Olema Pharmaceuticals, Inc. (OLMA) Earnings Call Transcript & Summary

You, you. Thank you. Thank you. We're going to go to this next gate. Well, the plane that was at the other gate was unloading. Yes.

just sat there while they unloaded the planes. It's ridiculous.

Alright. Well, thank you all for coming today and coming to the Morgan Stanley Healthcare Conference. My name is Ruck Byun. I'm head of West Coast Healthcare Investment Banking for Morgan Stanley. I have the pleasure of hosting Sean Bohen, CEO of Olema. I think as we kind of think about, kind of just to jump right in, the market keeps lumping every oral SERD together, but you know, you have to, you know, forecast a molecule by its own data and from standpoint, palbociclib doesn't look like everyone else's SERD. Why is treating this as one monolithic class the core mistake that investors are making?

Yes, so thank you, first of all, for the opportunity and the people here for listening to the Olema story. I think that there are a number of reasons why lumping these molecules together is... is a mistake, I mean, beyond an oversimplification. First of all, mechanistically, there are key differences. So fulvestrant is a complete estrogen receptor antagonist. Binds, shuts off the estrogen receptor signal growth and proliferation signal completely. That's true in the context of the wild-type receptor, where it competes for estrogen, but also the ESR1 mutated receptor, which is turned on without estrogen needed. Some molecules are SERMs. They're agonists in some contexts, and you do not want to turn on the estrogen receptor in these cancers. In other cases, it's really got to do with exposure and pharmacology. You need to have a high level of drug to force receptor binding and keep the receptor off all the time because estrogen receptor is a transcription factor. So its signal is highly amplified, and we have by far the best exposure within the class. And then finally, combinability becomes very important. I mean, our first Phase 3 to read out in Q1, OPERA-01, is a monotherapy trial, but the larger opportunity with this class of drugs is in combination with other targeted agents. And we have found, uniquely, that we are able to combine with multiple different modalities without having to compromise our dose and without enhancement of toxicity. Why do investors do this then? Why do they lump it all together? Well, I think the simple answer is it's complex. But if you decide to do that, then you decide to ignore some very important data that can distinguish the molecules.

Got it. Thank you. Maybe just since it's topical, um, on the SERENA-4. It seems like, from my perspective, the first line misses are about exposure and dose, not necessarily mechanism. camizestrant carries, as you said, the lowest exposure of all the oral SERDs because of its tolerability ceiling. Yes. Isn't that the very problem palazestrant designed out from the very beginning?

It is.

It's one of the many. So when the founders of palazestrant really took this on a little less than 10 years ago, other companies were working on this. And by the way, they were addressing as well that they had also — to the concept that you want a complete antagonist. camizestrant molecularly is a complete antagonist. Giredestrant, the Roche molecule, is a complete antagonist. So they had made that observation as we did at Olema. But in looking at those molecules, it was seen that there were these liabilities, that they weren't, they didn't have high enough exposure. They had tolerability problems, and then those tolerability problems are exacerbated in combination. And so the objective in designing palazestrant was, let's address all these things. We have successfully achieved that, that's all demonstrated, but one of the obvious conclusions of wanting to address those things is also saying, hey look, these liabilities are going to turn into limitations when you test these things for efficacy. Now, we don't know what the data is from SERENA-4. All we have is this cryptic, numerically different data for PFS. We do know what the data is from PERSEVERANCE, from the Roche compound in its first line Phase 3. And that data clearly indicates that there is better activity of an oral SERD than an aromatase inhibitor. They had a 5-month delta-PFS. They had a clear trend for hazard ratio, but didn't quite make statistical significance. And that in spite of some design flaws. And so I think those molecules demonstrate two things. One, we're leaving efficacy on the table with AIs. We can improve upon it. And two, these liabilities in terms of exposure and combinability are turning into liabilities when you look at achieving a clinically meaningful outcome.

Thank you. Since you mentioned it, PERSEVERANCE, what kind of folks may flippantly call a negative trial, still delivered about a 5-month PFS Delta, while they were carrying about 10% endocrine resistant patients which diluted some of the treatment effect and efficacy signals. In terms of your own trial, OPERA-02, can you tell us about how you've thought about patient selection exclusion and also your decision to run a ribociclib backbone? I guess as we're looking at what happened with PERSEVERANCE and even SERENA-4, a competitor miss may actually potentially raise your confidence given your trial design and some of these molecule differentiations, but would love to hear your perspectives.

Yeah, so there are multiple questions in there, which are all good ones. So the first one has to do with the patient-select. I must say, OPERA-02 is not the unique trial in terms of how we've defined our patient population. That is to say, the endocrine-sensitive patient population. What that looks like in these first-line trials is for the patients who have had prior adjuvant therapy. In order to be eligible for OPERA-02, you have to have completed your adjuvant therapy and had at least a 1-year of treatment-free interval without recurrence of disease. And that's the clinical definition of endocrine sensitive. And so that's what we do in OPERA-02. By the way, that's exactly what was done in SERENA-4 as well. AstraZeneca used the same criteria. The unique thing was that Roche decided, for some reason, to include patients who had progressed within that year of completion of adjuvant. As you said, it ended up being about 10%, but that was after they stopped that in their First Amendment. So it must have been much higher at the beginning of the trial. They did very poorly, right? So this is a trial where, yeah, 28 months in the control arm, letrozole plus palbociclib, 33 months, that's the 5 months you were talking about, median PFS, giredestrant plus palbociclib. If you go to that subset of patients that had that shorter treatment-free interval, the control arm wasn't 28, it was 19. The giredestrant arm wasn't 33, it was 14. They did very, very poorly. And so this actually not only diluted the treatment effect, it actually fought the treatment effect in terms of the hazard ratio. So this is a really... I would just say mistake to include those patients. SERENA-4 didn't do that. We haven't seen the data. I think SERENA-4 is very well designed. The challenge with SERENA-4 is camizestrant is not... it's got the lowest exposure. I think it's got limitations as a drug. Both of those trials, as you mentioned, use palbociclib as their CDK4/6. Now, that was not a mistake. At the time those trials were designed, Ibrance dominated the first-line market and that continued to be the case until Overall Survival started to read out from the CDK4/6 inhibitor pivotal trials and what happened there is the standard of care got flipped on its head. The third most used drug was Kisqali, ribociclib, but it had three out of three trials positive for Overall Survival. The delta, the improvement in Overall Survival, is about a 1-year. That's significant obviously for these patients. And so oncologists did what oncologists do. It's very simple. I am one. It's not being pejorative. I'm a doctor. They just follow data. And so when they had a survival benefit, they said, okay, this is it. We're switching our new patients onto this. We designed OPERA-02 with that knowledge in hand. We actually had done the palbociclib combo first. We thought that was going to be an Ibrance trial, but we were able to capture that change in standard of care.

Thank you. Now, giredestrant, going back to that, my understanding is that they cut from 100 mg to 30 mg such that they can manage bradycardia with palbociclib. You have an 8-day half-life and high exposure. And so I guess... as others may have to make some trade-offs on the exposure versus tolerability equation, you think based on your PK, looks like you have a, what seems like a durable moat. Can you elaborate on kind of that a little bit more?

Yeah, so it turns out our half-life, and we'll publish on this here, our half-life is even longer, that's about 14 days. And so our steady state exposure is even higher than we have published. And that 14-day half-life really allows us with daily, once daily dosing to achieve this very high exposure. Again, the objective with this treatment modality is to shut this signal off completely. And the way you do that is by forcing the binding of palazestrant onto the receptor. We had predefined the exposure threshold we felt we needed, right? We weren't first. We saw liabilities in the others and we thought, okay, this isn't worth doing these Phase 3 trials if you can't address those liabilities. We're able to easily cover 24/7 exposure threshold. Now, the giredestrant had pretty nice exposure, not as high as ours at 100 milligrams. And that was their original recommended Phase 2 dose. The problem was they combined with palbociclib and they found, as you said, they had this with this, this increase in rate of bradycardia. And they decided that that was not acceptable and so by mitigating their, their dose going down to 30, a threefold decrease in exposure, as you might expect, that they did actually control the bradycardia problem. We think, however, they sacrificed their receptor occupancy. And that is the kind of flaw that we were seeking to address.

Mm-hmm. Yep. I mean, even a great degrader leaves plenty of receptor behind. What controls the tumor is really keeping the residual receptor completely silent and believe that's what you set out to do with designing and developing palazestrant and so now does that reframe the next-gen SERD debate away from the degradation depths?

It should. Unfortunately, we have, it goes actually back to fulvestrant, this this sort of red herring that degradation could be a mechanism of action. Look, if you could absolutely eliminate all the receptor from the cell, probably that would be a reasonable thing to do. But in the very best circumstances, you still have 20% intact receptor. Now, remember, this is a protein versus the DNA binding sites, the promoters. It is in massive access to its binding sites. So going, you know... from estrogen receptor positive to estrogen receptor positive is not a significant therapeutic effect. And that's what happens with the degraders. Look, palazestrant is as good a degrader as anything. It's just that there's all that receptor remaining. You have to turn that receptor off. And that's the exposure. That's the complete antagonism that really adds up to what is, what is the formula for doing this successfully.

Thank you. Um... Now, talking about going to OPERA-01 dosing choice, you know, it seems like, you know, your dose selection is, you know, almost like a non-issue for you. Your 90 mg, 120 mg have equivalent exposure and both clear your preclinical target. Right. Can you talk a little bit around what drove and guided your dose selection and how investors should feel about that?

Yeah. So the first thing that you said is right. When we show everybody our exposures at 90 and 120, they far exceed the threshold, even with the 14-day half-life and the newer population. It's even more than what we had previously said with both doses. So 90 and 120 are essentially, from an exposure occupancy standpoint, the same. But it's a great question as to what drove the dose selection. How the dose was selected. I cannot tell you what drove it because um, we didn't decide. Our independent data monitoring committee, which was unblinded to the Part 1, 90/120 data, decided and then that data was provided to the, to the FDA, who concurred with their decision. But they were provided both safety and efficacy data at 90 and 120 from this 40 patients on each dose arm in order to make this assessment, made the recommendation of 90. Um, we had felt comfortable, comfortable with 90 all along because we knew the exposure easily achieved our objectives. And, you know, the FDA concurred with that decision. But as to the specific factors within safety and efficacy, I can't actually share what that was.

Talk about your ESR1 wild-type activity that you've seen. 5.5 months, which I believe is even competitive with your competitors post in the mutant setting. Yes. And I think that's really the part that the street might be not paying focus on. Can you talk about kind of write the condition that you have right, right? So, I mean, I think there's an underlying principle. It gets a little bit to your question at the beginning about...

comparing all the members of this class. This is such an obvious thing to say, but it does seem to need to be repeated. The best way to predict the future outcome of a drug or a regimen is to look at the past clinical outcome of that drug or regimen, not look at a different drug or regimen. Look at that drug and that regimen. So if you look at palazestrant, compared with the others in the class, we have outperformed in our Phase 2. No one has really seen any evidence of activity in the wild type. We had 5.5 months. Again, this is uncontrolled. It's Phase 2, admittedly. We had over 7 months in the ESR1 mutant. Again, 5 is about the best people do. It's usually more like 4. So again, in both cases, it seems like they're superior already. In the ribociclib combo post-CDK4/6, we had over a 1-year, which again really stands out. So of course, the question here is, and what we're answering in our clinical trials more often, approximately OPERA-01 is, do we duplicate the Phase 2 experience in the Phase 2 randomized setting? That is always a risk. I will say the OPERA-01 patients are somewhat less heavily pretreated than the Phase 2 group were. If we are able to duplicate that effect, then we really have a chance of being differentiated in the mutant subset by greater efficacy and being successful in the wild-type subset where no one has been. But that's the question, right? That's the risk that we're bearing as we go into that trial, and it is tested to answer those specific questions.