All right, thanks everyone for continuing to join us here at the Stifel Virtual Oncology Days. My name is Brad Canino. Happy to be joined on the next fireside with Relay Therapeutics. We've got Sanjiv Patel, President and CEO, and Peter Rahmer, Chief Corporate Development Officer. Thanks so much for joining us.
Thanks, Brad. Thanks for the invitation. Look forward to it.
Can we just start off with a brief intro and priorities for Relay into the second half of this year?
Yeah. I mean, we've been hard at it for the last few years, and hopefully the second half of the year will be a busy one for us. Biggest priority for us, obviously, is continuing to gather data for our mutant selective PI3K alpha inhibitor, RLY-2608. And we've committed to sharing a data update in the second half of the year from both our doublet and our ongoing triplet trials. And we're excited about showing that data as we go through the second half of the year. Second priority for us continues to be gathering the tumor agnostic data for RLY-4008, engaging the regulators, and then making a disciplined data-driven decision on the path forward there. We continue to work with our partners at Roche Genentech on RLY-1971 with its three ongoing combinations. And then our platform continues to be productive.
You know, we saw the mammoth Stifel report at the weekend around the power of the use of computation in drug discovery. And so we're excited about demonstrating our own approach to that by showcasing at least one new program, as we come into the second half of the year. So it should be an exciting set of updates from us next six months.
Yep. Yeah, it's not good when the banking team one-ups the research team with better reports, but we like the healthy competition here.
That was a great report.
Can we just kick in and talk about the expansion cohorts that are ongoing for the mutant selective PI3K? You've got the second one that I believe is still enrolling. You enrolled the first of 20 patients from, I think, July of last year to January, if I'm correct on that. So where are we with the second now?
Yeah. So as you may remember, obviously, we did the dose escalation portion of the trial, which is complete. And inside of that, we had 17 patients at 600 mg BID. And then we opened an expansion cohort, you know, in the summer of last year that enrolled another 20 patients at 600 mg BID . So as we came into the year, we've got roughly about 40 patients that have the opportunity, as we come into the second half of the year, to have at least six months on therapy and therefore will be CBR available. And then obviously, the 600 mg cohort was expanded and continues to enroll, and we're very happy with that enrollment. And so we should have a robust and interpretable end as we enter the second half of the year to give us all comfort that RLY-2608 is what we hoped it would be.
Yep. Okay. And now you mentioned CBR. Should we expect the PFS curve and the median to be disclosed? And I guess what median could even be possible given your kind of enrolled, I guess, half of the patients already that would have a decent follow-up, I guess? Talk to me about that.
Yeah, you're right. I mean, given that this study continues to enroll, even as we speak, we're not going to have a mature PFS curve to be able to report in the back half of this year. Well, hopefully not anyway. Our presumption is we will not. So I think what we'll be able to do is, as Sanjiv alluded to, roughly 40 patients that should be CBR available or at least have had the opportunity to be on for six months or more, we'll endeavor to try to do some PFS landmark analysis, most likely, maybe six and nine months. Obviously, six-month landmark analysis on PFS and CBR are going to be they should be closely related.
But I think those would be the key kind of efficacy outputs we would intend to try to generate, in addition to the, you know, traditional waterfalls and swimmers, but try to give a clear CBR rate based off that end and some potentially PFS landmark analysis.
Will you pool the dose-optimized patients at the lower 400-mg dose into this analysis?
The current intention is to focus this disclosure on the 600 mg twice daily. How we do or don't layer in other data in and around that 400 mg or otherwise, updates from the dose escalation is to be determined.
Okay. And the benchmarks we should think about when it comes to CBR and PFS context of the wild-type inhibitors?
Yeah. Fortunately, on the CBR side, we know from analysis of alpelisib, you know, always got to caveat their cross-trial comparisons here. The differences in patient populations are quite important, especially in HR positive or HER2 negative breast cancer. But the current benchmarks for CBR are in the mid- to high 40s for those two agents. We have not seen a CBR analysis broken out for capivasertib yet from the CAPItello-291 study. That's the phase III study. So mid- to high 40s is the current CBR benchmark, again, with the proper caveats around it.
Okay. Okay. Now you mentioned the patient characteristics. How well do you think your characteristics will match the data we're all going to look at for alpelisib and inavolisib? You know, I'm thinking about certain things like types of mutations, chemo pretreatment, etc.
Yeah. I think one of the, you know, as we take a step back and think about benchmarks and how we're going to what the relevant things to compare it to are, the first thing I would say is, as we think about what could be next steps after the study, so what is the first phase III study we could run, the two regulatory standards of care there today in the second line are capivasertib and alpelisib. And the current PFS benchmark from those two agents are somewhere in the neighborhood of 5.5-six months.
So as we look at these data and interpret these data in the second half ourselves, we're going to be looking for data that demonstrate confidence to us that if we were to move into a randomized study, potentially if we chose to go into that setting first, that we would be able to have a superior trial against one of those two agents. And so that's the main thing we're looking to try to robustly and interpretably demonstrate with this data in the second half. On the pretreatment side, the CAPItello study. It does a good job of subsetting. That study enrolled both CDK4/6 pretreated and CDK4/6 naive, but they've done a good job subsetting out the CDK4/6 pretreated, which is really the germane comparison for our dataset. It is otherwise, once you get into the CDK4/6 pretreated subset, it's a pretty representative patient population of the real world.
Chemo pretreated, you know, a quarter to a third chemo pretreated, probably about half have seen prior fulvestrant, and you're going to have some ESR1 mutated patients. Certainly, the one thing in a phase III study like that that's well controlled is the number of CDK4/6 therapies seen in the metastatic setting. That's an important parameter that impacts durability for subsequent therapies. So that's one thing that we'll, you know, we'll be continuing to look at in our study as we get closer to reporting data.
Okay. When do you have regulatory discussions about these data to inform the pivotal development in second line? Will that be before or after the data are presented to investors?
I mean, I think our goal at the moment is just to gather the data, let it mature, both at the 600 mg and at the other doses that we're enrolling, and when we feel like we have a robust dataset to engage the regulator and get that feedback. I don't think we'll have it in hand as we do our data disclosure in the second half of the year. I think we'll let the data mature before we engage the regulator.
Okay. Thinking ahead to that, though, do you think it's reasonably likely that capivasertib is the new regulatory bar for second-line mutant HR positive metastatic breast cancer?
I mean, obviously, we engage with a lot of treating physicians in the field, and I think the answer probably to that question is our best guess today is yes. There seems to have been, you know, a broad and rapid uptake given the fact that it does avoid some of the challenges of the current approved therapy, alpelisib in the second line, and the kind of high rates of grade three hyperglycemia. I know the treating physicians really don't want to be involved in the treatment or, you know, taking insulin with these patients. And they trade that for potentially the lower PFS that capivasertib has, but it obviously comes without the grade three hyperglycemia.
Now, in the real world, we're also hearing that the rates of hyperglycemia in grades one and two are significantly higher than were seen in the study, given the four days on, three days off regimen. And obviously, the high rates of diarrhea are a challenge. And so I do think for us, the goal would be to obviously have a more tolerable safety profile and keep that grade three hyperglycemia down and obviously try and beat that five and a half months PFS that we see with capivasertib. And that's probably the most likely trial comparator for us as we think about a pivotal trial in the second line.
Okay. Now, there is an ongoing second-line study of Roche's inavolisib versus Novartis' alpelisib. So Roche's suggesting that control under still feasible. Now, it started before capi was available. But what are your thoughts on this study and any commentary on the trial design, patient criteria considerations, etc.?
Roche was, I think, very smart in the way that they progressed development of inavolisib, focusing on a frontline, you know, higher-risk patient population first, where you could do some metabolic entry criteria that really only favored the inavolisib arm because you didn't have an active pathway comparator in that arm. So it was very, it makes a lot of sense to us as to how they progress these studies. And then, you know, bring the riskier second-line study forward where any of this metabolic subsetting that they're doing will now favor both inavolisib and alpelisib. And so I think the sequencing that they've chosen to do was made a lot of sense given that when you looked when you did cross-trial comparisons of the data that existed, it didn't look that different from alpelisib, generally speaking.
So I think that we, you know, that this second-line study probably has, you know, more risk to it than that front-line study had. But, you know, alpelisib, you'll have a period of time where alpelisib is approved outside of the U.S. and capivasertib is not in all geographies. And so I do think they'll be able to enroll that study. It may, again, be leaning towards ex-U.S. patients, but that's pretty similar to what they did in INAVO122 to get into healthier patients and try to mitigate some of the wild-type toxicity.
Got it. Okay. And can you expand a bit more on the inavolisib front-line success that you mentioned? I mean, what has that validated to you? And what is the residual opportunity that's left for a mutant selective inhibitor?
I mean, we were thrilled to see the data that Roche presented at San Antonio just before the end of last year. I think if you go back in time, there were plenty of questions around, is this a therapeutically relevant target in the front line, you know, given that the agents already in the front line use have been very effective. And I think this study, when it reported out, showed clear effectiveness of almost doubling the PFS to 50 months in the treatment arm versus the control arm. So I think that question is unequivocally answered, which is mutant selective PI3K inhibition is important.
I think for us, obviously, there is also a clear window here, which is the way that Roche designed that trial was obviously in endocrine-resistant hormone receptor positive HER2 negative patients, which we think is about 40% of the patients that are in that patient segment. They chose metabolically very healthy patients in the criteria to enter the study reflected that. So we, again, you know, as they presented the data themselves in San Antonio, I think one of either the discussant or the presenter showed that that was probably close to excluding, you know, close to 50% of patients in the U.S. So if you put these two things together, you probably exclude up to 80% of patients.
And so I think it provides us with confidence that the target is now clearly therapeutically validated, and there's still a large segment of patients that will not, unfortunately, be served through the trial that was designed. And so I think we feel confident that, you know, with RLY-2608, we may have a solution for these patients that will remain having a significant unmet need.
Okay. And how are you approaching early-stage clinical development as a triplet with a CDK4/6? And what's your timing of first data?
Yeah. So we, as we announced at JPMorgan earlier in the year, we did say that we've initiated the triplet combination, so ribociclib plus fulvestrant plus 2608. It's early days in that study, as you can imagine. The goal would be to try to identify a safe and tolerable dose to combine with ribo and fulvestrant. None of the other non-selective inhibitors have been able to establish a dose at ribo's full dose in the metastatic setting, 600 mg once daily. So we are going to be pretty deliberate and diligent about, you know, exploring that in the triplet setting. And we've guided to showing some initial data from that exploration in the second half of the year. But that'll certainly be focused on safety tolerability and will inevitably be small end.
Okay. As we all rush to, excuse me, compare the safety, how do the metabolic parameters of the different studies impact how we should be looking at hyperglycemia rates?
You know, as Sanjiv alluded to, at least as it pertains to INAVO120, you have to be careful about trying to when you start to do cross-trial comparisons, the metabolic entry criteria is very important. In an INAVO120, the HbA1c entry criteria was less than six, and the fasting blood glucose was less than 126. They held that parameter essentially constant as they moved into the second-line phase III too. They relaxed the HbA1c to 6.4. You know, our current entry criteria for ReDiscover is 140 on the fasting blood glucose and seven on the HbA1c side, which we feel is going to capture a large portion, if not all, of the non-actively managed diabetic patients in the U.S. with metastatic breast cancer with PI3K alpha mutations.
So it's going to be very important as we compare these datasets, these cross-trial datasets that we understand the metabolic entry criteria. I think, again, our goal here is to push the envelope on that balance of driving efficacy while minimizing grade 3 hyperglycemia. That is really the dose-limiting toxicity on the hyperglycemia side, which is when you fall over into the grade 3 side. And, you know, as you saw with the INAVO120, they were able to minimize that grade 3 hyperglycemia.
But again, we think that had a lot to do with not only the metabolic entry criteria, but general better fitness of patients that they were able to enroll into the study younger than what you would see on average in the Western patient population, BMI being lower, all of these parameters influenced the ability for Roche to generate the data as you can see in INAVO120.
Yeah. Okay. Now, do you hope to be able to interpret any initial efficacy data from this triplet being that it will be in these second-line patients?
Generally, the answer to that is no, not really. I think, you know, these are going to be second-line patients who have seen prior CDK4/6. Most likely, a lot of them will have seen ribo given ribo's increasing uptake in the front-line setting. And they're going to be essentially rechallenged with ribo again. There's some data suggesting that you can get efficacy from that in rivo retreatment. But I wouldn't say there's a lot of robust interpretable data suggesting what the understanding of that additional benefit is. So I don't think, you know, you'll be able to make some initial cross-trial comparisons, but the end probably won't be the end and follow-up probably won't be robust enough to really interpret efficacy that meaningfully.
Okay. You mentioned the patient selection element of the inavolisib study. I'm really looking at a younger patient population. I think everyone was struck by how poorly the control arm did, just being seven months. If you enrolled a more representative U.S. patient population into a front-line study, even if it had the endocrine-resistant parameter to it, would you expect seven months, or did the data point to that potentially being a bit higher?
The current data point to that being a bit higher. We would have the palbo/fulvestrant precedented data, which there's not a ton of, admittedly, but there is some that would have suggested 10-12 months median PFS for that palbo and fulvestrant regimen. As you move into what we think is probably more commonly used in the U.S. in those patients, which is ribo plus fulvestrant, you know, there you have some studies suggesting 18-20 months of median PFS. So again, you know, very, very smart selection of control arm. We'll see how the FDA treats that in a labeling conversation. But I think that is probably the most interesting next step actions to see is how the FDA, again, you know, interprets that dataset when it comes to labeling.
Yeah. And if you're to follow in the path of endocrine-resistant front-line population pursuit, do you see this aspect of potentially having a higher control arm hurdle as a positive or negative for the opportunity of your drug?
So I think the fundamental thing we continue to gain confidence in is that however contrived a patient population that an INAVO120 went into, you can't argue with the magnitude of benefit seen on top of CDK4/6 and endocrine therapy. And I think that tells us that as long as we maintain a meaningfully differentiated safety profile to allow patients to comfortably and with good quality of life stay on a triplet regimen for longer periods of time, you know, that should result in that same mechanism benefiting us just as much, if not more, by the safety or benefit where we should continue to show against an INAVO120. So whether that control arm is seven months, 12 months, 15 months, it certainly influences trial sizing considerations and length of time and cost.
But I think that fundamental learning from an INAVO120 stands no matter what, which is that this mechanism on top of CDK4/6 and ET looks to be pretty meaningful for patients.
Got it. Okay. Then given you might not be restricted by the metabolic considerations, do you think that provides any flexibility when it comes to regulatory discussions of how that front-line trial is designed down the road, you know, thinking that inavolisib will become a U.S. standard of care likely this year based on Roche's guidance?
I think that last piece of your question is critical in understanding to what extent and in what set of subset of front-line patients is an INAVO120 going to be labeled for that front-line patient population. I think that's the big piece we need to understand. You know, we have a little bit of precedent of how the FDA has treated this in the capivasertib experience. You know, they showed a benefit across all patients, but they were given the narrower label on AKT pathway patients only. And so clearly, the FDA, we think, is going to be, you know, is going to look long and hard at the entry criteria. And we would expect a label that is more consistent with that versus just getting a broader generalized label. But that's the biggest open question we have right now.
Okay. I know we're awaiting data updates for your selective FGFR2 inhibitor as well. That strategy has evolved over time. I'd love to know comments from you about how that's evolved to be synergistic with PI3K as you think about moving towards a potential commercial company with those assets down the line.
I'm saying that one.
Yeah, sure. So I think, you know, there's no, obviously, for 4008, that is largely tumor-agnostic approach. Obviously, breast cancer, HR-positive breast cancer, we saw an interesting signal in early data there. But I think, generally speaking, there's not a lot of overlap between the commercial footprint or commercial strategy for these two agents, which probably really leads to, you know, the reason behind the commentary we've been consistently giving, which is we're going to take a disciplined approach as to what to do next with 4008, given the, you know, the recent history we have with other agents in tumor-agnostic labels and the challenges that exist there commercially and the size of the opportunity in a very focused and well-precedented commercial footprint in breast cancer.
You know, these two things, in addition to the rest of the pipeline, are going to heavily weigh into our decision as to what to do next with 4008 as we get into the back half of the year.
Okay. And you have a commitment for new target disclosure this year as well. I guess what flavor of target disease type should investors expect?
I mean, look, we've been saying for a few years now, we've been working on at least seven other preclinical programs. And we've kind of hinted to see that we've moved from oncology into genetic disease and from small molecules into chaperones and degraders. And so, you know, our hope is, obviously, that's the envelope of things it could come from. And we hope that it is, you know, an exciting target that I think investors will be pleased that and we've used our platform to solve a problem that others have not been able to solve. And so we look forward to the disclosure.
Okay. Any preliminary thoughts on timing or form for this relative to the big data updates we've got coming from in the second half?
Yeah. If we had our druthers on it, we would want to keep them decently spaced out from each other. And, you know, most likely, some kind of focused company event would be the best form to be able to disclose a new program so we can put some context around it.
Okay. Great. Well, I look forward to that. Sanjiv, Pete, thanks so much for joining us. Really appreciate it.
All right. Thanks very much. Bye.