All right, good afternoon. Saving the best for the last here. Thank you for joining. This is Debjit from Guggenheim Therapeutics team. Thank you all for joining our 3rd Annual Healthcare Innovation Conference. My privilege to welcome Sanjiv Patel, President and CEO of Relay Therapeutics. Joining Sanjiv are Don Bergstrom, President of R&D, and Pete Rahmer, Chief Corporate Development Officer. Thank you all.
Thank you, Debjit, for the invite. It is a pleasure to be here.
I do not want to insult your intelligence by asking you to do an intro.
No.
Why don't we just get straight to the Q&A?
I am happy to do an intro.
No, I think anybody on this, anybody listening in probably knows enough about Relay.
Okay.
Let's skip the intro—
All right. Thank you.
—because there is a lot happening with the company.
Yeah.
I want to start on the ASH update.
Yeah.
Let's talk about the data first. We are expecting somewhere in the ballpark of 60 patients. Is that sort of a right estimate?
I think at the moment, all we're saying is that the ISSVA data that we showed back in May of this year, we showed data on the 32 patients in the dose randomization portion. Of those 32 patients, we showed safety data on all 32, and we showed efficacy data on 20, and that was because 12 of those patients hadn't yet had a scan at the data cut. At ASH, what we'll show is those 32 patients again, and we'll show now efficacy and safety data on all 32 patients. As you may remember, back in ISSVA, we showed across those three doses a 60% volumetric response rate. I think the question for investors is, how has that evolved as we go from this N of 20 to 32?
Our hope is that it stays in and around that, and that we can show some consistency to make sure that investors understand how differentiated our therapy is. Obviously, looking at the much longer follow-up on that, how the safety profile evolves, we would obviously want it to stay consistent to what we saw earlier in the year. Then we will show additional data from the dose expansion portion of the trial. There, the majority of the patients that we have enrolled in those dose escalation cohorts have been at the 400 mg once a day. We are not going to say today how many patients we will show there, but those patients will predominantly have had only one scan. So, the efficacy of earlier patients will only be at the 12 weeks. The goal there is to show two things. One, we want to show a very tolerable profile.
As we came out of ISSVA, the real predominant feedback from the community of treating physicians is tolerability is key. The volumetric response rate is interesting for the FDA, but what patients want is a tolerable therapy they can take for life. So, the focus of us in this, choosing the dose to go forward with in the pivotal trial, is going to be what is a dose that a patient can take forever? So what we want to show is low levels of hyperglycemia, diarrhea, rash, dermatitis. Obviously, the efficacy profile has to be differentiated, so it has to clear the bar of alpelisib in the kind of high 20s. So as long as we can do that with a tolerable safety profile, I think investors should feel very confident that we will produce a differentiated therapy that should be commercially successful.
So that is what we look forward to at ASH on the data. We will also show a regulatory update. So, the path forward there around what is the indication statement and what is the pathway towards accelerated approval, we will provide color on at ASH as well. So kind of lots to think about as we go forwards, but that is kind of the overview of what we hope to show.
Got it. On the regulatory side—
Yeah.
—at this point, I'm assuming the meeting has already happened or about to happen?
We're not commenting.
Got it. Do you expect to have, or are you hoping for a broad label right off the bat from a filing perspective?
Look, there's a precedent being set here by Novartis where they have a label for PROS, and they got an accelerated approval, and then they're running now a separate trial for lymphatic malformations. So the precedent is out there of having separate labels and separate indications and separate trials. Our approach will be to obviously go for the broad pooled population of having PROS and LMs together, and obviously, that is the question for the FDA to see. Our logic is that we believe both these conditions are predominantly caused by a PI3Kα mutation, and so there is some commonality in this group. You do not need to test these patients, and so we'll try and go for a broad label. But I think we'd be happy with the base case that we have of having exactly as Novartis has, two separate labels and two separate trials.
From the FDA perspective, what would prevent them from going forward with a PROS plus LM? You might have enough patients, and the safety is probably going to look far better than what the commercially approved drug is. What is the holdup there?
I think all you are hearing from us is that there is not a lot of precedence in this space. We do not want to get out over our skis on trying to predict what the FDA may or may not do in a context, in a disease context in which we cannot point you to six to seven, a dozen different instances of precedence to support our hypothesis of what to do or not to do. So far in our data, the performance of zovega in LMs and PROS has been extremely consistent. We do think the underlying disease biology is very similar. A lot of the morbidities that exist with these patients is very similar. The treatments that you see these patients are on before they come on to zovega, very similar, sclerotherapy, surgery, sirolimus, or alpelisib.
There is a lot of good scientific justification and rationale as to why you can pool these patients together, but that is ultimately going to be a decision for the agency to opine on.
Got it. Thinking about the efficacy data, although you will have just 12 weeks of follow-up, is there a bar? The reason this comes up a lot is there is a bunch of copycats coming through quickly. If the label does not get expanded or if the efficacy is, let us say, falls short of expectations, that opens up an avenue for these copycats. Your take on those?
I mean, I think as we've said, look, we believe that the bar really here is around safety and tolerability. That's what people want. They want a tolerable therapy that they can take for life. As long as it clears the bar of alpelisib, which is the currently available therapy, it will be commercially successful. That's really our approach to this. Obviously, we'd like it to be differentiated from alpelisib. We've shown already that we can achieve 100% response rate with the 300 mg once-a-day dose. That won't stay at 100% as you get more patients. We've shown that you can really max out the efficacy. The real question is, should you? You could take your approach around, you don't want to have any kind of competitive challenge in the future, so just max out the efficacy.
That will actually result in a commercially unsuccessful therapy. You could actually see that with alpelisib, right? Which is, they have a therapy which they really tried to max out the efficacy with their 250 mg dose, but it's just not tolerable. Patients cycle on it, and then they cycle off it, commercially unsuccessful. I think for us, the choice is around threading the needle between having enough efficacy to make sure that we remain differentiated, but it really is around the safety and tolerability profile that's important. The second point around, yes, we imagine that there'll be entrants that come behind us with PI3K mutant-selective assets. It's not so easy to enter this field as a fast follower for lots of different reasons. The reason that these patients and physicians are very cautious.
The thing that made us able to recruit the patients we did in the timeframe we did was the several hundred patients that we have over multiple years now from breast cancer. Again, that took us four years to accumulate, the chronic tox that we have. Obviously, the sites that are out there aren't that many sites across the world where these trials can be conducted. Obviously, we have a very robust site footprint now, and many of these sites can only run a small number of trials at any one time. I think that coming after us is going to be tough.
Rather than us try to maximize the response rate to try and shut the door, I think the best way to shut the door is to produce a commercially very attractive profile that physicians want to use and patients get comfortable with. That's our approach to choosing the dose.
Got it. On the commercial side then.
Yeah.
I mean, you are at the cusp of getting some FDA feedback, so it is almost time to talk commercial. Where do you think these patients are? I mean, from a percentage basis, are located at vascular centers, pediatricians, dermatologists? How should we think about the market?
First thing to say is, this is a very nascent disease space, and so the direct answer to your question is no one really knows. As we now have dozens of sites across U.S., Europe, and Australia, we have been out seeing in person all the leaders of the multidisciplinary teams at all those sites over the past three or four months. As we see the landscape and have done the work that we can do, either through primary market research, talking to our investigators, or what is in the academic literature, it looks like a lot of the moderate-severe patients who are seeking more complex treatment, serial surgeries or serial sclerotherapy embolization, or contemplating one of the two systemic therapies available, they do find their way to the vascular anomaly centers, where you have this multidisciplinary team that can treat the totality of the disease for the patients.
We think there are a few dozen of those in the U.S. and probably a few dozen of them in Europe. Between that number of centers, you are probably capturing a large portion today of the known addressable market. Do patients originate in some of the primary care areas or dermatology and other disciplines? Absolutely. But I think as they progress towards contemplating systemic therapy, they are generally making their way to the vascular anomaly centers today.
Got it. Closing this regulatory discussion here, there are obviously PROS plus LM debate, and then the scope of the label in terms of adolescents versus pediatrics. Can you sort of give us a sense of how many of these 12 and under patients are diagnosed and known versus adolescents? If the label were just an adolescent versus an adult, what happens to these younger patients, and how many are there?
I don't think there's a disparity between the diagnosed and known across adolescents or adults. There are gaps in adult care, so a patient that was diagnosed as an adolescent or pediatric, most of these vascular anomaly centers are at children's hospitals, and so some of them will age out of the children hospital care and then aren't cleanly passed off to adult hem-onc and may get lost in the system a little bit today. That is a known problem, something we pay close attention to and have a lot of conversations about how do you fix in the future. Specifically to your question, it's not a diagnosis problem. It's really a continuum of care problem that is readily acknowledged across all professionals in the space, and there's a lot of activity and effort to go against solving that problem.
So what I'm hearing for a lifelong therapy, it's going to be safety, safety first. If you were to rank order glycemic issues, rash, diarrhea, which ones are the most important for patients to stay on therapy?
Yeah. In any context, these patients come on with morbidity of their disease, and the reason why safety and tolerability is so important is because as that gets too severe, that then becomes the other side of that morbidity coin, which is really hampering their quality of life, which is really what we're ultimately trying to talk about, is symptomatic improvement and quality of life improvement. If that comes with volumetric reduction, that's secondary to these patients and physicians. It's the totality of the safety profile, the severity of any one event. I think one of the things we will focus a lot on is minimizing the number of grade 3 AEs in totality overall.
The other thing we are going to be watching for is, are these events happening iteratively over a longer duration of treatment, or do they happen earlier in cycle one or cycle two? Are they largely self-resolving? Can they be easily managed, and do you see them dissipating over time? Those are the concepts that we will be looking for from a safety and tolerability standpoint.
Got it. We could go on with this, but given the scope of the pipeline here, let's switch to the second-line breast cancer study. The ReDiscover-2 is going to enroll sometime before the first half of next year or around the first half of next year. The obvious conversation now has become, how should we model it? What is the most reasonable expectation for capi? Given one of the sponsors recently had a significant compression in their PFS, how should we think about where Relay is going to land up?
Yeah. I think on the capivasertib, there is enough data out there now that makes us confident to model five and a half months. That is what we are seeing both in the kind of the subcuts we saw in the pivotal trial as well as the real-world evidence. I think in terms of how you model zovega, obviously we have shown 11 months worth of PFS in the data that we presented, but that was in a kind of mixed second, third line population. On one side, you see potentially, as we go to 200 sites across the world, that investigators are less familiar with using zovega, that you could see a compression. On the other side, it is balanced by the fact that you have now a pure second line patient population.
We do think the two things will probably balance them out, and so we feel confident that we will end up back where we started. That gives us a significant buffer against the five and a half months, and so we expect a high confidence of a positive trial.
In case of a compression, how much compression have you modeled in from a stats perspective?
We haven't disclosed that, but suffice to say, the two barometers here obviously are our 10.5 - 11 months median PFS that we've seen across our two independent phase I single arm experiences. On the capi side, as Sanjiv said, a couple of different subcuts of CAPitello-291 both land you at about five and a half months of median PFS for capi. You can imagine, as we designed the study, we did it with some conservatism on both sides, anticipating the possibility for some compression for zovega and maybe even a little bit of over-performance of capi and still have a buffer to have a successful study.
Got it. The other sponsor will have gedatolisib in the market at some point middle of next year. How does that change your launch assumptions, or it's completely independent given that yours is an oral versus an infusion product and with a little bit of a safety baggage or more than a little bit of safety baggage?
Yeah. I mean, I think the assumptions that are probably now we know the gedatolisib data much more kind of robust and positive than when we didn't know. Obviously, we did a lot of thinking around the scenarios of what gedatolisib triplet could look like and a lot of market research. Obviously, it's a weekly IV regimen that comes with prophylactic steroid mouthwash and potentially other prophylactic measures. So it's a multi-drug regimen that requires a patient to come to an IV infusion center once a week. The real challenge for us commercially is it's going to be on the market before us. So for it to really establish a dominant position, our modeling was that it has to show a significant benefit over an oral that is in the 10 and 11-month range.
It would have to show over 15 months of PFS, if not closer to 18 months to be really difficult to dislodge. Those were our kind of most pessimistic assumptions, which is if it did show 18 months of PFS in this space, even though it had all these constraints, it would be a difficult commercial proposition to go up against. Now we know the answer is 11 months. I think we feel much more confident that as we launch into this market, irrespective if they've got a head start of multiple years, that this should be a market that we can take share from rapidly.
There's an interesting analog here when you look at alpelisib and capivasertib. Alpelisib was launched in 2019. Capi launched in late 2024, and you had. The context there was actually capi numerically had a slightly less median PFS in its registrational studies compared to BYLieve at that time. It just had the perception of a better tolerability profile, when it really probably just a trading of some toxicities for other toxicities. At that time, you had alpelisib had plateaued at about $350 million-$400 million a year run rate. Within six months, or sorry, six quarters, capi had completely taken that market share and grown the market. That's in the context of slightly worse PFS and a trading of toxicities.
Here, our proposition is going to likely be if our phase I, II data play out and are recapitulated in ReDiscover-2, you'll have probably similar, if not slightly better, median PFS and a dramatically better safety and tolerability profile, and that's before you even get into the convenience of an oral regimen versus a weekly IV.
You're 12, well, 10 months roughly from the time the trial study enrolls to where we could get the data. Obviously, it's time to almost start building your commercial organization. How are you thinking about the scope and size of that as you prepare for launch?
Yeah, it's definitely on our mind now. We announced last week a Head of Commercial, Shannon Campbell. She was the ex-U.S. Head for Novartis Oncology. She was there when they were with ribociclib, there for the launch of alpelisib, both in breast cancer and vascular anomalies. So she knows both breast cancer and vascular anomalies markets very well, and I think she'll get to work now on doing all the preparation and sizing that we need. The breast cancer market is very well established, and so we know from many companies of our size and scale how they have invested and sized their sales forces. I think there's a lot of knowns on the breast cancer side.
I think the unknowns for us are just going to be how we think about this kind of total green field of vascular anomalies and how to do a lot of work that no other companies have ever done. In breast cancer, there's plenty of companies that have laid the groundwork for lots of medical education, patient advocacy, that all that work needs to be done on the vascular anomaly side. We will do all the work that is needed to make sure that we do the best job we can as we launch zovega.
Let's switch to the frontline because we can go on with this, but—
Yeah.
—limited time here. There's been a lot of noise around the new Pfizer PI3K. When we looked at it structurally, it doesn't look any different than some of the prior non-viable candidates, if I might call it. Thoughts as to why Pfizer might even move this into the clinic?
Look, they are a breast cancer company, and they currently do not own a PI3K inhibitor today. We've known gedatolisib came from Pfizer. We've known that they have active research against the target. I'm sure they still have active research against the target. It is not surprising to us that they are trying to find a pretty low-cost way to have some presence in their pipeline with a PI3Kα inhibitor. Today, it has no bearings on what we're doing. We have a clinical trial supply agreement in place with Pfizer to where they supply atirmociclib, their selective CDK4 molecule, for our frontline trial that we're on track to start early next year, and they've been absolutely great to work with in that context. We remain operationally on track to move as fast as we can, and they are supporting that without question.
That's all we care about at this point. We agree with your assessment of the molecule. We've seen a lot of these come through, and I'm sure we'll see more try, but we are very confident in the bar that we are setting to maintain our commercial superiority for a very long period of time.
When you think about your frontline study, is 20 months the right kind of PFS for the control arm?
No.
Or it could be slower because patients don't trust or mutant PIK3CA kinase patients don't respond to CDK4/6 as well in the frontline setting?
That's exactly been the experience, that the PIK3CA mutant patients do benefit from CDK4/6 inhibitors, but not to the same extent as PIK3CA wild-type patients. What we've seen in the case of ribociclib in the MONALEESA trial in endocrine-sensitive patients was while the blended population overall was 28 months, the PIK3CA mutant population was 19.5 months, and the PIK3CA wild-type population was over 31 months. There's about more than a year difference. We feel comfortable with this 20-month assumption roughly in the PIK3CA mutant population based on some of these retrospective analyses from the CDK4/6 inhibitors.
The dose selection, this comes up a lot in the discussions because you're going forward with the 150.
Yeah.
But the way I think about it does give you the flexibility to go down to 100 mg for a triplet that's going to be used for almost three years. Is that the right way to think about it? Because even with the 100 mg, you have enough target coverage?
Yeah, exactly. Just to remind you, because atirmociclib does increase the blood concentration of zovega, that 150 mg dose gives us approximately the same target coverage as our 400 mg dose in the doublet with fulvestrant. So we're essentially using full- dose zovega, even though the actual dose the patient's taking is lower.
Got it. Unfortunately, we kind of run the clock here. I am really looking forward to the ASH update across the board and then take it from there. It is going to be an interesting three, four years for you guys.
Great. Well, thanks again, Debjit, for the invitation. Thanks to Guggenheim.
Thanks.
Appreciate the time. Thank you so much.
Thanks.
Thank you.