Welcome to our Fireside Chat with Curis. I'm Li Watsek, a biotech analyst at Cantor, and very pleased to have CEO Jim Dentzer joining us for the chat. So Jim, why don't we start with an overview, and tell us what's coming up for Curis?
Sure. First, thank you for having me, really appreciate it. Curis is a company really focused on a novel target on IRAK4. So Curis discovered this target a few years back, and it's now been building a lot of enthusiasm in academic centers, NCI, and now some other companies as well. It plays a very important role in multiple different types of cancers, so AML, NHL, and solid tumors. The data we presented earlier this year and through ASH of last year was in AML and NHL.
Mm-hmm.
We've been making some terrific progress. I think at this point, what we're looking to do is identify the fastest path to regulatory approval for both for AML and for NHL, and also produce what I'll call a proof of concept data set, so a roughly 15-20 patient data set-
Mm
... in both cases. We think we got partially there by midyear in AML. We'll be fully there by year-end, and in NHL, we had five patients at ASH. We should have 15 to 20 patients on drug by the end of the year, and you know, somewhat lower number than that for data, but at least a doubling and hopefully a tripling of the data set that everyone has seen so far, which would be very exciting.
Okay, maybe let's start with your AML study. Obviously, you shared some data earlier, this year, obviously in different subgroups. So maybe just review those data for us, and in terms of, you know, monotherapy path or how should we think about in a, FLT3, you know, mutated population?
Sure. So, within the context of AML, IRAK4 plays a unique role that it doesn't necessarily play in other cancers. In other cancers, IRAK4 is a very convenient target for shutting down the Toll-like receptor pathway, which is itself very important. In AML, IRAK4 is the problem, or to state more specifically, the long isoform, the oncogenic isoform of IRAK4 is the problem in AML. So, we know that when you look at the Cancer Genome Atlas, unlike the broader population, so all of us, we have IRAK4's wild type isoform, IRAK4 short. AML patients express IRAK4 long. IRAK4 long is a problem because it forces constitutive activation of the Toll-like receptor path, and that's what's causing disease in leukemia.
So what we have first done is to demonstrate that theory in preclinical data and clinical data that show that knocking back this target in monotherapy will yield responses in the patient population. And then the next step is to figure out how do you maximize that economic opportunity, and that's with combination therapy. So in monotherapy, we put some data out this summer in spliceosome patients and FLT3 patients. The spliceosome data set was encouraging, but it wasn't a slam dunk that this clearly makes a lot of sense. FLT3 data were really encouraging. The data set wasn't quite as large. It was 11 patients. We really were hoping for 15-20. We'll have the FLT3 data set at 20 patients by the end of the year, so that will be coming at ASH, and we look forward to showing that.
And then the long term, of course, how do you go from a relapsed refractory setting from first label-
Mm-hmm
... to including IRAK4 in frontline as part of a combination therapy? We know IRAK4 is important. We know it is a significant driver of disease, not the only driver, and we know that standard of care, Aza/Ven, doesn't hit it. So we just started the study in frontline patients. The first thing we need to show is that it's safe, presuming that we have those data, which we expect to have sometime around year-end. Then we would go headlong into efficacy in the frontline setting.
Okay, maybe let's dive into the FLT3 monotherapy a little bit.
Mm-hmm.
So it sounds like you're gonna share more data by the end of the year with more patients, so-
That's right
... tell us a little bit about, you know, what you need to see from the CR/CR rate perspective?
Mm-hmm
... for you to have sort of a monotherapy path forward.
Sure. Well, so it's monotherapy path forward and also confidence that as a monotherapy, we could get approval. What we really need to show is that we can beat the best FLT3 in the market, and the s-
Mm
... the standard for FLT3 is right now gilteritinib. We know that in their label, gilteritinib can get a complete response or CR/CRh rate of 21.0%. That's in the FDA label.
Mm-hmm.
In our data that we released this summer, it was only in 11 patients, but we had a CR/CR rate of 30%, and on top of that, we know that the gilteritinib patients, the ones they studied for their label, those were primarily patients who had never seen a FLT3 inhibitor before.
Mm.
87% of them had never seen a FLT3 inhibitor before. And in our patients, it was the opposite. The majority of our patients had been on a FLT3 and failed, which means they shouldn't have even gotten to 21%, and yet we got to 30.
Mm-hmm.
So what we need to show by year-end is, does that, does that hold when you get a sizable group? Because in a small patient population, really anything can happen, of course, right? And 11 patients is encouraging, but if we can show that we can hit that, that kind of data set when we get to 30 ... I'm sorry, to 20 patients, that gives us a lot more confidence that when we go to FDA, this drug is approvable, that it is the best FLT3 drug on the market.
Mm-hmm.
And we believe the reason why it should be, at least the thesis for why it should be, is the other FLT3 drugs don't hit IRAK4, and we know IRAK4 is an important target.
By the end of the year, would you have enough of durability data?
That's what we're hoping. So most of the patients, the answer is yes.
Mm.
Obviously, for the newer patients, less so. But if... As I say, if we can beat gilteritinib's data set-
Mm-hmm
... which is, as I say, roughly a 20% CRh rate, roughly a four-month duration. It's a little bit longer. It's 21 and 4.3, I think, were their numbers. If we can show that in 20 patients, fantastic. Obviously, the patients we're just putting on now, their opportunity for duration's gonna be less because they've just got on study, but we should be able to see median. And we will certainly show in a swimmer plot data by, you know, patient by patient, so people can get a sense of, first, does the drug hit the target? Can we get the patients to response, and are these responses durable? And I hope we should be able to answer all three questions in that data set at ASH.
So, obviously, as Jim, you mentioned, combination EML-
Mm-hmm
... is really important, and that's where the commercial opportunity is, and I think for a lot of AML agents that we, we've seen, safety is really, really important as well.
Mm-hmm.
So tell us a little bit about, you know, the triplet combo that you're doing. What do you hope to see, I guess, by the end of the year, and what are some sort of the next steps?
Yeah. So, to put that into context, the... So the gilteritinib today is a $350 million drug.
Mm-hmm.
So obviously, the low-hanging fruit would be if the data continue to look superior. We know it's a $350 million drug, which is interesting.
Mm-hmm
... but it's not blockbuster. What of course we're after is something much larger than that. Much larger than that would be front-line setting, changing standard of care from AzaVen to Emma AzaVen, right? Emavusertib plus azacitidine plus venetoclax. That's where you'd get the blockbuster opportunity. So first things first, let's show it by year-end that we truly do have the drug that it's a better FLT3 than Gilt, right? Than standard of care. That establishes the low bar of, "Okay, this is a $350 million drug," and then chase after in the combination the potential to change standard of care. In this first study, we won't have efficacy data. It's just a handful-
Yeah
... of patients, but as you sort of implied, the issue with Aza and Ven as a treatment is that both drugs are pretty tough to tolerate. They are standard of care. They do work, but both Azacitidine and venetoclax, in their label, the dose-limiting tox is myelosuppression.
Mm-hmm.
We need to make sure we don't have any reason to believe, based on the preclinical data and based on the clinical data that we've seen so far, myelosuppression was not a dose-limiting tox for emavusertib, but we need to demonstrate that these three drugs can be administered safely first, and once we've shown that, we give ourselves comfort, we give the regulators comfort, and of course, we give the investigators comfort that this is a drug that should be moved more broadly, and that's when we go full bore into the study to look for efficacy and see if we can actually improve on AzaVen.
How many patients do we expect to see from the triplet by the end of the year?
Oh, small. We've guided in the past to five or six patients. I mean, there's a small data set by year-end, and again, I don't know that we'll have all of the patients on long enough, but somewhere in that period, call it Q4, Q1, we hope to be in a position where we can say that the triplet looks safe, and that therefore we can pull the switch on the broader, heavier study to go into a hunt for efficacy. You know, five or six patients is probably enough to get a handle on whether or not we can do that safely.
Okay, so in terms of your front line, sort of, I believe you're going after all comers?
Mm-hmm.
I guess, what the study design would look like because a lot of front-line trials probably gonna be looking at some survival-based endpoint. Is that something, you're considering?
Good question. We're thinking of two separate designs-
Mm
... in that front-line setting, so the first study that we're working on, this safety study, we're actually taking patients who are already on AzaVen and have gotten to complete response, but they still have residual disease, that we're then gonna add Emma to it.... hopefully we'll get them to MRD negativity-
Mm.
And get them to no detectable residual disease. But what we're really trying to do is try to isolate the safety effect in that design. So by definition, we're taking patients, you know, four or five cycles in, you know, much later. Assuming we show safety, what we would then do is change the design to go from cycle one, day one, you go on all three drugs right away.
Mm.
And then at that point, what we'd be looking for is responses, durability, and then eventually, yeah, absolutely progression-free survival. But, you know, we think we should be able to increase the response rate of Aza and Ven alone. That's certainly what the preclinical data would suggest, and that's what we'll hunt for.
When would you be in a good position to sort of approach the FDA to talk about some of the registrational pathways, whether before your monotherapy FLT3 or for your combination?
Yeah. So I think. Well, first things first, we're already talking to the FDA about registration, but it's on the- on the-
Yeah
... NHL path, right?
Yes.
And that's where most of investors have focused just because the economic opportunity is so much larger than in AML. But assuming that we have success there, we will have had at that point ... We already have over 200 patients worth of data between leukemia and lymphoma combined. If we have an alignment with FDA on the pivotal design for lymphoma-
Mm-hmm
... we think we would be well-positioned then to have a similar discussion about the setting within AML. Certainly the safety would be well-established. Both monotherapy combination with BTK, and by that point, combination with aza and ven. And then I think what we would do is look in that conversation with FDA to design a pivotal study that would look for ORR as a primary endpoint for accelerated approval. And then we would have to, of course, have a survival endpoint for a confirmatory study, but it would make sense that within the context of AML, responses have generally historically-
Mm-hmm
... correlated with survival, so we think we'd have a strong argument for accelerated approval.
Now, maybe let's switch to NHL. Obviously-
Mm-hmm
... you generated some pretty nice data there. So, maybe just walk us through sort of your strategy there, and I believe you got some, you know, nice data in primary CNS lymphoma?
Mm-hmm.
Maybe talk to us about that opportunity.
Sure. So let's start with the rationale for NHL, and then put it in the context of the data we got, and then why we're having the discussions with FDA. So the rationale in non-Hodgkin's lymphoma is that the problem for patients with non-Hodgkin's lymphoma is really NF-κB complexes overactivity. When NF-κB is overactive, it decreases the body's ability to fight cancer, to kill off the cancer cells. So the apoptotic process for the malignant clone starts to drop off. The traditional way to treat non-Hodgkin's lymphoma or to down-regulate NF-κB is to use a BTK inhibitor, and it's quite effective. It blocks the BCR pathway.
Mm-hmm
... which is one of two pathways that's driving NF-κB, and it works. There are, I think, four or five different BTK drugs now that are approved. Five that are approved. And it's an $11 billion market.
Mm-hmm.
But none of them address the toll-like receptor path. We have the only drug that does. So our view would be, if you want to treat non-Hodgkin's lymphoma, you definitely want to block the BCR pathway, but you also want to block the toll-like receptor path.
Mm-hmm.
You block both pathways, it's better than blocking either one alone. That's the thesis for the drug. The data we put out at ASH last year demonstrated that that was true across all, many different types of NHL. Among those types, there are five different indications that constitute the $11 billion in revenue for BTK players. One of those indications in particular struck us as very attractive from a regulatory perspective. That was primary CNS.
Mm-hmm.
It's ultra-orphan, it's very rare. There are no drugs approved. BTKs get used, ibrutinib is in the NCCN guidelines, so it gets reimbursed, but it's not formally approved, and furthermore, there are no treatments for relapsed refractory patients once they fail BTK, so the first line of therapy is typically chemo, radiation, and methotrexate. It works pretty well, high ORR-
Mm
... but very low duration. Once they fail, and they all fail unfortunately, they go onto a BTK. And again, you can get a reasonable response rate. You can get a complete response rate of 20%, ORR of 52%. Once they fail that-
Mm-hmm
... there is nothing. You can imagine if you ... Once you fail a BTK, if you went back on a BTK again, your response rate is zero. So what we were hoping to do is, can we go into that setting and get the classic, you know, ORR above 20? We know chemo's duration is two months.
Mm.
We know ibrutinib can be a little longer. Could we get two to four months duration? We were very pleasantly surprised that we exceeded both of those in the first five patients. So we're getting a response rate more like 50%. We're getting four months duration plus.
Mm-hmm.
It was early days, as I say, only five patients. But as we got more patients into the study in the first half of the year, we haven't published those data yet, but we obviously liked what we saw, and we decided to go directly to the FDA. Normally, you'd wait till the end of phase to have that meeting to talk about pivotal design, but we thought our data merited talking about this study being the pivotal design. So, we've reached out to the FDA. We were pleased that when they saw the data, they agreed to take the call.
Mm-hmm.
And we're now in discussions with them, and what we're hoping for is we think this drug merits accelerated approval, not full approval. We don't need to go to a survival-based study. An ORR-based study should be sufficient. We think a single-arm study is appropriate, not a randomized study.
Mm-hmm.
It'll cut down the size and the speed. We think that the size of the study really ought to be somewhere in the 45-60 patient range, not a 2-300 patient study. And then most importantly, we think our existing patients ought to count towards that total. So if we're gonna have 20 patients on study by the end of the year, and this is a 45-60 patient-
Right
... study for accelerated approval, we could be 25 to 40 patients away from filing our NDA. Obviously, we're still in discussions yet. Those haven't been completely ironed out. But we are very encouraged that the FDA is interested in helping us on that path, and I think it's because, nothing works, right? Response rates are zero for those patients in this setting. And, assuming that our data hold, and we'll have more data at ASH, I'm very excited about the ability to bring something to the market that could have an outside benefit.
When do you think you might be able to have that alignment with FDA? Do you have to wait for the ASH data, for that to happen?
We're discussing the confidential data now with them.
Mm-hmm.
So I don't think we need the ASH data. We're showing them data in real time.
Mm.
We sent them data in the first half. That's when they took the call, and we're having the discussions now. I don't know when we'll be able to say with-
Right
... definition exactly what this is. You know, the FDA always wants to reserve the right to have some wiggle room. They're very quick, though, to tell you when they don't agree, and that's the good news.
Mm.
So if we say, for example, "We think this merits this accelerated approval with an ORR endpoint," if they disagree, they'll say, "No, no, no, it doesn't, and you really need to run a survival study." Okay. If we say, "We think this is a single-arm study," they could come back to us and say, "No, it needs to be randomized, and here's what we want you to randomize to." If we say that we think it's a 45-60 patient study, that's their opportunity to say, "No, no, no, no, no. You, you need to run a study that's two or 300 patients." Okay, you know, all, all of those things are opportunities for them to course-correct us. So we may not get them to, you know, lock themselves into a box that, "Here's...
If you do these things, we guarantee approval," but we do gain alignment.
Mm.
And I think gaining that alignment-
Mm
... is really the important part, and I'm hopeful that by year-end, we'll have an opportunity to have a better sense of that direction.
In terms of the data expectations, I guess, at ASH, so you're gonna have about 20 patients.
What we've said consistently is we expect to have 15 to 20 patients on study by year-end.
Mm.
Now, just because of patients on study, that doesn't mean that we have the data for all of them, right? First, a number of the patients won't have gotten their assessments yet, so we won't have the response rate data. And for those patients who have response rate data, of course, you need to follow them long enough 'cause you wanna know if you get duration.
Yeah.
But we will be able to say definitively that it is a significant increase in size from what we had at ASH, which was five patients. So for a number of the additional patients, we'll have responses and duration. For some of them, it'll be response only.
Mm.
But I, I do believe that we're on track to have 15 or 20 patients on study by the end of the year, for sure.
How should we think about the market opportunity here?
Oh, yeah, I mean, it's pretty straightforward. I, I think this is one of the reasons why the investors are so interested in NHL. I, I don't get very many questions about AML, even though the data look, from our perspective, quite compelling. We don't get questions on solid tumors, even though that we've got five ISTs ongoing with support of the NCI. I think the reason for that is because everybody knows today that the 2023 revenue for the BTK space was $11 billion. Most of that's ibrutinib, but acalabrutinib is very large, so is zanubrutinib. Pirtobrutinib is early in the days but making fast ground. Ono and Merck both have next generation BTK inhibitors that look like they're on the path towards joining the market as well.
And with all of these competitors fighting for non-Hodgkin's lymphoma patients, we're coming in at the same time, and we're having a conversation with the clinicians today that we would have tomorrow when it's approved. We don't really care which BTK you wanna put your patients on. Blocking the BCR pathway is an important way to treat non-Hodgkin's lymphoma. Pick your favorite one, reversible, non-reversible, covalent, non-covalent. We don't really care. Pick your favorite BTK, add emavusertib to it, and it should make it better. Is it the same in every patient, in every indication? Well, no, of course not. But again, it gets back to the scientific thesis of the drug. You're gonna block the BCR pathway with one drug, the toll-like receptor pathway with the other.
Mm-hmm.
Blocking both should always be better than blocking either one alone. So $11 billion today, and we're just gonna sell the add-on therapy to it. You get a 20-25% market share, I mean, I don't understand the rationale for why you'd be on a BTK and not add on the therapy, but even just say for sake of argument, that you only get a 20-25% market share, that's a $3 billion drug. Our market cap today is $40 million. It's a pretty outsized-
Yeah
... opportunity.
So I guess besides primary CNS lymphoma, what other indications that you're looking at?
So all of the indications, wherever BTK would be approved, all of them are sensitive to varying degrees, to Toll-like receptor activity. There's been a lot of literature, and we talk about this a little bit in the corporate deck that's on our website. So if you went to curis.com and just scroll down, you'll see the corporate presentation, and I would encourage everyone to do that. So the diseases are diseases like Waldenstrom's, Marginal Zone, Mantle Cell Lymphoma, CLL, DLBCL, the ABC subtype. All of these indications have publications, not by Curis, but by academic centers throughout the world, that show that IRAK4 signaling plays a key role in these diseases. It's not the only driver, of course, but it has been shown that knocking down IRAK4 should provide benefit in combination with BTK for these patients.
So we would look to go after all of them. Right now, it's just a resource question. We have a limited amount of cash, a limited amount of employees. We're gonna nail down primary CNS lymphoma first-
Mm
... and we're gonna look to talk to a partner about funding those other indications. And with the data that we have at year-end, as I say, a safety database of a couple hundred patients, efficacy data in PCNSL and AML, and hopefully some increased guidance or clarity on the regulatory front, I think it makes a pretty attractive case to a potential partner. That with certainly less risk than the conventional program, we have the opportunity to have a real novel therapy that could be very important to add across the non-Hodgkin's lymphoma space.
So I guess on the partnership front, so do you get a sense of, in terms of what's gonna move the needle for them? Is it the regulatory piece? Is the data piece, or it's sort of the combination of both?
Yeah, it's really both.
Yeah.
Right? So, you know, pharma partners want what all of us want, right? They wanna make sure that it's been de-risked as much as possible, so they want to know that the scientific thesis makes sense. Is this a good target? They wanna know that the drug is an effective inhibitor of that target. They wanna know that the drug is safe, and that you've got clinical data that shows all of those things are true, and it adds efficacy that is approvable. As you can show that in sizes that matter, I think that starts to build the compelling case. So I'll start with, what's a size that matters? What most investors tell me they're looking for is 20 patients.
Mm.
That's a proof of concept database. So they would look at our NHL data, for example, of last year and say, "Okay, you've got five primary CNS lymphoma patients, and you get a 50% response rate. That's really nice, but until you get it in 20 patients, I'm not sure it's a fluke." Right? "So just show me that." Investors want that. Partners want that. The regulatory piece sort of layers on top. To the extent that we can say with confidence that we've had discussions already with FDA, and we know directionally-
Mm
... where we need to go, that de-risks it even further, and of course, makes it that much more attractive, both for investors and partners.
Okay, great. Thank you so much, Jim.
Thank you, Li.
It was a great discussion.
Thank you. Really appreciate it. Again, thanks for having us.