Relay Therapeutics, Inc. (RLAY) Earnings Call Transcript & Summary

Going here with our next company presenter at the BofA Annual Healthcare Conference. Pleased to be introducing Relay Therapeutics and Pete Rahmer, Chief Corporate Development Officer. My name is Jason Gerberry. I'm one of the SMID cap biotech analysts. And Pete, thanks for joining us once again at the conference. And it's timely that we have you here in terms of Relay, its story and some recent developments in terms of the strategic cost structuring of the business and prioritization. So maybe we can start there, as we think about where you guys have come as a company. There's been some challenges, I guess, over the years with precision oncology and changing kind of regulatory landscape, which has ultimately led you to prioritize and focus on your PI3K program for breast cancer. There are some other opportunities behind that, but it does seem like the lion's share of your capital now will be devoted to this program. And maybe we can start there, and then we'll jump into more specific questions.

Yes. First off, thank you to Bank of America, Jason and [ Chi ], for having us. We're really excited to be here. You're right. It's been a pretty interesting period, not just for us, but for the entire sector and obviously a lot of macro events shaping how companies are operating today, which is a bit unusual for our sector. And for us, had to make the difficult decision to dramatically reduce our research footprint. Made that particularly difficult was the fact that research organization, I would put forward, be one of the most productive research organizations in small molecule oncology development in a long time. But we find ourselves in an environment where we can't rely on the normal course access to capital and the equity capital markets anyways. And so as we looked at the ways to create value with our balance sheet over the next, call it, 3 to 5 years, it was clear, like you said, that 2608, our PI3K-alpha-immune selective molecule was going to be a key driver to that. And we really needed to focus our effort and our capital against that. And one of the key objectives we were trying to achieve in this restructuring was to make sure we are able to get meaningfully past the top line data of our upcoming rediscovered 2 Phase III study for RLY-2608, and that's what really drove us to make the decisions that we made and now sit here with $710 million in cash on the balance sheet. Cash guidance into 2029, which should get us meaningfully past top line data, the Phase III, that will also allow us to run additional support of breast cancer studies as we think about potential other treatment regimens already even moving into earlier-line settings within HR-positive, HER2-negative metastatic breast cancer. We also recently started a vascular malformation study, which is a PIK3CA mutant-driven genetic disease. There's about 170,000 of those patients in the U.S. And so we are just in the very early innings of that clinical exploration. But it's a very interesting area where the biology has been derisked and we have a lot of the same tenants and hallmarks of target oncology that we can draft off of there. And then we will also continue to progress our en masse Fabry molecules preclinically to at least IND readiness and kind of at that point, assess the environment and the landscape. And we also have one unnamed preclinical program that the research team, the core research team that's left will focus on. It's a very exciting target that has to date been pretty undruggable. So excited for the research team to try to tackle that.

So maybe talk about what the next 2 years for the company looks like because you've got this -- your pivotal study, and that will take some time. I guess we can look at trials like CAPI and things like that to sort of form some sort of benchmark for how long it takes to enroll studies and get to data. So imagine what are the sort of data derisking milestones and things that investors can look to, to get more confidence, either in things like vascular malformation or some of the alternative data sets you might be generating in, say, the breast cancer setting between now and ultimately when you get through some pivotal data that's really going to be the key thing for the company.

Yes. I think one of the things that we are most excited about is that these opportunities have been largely derisked by the existing data. And so we're fortunate in a sense that the data has played out the way that we had hoped it would. In the case of the breast cancer opportunity, the ReDiscover data that we presented last San Antonio Breast Cancer Meeting in December, and we'll update that data a little bit further in ASCO coming up in a couple of weeks. Those data, in our view, look extremely strong in -- and the potential for success in the Phase III study against CAPI and fulvestrant. They're pretty mature data at this point. So at ASCO, we'll have median follow-up out past 12 months. And we're just looking for a consistency of that data from what we presented at SABCS. The timing of that is really helpful, given that we're going to be on the precipice of launching that Phase III study, and we'll have an audience with global investigators to be able to share that updated data with at ASCO. And then in the case of vascular malformations, we're fortunate that we've been able to see that biology get validated to the clinical data that's been generated by alpelisib in that setting. And they've shown clearly across the accelerated approval data set and now in the recent EPIK-P2 confirmatory study that unfortunately failed, but still provided clear clinical proof of concept that through different dose levels, you're seeing at least a dose response with alpelisib, which plays into our thesis that if we can drive to greater target coverage we should be able to see -- to drive to greater benefit for those patients.

Okay. So yes, the Phase Ib data at ASCO, I think you referenced, you saw on a median PFS basis, almost a doubling, I think, of what is sort of the competitive benchmark in the space. And so I guess like sort of the preview here is look for some consistency and more robustness of data, data follow-up. I know that, I guess, some of the questions when you last updated The Street, maybe it was around some patient sensoring and duration of follow-up. Do you feel like those questions will be largely addressed as we come out of ASCO and get that installment?

I think so. I think we'll be able to continue. At the time, as most folks know, when patients are still ongoing, the way they're treated in the Kaplan-Meier curves, they're sensitive. At SABCS, we said that the majority of our patients that had been censored were -- remain on treatment and a good number of them remained in response on treatment. So I think we'll be able to continue to address that with median follow-up now out past 12 months, which we think should probably start to be out further than the current median PFS, which is always a good barometer of data maturity.

Okay, which is like a competitor benchmark has been like 5 or 6 months, right?

Yes. The CAPItello-291 data in the PIK3CA mutant subset is 5.5 months. And so that's -- I think that's kind of -- that's the benchmark we use to guide us as we moved into the Phase III design for ReDiscover 2.

Yes. So in terms of the PI3K competitive landscape, how do you sort of characterize -- your clinical data is obviously very, very intriguing. And there's earlier-stage competitors that you probably -- people compare you to, one of which got acquired by Eli Lilly. How would you kind of frame for investors how you think your molecule 2608 is different than the now Eli Lilly drug?

Yes. Specific to the now Eli Lilly drug, I think the -- the key conclusions that we have based off of the data, so you have both molecules have a preclinical manuscript published in cancer discovery, and then the Scorpion data set was about 60 patients of Phase I monotherapy with 1.5 months worth of followup. That's not a lot to actually evaluate to be honest. There is no monotherapy regulatory or commercial path here in this setting. So really, the onus is on now Lilly to -- as it moves into combination data to be able to meet or, in this case, have to substantially beat our -- the benchmark that we have now set, closing in somewhere in the neighborhood of 10 to 11 months in median PFS in combination with fulvestrant. I think what we would say is these molecules are likely not that different, and that's what both the preclinical and early clinical data would suggest. You did see some idiosyncratic toxicity with the Scorpion molecule, and it's somewhere in the neighborhood of 2 years behind. Just there are a number of other competitors further behind. There was a number of them profiled at AACR a few weeks ago. And I think we continue to see molecules that just look like fast followers to either the block 47 selective approach or to our pan-mutant selective approach.

Yes. The 2 years behind undifferentiated is always a tough sell in oncology.

Yes. I think we've seen this playbook before. In targeted oncology, it is very clear. You need to put up the next-generation profile to the existing therapies, which we believe we've done in comparison to inavolisib, alpelisib, capivasertib. And then once you get into that next-generation class, you have to be first to market. And that is what we are focused on, the execution to make sure we get to be first to market in a very large patient population in the CDK4 experience setting.

Okay. So as you're about to embark on starting the Phase III ReDiscover 2 trial for 208 in metastatic breast cancer, maybe how are you thinking about the types of patients that you'll end up enrolling and prior exposure to CDK4/6, some of which could be thinking the adjuvant setting, some of which could be frontline setting exposure and so your trial could it have a meaningful proportion of frontline patients. And obviously, the proportion of frontline patients could drive a line-agnostic approval. It's more of a CDK4/6-exposed patient population as standards of care shifts.

Yes. I will -- I'll probably fall short of trying to predict labels and regulatory decisions. But what we are endeavoring to do here is if you observe the evolution of the landscape right now, you've got both ribociclib and abemaciclib adjuvant labels. The ribociclib label, the label would address about 40% of that adjuvant population. So there's going to be a growing number patients in the first-line metastatic setting that have seen CDK4/6 in the adjuvant setting. And so we wanted to construct a study that had the ability to enroll those patients -- with the potential to think about being able to have a CDK4 experience line-agnostic conversation with the agency should we end up with a positive study and have a sufficient data set there.

Yes. Yes, we see this with a competitor in the -- post the HER2 setting, like running a post in HER2 trial with zanidatamab in breast cancer and sort of a similar goal of having a therapy evolves and shifts sort of validated post standard of care option. So it sounds like that's kind of a similar strategy you're employing.

Yes. Yes. I think that's a good analog there. The goal is to try to be running the right contemporary study but also skating to where the puck is going a little bit.

Yes, makes sense. Maybe can you talk about the hierarchy testing for the Phase III primary endpoint as you design the trial and look to maximize success for the Phase III outcome?

Sure. We had seen very exciting data across all means. And so we're sitting, like we talked about earlier, near doubling of the existing standard of care today across all mutation types. And within the kinase subtype, we seem to have, at least in the existing data set, a bit of a super performance, if you will. And we also, in the competitive landscape, there are a number of players coming behind us with H1047 selective molecules. And so given the slight even outperformance of kinase and these potential competitors coming behind us, we felt it was prudent to be able to have the kinase evaluation in the statistical analysis so that we can make label claims to the kinase population once we get there. And so we will test the kinase population first, and then we'll move to the ITT population.

Do you -- would you anticipate that happy performs differently in the kinase versus the nonkinase populations based on kind of what you know? Is it well understood?

They've never broken out that level of subset analysis. The data-driven answer is we don't know biologically and mechanistically from the molecule. We would not anticipate that it performs differently because of the molecule based on how we think we tell looks to be equipotent between the kinase and nonkinase population along with wild type and a lot of other kinases.

Okay. And then in terms of your Phase I efforts with triple combinations, maybe just remind us sort of what the current objectives are with triplet cohorts and when we might start to see some data generated on that front?

Yes. So we currently have ongoing 2 triplet regimens. One with atirmociclib, Pfizer CDK4 and with ribociclib. Those -- both those studies continue to remain ongoing and kind of a dose-finding portion of the study. And we have not guided to when we will share data from those regimens, likely not in 2025. The goal there is to, again, thinking about that eventual CDK4/6 frontline metastatic population, what's that future regimen look like? And what's the registrational study to eventually run there? The thing there is it's a bit of an unsettled regimen in that you have some oral SERD studies ongoing in the front line. Early evidence has not been playing out that favorably. Those molecules being effective in the ESR1 wild type population. And we -- whether it be Pfizer CDK4 or Beijing CDK4 and others haven't seen a lot of data yet. I think we're going to see some updated data at ASCO. And so you'll likely see us continue to do those studies and potentially some other oral endocrine therapy studies. -- but still kind of weighing with some of the options there and where is the right place to place resources.

Okay. And with those triplets, are you interrogating the 400 mg, I guess, Fed state regimen?

Yes, all of our ongoing studies are with -- in the Fed dosing regimen.

Okay. Any additional combination strategies? I think you outlined some of them that are, I guess, like contemplated at the moment.

Yes, I think the most interesting ones would be oral endocrine therapies, whether it's just establishing combinability and just some of dose confirmation with as simple as doing some aromatase inhibitor work or potentially some of these next-generation endocrine therapies if it makes sense there. Although it's with some of the recent data, I think it's been tougher to get too excited about putting resources against it.

Yes. Just like one macro topic on this point, right? I mean you guys would have presumably always, from an IRA perspective, fallen within the biotech exemption, single product, right? So the nine-year fill penalty possibly changing and going to 13 years is probably not immaterial from your perspective, although if Relay was ever to be acquired and that pill penalty changes that could change the attractiveness, I suppose, of the asset in the eyes of external eyes. Is that a fair way of putting it?

Yes, I think that is fair of analysis as any of us can make today in this volatile unpredictable environment we existed. Yes.

Yes. Well, at least it's not speculative at this point. Like now we've finally gotten this put into legislative proposals and has clearly indicated that this is a distortion that needs corrected and fixed.

Yes, it still requires, in normal times, legislative action, but yes.

Okay. Maybe shift gears to genetic diseases in the final 10 minutes here. So you've initiated a Phase I evaluating 2608 in vascular malformation. I guess we -- we know the mechanism, right, has the potential to -- it's driven off of a competitor data that we've seen with alpelisib. They were able to get accelerated approval for a subset of vascular malformation. So as we think about what aspects of 2608, might be an improvement on alpelisib beyond just the broader development approach and lens. What are the inherent aspects of the molecule? Is it really just the impact on hyperglycemia that allows you to study maybe more broadly?

Yes. It really is pretty simple and that a lot of the same reasons as to why we've been able to demonstrate differentiation against alpelisib thus far in oncology, why we think that we should be able to do that also in vascular malformations. It's a more selective molecule, which should let us drive to greater target coverage, better dose intensity while sparing the toxicity profile of a nonselective molecule. And so as you know, that in targeted therapy efficacy and safety are 2 sides of the same coin. And so we are -- our hypothesis is that should also benefit vascular malformations with PIK3CA-driven disease.

Right. So I guess like sort of dose-limiting toxicity could be speculated at least maybe in terms of what limited developmental efforts versus how you're able to interrogate the broader landscape of vascular on formation. Is that...

Yes, yes. I think that's it. The alpelisib -- Novartis and alpelisib, I don't think we have strong corporate focus in this area, too. And so they focused the initial development in PROS or PIK3CA-related overgrowth spectrum. They also have a Phase III study going on in lymphatic malformations, one of the larger PIK3CA-mutant-driven subpopulations there. So I don't think -- I think the Apollo, the shortcomings in pros would be the shortcomings of any of these other subsets. You'll see us run our first-in-human study here in both the mix of PROS and lymphatic malformation patients. And the goal is to be able to meaningfully differentiate on both efficacy and safety.

Do you feel like the work you've done in oncology pretty meaningfully derisks safety such that really what you need to learn is sort of the relative efficacy profile here and where you can take the drug from an efficacy standpoint? Or there's just so many confounding variables in the oncology trial and differences in dose that it's a little bit difficult to extrapolate the safety?

No. I think, proportionately, you can look at the difference in safety in alpelisib oncology patients, 2608 oncology patients. If you -- the alpelisib doses that have been studied in vascular malformations are lower. And so the safety profile does not look as pronounced as it does in oncology. But similarly, we would -- and it's just -- it's not a late-stage cancer patient population. These are younger patients certainly burdened with this disease that has a lot of issues that you're both symptomatically that you're trying to resolve with inhibiting the driver of the disease. But otherwise, they're not as burdened with like years of cancer therapy. So not -- there should be a little bit more fit. And so therefore, you don't see as quite the level of toxicity. They're also on the younger side. So you may not see as pronounced hyperglycemia. But we still have seen both with the results that have recently been come out and presented for the EPIK-P2 and anecdotally talking to physicians that tolerability is one of the key issues with alpelisib in this patient population.

Is the dynamic -- I remember in, I guess it was in breast cancer, the screening criteria ended up being a lot stricter for who would get studied for alpelisib to mitigate some of the AEs like hyperglycemia in the vascular malformation setting, is there -- is it similarly like restrictive in terms of who they've studied to see that relatively better side effect profile in that setting?

Yes. They had been similarly restrictive in the metabolic criteria. But again, this is a disease that this mutation happens in neuro. It's a somatic mutation. And so the ideal point in time of clinical intervention is as early as you can detect a mutation in the disease in these patients. And so therefore, you don't have a lot of 6- to 12-year olds walking around prediabetic as you might in a breast cancer patient population.

Okay. And so as we think about the Phase I study that you're embarking on here, how are you designing this to get answers to kind of proof of an efficacy signal?

Yes. So one of the benefits of taking 208 into the setting is because of all the oncology experience, instead moving into a traditional dose escalation Phase I. We are moving into a concurrent randomization of 3 biologically active doses. So the highest dose will be the recommended Phase III dose in breast cancer, and then there will be 2 doses below that, all biologic, what we believe to be biologically active. So that in and of itself should help us move a little bit quicker than the traditional dose escalation. And then we're going to try to -- we -- our inclusion/exclusion criteria. We intend to focus a lot on pros and lymphatic malformation patients. Obviously, with no PIK3CA mutant disease. And so we will be able to get, hopefully, early indications of efficacy in the early dose escalation and expansion portion at the same day.

If you see a signal -- and I know you've had backup molecules in PI3K and I don't know if it makes sense if you see a signal to approach this molecule with different drugs, different pricing structures, things like that to maximize the opportunity in a rare disease setting as opposed to sort of the oncology price kind of bandwidth that you'd fall within? Or maybe do you think they kind of like overlap?

I mean that is the reason for having the additional molecules to provide us with that optionality. And so that -- we have another pan-mutant selected molecule that is at such a stage to where if we see that early clinical proof of concept that we hope we will, will have the ability to make that strategic decision to bring this additional molecule into this setting. And bifurcated for all the reasons you just highlighted, the ability to have separate brands and IP and distinct chemical and disease in these 2 different settings.

Yes. How big can this market be? I know you guys provided some insight into this when you did an R&D Day about a year ago and..

Yes. So the current prevalence in the United States, vascular formations with PIK3CA-driven disease is 170,000. Now the question here is, what proportion of those patients over the course of their life? Will they would they pursue systemic therapy for the -- to control the disease? That's a little bit unknown. But when you're starting at a prevalent pool that large, it doesn't take a large percentage of those patients to seek chronic systemic therapy for it to be a very meaningful commercial opportunity for us. We have done some market research suggesting that depending on what subset of vascular malformation you're in, anywhere from 10% to 40% patients would seek systemic therapy for the disease.

And alpelisib's label speaks to what proportion of that?

They sit in a small subset of PIK3CA-related overgrowth spectrum patients. That's about 5,000 to 15,000 of that 170,000. That is one of the higher -- it's one of the more severe subsets. And so that's where you get closer to about 40% of those patients seeking systemic therapeutic controller disease and lymphatic formations. It's probably 20% to 30%. But there that there's about 80,000 of those patients and the vast majority of them have PIK3CA-driven disease. And it is diagnosed clinically, not requiring a genetic test.

Okay. Maybe in the last minute here. So in lieu of the, I guess, the strategic cost restructuring aspect of things, does that impede you at all to do kind of the earlier stage stuff like potent Fabry or NRAS or any additional preclinical programs? Or are you still able to sort of move those forward depending upon external environment or just company catalyst, then you can kind of advance those forward at the right time?

Yes. So the current cash envelope contemplates taking Fabrys and NRAS at least to IND readiness, and the ability to potentially bring 1 into early clinical studies. But you're right, I think what we'll do is when we get to the IND stage. We'll do an assessment of where we are, the macro environment and kind of make a business decision from there on those 2 programs. And then we are -- the small focused research footprint we have left in the company is focused on 1 very high-value novel oncology target.

Yes. All right. Well, looks like we're out of time. So Pete, thanks so much for joining us.

Yes. Thanks again for having me. Appreciate it.

Of course.