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

And thanks for joining. My name is Jason Russell with the biotech investment banking team. And it's absolutely my pleasure to host this fireside chat with Jim, the CEO of Curis. Before we jump in, a quick disclaimer. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. With that, Jim, welcome, and thanks for being here with us today.

Thank you for having us. Really appreciate it.

Yes. Listen, it's -- maybe just jump right in. You've had a really incredible 12 months as a company.

It's been amazing. Absolutely amazing. Thank you.

Yes, absolutely. I mean I think most of the focus, at least is as far as most groups will be concerned is around your lead candidate. I want to get to that very quickly. CA-4948, a small molecule, IRAK4 inhibitor, let's start there. But first, let's just jump in on IRAK4. Tell me about IRAK4. Why is that target so interesting? And potentially not only in AML/MDS, but also NHL, and then we go from there.

Yes. So AML/MDS and NHL and, frankly, we are finding new indications constantly. The literature in the community is really exciting. We are doing more research. And of course, in the last 9 months, our partners in NCI are doing more research as well. IRAK4 is just a really important target in the context of cancer across the spectrum. Curis is the company that first pioneered IRAK4 as a target in cancer. Its role specifically in NF-kappaB driving B-cell proliferation was known, and that was the focus of our original efforts. And then just 1.5 years ago, there was a novel paper that came out, was presented at ASH, not our paper. It was presented by Dr. Berman and Dr. Starczynowski at Einstein and Cincinnati Children's, respectively. But it highlighted the role of IRAK4, in fact, the long isoform of IRAK4 in AML and MDS. It's now not just the newest genetic driver of disease in AML/MDS, it's the largest driver of disease. It looks like somewhere just north of 50% of AML and MDS patients overexpressed this long isoform, this oncogenic isoform of IRAK4. So really exciting target. As I say, the data that we've had so far, we are thrilled. It has really shown to be a novel effective drug in ablating cancer in a patient population where no other treatments work. So a really exciting work in progress.

That's great. Helpful background. Certainly, this is all new to a lot of folks, and everyone's getting up of the curve very quickly. So let's jump right into it. Your candidate, you said you're the pioneer in IRAK4. You're leading the clinical charge there in cancer. So CA-4948, give us a brief overview, talk to me about the design, maybe the intended MOA. How -- and also how that might differ across these different subpopulations, which I know is topical. And then also, if you could just comment, Jim, on where it came from.

Sure. So this is a really interesting story. It goes back a couple of years. It was in the lab at Curis where we identified that IRAK4 was a terrific target. In fact, we identified it as a target in cancer before we had an IRAK4 program, which is really interesting. It doesn't always work that way. Usually, it's the other way around, right? You find a molecule and you figure out where it can work. We found this opportunity of this really compelling new target. And before the rest of the world caught on, we canvassed the world to see who had the best IP, who had the best library of candidates that we could try to bring in. And we looked at a number of different companies that had IRAK4 programs. Most of them have been geared towards autoimmune or inflammatory spaces because that was -- up until Curis' work, that was the thought of where an IRAK4 would work. We found that the best IP and best library of candidates and best research capability was actually at a subsidiary of Dr. Reddy called Aurigene. We licensed in their portfolio and their IP interestingly before we had a candidate. We signed the deal first and then spent the next 2 years doing the medicinal chemistry work to develop the lead. And that's really important and pretty rare. And it's why Curis was able to get it. Why did big pharma not go in? Big pharma is not interested in things 2 years before you have a candidate, right? They're looking at licensing and something that's about to get approval, that sort of end. But it gave us the opportunity, given our unique knowledge of IRAK4 and cancer to specifically design the molecule that would be optimal. And we took advantage of that 2-year time frame. You'll see, there's a kinase map of sort of a molecular footprint -- a fingerprint, if you will, of the drug that's in our corporate deck. It highlights the targeting that was designed hit specific receptors or kinases that are drivers of cancer. Two of them that make the most -- or that are most important in AML and MDS context are IRAK4 and FLT3. And we now know -- we put this drug into the clinic last year, and we got our first readout at ASH. We got some more data at June this year and more is coming every day. This drug so far, knock on wood, it seems to do exactly what it was designed to do. So in patients with a FLT3 mutation or patients with lots of IRAK4, even though their prognosis is really grim, we can completely ablate their cancer. That is really exciting. I know there are now a bunch of other companies that are trying to repurpose their IRAK4 programs to come into oncology. We welcome that. We like that we've got a 2-, 3-year head start. We've got 9 active programs in AML and MDS and B-cell cancers all running simultaneously. And we think having the lead program with a novel target that can, frankly, transform the lives of patients with cancer is really exciting.

Yes. No, very exciting. I mean let's peel back the onion a little bit there, Jim. You mentioned the kind of initial data. I think it was ASH 2020. And then as you noted, you had an updated data set here back in June at EHA. Give us a flavor of what you're seeing -- in what populations does the drug appear to be more active? Or is it truly going to be in the 50% of all AML/MDS patients that you stated? How should we interpret the data that we've seen thus far, recognize it's still early and it's still growing?

Yes. So I think the drug is going to have a space across the entire spectrum of disease, whether you're talking about low-risk MDS, high-risk MDS or AML and whether you're talking about monotherapy or combo therapy. Every patient in every indication is going to be slightly different, and a different approach may matter. But we think this drug, because of its novel targeting, because it's an oral, because it's a small molecule with a great safety profile, it really is amenable to either monotherapy where monotherapy makes sense or a component of a drug cocktail where combination therapy makes more sense. So let's talk about where the monotherapy opportunity is because I think that's a faster regulatory path. Long term, we go after all of it. But in the short term, the 2 areas that seem to make the most sense are IRAK4 and FLT3, and let's talk about those separately. With IRAK4, the new finding from 1.5 years ago was that there are specific genetic mutations. These are spliceosome mutations that alter the splicing process and result in the creation of a long form of IRAK4. So the normal form is a shorter form. If you have one of these splicing areas, additional material gets spliced into it into the kinase. That long isoform, that extra material is, in fact, oncogenic. And it's because it causes some of the machinery in the cell to become constitutively active. So it's on all the time. That is what's driving cancer in these patients. And it seemed to happen...

So it is genetic in that matter, Jim, or...

Well, so the genetic component is the spliceosome mutations. As we said, more than half of all patients have IRAK4 overexpressed or at least that's the data that exists so far. And that is what's driving their cancer. The most recognizable cause of that is the spliceosome mutation. But there are lots of other things that might cause you to overexpress IRAK4-L. So the real thing -- the real use of the drug is knocking back IRAK4-L. The more IRAK4-L you have, the more you would need a drug presumably to knock it back. The reason we're focused on the genetic mutations is because this is a novel target. If we were simply looking at IRAK4-L over-expressers, we would need to have an addition to a clinical program and a registrational study for the drug. We would need to have an approved companion diagnostic to show which patients have lots of IRAK4-L. We know that one of the leading causes for having this long isoform is these genetic mutations. And those genetic mutations are already known and looked for in existing panels that you can get from Illumina or Foundation Medicine that are used ubiquitously. So our view would be we don't need to have our own asset. We can piggyback on the existing genetic panels and still get a very significant chunk of the patient population. So that's why that is likely to be the fastest path to registration. Longer term, of course, we may add an assay and go after those patients that have IRAK4-L for other reasons. And in fact, as you saw in the clinical data set so far, one of our full CRs was in a patient that did not have a spliceosome mutation. It underscores the idea that there are other drivers of IRAK4-L. And simply hitting IRK-4L is the driving force of the drug, and it does seem to -- can benefit to patients. FLT3 was the other one. In the literature, there are FLT3 inhibitors, but the literature suggests that the reason the inhibitors don't work all that well, the CR rate for gilteritinib is something like 15%, somewhere in that area. The reason they don't work all that well is because IRAK4 is the escape mechanism for FLT3. That's what the literature suggest. Now it's early days for us. We only have 1 patient that we've made public so far, and that was at EHA. That patient is a FLT3 mutation patient, had been on gilteritinib and, in fact, didn't get a response, flew right through, which is, of course, what the literature suggests could happen. When the patient came on to our study within 2 cycles, not only was that patient getting a response. But when we did the genetic analysis, there was a complete eradication of the FLT3 mutation. So only in a 1, but it is exactly what the latest literature suggested would happen if you hit IRAK4 and FLT3 at the same time. So we're really excited. Those are probably the 2 most likely monotherapy paths to lead the first label. But of course, longer term, our view would be we go across the spectrum of disease, and we introduce combo therapy as well.

Got it. Yes. I mean and the landscape is kind of changing so rapidly. It has been so stagnant for so long. And then, I guess, with venetoclax around us and a lot of other activity, I know that's topical for people. But it's great to hear that you have that path kind of near term to get the drug into some -- into patients quickly and then longer -- build out the funnel opportunity for these patients to really help. So it's great. Let's go there maybe just quickly. So we touched on the data you have. I would assume data update towards the end of this year given the kind of cadence of heme updates. Can you tell us what -- frame for us what you're doing right now on the trials, multiple arms, multiple trials and what we should expect towards the end of the year?

Sure. So we've got 9 separate studies or actually not technically clinical studies, 9 separate pockets of patients that we're exploring with either monotherapy or combo therapy with this IRAK4 inhibitor and FLT3 inhibitor. The 2 monotherapies are likely to generate data more quickly. That's the spliceosome patients in MDS and the FLT3 patients in the AML. But we also have combination therapy with aza, combination therapy with venetoclax. And increasingly, it's not yet approved in both AML/MDS, but increasingly, people are getting excited about the potential of combining aza and ven. Well, when you look at our preclinical data, whether you use aza or venetoclax or the doublet of azacitidine and venetoclax, in all cases, adding 4948 was better and statistically significant better. So our view would be -- we're running those studies simultaneously as well. But their combination studies, they're going to take longer to produce data. Our original -- we got a lot of questions when the spliceosome data first came out. It was stunning. We went 3 of 3. That shouldn't happen. Nothing works in this population. Two things maybe to keep in mind. When azacitidine was approved in MDS front line, they got a CR rate of 5.6%, really low. So in front line all comers, it's really hard to move the needle on these patients. Let's add 2 other pieces of information. First, what about post-aza, when you look about second line, although literature suggests your survival is 4.3 months at that point, and that's all comers, spliceosome or not, 4.3 months, really grim. Add in now the impact of spliceosome mutations. The data that's out there now -- there's a recent publication where they looked at chemotherapy used frontline. Chemotherapy, as you know, can generate a response very quickly. It doesn't always hold, but you can get a response fairly quickly, which is why it's often a first attack on the cancer. 80% of all comers got a response to chemo. Only 22% got a response if you had a spliceosome mutation, 4x worse. So now let's go back to...

That's pretty telling to the initial label opportunity here.

Yes. So look at -- so again, comparing to aza, aza only got a 5.6% CR rate and really hard to move the needle. And we know it gets worse if you're refractory and even worse if you're a spliceosome patient. That prognosis is really, really grim. That's why we think the data that we had were so shocking. We had an ORR of 100%. It's only 3 patients. So let's not get ahead of ourselves. It's not going to be 100% as we, of course, expand the data set. But in all 3 patients with the spliceosome mutation, we got marrow CR or better. There were a lot of questions about why don't you just go right to the FDA. Well, the fact of the matter is we are encouraged, we think the data are fantastic. The patients are really excited, but 3 patients is not enough to go to the FDA. We probably need more like 10 to 20 patients. So our goal is we had those 3 patients at June. Let's try and add a bunch more and try to get 10 to 20 patients on study by end of the year. If we can do that and if the data are at least consistent -- they don't even need to be as good as we've seen so far, they're simply consistent, we think we've got a drug that will provide benefit where no other drug does. And that should lead to a really compelling discussion with FDA in the first half of next year, we hope.

Very, very interesting and exciting. I mean maybe that's a good segue, Jim, to -- you're actually also looking at the same drug in lymphoma patients. I think it's actually been in patients longer in NHL than has been in AML/MDS. Tell us about that. Why, again, is IRAK4 relevant in that subsegment patient population? And what are you doing there in the clinic?

Yes. So thank you. It is the one we went after first. We've got patients that have been on drug now for over 2 years, which is fantastic. This was really the original scientific work that happened at Curis that led us to get excited about the potential for IRAK4 as a target. We knew that in the context of B-cell lymphoma, so the over-proliferation of B cells, that was being driven by NF-kappaB. And then we work backwards from there. What is driving the overactivity of NF-kappaB. And it turns out there are 2 pathways that drive that. One is the BCR pathway. That's where BTKs live. And the other is the toll-like receptor pathway. That's where IRAK4 lives. All to date, all of the drugs going after B-cell cancers is dominated by BTK inhibitors. It's a $10 billion space today. But nobody was going after the other driver. Nobody was going after the toll-like receptor path. We decided we could. And it's as simple as your sink's overflowing water. You've got 2 faucets, hot and cold. The answer isn't decide which one you want to turn off. You want to turn off both of them. Now as a practical matter, we needed to study the drug as monotherapy first, just to prove the thesis that this drug, this pathway, this target does, in fact, matter, and we've done that. We're getting responses in these patients as a monotherapy. But we are now transitioning to where we think the drug will go long term. We don't expect this is going to be a drug being used in monotherapy. If you have a problem where you've got overproliferation of B-cells driven by NF-kappaB, you want to stomp on NF-kappaB as hard as you can, every way that you can, and that means turning off both pathways that are driving it. So that's the study that's ongoing now. It's ibrutinib plus 4948. We expect to have those data certainly in real-time coming, but it will probably be an update at the medical conference in 2022. And we're -- as I say, we're really excited about that. Having patients on drug over 2 years that are not just getting better over time, but showing a terrific safety profile as well is tremendously exciting.

Yes. Absolutely. And maybe just briefly, Jim, in the aggregate here, when we think about both the AML/MDS opportunity, supersafe patients, the NHL opportunity, obviously, still sick patients, probably not quite as sick. But in the aggregate, on a safety basis, you touched on, let's say it's been pretty safe in NHL patients, what has been the safety profile of the drug in the aggregate? And are there concerns? Or are there things that you're doing to mitigate any safety concerns?

Yes. Safety profile has been terrific. So with any small molecule, the whole point of your Phase I study is to dose escalate until eventually you see toxicity. And what you hope to find when you see toxicity is that ideally, it will be something that is predictable, something that is easily detectable and something that is reversible. That's exactly what we have. So we found that we went at really high doses beyond our recommended dose. We found that, in some cases, we did see CPK elevation. It was typically seen only in the first 10 days on drug, so it happens right away. It's easy to identify. It's a simple blood draw. And it looks like it's happening in patients that have confounding factors that are contributing. So for example, you're on a statin. Statins have a black box label for this. Or if you're a heavy exerciser, that drives CPK elevation. It looks like if you have one of those things and then you add really high levels of 4948 to it, you can push someone into CPK elevation. But in all cases, we found that, frankly, not only was it quickly detectable. But all we did was take a quick dose reduction or a dose holiday and it went away, it was reversible. So given that it was easily detectable, easily reversible and only at really high doses in patients that had these exacerbating factors, it frankly was a terrific safety profile critically or, I should say, really importantly, there does not seem to be any overlapping toxicity with any of the commonly used therapies in the space. So that tells us that we know we've had patients on drug for 2 years. So we know as a monotherapy, it clearly has a terrific safety profile, but it gives us confidence that as we move into combination therapy, that we're unlikely to see any issue to prevent combination therapy use as well. So I'd say right now, early days, but really encouraging signs.

Okay. Great. Well, I want to leave a few minutes here for -- you actually do have another clinical-stage drug candidate, right?

Yes.

0 And I think it often gets overlooked just given the excitement around IRAK4 and 4948. So can you talk to us about 8993 and VISTA more in the solid tumor arena, potentially another kind of checkpoint size candidate. So talk to us about the pathway and your candidate there.

Yes. Well, let me preface that with a year ago at this time, everyone had overlooked Curis, period. As you know, we went from a market cap of $22 million to $800 million in 12 months, right? It has been a remarkable ride. I would tell you that even today, the IRAK4 side is completely unappreciated. Folks are looking at this very narrow spliceosome. They're saying, maybe just 30% of the MDS path. They're not seeing the full application of the drug in AML/MDS, let alone the additional indications. On top of that, there's no way there's an appreciation yet for VISTA, and VISTA is really exciting. So we've been working with Randy Noelle up at Dartmouth. Randy is the guy who discovered VISTA. We licensed in his IP, and we've been working in the clinic with this molecule, 8993. It has the potential to transform the way immune checkpoint inhibitors are used, which I do not say lightly.

Transform is a big word, Jim.

Yes. Well, I mean let's look at the data that have been shown to date. We're still in the early days, and we still need to prove out safety levels. We do know that VISTA has on-target tox. It has been known historically to be associated with CRS, with cytokine release syndrome, which in years past, going back 6, 7 years ago, that was a really scary thing. Now of course, in 2018, the NCCN came out with guidelines. So there are ways to manage it. And in fact, it's the key toxicity associated with CAR-T therapy. So of course, the whole world has gotten comfortable with it. But it was -- in the early days, it was sort of a scary side effect. But the efficacy was -- and prospects for efficacy was never in doubt. It looks to be -- there are 3 primary immune checkpoints that probably matter more than the rest. And in order of sequential activity, it's VISTA, then CTLA-4, then PD-1 or PDL1. VISTA is responsible for quiescence. It controls the percentage of your T cells that are going to be active and sent out in periphery looking for tumors. CTLA-4 controls the priming of those cells, and PD-1, PDL1 controls the effect. But let's go back to that primary role of VISTA. VISTA's primary role is it keeps T cells quiet all the time and then releases a very small percentage of them. The preclinical experiments and why so many people are excited about VISTA is that it looks like, when you examine the data in CTLA-4 and PD-1 to date, these drugs are novel, but the response rates are underwhelming. In most of the approved indications, the response rates are 10% to 20%. 80% to 90% of the patients on these drugs are not getting a response. That's incredible. And if you're 1 of the lucky 10 to 20, you often relapse. Well, why is that? Well, the work that's gone on up at Dartmouth that is so exciting appears to point at VISTA as the cause for both of those things. In fact, that has now been repeated multiple times in the literature, interestingly, by Jim Allison's lab down in Texas, Nobel Prize winner from last year. It appears as though the issue, when you treat a patient with a checkpoint inhibitor, a CTLA-4 or a PD-1 or PDL1, you get this mass explosion of VISTA, up to fivefold increase of VISTA. It happens on T cells, it happens on tumor cells and it happens critically on the myeloid cells. And then if you remember, the primary role of VISTA as being quiescent, what's happening when you get this VISTA explosion is that VISTA is putting the T-cell back to sleep, back into its quiet or exhausted state. What we want to do is take the next step, which is if the problem is, VISTA is restraining your T cells, why not block it. And when you do that in the lab, you get a transformative effect on the ability for CTLA-4 and PD-1 to have an impact on cancer. That's where we're headed in the clinic. First things first, we want to prove it safe, as I said. There are some well-known side effects, CRS-associated side effects of VISTA treatment. We need to make sure that we can manage those effectively. I want to be able to be in a position by year-end that we've had enough patients on drug for a long enough period of time that we can say definitively, we can manage it. And if that's true, that's a big value-creating event, and we start the hunt in 2022, looking for efficacy. That was a pretty long answer to a pretty straightforward question.

You covered off the rest of my questions, Jim. So it's great. So listen, we're right at time. A lot of very exciting things happening at Curis, and thank you for joining us. Everyone, Jim Dentzer, CEO of Curis.

Thank you very much for having us. It's been a pleasure.

Thank you.