Here. I'm one of the senior biotech analysts at TD Cowen. Thanks so much for joining us for Cowen's Fifth Annual Oncology Innovation Summit. So for this session, we have a fireside chat with Tempest Therapeutics, and it's my pleasure to introduce Steve Brady, who is CEO. It's a privilege to have you here, and thank you so much for joining me.
You too.
So before I get started, thanks to those in the audience, feel free to email me any questions that you might have or use the link that was provided in your invite. So now, Steve, to begin, why don't you start with a general overview, any updates that you wanna highlight for us before we dive into the programs?
You know, thanks, Tara, and I appreciate your invitation to you and the broader TD Cowen team. It's great to talk to you today. So Tempest is a novel biotechnology company, in some ways a very traditional biotechnology company. We have a diversified portfolio of assets. Two are in the clinic. The main one, the thrust of everyone's attention, is called TPST-1120. We do have our INN name. It's amezalpat, which I love because it of the beginning of that word, we just got that. And that's the lion's share of the interest that everybody's looking at because we released some very exciting data in fall, in October, in a randomized phase II, where we added amezalpat to standard of care in first-line HCC, and it was compared to the standard of care.
As you know, the stock went on a tear. I think we broke a Nasdaq biotech record. It went up almost 4,000%. And that was a wonderful, wonderful validation for... You know, 'cause we were one of those companies that's been caught in a dislocation from a market cap perspective, given not only the fact that we're an oncology company, but also because of the biotech slump that had gone on for over two years. So that was a wonderful bit of response from the market, and this is a year where everybody's not only waiting for the key piece of data from that data set, but also about what we're gonna be doing moving into a pivotal study. And but because that gets the lion's share, we often don't get to talk about what's also in the company.
There's a second program in the clinic. It's a dual EP2/4 antagonist. We're probably not gonna have time to talk about that today. It's called TPST-1495. Also interesting, but behind 1120 and amezalpat. And then we have—we are an integrated wet lab company. We have a research group, working on novel programs and everything that's public and also what's not public. Other than one program that we in-licensed from Cal that's public, the target's not, but that we in-licensed, everything is invented inside Tempest. So, that's why I call it a traditional biotech. It's an exciting place to work with real novel science.
Great. Thank you for that, for that overview. So, I do expect that we will have plenty of time to get to 1495, but definitely, of course, want to focus on 1120 first. So, just as to give us some background context for it, can you tell us more about why you chose the, like, this particular mechanism, and what role does it play in cancer, in particular in HCC?
Yeah, it's actually one of the reasons I joined the company. The program was decided upon, for lack of a better term, to target PPAR alpha. And as you know, the pathway is well known. The agonism side of PPAR alpha is well known, but antagonism has been a largely unprosecuted area of drug development. Not completely, there were with some big pharma programs in the past, but we believe, to our knowledge, we're the latest stage clinical program, and I think the only clinical program targeting it. So the original idea, and this is what I love, was to try to have a dual mechanism of action. So PPAR alpha is expressed both on certain cancer cells and on tumor-suppressive cells, again, M2 macrophages and T-regs, right?
The idea here was, if this works, we could be working directly on the cells, so a targeted therapy, and also have an immune-mediated aspect. Fast-forward, that's what we actually think is happening now that we have the phase II randomized data that were released in October. We think we're actually, you know, seeing that, which is wonderful when it dovetails back all the way to the beginning. The idea also was that there are some cancers that, you know, heavily express this target, HCC being number one, all the way on the right side when you do a genetic profiling analysis of PPARα. But it was important to us not to just be an IO therapy, and this is before IO became, you know, unpopular, right? We just wanted...
We loved that idea that even in patients where, you know, there's an immune-suppressive environment, you know, that we could target the cell directly. But then there was this hope, right, and part of science, novel science is a hope, that if we could knock out, given the target expression on M2 macrophages and T-regs, if we could knock out those immune-suppressive cells, could we sort of make a cold tumor hot, so to speak, right? One of the holy grails on that side of the equation. But at its core, it's probably a next-gen on the cancer cell side, actually, on any of these cell side, it's a reliance on FAO, right? These cells go through their life cycle, and they become reliant on fatty acid, you know, metabolism to live after they've used up sugar.
And the idea was that they're gonna become so reliant that if we hit this target, especially in the, the cancer cell types like HCC, that we're gonna see therapeutic benefit for these patients. There are mutations that are relevant as well, like beta-catenin mutation, a very popular topic right now in the industry, increases reliance on FAO metabolism. So we expected before we got the phase II data, we were really interested to see, would beta-catenin patients do better? It's important to note that we are not this is not a beta-catenin-targeted drug. This is an FAO-targeting molecule. But we thought it would be interesting, one, in those patients, would they do better? And two, would we be effectively indifferent in an immune-suppressive environment? And so because of that dual mechanism of action. So HCC definitely being the poster child.
In our phase I, we didn't have many HCC patients, largely 'cause of the pandemic, right? We were in academic centers, and they retooled, and you didn't really see many of those patients. So we released some interesting data in RCC and in cholangio, in an oral presentation at ASCO two years ago. Also, to the far right of the PPAR alpha expression profile, so again, it's that one of the nice things right now in this program as we're moving towards a pivotal study, is that the hypotheses in the beginning are all bearing out with the data.
Right. Okay, so, yeah, I guess you've gone into a lot on the design and the advantages over beta-catenin inhibitors alone. So I guess let's move now to the phase II, because-
Yeah
... as you said, that's, that's obviously very exciting, the data that you put out last fall. So can you, can you start by describing the design of that study and why you chose the frontline setting?
Yeah, yeah, absolutely, happy to. So, as you know, Roche is the dominant first-line HCC player. I think, you know, depending on, I mean, well over 50% of the market is dominated by the standard of care, atezo- bev. Multiple people have tried to get into that market and haven't, you know, really taken significant market share from Roche. So this is relevant to how we ended up in this position, right? 'Cause we established a relationship, a clinical collaboration with Roche. They took 1120 or amezalpat, combined it with atezo- bev, and compared it to atezo- bev, in a randomized study, 40 versus 30 patients. 40 patients receiving the 1120 combo, 30 patients receiving the standard of care. And what that...
From a timing perspective, that study started while we were still in our phase one. So it actually leapfrogged the 1120 program forward years, and what's wonderful is Roche ran the study. And so when we talked about it, and they said: "There's an opportunity here for you to go," instead of having to start all the way in late line, which most of us do in oncology, to go all the way to first line, we jumped at that opportunity.
Yeah, I wanna ask a follow-up question, 'cause, you know, the fact that Roche is running this trial.
Mm-hmm.
So you ship 1120 to Roche to support the trial. So what can you tell us about the amount of 1120 that is currently shipped, with respect to the number of patients still in treatment?
So we haven't, we haven't released an update on the number of patients still in treatment. That's a great question, though. As of the October data update, I'm searching my memory here, about 40% of our patients were still on therapy. And then a larger percentage were still in survival follow-up. Actually, some of the data, one of the beautiful things about the way that this combination has been revealing itself, like, for lack of a better term, 'cause as you may remember, last year, there was the big data set that everybody went crazy around in October. But we actually had a very early look where we held a call in October, and we didn't have the primary data cut from the study. All we had, Roche did us a favour, actually. They looked, and we just had...
It was a terrible market, right? And we said: "Can we get an early look? We, you know, we wanna know what's happening with our lead program." And everything was directionally correct. All we got were some values, like early ORR, number of patients on study, but everything for the 1120 arm versus the control arm looked better. That all became even more apparent in the fall when we and although, again, we didn't have OS, but every category, whether it was ORR, PFS, the hazard ratio for OS, subpopulation data in PD-L1-negative patients or beta-catenin immune, everything was the 1120 arm was doing well and/or doing much better. And so in terms of drug, yeah, 40% of our patients still being on, I think it was versus, I can't remember the number of patients on study.
I think it was under 20% for the control arm. Again, all pointing towards the most important endpoint for first-line liver cancer, which is overall survival. That's the regulatory endpoint, and that's the one data point we didn't have in fall, for a good, good reason.
Mm-hmm. Okay, yeah. You know, I, I feel like for those who are new to this story, I think, I think a recapping of what you did show in, in the fall would, would be really helpful because those, those data are really important.
Absolutely. Yeah, no, happy to. So this was the primary data cut from the phase II, per protocol data cut. 40 patients in the 1120 triplet arm versus 30 patients in the standard of care arm. And what we presented, there's if you go on our website in the corporate deck, there's a summary slide in a tabular format that breaks this all down for you. But we showed a 30% confirmed ORR in the 1120 arm versus around 13% in the control arm, so significant, you know, improvement for those patients. We'd had reached a median PFS in both arms. It was seven months for the 1120 arm versus about four and a half months in the control arm. The atezo- bev standard of care control arm had reached its median OS.
It was around 15 months, and we hadn't. But importantly for us and for our statistician and for our, the team here, we had a hazard ratio in OS of 0.59, which was directionally very exciting for us, 'cause again, this, this triplet has been everything that has been directionally correct has borne out to having been correct, from the first data cut to this one. So we were excited about that. And then, because of that MOA that we talked about a few minutes ago, we were interested in both PD-L1 status, kinda, and immune, you know, desert status, and beta-catenin mutation. Our strategy, and this is important to repeat, is to look in the overall population. That 30% versus 13 was in the overall population, regardless of immune status and regardless of beta-catenin mutation.
So we think we're winning in that population, but we wanted to know about the subpopulations. So in the PD-L1-negative population, the ORR of the control arm dropped to 7%, and if Sam were here, he would say, our Chief Medical Officer, Sam, would say, "That's okay. Atezolizumab is an immunotherapy. It's not supposed to do as well in those patients." And what was exciting for us, going back to that dual mechanism of action, was that the 1120 arm was effectively unchanged at 27%. So that was. And there's a really nice, there's these two really nice waterfall slides in our deck that's publicly available, excuse me, where it shows. Well, it's a waterfall, so it shows all the tumor reductions, but on, it also matches those bars with the immune and beta-catenin status of the patients.
And so you can see there's just this, the word that comes to mind when you look at the 1120 arms, activity is indifferent, right? We do well in PD-L1 negative, PD-L1 positive, and we do well in beta-catenin wild type and beta-catenin mutants. In the beta-catenin mutant population, we only have this for, Roche only gave it this, this for the 1120 arm, but our disease control rate is 100%, and our ORR goes up from 30%-43%. So again, mechanistically dovetails wonderfully back to the original hypothesis of the 1120 program.
Okay, perfect. So big question, thinking ahead to MOS. What are you hoping to see? Like, what would be really good for you, in that data, considering it was not reached in the last update?
Right, so two things would be good for us. One, a strong maintenance of a strong hazard ratio. You know, we wanna look, when we look at the phase III, we wanna be confident that we're gonna hit our targets in the phase III, so that would be number one. From an absolute median OS perspective, anything above three, you know, three months or higher would be great. If you talk to our clinicians, they'll say three months is clinically meaningful to them. Especially, remember, liver cancer is horrible, right? I mean, the high watermark from the atezo-bev phase III was a 19-month median OS. There are multiple real-world studies that have that at 15 months. Notwithstanding all of the progress we've made in other histologies, I mean, there is significant room for patient benefit improvement here.
Aside from the commercial, I mean, it's just... there's just a massive quality of life and extension of life aspect to this.
Okay, fantastic. Okay, going back for just a moment, to the analysis from the fall. So, we have investors asking about any biomarker analyses that you've done. I'm interested to hear, especially, you know, considering that you presented some data at SITC Spring. So any insight from that that you wanna share?
I have to think about the data that we presented at SITC Spring. The two main biomarkers are beta-catenin and the PD-L1 population, but we're not... The strategy that we've discussed publicly moving forward would not be a biomarker-based strategy, right? We are not planning in the phase III to do a companion diagnostic looking at anything because we think that would shortchange the addressable population and the people who could get benefit from 1120 . It's not... Even though it looks like it's doing better in the beta-catenin patients and we are indifferent to PD-L1 status, that doesn't argue that we should subdivide the population if we think that we'll just broadly provide benefit.
The SITC Spring meeting, I think, was, in addition to some beautiful structures that were in there from our lead biochemist Valerie, there was interesting RCC data, I think, in that presentation, in the model, not in human beings, that showed benefit, 1120 benefit in RCC models that correlated with CD8-positive T-cell infiltration. Now, translating that into the clinic, in our phase I presentation, in the oral presentation at ASCO a couple of years ago, we had interesting, really interesting RCC data. And so when we get asked by investors and also KOLs, you know, "What's the next place you guys would go?" It's always RCC and cholangio. Those are always right, right next.
Okay, so, next, let's move to pathway forward. So, you know, obviously, there's the atezo plus bev combo, but curious if you would consider maybe other PD-1s and/or other, kinase inhibitors, and just what are your thoughts on, on the direction you plan to take, with these combinations?
Right. So the easiest path forward in HCC would be to do exactly what we just did in the phase II and expand it. Even for the detractors of atezo, who think, you know, you should go with a PD-1, and you bring up the ADA question and so forth and so on, the data are the data, right? And so one of the things that we believe as a management team is let's not make the same mistakes where the growth stage biotech has good news and then changes a bunch of variables when they go into their phase III. Right now, our base case plan is to take what we had in the phase II and expand it to just the right statistically significant number size to find that clinically meaningful benefit in an approval study. And there's another aspect to that.
In addition to just probability of technical success, there's a regulatory smoothness of that path, right? When you think about going globally and talking to regulators and saying, "I'm adding to standard of care, compared to standard of care," it's just a smooth, easy path. You don't get blowback. And our KOL conversations to date have been nothing but positive, right? They love the idea. And I think if, unless, I mean, I'm pretty comfortable saying that aside from the Roche, the ongoing Roche TIGIT triplet versus their standard of care, we would be the other phase III globally against standard of care.
... Yeah, absolutely. So do you, do you have any more updated thoughts on, on when you might be able to start a pivotal trial?
Yeah, so we're still on course. Now, this goes back to, and I don't mean this apologetically 'cause it's great, we jumped from phase I data to randomized phase II data in a really short time period because of the fact that we started this collaboration with Roche at the same time we were in the phase I. So we have been doing catch-up and, you know, CMC work, everything, you know, moving from capsules to tablets, and all the things you do when you, if you had been in a longer situation. So we have guided to end of the year/first quarter, and that's still absolutely the plan. And so when you backtrack from that, and you do the math, our plan is to talk to FDA this summer.
Okay, and so ahead of those conversations, do you have an idea in mind of what you want the design to look like? You know, if you haven't talked about that with the FDA yet, what would you consider the ideal design?
Yeah. It's to take this phase II, expand it to the—what we haven't talked about are actual patient numbers, but expand it to the right size to, and based on the hazard ratio that we're seeing back then and what we'll be, you know, talking about in terms of an updated number when we finally discuss OS. And then, no companion diagnostic, no. You know, we, we're gonna look at the overall population just like we did in the phase II. It's literally expanded. And then, you know, we have discussed internally and externally doing an early efficacy read, right? With obviously there'll be a futility analysis. I mean, that's a given. I think that's just diligent. But then the question is: Will the FDA agree on an early efficacy read?
Because notwithstanding the real-world data that I mentioned earlier, we're gonna power the phase III to the control arm's, you know, high-water mark, right? And so it is worthwhile, sorry, depending on what your hazard ratio is and how you think things could go in the control arm to have an earlier look. And so once we have FDA feedback, we'll give more details on that. But that would cut down the time, right?
Yeah. Mm-hmm.
And now to go to the dark side and talk about commercial, I mean, when you think about the size of this market, this phase III is neither as large as a lot of large indication phase IIIs, nor would it be the, you know, ridiculously long. We're talking about three years, right, without early reads.
Mm-hmm.
It's just not, it's just not that long.
Yeah. Yeah, exactly.
If the TIGIT phase III that Roche is running is positive, it's gonna be replacing their existing standard of care, so it can, from a competitive dynamic, right? You're not actually changing the competitive dynamic of the landscape, which is interesting.
For sure. Yeah, so I guess that was my next question on the market. But I guess aside from HCC, would you consider a pathway in renal or prostate or cholangio?
Yes. Cholangio is tough, only just from an allocation of scarce resources perspective. If we could do maybe something, you know, one of our investigators is at Hopkins, is a hepatobiliary cancer lead there, right? And so we've discussed with him cholangio often. Those, talk about a horrible prognosis, right? It would be great if we could do cholangio. It's just a really small market. RCC, definitely with capital. In fact, we've discussed: Could we replicate what we just did, right? Or there's an argument that RCC is moving to first line, is moving to a triplet. Why not do what we just did again? Why just start late because that's what we've always done?
Yeah, that's great. Okay, I told you that we would find time for this, so how about let's move on to 1495? Can you tell us a little more details about the molecule, the pathway, and the status of where you are now?
So like amezalpat/1120, it's a small molecule. The two publicly known programs are small molecules. The idea here was to be highly specific for two targets in the prostaglandin pathway, EP2 and EP4. So this was an elegant bit of chemistry because the main receptors, EP1, EP2, EP3, and EP4, are homologous. And so we designed this to be, specific for EP2 and EP4, for the main reason, that they are tumor-promoting, and EP1 and EP3 are, you know, are friendly in terms of immune system anti-tumor effect. So the hypothesis there was hit those two, and then you will see a anti-tumor effect. This is, I would say, a next-gen version of an old idea. If you talk to anybody who's been around oncology for a long time, the prostaglandin pathway, a la NSAIDs, ibuprofen, and everything, has been of interest in cancer for years.
There just hasn't... You remember Celebrex? You know, there hasn't been success with molecules who have come out in this space. Either the therapeutic index wasn't there, or the efficacy wasn't there. This program has evolved. We presented phase I data at ASCO. We were very interested in prostate here because of the implication of the pathway, and the evolution has actually gone towards FAP, which is a hereditary indication that's almost like it's cancer-adjacent in that it's pre-colorectal cancer. It's related to it. You get hundreds to thousands of polyps, and it appears in young people. And so what's happened here, and we've publicly announced this, is that we've been approached by, consortium is a strong word, a collection of FAP physicians, 'cause there is no therapeutic option right now in FAP. It's surgical only, right?
That's dramatic for young people. So the idea here is they said, "Could we take your 1495 molecule and do a phase II in FAP, and we will fund it?" Where we are right now is we have approval for that study. We expect it to start this year, and we're just waiting, 'cause it's the government, right? So we're just waiting for formal approval after you have approval, but we expect FPI on that in the second half of the year.
Great. Okay, so I know, I know we're up on time, but I really do want to get the chance to ask one more question. If you could talk about IP for each program and where you stand there?
Yeah, so the earliest expiration for amezalpat is 2033, but if you were to talk to your lawyers, they would say it's more like 2036, 2038, with patent term extension. One of the beautiful things that just happened for that program is that, China just enacted a patent term extension law that mirrors the EU and the U.S.'s. And so you're talking about the same effective extension, 2036 to 2038 in China. Why is that beautiful? Because it's an enormous HCC market. I mean, it's a... And Roche, you know, has published a number. I think they treated over 200,000 patients first line in China, in their 2023 Pharma Day. So it's an enormous market, and of course, you can ask me about 1495. I think it's 2043-ish.
Right.
When that would, the term extension would be in the late 2040s.
Wonderful. Okay, we are a minute over time. I don't wanna keep folks. You know, I could talk about this all day, but thank you so much, Steve, for your time and this great discussion. Appreciate it.
Thank you, Tara. Appreciate it.