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

I mean I think you'd rather see what happens in terms of practice patterns to say what's exploring the opportunity. I would say in the first-line setting, I don't think the stomatitis, the mouthwash and the IV infusions are going to be viewed as very attractive compared to what is considered a relatively easy to take well-tolerated CDK4/6 plus AI or potentially something like palazestrant. But again, remember, in either case, you're talking about 2 oral daily medications with the CDK4/6 and the endocrine agent, be it AI or for instance, palazestrant or if camizestrant is successful, SERENA-4 is another one that's out there playing. So I just don't think IV in that setting is really going to be very attractive.

I see. Is it just the IV? Or is it like a triplet just not...

I don't think -- I think of a triplet where if the quality of life were -- and IV here is a quality of life issue as is the stomatitis. I think if it well tolerated and it didn't disrupt life like IV daily, I think more efficacy will be attractive to patients to have a triplet. On the other hand, I do think that there are liabilities that are quite over.

Got it. Okay. You guys presented the Phase II Pala/Ribo study a year ago, and it showed very impressive 14 months in all patients and 13 months in patients after CDK4/6 use. So these were, to me, like they look very competitive compared to what VIKTORIA showed either with the wild-type or the mutant population. So why not just given that, why not do a Phase III combination with Ribo in the second-line setting? That way, like you don't have to worry about ESR1 mutant or wild-type. I mean just is it going to be all covers?

No, it's a very good point. So I'll explain it in 2 ways. I certainly agree with you that the data is very compelling, and it was a compelling in the mutant and the wild-type subsets. Here's the thing -- and we know for instance, that prescribers gave CDK4/6 after CDK4/6, that's a common existing practice pattern. Now it's done with fulvestrant, but Pala has -- certainly, if it has more efficacy, but also then it has the daily oral convenience. The one thing they don't like to do is they like to switch the CDK4/6. And so while that data is compelling and some of it is post Ribo, patients and oncologists tend to like to switch the CDK4/6 if they've progressed on it. And almost all patients are getting Ribo now in the first line, right? So Ribo after Ribo is less attractive. So I think we're thinking about do we want to do a combination in the second and third-line setting? And if so, it's probably not ribociclib because of the first line.

You leave it up for physician choice.

There are multiple ways to do it. The thing right now where we are it's probably useful to see how OPERA-01 reads out. It's not that far off.

Right. I see. Okay. So sort of let that lead the way in terms of the fulvestrant.

Yes. And then if we get the wild-type there, that sets obviously a very different situation. It's very differentiated. And then I think we can sit down with a different -- with a more full picture in mind about what do we want to achieve with combination in that setting. It is absolutely true that there are physicians and patients who say, "Geez, I don't want -- I'm not ready to take just a monotherapy now. I want to get a targeted agent. Do I want to get the most active things I can. And so you want to be able to provide both.

Right, right. Okay. So let's move on to persevERA. That's another big trial, a lot of redo from that. So me and you discussed this trial many times well ahead of ASCO. And I was fairly bearish about how this trial would read out. On day 1 when lidERA read out, I was -- we put a notes saying that I see high risk for persevERA. So we finally saw the full presentation at ASCO. What is your overall impression? And is there any concern to OPERA-02?

Yes. So my overall the trial was negative. That's absolutely true. In terms of its hazard ratio of 0.89 did not achieve statistical significance. On the other hand, there's very clear evidence of activity. In other words, the underlying hypothesis of these first-line therapies, persevERA, but also OPERA-02 is that aromatase inhibitors are not completely suppressing the growth and proliferation signal from the estrogen receptor and that you can do better by hitting the receptor harder with a complete antagonist and ideally by doing that with a higher exposure where you completely shut off the receptor all the time. I believe that giredestrant lower exposure probably was a liability there. But even so, they had 5 months delta in median PFS. That's meaningful. It didn't hit statistical significance, but it's meaningful. And so I think where we have better Phase II data than giredestrant had, it really reads through nicely to OPERA-02. My opinion, it increases the probability of success of OPERA-02. Now on the other hand, you asked about things to learn or concerns. One thing that happened in persevERA that I think is hard for me to understand, but I think it was definitely a liability was they enrolled 10% of their patients were endocrine-resistant patients. And those patients didn't do well. And I think they would have deleted -- I'm sorry, diluted the treatment effect in the endocrine-sensitive patient population. Now we don't enroll any of those patients in OPERA-02. I should say SERENA-4 doesn't either. So I think that's one thing we suspected you shouldn't do, but now it's confirmed. Now we didn't do it. There's another thing that's interesting to watch, which is that the control arm, palbo plus AI outperformed historical a bit by about 3 months. That's normal in oncology. Over time, a regimen often gets better. Oncologists get better at giving these medications. I think we will want to look at that and ask ourselves, do we want to make any changes to OPERA-02 going forward. Now we won't -- if SERENA-4 stays on time in the second half of 2026, we'll wait for that because then we'll have 2 data sets, and that's fine. But if that gets pushed out, we might not be able to.

I see. Okay. Yes. So I think when we also looked at that trial, besides the PFS sort of missing with that hazard ratio of 0.89 that you mentioned and then the p-value being 0.156. I think the ORR, CR, PR, SD, CBR were basically identical between the treatment arm and the control arm. So I agree with you, there's a 5-month improvement. And I think in the past, you guys called out that 6 months would be static. And then you look at the Roche's plan that stats plan where they were powered to like 89% to show hazard ratio of 0.77, but then they set the threshold -- the boundary for success at 0.85. I mean, do you agree with that statical plan to begin with? Maybe I just wanted to just ask you that. And what is sort of your hazard ratio and powering assumption for OPERA-02.

Yes. So we haven't discussed our plan for OPERA-02. The biggest reason for not discussing the plan is because if we see data from SERENA-4, we've already seen persevERA, we may change it. So you don't go out and say something and then go change it. But we do -- so 6 months is a slam dunk, right? It will change the practice pattern. Five months, if you were statistically significant, would be -- if you go ask the investigators, the breast cancer docs, they'll say, okay, 6 months, if we see that, we change our practice pattern 5 months. Yes, that would be good, too. You get down to 4, they start saying, well, tolerability. So definitely, 6 months is clear. I go back to this plan. I really wonder -- and we can't do it, but Roche's could do it. They could remove those patients who were endocrine resistant and say, what does the curve look like? And I think it would separate earlier. The separation was then maintained. So I really think that it's not necessarily the stats plan that I wonder about that trial. It's the inclusion criteria in the patients they studied. I think that trial design might have been perfectly fine, if you didn't put on endocrine-resistant patients.

I see. Yes, it's hard to tell because they never -- we didn't see...

Well, yes, they get 12 minutes. So maybe down the road, we will see it. But now, we haven't seen that so far. You're right.

Yes, exactly. So you mentioned about that control arm being a little higher than sort of -- it was basically the same. I think same with that MONARCH 3, about 28 months. I think it was higher than the PALOMA-2...

Yes, by about 3 months.

By about 3 months. Exactly. So -- but when you think about sort of if the control arm is now, I think like what you said with the modern-day treatment, better care that you're now going to see more of that higher or higher end of that range. Do you believe that now you need more than 6 months to be static just because the control arm is stronger?

No, it's not the more than 6 months. It's just that you change your statistical assumptions around how many events you need, right? And so -- and obviously, when you're doing something like that, you only want to do it once, if you're going to do it, and you want to be as well informed as you possibly can be. So if there's a second trial that you can get, that would be huge. But what you do is you just go through and you calculate, okay, here's the irony, right? Hazard ratio is how the primary endpoint is done. Hazard ratio is how the trial stats plan is designed. Decisions about treatment are made by delta and medians. So you are kind of extrapolating from the one to the other. And what you do is you just take the delta you're shooting for, you take what hazard ratio you get and then you decide how many patients do I need, how many events do I need? How long do I have to wait? And those would be the calculations.

I see okay. Okay. Got it. So I think you mentioned in the past that after looking at persevERA, you may go back and change the size of the study. Is that still on the table that you guys are evaluating on the side? Okay. So based on that, how -- I mean...

We'd like to have SERENA-4 though to complete that evaluation.

We should just wait for that. You're not.

Yes, yes. No, you don't do these kinds of things twice. In particular, if -- again, if -- and I'd just take them at their word, if AstraZeneca is able to stick to their time line of half 2, we have time to use both.

Because SERENA-4 is a larger study, so you kind of see that.

It's a larger study, which we like, but the point is that the patient population is more appropriate to OPERA-02, because like OPERA-02, SERENA-4 does not allow any endocrine resistant patients on the trial.

Right. Okay. Got it. So now given that higher performance of that, that placebo arm into the better care out there. How would increasing the size would help mitigate some of this risk?

Yes, it does 2 things, right, which is a 6-month delta on top of a 25-month control arm and a 6-month delta on top of a 28-month control arm have slightly different hazard ratios. So that's the first thing you do is you want to say what hazard ratio am I trying to detect in order to see my clinically significant benefit. So that obviously changes the stats plan, right, because that's your primary endpoint. The second thing it does is it changes the time, changes how long you need to wait to get the events you need. First of all, your event rate is slower because you're control arm has got a longer median. But then as well, you're kicking out what 6 months is. And so there's another calculation, which is how long does it take for the trial to mature. There are 2 ways to get those events, right? Wait longer or have more patients at risk. That is to say more patients treated. And so you put both of those factors into the calculation.

I see. Okay. Got it.

And then you go to regulators, then you go to the steering committee and it's a real process.

Right. So now you look at the persevERA trial, I think you pointed out that 10% patient progress, right. So there's other stuff to it, too. I think the de novo disease and then there's higher an alpha-2 -- how is it designed differently that you think? And what are other ways that you believe besides that 10%?

That's the main thing, actually. I don't know why they can't de novo disease. That's interesting. So it was a bit lower than you would see. I don't know why they had more visceral disease. I didn't see anything that Roche's did in its design or execution that would cause that to happen. They stratified for them. Yes. I don't know. It was -- you're right, it was a bit on the high side. So again, these are aspects that I think with 2 data sets are kind of things you look at and say, are we seeing a change in who gets enrolled in these trials? Or is this an outlier versus what other trials obtained.

I see. So SERENA-4 will be reading out probably could be around the same time frame as OPERA-01.

Could be.

How do you -- is there any reason to believe that trial could succeed? And then if it doesn't succeed, what does it mean for you guys?

Yes. I think -- so first of all, it doesn't succeed, it really doesn't mean anything for us. The control arm will still be interpretable. It's palbo plus AI and Letrozole in that case. Secondly, look, OPERA-02, and you mentioned the data already, the long period of progression-free survival post CDK4/6 with ribociclib and palazestrant. Our trial is based on our data. And our data uncontrolled Phase II looks better than the others have. And so I think it doesn't change our view of OPERA-02. In fact, as I said, persevERA gives us higher confidence because of the level of benefit they did see without demonstrating statistical significance. Your question was, could SERENA-4 be positive? It's interesting. I think if SERENA-4 without the endocrine-resistant patients with the higher end with the same level of benefit maybe could be positive actually. The concern I have is that camizestrant has the lowest exposure of all of these agents, because of the tolerability profile. Is that enough or is it not? The...

I see. Any learning from SERENA-6, because that was -- there was an updated presentation at ASCO with the updated PFS2. Is there anything there that you feel like you can apply to either SERENA-4 or OPERA-02?

Nothing. Nothing. Yes. Yes, the learning is don't do that.

Fair enough. Okay. Let's spend the last 2 minutes on OPERA-01 trials reading out in fall. This's very -- designed very differently from VERITAC-2 in some way, right, even though you guys both follow the EMERALD data. So one big difference is that you guys allow prior fulvestrant and then also adjuvant therapies where VERITAC-2 did not. So as you think about some of these differences that could either benefit or not benefit OPERA-01, how do you think about that compared to VERITAC now after you see that data? Do you think you made the right choice to not allow -- sorry, to allow fulvestrant and also adjuvant use?

I do. And the reason I say that is because I think that more mimics the standard of care. It also -- you're absolutely right that we're different than VERITAC-2. We're much more similar to EMERALD in that respect. So we do have some differences. We didn't allow prior chemotherapy. EMERALD did. We require 6 months on your prior endocrine regimen before you get on the trial, and there was no limitation on that in EMERALD. But we feel like EMERALD was the right trial. It was in this patient population, it was best mimic the practice pattern that we're seeing. Now you control for some of these things, right? You stratify by if they had fulvestrant or not, you do things to make sure you're equally distributed. But we really felt like EMERALD was very informative.

I see. Okay. Final question for you. Your confidence level in that ESR1 wild-type. If you don't succeed, what does it mean for Pala in that second-line setting? What does it mean for OPERA-01? Do you must need it to succeed?

I don't think we need it to succeed. I think that there's a way to differentiate just in the mutant potentially by having more than 2 months extension of PFS, and that would be quite meaningful. It is -- I don't mean to downplay. It is a really meaningful differentiator if we're able to get it. We saw 5.5 months in the wild-type in our Phase II setting. If we can recapitulate that, obviously, uncontrolled error bars, everything. If we can recapitulate that, it should be compelling. And then that gives a group of patients who are currently have an unmet need that's unaddressed a treatment option. The trial is well designed to address that question.

Fantastic. Thank you so much.

Thank you.

Pleasure hosting you. Always an interesting discussion. And I'll turn it to you for final remarks.

It's a busy year for Olema and for this space, and we look forward to our first Phase III trial readout in the fall to updating on KAT6 with combos as soon as we can with palazestrant and fulvestrant. And OPERA-02 is enrolling well. So a great opportunity in the first-line setting, I think, supported by the data we've seen recently.

Great. Thanks so much.

Thank you.