UBS Global Biopharma Conference - OLEMA PHARMACEUTICALS INC
Thank you for joining us on our next session. I'm Michael Yee, Global Head of Biotech Research at UBS. Very happy to have on the stage with me the President and Chief Executive Officer of Alema Pharmaceuticals, Sean Bowen. Thank you, Michael. Thank you. Thank you so much for being here, Sean. It's a pleasure to see you. It's great to see him. Thank you very much. Thank you all for hosting us at this beautiful venue. Absolutely. So talking about beautiful things, you guys have done some beautiful execution because you're executing across your Phase III studies, put up a bunch of data over the last year we've watched a lot of data you've got cat six talk about that too uh and obviously you've Got some phase three data coming up uh in the first part of 2027 so maybe we could just take a step back and you could offer us um and talk about and offer some opinions about the phase three opera one study for palisestrin and second line metastatic breast cancer. It's your first phase three study and a billion dollar indication. I think it should work, but it's to the degree of how good does it work? That's exactly right. So that investors can understand and have a better flavor for your drug versus the other surge or syringe or however you want to say it. So talk to us about this phase three result and what you think is going to happen. Yeah. So thank you again for the opportunity to talk about Olima. O2, O1 is our first phase three, as you've said, we have a second one in the first-line setting, APRO-2 ongoing and rolling well. APRO1 will read out in Q1 of next year. It is in the second, third-line setting. It's a monotherapy palazestrin versus fulvestrin or eczemestane, again, as a monotherapy with a progression-free survival primary endpoint. One of the interesting things about that particular indication is that it is really two populations, molecularly defined. So, about 40% of the patients, the tumors harbor activating mutation in the ESR1 gene, the gene that encodes the estrogen receptor, which is the target of palazestrin. The remaining patients do not have that mutation, but all of these patients will have progressed on their CDK4-6 plus AI therapy. So they have demonstrated endocrine resistance. And if we go to the others in the CERD class, what has been shown is there have been a few that have demonstrated activity in that ESR1 mutant subset. So clearly estrogen receptor driven, you turn off this activated receptor and you get some benefit in terms of controlling the disease. Much more difficult has been to get activity in the ESR1 wild-type subset, where there have been efforts, but they have not succeeded. So if we go back to our Phase II data, we have compelling data in the SR1 mutant subset. We had, in a fairly similar population, 7.3 months of medium progression-free survival. You expect 2 to 3 in the control arm, so that would be a significant prolongation if you were able to duplicate that result in phase 3. In the wild-type setting, we have 5.5 months, which, again, if you Were able to Duplicate, would be significant and would be unique. So we see two ways in APRA-01 to differentiate. One is success in the mutant, and perhaps with better prolongation, or delay of progression, but prolongation of PFS. And then to have success that shows some benefit in the wild type would be unique, and we'd be there alone. The mutant is a $2 billion a year market opportunity. The wild type is about a $3 billion ayear market opportunity, so these are significant on their own, but together would be really quite significant. So let's unpack that a little bit. So in this phase three result, where you said approximately 40% would be mutant. That's right. And 60% wild type. We should start with the fact that other FDA-approved drugs, including Osirdu and Lilly, I think, have modest benefits but statistical significant two-month benefits on ESR mutants. Those drugs are approved in that subgroup, and it shouldn't be shocking that this drug also should show these types of results, because actually you think you're best in class, you're more potent, you're a more complete blocker, and all your phase 1-2 data has been better than theirs as well. So I hope that, and would have high confidence that your drug will be at least two months in the mutant population. You agree? It should be at the least two months. Definitely have higher probability of success in the mutants. that would be awesome and by the way even if it was three months i mean you said 8.5 versus two to three was that the number i quoted uh 7.7.3 versus two and a half i mean if it was three weeks three months that's already better than lily so maybe you get hundreds of millions from that uh there's four months my goodness sean that wouldbe pretty easy yeah and that's great so we could have a best in class esr1 mutant now where there could be significant upside as a read-through before we start getting into the first lines, which we'll also talk about, is the fact that you might have a statistically significant, clinically meaningful benefit in the wild type, because the others have failed. They show zero. That is correct. Now, just because you've shown five and a half versus two to three cross-traumatic comparison, but five and a half for sure in your phase one what is your study power to show what do you think it could show what is meaningful yeah because they're being evaluated independently other studies they did intend to treat and they kind of look at it yours is kind of interesting and a positive sean because you're doing esr1 mutants and then you're independently looking separately that is statistical significance as a primary endpoint for the mutants. So let's do two things. I'll talk a little bit about that last point. The others weren't necessarily power for that. That's my point. That is right. Talk about what we are testing. And then I will also talk about your question about what is significant. What would you be able to detect? Three months. So recognizing that this is two different populations and obviously the outcomes have been different in these populations. The way that we designed the trial going into the trial as a patient or investigator it feels like one second third line trial progressed on your CDK4-6 plus AI you are randomized you either get palzesterone one pill a day you get fulvestrin or exomestane we follow for PFS now in that process we test you for your ESR1 mutational status and we stratify so that they're evenly distributed when you go and look at the analysis plan it almost looks like two trials And it almost looks like a trial to evaluate ESR1 mutant and a trial to evaluate ESR2 wild type So the net effect is that there are three potentially positive outcomes to the trial Obviously if you hit in mutant and wild type that's not hard to figure out If you hit it in mutant only you don't hit in wild type that's a positive trial but obviously you've only hit in that one subset Biologically I don't know how this would make sense but statistically you could hit in wild type and not hit in mutant, that would also be a positive trial. So we really are independently testing these populations. Can I ask a question on that? So I agree that that seems unlikely on the last scenario, but the reason I say that is because have you said, clarified, or have you said whether these are each co-primary endpoints. Is it hierarchy? When you run a study, there's a primary endpoint. And when you hit the primary endpoint, it's testing a specific test. In this case, intent to treat? Probably not. It could be ESR1 mutants first. If you hit that, you get to test the second one. You haven't said if you're splitting alpha. So if you split the alpha, you get to test both. If either one is P.025, it's a win. If it's a hierarchy, you get to P.05, and then you get use all P.5 and recycle that for the second one. We have not disclosed the analysis plan. There are other ways to do it than the two very valid ways you've described, but we do test them independently. You meet the critical value with either, and that's a positive trial. Both is obvious, right? You're in a good space. To the second, actually, it was the first part of your question, which is what are you able to detect? Can I ask that one more time? Do you split the alpha, or do you get to test at least the wild type with P.05? Can I asked that? Do you get the full P.0.5? We get to we get to them both. OK. And we have a critical value for both. The important part is that we are powered to detect any clinically meaningful benefit in either population. Okay. So it goes a little bit to your first question, which is what could you see and be positive? If there is a clinically meaningful beneficence in either population, that population will be statistically positive. So what is clinically meaningful? So it might be a little different in the two populations, And part of that has to do with the existing standard of care, as you said. There are drugs that have achieved about two months, you know, imlunestrin 1.67, a little less, but in the range of two months in the ESR1 mutant population. Those are approved. So obviously, to some extent, that's the bar. You don't want to be worse than that. Just to be clear, like the control arm, literally the tumor is enlarging, it's progressing in two months. That's correct. By adding another month and a half or two, you're literally increasing by 50% or 100%. You could be doubling. In fact, that's meaningful for patients who are on last line. It is meaningful for patience, and I think people have to understand as well, you're really comparing most of the control arm is fulvestrant. So what you're doing is you're comparing two large-volume intramuscular injections every 28 days to a pill you give once a day. So in addition to the delay of progression, which is meaningful, you also have a quality-of-life benefit of not having to get these injections. Intramuscular, I think it's in the buttocks, by the way. Yeah, on one each side. This is meaningful, and therefore we can take one daily pill rather than two intramusular buttock injections and double the PFS and have a better side effect. That's a win. Yeah, so the other question you asked was in the wild type. The wild type has obviously not been addressed at all. So we would see benefit, you know, there isn't an external bar yet. It's got to be significant enough that, you know, first of all, it's statistically significant. But, you know, it has to be some prolongation of PFS. Is PFS for these patients basically the same two to three? Trials have varied a little bit, but for most of the trials, it looks like they're about the same with the control arm. Obviously, they weren't the same in the treatment arm because the other drugs prolonged for the mutant, but didn't for the wild. Yeah, that may or may not. It's not like it's a negative prognostic indicator. No, no, no. That is definitely. Except there's a mutation on the receptor. If you have a drug that binds the receptor and conforms, then you will have a benefit. But that's right. It's a drug. It's, it's not a negative. Prognostic. That is, that is. The reason why it doesn't work in mutants is because there's your mutation in the, in the, and the fulvestrin is not binding. That is exactly right. In the prior trials with the other agents, the diagnostic has not been prognostic, meaning it tells you whether your disease is better or worse in general. It has been predictive of whether you use it. So we would expect two to three in the wild type. Yeah. So what do we think is going to happen with your drug? Two to three months better? Two months? So like I said, the question here is the classic oncology question, does phase three replicate phase two? What we saw in phase two, the second and third line setting, was 5.5 months in the ESR1 wild type. So obviously, if that replicates, we think we have a pretty decent chance of saying, hey, this is— 5. 5 minus 2.5 is 3. Yeah. That would be a massive home run. That would Be huge. Huge. That would Now, the issue is that on one side, we have this phase 1, 2 to phase 3. Yep. So whenever you go from, like, help me with the end, 20 or 30 patients, more? Yeah, a couple dozen. Okay, then you go to hundreds. Yes, that's correct. Then we could have a change. Rarely in biopharma does it get better. It only comes down a little bit. But if we go from three and a half months to two and a half, that is great. Now, if I may say a positive, one side is that in those patients that you tested, correct me if I'm wrong, but they were worse, more beat up patients, or are they pretty much the same patients? So there are a couple of differences that suggest that the phase two patients would have been more heavily pretreated. Okay. So one is that they were allowed to have prior chemotherapy in the advanced for metastatic setting, and a significant percentage did. In the OPERA-01 trial, prior chemotherapy and advanced for medistatic disease is excluded. How many patients had prior chemo in the phase one, two? 30%, I think. So 30% had an additional line, which includes chemo. Yeah. Here, they're not getting chemo, It matters. If you've got chemo, that's more pre-treated. Yeah, and they tend to be prognostically worse. The second is that in the context of the upper O1 trial, your prior therapy, you had to be able to stay on it without progression for at least six months. So you couldn't rapidly progress on your prior therapy. We didn't set any time limit as to progression on prior therapy in the phase two context. So those are really the two differences that would be a positive yes because if you're getting patients you progress within six months they're clearly bad patients who are very sick and they're not going to do well on your next drug yes they're good patients how many patients had not gone to six months in the phase one two so i don't even know that i don'T know the answer to that handful yeah probably not the majority, but... So that would be another group that means that the phase 1-2 is a little bit sicker. Okay, so you could have a better result. Yeah. Now, of course, when you do these, when you change these criteria, you change them for both arms of the trial, just to point that out. It's everybody who goes on the trial. My point is that 8.5, you know, again, we're finding reasons why the 8.1... 7.3. 7.5. Sorry. Could be... I like 8. I do have to say what we actually got. We're going to check the transcript. Mike says 8.5. We're gonna see 8. 5. I like your version much better, but I do want people to know. But the point was is that the 7.5 could be under-representing because they were slightly sicker. That said, we're going from 25 patients to a couple hundred. US and Australia to across the globe. Yeah, there are differences. Any other reasons how to think about this result? Because this data is coming in Q1. it's a big readout and the reason I say this is because if this drug works in ESR1 wild type it's the only drug to ever hit this it's 60% of the population it's multi-billion dollar situation and no one has this, not Lilly, not Osiru not Astra, nobody okay I'll see you in Q1 for this it's big deal it's for sure and again the read through Now we're going to get to the meat of this in the last half of our discussion, is because if it works in wild type, I will believe, like the lung cancer drug Tigriso, that by blocking the receptor before the mutation happens, we will not develop the mutation. That's why working in the wild type would work, because you're wild type in first line. Follow me? Yeah, so a couple things there. This is important. Yeah. It's a read-through. So it is, but I think the point is that I think it is important to separate these populations out a little bit. So if we look at the OPERA-01 trial, we have the mutant and the wild type, but we have one other property that's different than OPERa-02. Okay. In OPERа-01, the patients are endocrine-resistant. Okay. They have progressed on an endocrinotherapy. That's right. And an endocrinotherapy plus a CDK-46, so effective therapy. in opro 2 they're endocrine sensitive okay so endocrin sensitive endocrino resistant doesn't necessarily predict what happens in endocrines sensitive and let me i've been saying this for a while but now we have evidence okay let me give you the evidence if you look at the roast drug, duragestrin, they did a trial called Axolera. It was kind of like Opera 1. That trial failed. Hint of activity in the mutant, didn't see activity in the wild type, didn' t lead to registration. Oh yeah, I know where you're getting that. But it worked in adjuvant? Later on, they did a very large trial called Ladera in the endocrine sensitive wild type population, the adjuvant setting, as you said, was very clearly positive. So very clearly, that drug was unable to make a difference. What's the reason for that? Why did Roche not have any benefit there, but it works in Lidera? Well, the reasoning is that when you progressed on the CDK46 plus AI, you somehow or another accumulated mutations that developed resistance to that therapy. The most common of these we know. It's the ESR1 activating mutation. But in the wild type population, there are other mutations clearly accumulating, and the drug isn't good enough to overcome those in its targeting of the estrogen receptor. We'll see if ours is or not. So when you progress on a CDK4-6, you're going to develop other mutations in the endocrine receptor that are, well, outside even, and even outside, right, and therefore that when giving a SIRT or Saran, you're not going to have any effect, because there's other mutations ongoing, well, with endocrine sensitive earlier, that's different, it's definitely more sensitive, yeah, right, now we know still, even in that endocrin resistant setting, the receptor is doing something, right, the reason we know that is because what you often give there is another targeted therapy, the mTOR, everolimus, you know, if you have an ALK mutation, you may give capivacertib, PI3 kinase, alpalisib. The thing is, those drugs don't do anything without an endocrine therapy. So very clearly, you have to give them both in that setting. The question for us in APRA-01 is, can we get enough activity just by targeting the estrogen receptor in the wild type to get a benefit? APRA2 is different. That's the endocrin-sensitive setting, right? So in that setting we know the estrogen receptor is still working in all of these people and can you prolong freedom from progression by turning off the wild type receptor and as you said suppressing the growth of so in one scenario where esr1 wild type works boom right away you have a multi-billion dollar drug right away everyone can say it's a better surge saran than everybody else that's because it even works on wild type, and no one has done that. You immediately go to the top of the food chain in terms of which one looks most promising. However, if it doesn't work in the wild type in Q1, we will say, you will say. The market may say, okay, but giridestrin from Roche also did not work there, and yet is, what is Roche saying now? $5 billion plus drug and adjuvant? They put a lot on it, yeah. And it's probably accurate. Okay, and they did fail in first line, but they did work in adjuvant. Well, they had evidence of activity in the first line. They failed narrowly. The hazard ratio, 0.9, 0,89? 0.86, 08, 07. 0.87, okay. So it was there, barely missed, so you can learn from that, and it works in adjovant. Right. So that's the setup in terms of, A, there is still lots of potential because giridestrin was a mod, well, did not work in the SO1 wild type, but it just barely missed in front line, and it works in adjuvant. AstraZeneca, excuse me, Lilly is now running a study. Let's get to this one. And this is important because it's another early. The adjuvants. Yes. Yes, and what does that, what does that mean, if anything? Is that a surprise that that would work or not work? Do we expect that's going to work? Well, again, remember, the difference, if we were, if it were to look at imlunestrant, and we were looking at geradestrant in the second-third line, imlutestrant did work in the mutant setting, didn't work in the wild-type setting, whereas XLR didn't work in either. So I don't think it would be too much to think that, gosh, they should have activity in the endocrine-sensitive wild-type setting. Now, the trial design, the trials called EMBER4, the trail design is a bit different than LIDERA. LIDera is a unique design. It was appropriate at the time it was done. It's single agent aromatase inhibitor versus single agent diridestrin. But that's not the standard of care now. So Amber 4 anticipates that a little bit. There's CDK4-6. Because while Ladera was being run, the standard of care switched. And that switched to for the moderate to high risk adjuvant patient to give a period of CDK-4, 4-6. You could either give Fresenio, Abamaciclib, the Lilly drug, or Ribociclib. So are doctors going to use that on top of a CDK 4-5 or they're going to give Girodestra without the CDK4-6? To be perfectly honest, I think they're struggling a little bit. And at the presentation of the data, the LADERA data at the San Antonio Breast Cancer Symposium last year, You heard the discussant struggle a little bit with it. Do we give CDK4-6 plus AI, then switch from AI to geronics? What do we do with this? That isn't what was tested, but they clearly, and this makes sense, they don't want to give up the survival benefit of the CDK-4-7. Right, so we've got to use CDK 4-6. But they also believe, and the data really suggests, that in the monotherapy setting at least, geronic is better than the AI. And so I think they are struggling a little bit. Ember 4, if positive, helps them a bit with that because it did incorporate the CDK4. If Ember4 is positive, by the way, I cover Lilly, we think it can be positive. Is that not bad for Ghirardesher? Because then they're going to go right to the CDKS4-6 SIRD. Why would they use that? Let me put it this way. I do think what it does is it gives these treating oncologists a clearer path to using the CDK4-6 that they want to use because of the survival benefit. And by the way, both Kiskali and Verzenio showed survival in this setting. So the CDDK4.6 helps. They might even switch up the CDG4-5. At least it's in common. They do kind of do their thing at some point. But I do think it's informed. I'll only say that you know, Roche then announced a plan to do a CDK4-6 adjuvant trial as well. I think that makes perfect sense. The only challenge with that is they take a really long time. Years. By the way, that's not why, that is why Aleema is not running that study right now. Not right now, I mean, I think, yeah, it's very expensive, it is long. These studies started years ago from Big Pharma Company. That is exactly right, and I think for us, with a very proximal readout with APRA-01, we feel like it's better to have that readout and allow that to inform our future strategy. I will say, I think it does make sense to consider an adjuvant trial for palisastrin. Yes, it does. And I say that because, obviously, with your result, if you have better data in ESR1 mutants, if you Have positive data in wild type, I think any company could say the probability of success of that product in earlier lines It's higher than the other companies because they had weaker data. Therefore, since breast cancer is basically use the best drug and the other drugs eventually go generic, but if your drug is much better, it will be used. Therefore, there is certainly lots of timeline. If your drug's not going, it's not gonna go generic anytime soon. That's 2040 or later, yeah. Okay, so then if these readouts are coming and I think that these are more adjuvant situations, yeah, right, we talked those are adjuvant. Right. But the point was that it doesn't mean that SIRDs are no good in earlier lines if they don't work them out. That was the conclusion. We've seen that now. Now, ultimately for you, this is also a read-through in OPERA 1 because of OPERO 2. And this is the phase 3 that you're running in first-line metastatic breast cancer on top of a CDK416. Kizkala. The CDK46 of preference. Yes. And Novartis is contributing the drug. That's right. pay for that part of the drug so that they can test their CDK4-6 with yours. Two reasons. One is, hey, they're doing that to support broader use of that, but importantly, if your drug works, they want their CDA4-5 to be used with the new standard of care. I believe so. And it's safe and well-tolerated together. The phase two data really didn't show any difference in the tolerability between what you'd expect with an AI with Kizkali. Right. So then where are you with your first line study wall street is pretty bearish on first line these days yes because of the failure of the failures of everybody else where are you with this study and what are the timelines for this study yeah so where we are is it is it enrolling very well i mean obviously right now kiskali is the global standard of care it should be in the first line it's not hard to enroll it's because because patients the control arm kiskalli plus ai is the gold standard therapy and then you get the opportunity potentially to have palzestrin depending on where you're randomized. So it's enrolling very well. The one thing that we're going to look at very carefully in early December when we see the Serena 4 data from AstraZeneca is how does that control arm perform? Because there is a very well-established history in oncology that a given drug or regimen often is more effective over time. What is that? Mostly it's that oncologists are really good at managing toxicity. And over time, with that learning curve, they maintain dose intensity and they discontinue drugs more often, and that makes them more effective. So we want a more modern estimate of what the control arm is. If it's longer, we saw a little hint in Persevera, but we want this more comparable data set in Serena 4, If it's longer than what we had predicted based on historical, we might have to modify our trial size. That's right. We have time. We have times. All right. Near term, near term, San Antonio breast cancer cancer. Yes. You have a CAT6. We do. Pfizer, which I cover, has a decent CAT6, it's in phase three, it shows 37% response rates in combination with fulvestrant in second line plus metastatic breast cancer. Those were second line. And they moved it to a phase three. Yes. You know, Pfizer is reading out their phase three with their novel cat six billion dollar blockbuster drug in second half 27. Yep. So next year, Pfizer may have a positive cat six, but you have a cat six. Is it going to look better than the Pfizer data at San Antonio with a bar of 37 percent? Well, first of all, the 37 percent was only second line patients. OK, this is phase one B data. So you get more heavily pretreated patients. You have sicker patients. Sicker patients, okay. I also say that the responses improve over time with the CAT6s. So their initial presentation of that data was 30% overall response rate. Okay, 30. Following the same patients for a longer time, it went up. Got it. The way you said. And I think that that's a property of these mechanisms. Look, the Pfizer data, it validates this target. So you'll see the initial fulvestrant. You'll seethe combinability. You'll seethe tolerability. Look at the cytopenias. You'llsee the early evidence of activity in the more heavily treated patients. Ongoing is the palazestrant combination. Preclinically, palazetrant really seemed to potentiate the activity of CAT6, more so than fulvescent. Actually, it didn't matter which CAT6 we used. Right. So we're testing that. That data will be... Yeah, go ahead. The mid-year next year. Okay, mid- year next year, like phase one data. Phase one B, yeah. But the point is that the combination of a Saran plus a Cat6 is synergistic. Pre-clinically increased activity. Just to be clear, so 30% was the initial Cat6 Pfizer data in second line patients. You have slightly sicker patients. So it's 30% a bar? I mean, you'd want to be in the ballpark, I would say. Ballpark is good. Thank you very much, Sean. I appreciate it. Looking forward to San Antonio and all the results in Q1. Thank you, Mike. Thank you. Thank you so much.
- Company
- OLEMA PHARMACEUTICALS INC
- Keywords
- OLEMA PHARMACEUTICALS INC