug is hitting its target, it's hitting IRAK4. And that it's -- IRAK4 is doing its job, it's blocking the TLR pathway, which is, in fact, down-regulating NF-kappaB, which is really what we want to do. The reason we're going after IRAK4 is to downregulate NF-kappaB. So we now -- only 1 patient, but we still have a very clear sign that this biomarker works. When we look actually at the clinical results so far, again, and this is monotherapy. What we did find is, in fact, a very strong dose response. So this happens to be a patient with Waldenström's macroglobulinemia, which is associated with MYD88 mutations. We started the patient at 50 milligrams in the study. When we cleared the next dose for safety, 100 milligrams, the patient was eligible to dose increase and did. And in fact, when the patient increased to 100, saw a further reduction in tumor burden. Same thing happened at 200 milligrams and again at 300 milligrams. And as you can see at the bottom half of that chart, that patients is now a PR, a partial response, and has shown a 67% reduction versus baseline in tumor. So very exciting. We've got the biomarkers identified that show that we're hitting the target. And in fact, in this patient, we were fortunate enough to have a patient that had been at multiple dose levels in the study, and we can see that clearly, we're having an effect of not just hitting the target but a dose response. More drug is better up until we get to the recommended Phase II dose, which is 300 milligrams. This next slide, Slide 21, is merely a review of the data of what it looks like for those patients who were on at 300 milligrams. Again, this is monotherapy. This study is ongoing, and we are collecting more data, and we are following these patients for more time. The next step in the clinical program is, as I said at the outset, the best answer for these patients is likely combination therapy. If there are 2 ways to down regulate NF-kappaB, a BTK inhibitor and an IRAK4 inhibitor, rather than choosing one over the other, you really want both. But first things first, what we need to be able to do -- this is the preclinical data slide that highlights that same idea. What we needed to be able to do is show that, that red line worked. We already know the blue line works, this is an approved drug. But we need to be able to show that the red line works, meaning that if you block IRAK4 that, in fact, you will get benefit. And then the bottom line, as you can see, is combination therapy. This is in the lab, so the -- this is mouse data, but it's the next step in the clinic. We're enrolling those patients now, and we hope to have data from that study at the end of the year. The last program I'm going to touch on it, and I'm going to do it fairly quickly, is VISTA in the few minutes we have remaining. So there are 3 critical check points that everybody agrees on as being the most primary of the checkpoints. In this Slide 27, we took directly from the January issue of Science last year. Those 3 checkpoints are VISTA, CTLA-4 and PD-1. Of course, Jim Allison, won the Nobel prize for his work with CTLA-4 last year. PD-1 came on a little bit later, that was the second of the 3 major checkpoints to be addressed. And the third to be addressed is VISTA. The first program to go in the clinic blocking VISTA is our program, and I'm going to walk you through a little bit of the logic as to why we think this might be a great idea. We've got a 4-step diagram of what happens in the tumor microenvironment. So VISTA's primary role among the checkpoints to keep T cells quiet. What you see in this first cartoon on the left-hand side is a T cell and a tumor cell. One of the checkpoints, in this case, it's PD-1, but all the checkpoints work similarly. It's expressing on the T cell. PD-L1 is its ligand on the tumor. It's binding to PD-1, and the tumor is passing a signal to the T cell to tell it to go to sleep. Once that T cell goes into an exhausted state, of course, the tumor cell now grows unabated. The T cell no longer at tax it. What won the nobel price is the second graphic. The second graphic, that comes from Jim Allison's Lab, and it says, well, the problem is that tumor cells are communicating through PD-1 to put the T cell to sleep. What happens if you block PD-1? Prevent that connection from happening. Of course, when that happens, the T cell stays in the fight. It attacks the tumor. The issue is that when this happens, you only get a response 10% to 20% of the time, outside of Hodgkin's. Hodgkin's lymphoma is a bit of an outlier. But in all of the other cancer indications, these drugs only seem to work 10% to 20% of the time. Why is that? Also, if you're one of those lucky 10% to 20% of the patients for whom the drug works, you often relapse. Why is that? That's what leads to that third graphic. Again, this also comes from Jim Allison's lab, but now many people have replicated his work. When you treat a patient with PD-1 or CTLA-4, what we find is you get an explosion of VISTA expression. VISTA expresses on T cells, on the tumor cells and critically on the myeloid cells on the MDSCs. And as I said in the beginning, VISTA's primary function is quiescence. It keeps T cells quiet. So what's happening when you get this VISTA explosion? It's effectively overpowering the effect of the other checkpoints, and it's putting the T cell back to sleep, back into an exhausted state. And now of course, the tumor cells are for growing unabated. This answers the question of why are the response rates so low with these other checkpoints. And it also helps answer the question why are patients developing adaptive resistance. Well if this is the problem, then you see the graphic on the right, and this comes straight from Randy Noelle's lab at Dartmouth, the team that discovered VISTA, and certainly one of the leaders in VISTA work. And what you find is they took that obvious next step of once you've understood how VISTA works and where it is, if you design a molecule to block it. You can prevent VISTA from overpowering the other checkpoints. The T cells will stay active and they will fight the tumor. The clinical results from their lab, and this is the last slide I'm going to talk about, the clinical results from their lab are incredibly exciting. On the left-hand side, you see what happens when they treat mice in monotherapy. And this appears to be very effective in -- if you look at the Cancer Genome Atlas, in the types of cancer that are clearly driven by VISTA from the get-go. These are types of cancer like melanoma, [indiscernible] melanoma, triple negative breast. But what about those other cancers, cancers that are not driven by VISTA from the get-go? Like colorectal cancel, for example. Well, they uses the CT-26 model, which is a particularly aggressive model of colorectal cancer in mice. And what they found when they treated their mice with a combination of CTLA-4 and PD-1, which is a very effective combination. Even though that -- those are great drugs, at day 15, all of the mice were dead. If the only thing they did differently is give their mice anti-VISTA antibody. They find that at day 15, 60%, 6-0 percent of those mice are still alive. They go out to day 60, 40% of those mice are still live, even in this incredibly aggressive model. This is potentially very ground breaking. They're early days, and this is just mouse data and just in the lab. But as I said, Curis has the lead drug in VISTA. We entered the clinic in Q3 of last year and we hope to be in a position by the end of this year and have some data worth talking about. At this point in time, it's all, of course, excitement about theory and the lab and the lab work. But we're now testing in patients and hope to have results later this year. So 3 really exciting progress on Curis: an IRAK4 program in AML/MDS that caught the world's attention at ASH 8 weeks ago; an ongoing effort in NHL that's just entered combination therapy studies this quarter; and then the newest one, the anti-VISTA program, we started dosing patients in early Q3, and we hope to have data on those patients by the end of this calendar year. So 3 really exciting drugs. I want to thank H.C. Wainwright for having us here at the conference. It's a pleasure talking to you, and I wish you all a terrific day.

Excellent. I want to thank you, James, for a very informative presentation. Hopefully, our next conference will be in person. So...

I hope so. I hope so. These virtual conferences are awfully hard. And I feel for all the people who have to watch them.

Exactly. All right, have a good day.

Looking forward to meeting in person. Thank you again.