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

Good day, ladies and gentlemen, and welcome to the Relay Therapeutics Corporate Update Call. As a reminder, this conference call is being recorded. I would now like to introduce your host for today's conference, Mr. Pete Rahmer, Chief Corporate Development Officer at Relay Therapeutics. Sir, you may begin.

Thank you, operator, and good morning, everyone. Thanks for joining us during a busy SABCS. We are excited to share these updated data for RLY-2608 in combination with fulvestrant with you today. You can access the press release from today, the slides we are reviewing and a replay of this call by going to the Investor Relations section of our website. As a reminder, during this call, we will make certain statements that are considered forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995, including expressed or implied statements regarding our strategy, business plans and objectives, the expected therapeutic clinical benefits of our product candidates, the potential of our platform or product candidates and progress, timing and execution of our clinical trials. Such forward-looking statements are not guarantees of future performance, and therefore, you should not put undue reliance upon them. These statements are subject to numerous risks and uncertainties that could cause actual results to differ materially from what we expect. I refer you to our SEC filings on our website for a discussion of these risk factors. The forward-looking statements in this presentation speak only as of the original date of this presentation, and we undertake no obligation to update or revise any of these statements. I'm joined today by Sanjiv Patel, our CEO; Don Bergstrom, President of R&D; and Dr. Sarah Sammons, who is an Assistant Professor of Medicine at Harvard and the Associate Director of the Metastatic Breast Cancer Program at the Dana-Farber Cancer Institute. Dr. Sammons is a medical oncologist specializing in the care of patients with all stages of breast cancer, and she will be providing context to the treatment landscape for patients with PIK3CA-alpha mutated HR-positive HER2-negative breast cancer. I'd now like to turn the call over to Sanjiv.

Thanks, Pete, and thank you to those of you joining our call today. Before getting into the updated 2608 data, let's start by stating PI3K-alpha represents a very significant commercial opportunity and one of the largest unaddressed areas left in targeted therapies with more than 0.5 million PI3K-alpha-mutated patients in the U.S. alone, spread across hormone receptor positive, HER2-negative breast cancer, vascular malformations and a number of distinct solid tumor types. These areas remain poorly addressed despite several approved therapies. A significant toxicity of non-selective agents limits their efficacy and makes it difficult for them to be combined with other agents. We believe we will address these liabilities by using our Dynamo platform to create RLY-2608, the first allosteric pan-mutant, isoform-selective inhibitor of PI3K-alpha. Our initial focus for 2608 development is in breast cancer, where there are more than 140,000 PI3K-alpha-mutated patients in the U.S. alone. We believe the profile of 2608 continues to demonstrate that it can become the backbone therapy for all PI3K-alpha-mutated patients across all lines of therapies regardless of the combination partner. This represents a significant multibillion-dollar commercial opportunity. Today, we are presenting updated 2608 data at the SABCS conference. These data are in heavily pretreated metastatic hormone receptor positive, HER2-negative patients with a median follow-up that is meaningful at 9.5 months. These RLY-2608 data show a clearly differentiated safety profile that translates into greater efficacy. Across second line plus patients, we see a 9.2-month median progression-free survival, a 67% clinical benefit rate and a 39% confirmed objective response rate. And in true second-line-only patients, we see a double-digit median PFS of 11.4 months, all of these numbers compare very favorably with the approved nonselective agents such as capivasertib, where in an analogous population, you see a median PFS of only 5.5 months. Collectively, these data give us great optimism that 2608 will be able to demonstrate clinically meaningful improvement in PFS against the newly emerging standard of capivasertib. We anticipate the second-line pivotal study to begin next year. In addition to the second-line pivotal trial we plan to initiate next year, we also continue our efforts to demonstrate 2608's first-line potential by continuing to dose-escalate in the ribociclib triplet trial where we are dosing at biologically active doses of 2608 and in the newly initiated triple combination with Pfizer's novel selective CDK4. Earlier this year we had hoped to share data from the ribo triplet arm in Q4, but it's still too early to do so. We continue to push forward our triplet trials in both arms that will generate the data that can support the use of 2608 in earlier lines, and we'll give updates on the future when we have robust and interpretable data to inform next steps. Finally, on the genetic disease front, our vascular malformations team is on track to initiate the 2608 trial in the first quarter of 2025. Now, as we move to discussing our breast cancer plans and data in more detail, I'd like to hand it over to Dr. Sarah Sammons.

Good morning, everyone. My name is Dr. Sarah Sammons, and I'm a breast medical oncologist and clinical researcher. I started independent practice 7 years ago at Duke University, where I began caring for patients, and I also led the clinical trial portfolio there. I was recruited to Dana-Farber Cancer Institute in 2022 to oversee the metastatic breast cancer program. I do clinical research, clinical trial research and drug development. And I also see patients in the clinic. I do about 375 consultations for breast cancer annually, of which about 1/3 are metastatic patients that many remain in my longitudinal care. So really dynamic and exciting time to be a breast medical oncologist and to be treating patients with metastatic breast cancer as we have more options for them than we've ever had before. Roughly 20% to 30% of all early-stage breast cancer patients will become metastatic at some point. And the vast majority of these patients have hormone receptor positive, HER2-negative disease. Once metastatic disease develops, treatment is not considered curative anymore, but it's considered palliative. And patients with hormone receptor-positive disease live on average for about 5 years. The goals of treatment in these patients are to control their disease for as long as possible and in doing that extend their life, but also to preserve their quality of life, which is really key. Survey data in these patients with metastatic disease shows that they're just as concerned about quality of life and side effects of medications as they are about their disease progression. And so I really do feel strongly about that. Now to focus your attention on the slide, which shows the treatment paradigm for the first 2 lines of therapy for patients with hormone receptor positive, HER2-negative metastatic breast cancer, for the first-line population, treatment is bifurcated by the patient's endocrine status. And we either consider them to be endocrine-sensitive or endocrine-resistant based on certain criteria. Patients in the first line that are considered endocrine-sensitive are generally treated with a CDK4/6 inhibitor and an aromatase inhibitor, and this typically provides patients about 2 years of progression-free survival on average. This group is considered to have favorable prognostics, and many of them do quite well. For the first-line endocrine-resistant population, meaning that they've progressed already on endocrine therapy, probably in the adjuvant setting, CDK4/6 inhibitor plus fulvestrant, which is a selective estrogen degrader, is the general recommendation, and this gives patients about 12 to 18 months of progression-free survival. We recently had FDA approval of PIK3CA inhibitor in avolisib added to palbociclib and fulvestrant in the endocrine-resistant PIK3CA-mutant setting. And I think of this trial as more of a proof of concept and certainly the first approved triplet. But I don't think it's something that's going to acutely impact practice or the community at large for a couple of reasons. And avolisib still has a pretty significant toxicity profile. The triplet is fairly toxic, and the approval is in a pretty narrow subset of patients. Also, there are some issues with hyperglycemia, so it will be limited to a very metabolically fit population. So I don't think this will be a large portion of market share at all, but I do think of it as proof of concept for a triplet. Moving on to the second-line post-CDK4/6 setting. I think this really remains the largest area of unmet need. Treatment options here depend on genomic profiling of the patient's tumor. So once they progress on a CDK4/6, we're either sending circulating tumor DNA or sending their tumor for sequencing to identify mechanisms of resistance, namely a PIK3CA mutation, AKT mutation, PTEN alteration, ESR1 mutation. And then we're tailoring the patient's therapy based on our findings. Existing treatments in this setting are not optimal. Everything approved in the space that you see on this slide has a progression-free survival of less than 6 months, and the combinations are quite toxic. Focusing specifically on what we have approved for the PIK3CA-mutant population, which is about 40% of our patients, the 2 existing agents are alpelisib and capivasertib that are approved in the second line. And both of them are fairly challenging to manage in the clinic due to their toxicity profiles. And I think that the bar is actually pretty low here for improvement in a PIK3CA inhibitor that has better efficacy and tolerability for our patients. The most recently approved agent, which is an AKT inhibitor, was approved for PIK3CA, AKT and PTEN-altering patients is capivasertib. Everybody was very quick to accept this over alpelisib just because how toxic alpelisib is, but capivasertib still has substantial diarrhea, rash, nausea, mouth sores and is challenging for our patients. So all of this is just to highlight the low bar and the absolute hunger in the field for a mutant-specific, better-tolerated, more efficacious PIK3CA inhibitor, which would be widely accepted and appreciated. Moving on to the next slide. We just wanted to make note of the breast cancer therapeutic landscape and validate how fast pace it's moving with significant innovation almost on a yearly basis. And while this can make things seem confusing and raise the question about new treatments becoming quickly obsolete. One thing that I do think will remain constant in this evolving field is the need for a differentiated PIK3CA pathway agents. We always are going to want to identify the mechanisms of endocrine resistance and then target that. And PIK3CA is a foundational mutation in about 40% of these patients, and we know that it drives resistance to endocrine therapy and that if we can target it, we can prolong progression-free survival. And while the novel endocrine backbones for these PIK3CA inhibitors or even triplet or doublet strategies may very well come along and likely will. Local targeting of this pathway is not going to go anywhere and will always remain relevant. I do think that this is where combined ability of any novel PIK3CA inhibitor with either novel estrogen-targeting agents, which there are many coming down the pike or even triplet combination becomes critical because a PIK3CA inhibitor will always be combined with something. We never give it as monotherapy in hormone receptor-positive disease. So whether in the future, we want to combine it with a SERD, a SERCA, what have you, combinability of these agents will really differentiate them and provide an advantage moving forward. I've been really fortunate to be involved with the RLY-2608 journey as an investigator from the beginning and the early data that we're observing for RLY-2608 in the ReDiscover study are encouraging, both in terms of tolerability and efficacy. My patients on trial have generally done quite well. I'm optimistic that this compound could change the treatment landscape and progress things for patients in need of a better PIK3CA inhibitor. Thank you so much for having me on this early morning. And I will now pass it over to Don to review the data.

Thanks, Dr. Sammons, for providing that overview. I will now share updated results from the hormone receptor-positive HER2-negative breast cancer patients treated with 2608-plus fulvestrant in the ongoing ReDiscover clinical trial. As previously disclosed, in the doublet arm of the trial, we initially dose escalated from 100 milligrams BID up to 1,000 milligrams BID. No MTD was identified. The totality of the data, including PK, PD, safety and early efficacy suggested 600-milligram BID was a potentially optimal dose for the RP2D, and this is the dose we will focus on today. As previously reported, and as would be expected in a Phase Ib clinical trial, patients were heavily pretreated, with about half having received 2 or more prior lines of therapy in the metastatic setting. 2608 Phase III development will focus on patients with any PIK3CA mutation, either kinase or non-kinase, but will exclude patients with PTEN or AKT co-mutations based on our understanding of pathway biology in 2608 mechanism. At the RP2D, we treated 52 patients with this genotype and observed a median PFS of 9.2 months with a landmark 6-month PFS of 66% and a 9-month landmark PFS of 63%. With median follow-up of 9.5 months, we have increasing confidence in the robustness of this interim results. Additionally, CBR has improved to 67%. We have also observed a median PFS of 11.4 months as data mature in second-line patients who received a CDK4/6 inhibitor plus an AI or fulvestrant as their prior treatment in the metastatic setting. The 6-month PFS of 79% and 9-month PFS of 72% are indicative of sustained clinical benefit for the majority of these second line patients. Additionally, reductions in tumor dimensions have been observed in 74% of patients, and 12 of 31 patients with measurable disease achieved a confirmed objective response for a confirmed ORR of 39%. Tumor regression and multiple objective responses have been observed in both patients with PIK3CA-kinase domain mutations and patients with non-kinase domain mutations, consistent with the pan-mutant mechanism of 2608. Consistent with the observation of clinical activity across PIK3CA mutation subtypes, there was robust clearance of circulating mutant PIK3CA DNA in both kinase and non-kinase patients. Nearly all patients treated with the RP2D of 2608 showed reductions in ctDNA with more than half of patients completely clearing the mutant ctDNA clone. Given others' efforts to target PI3 kinase domain mutation specifically, we are also showing a subgroup analysis in these patients. As you can see on the waterfall plot, 2608's efficacy in kinase-domain patients was consistent with the activity observed in the overall population of PIK3CA-mutated patients with a confirmed ORR of 67%, CBR of 79% and median PFS of 11.4 months. One of the hallmark toxicities of PI3K-pathway-targeted therapies has been hyperglycemia. For 2608, we maintain continuous daily dosing and greater than 1/3 of study patients had baseline BMI of at least 30 and/or hemoglobin A1c of at least 5.7%, a population shown to be at increased risk of hyperglycemia in the Phase Ib trial of inavolisib. And we see very little perturbation of glucose levels at the RP2D as shown in this graph of being glucose versus time for each dose level. Overall, 2608 plus fulvestrant continues to be well tolerated. There were no observed Grade 4 or Grade 5 TRAEs. There were only 2 cases of Grade 3 hyperglycemia at the RP2D, and the majority of hyperglycemia AEs were Grade 1, requiring no medical management. Rates were low for AEs, commonly seen for the PI3K pathway class, including diarrhea, rash and stomatitis, with Grade 3 rates for these AEs in the low single digits or 0. The favorable overall tolerability profile is associated with the ability to maintain dose intensity, which is critical for driving long-term clinical benefit. Median dose intensity at the RP2D was 94%. Dose interruptions when they occurred were brief, and only 2 of 64 patients discontinued treatment due to TRAEs. The majority of discontinuations were due to disease progression. Putting these data in the context of existing nonselective PI3K pathway inhibitors, we continue to be encouraged by 2608's potential efficacy differentiation. As you can see in these cross-trial comparisons, 2608 continues to demonstrate the highest PFS, CBR and ORR, ever seen from a PI3K pathway inhibitor despite being in a heavily pretreated Phase Ib population. These data further improve when you look only at second-line patients, which, as we've shown, currently have an 11.4-month median progression-free survival, more than double the PFS, seen in an analogous patient population treated with capivasertib. The 2608 monotherapy arm in the ReDiscover trial has been a lower priority for us. So we have only dosed 3 hormone receptor-positive HER2-negative patients across all doses with RLY-2608 monotherapy, but we have still seen activity that is consistent with a very strong combination data we've shown, which is the relevant treatment setting for hormone receptor-positive breast cancer patients. While there are a number of other PI3K-alpha-driven tumor types, including head and neck cancer, where we have seen a confirmed partial response, we have prioritized development of 2608 in the largest patient populations, starting with hormone receptor-positive breast cancer, followed by vascular malformations. While there are starting to be other PI3K-alpha-selective agents entering the clinic, 2608 is the furthest along and is the only asset to have generated a robust clinical efficacy and safety dataset to this point. We will leverage our strong balance sheet, improving clinical trial expertise to move forward as quickly as possible, and we'll begin the second-line pivotal study in 2025 pending regulatory alignment. Next, moving to our triplet combinations, which have the potential to build on the second-line opportunity with the doublet and access a large patient population in earlier-line settings. We are currently enrolling patients across 2 triplet cohorts, 1 with ribociclib, where we are dosing a biologically active doses of 2608 and one with atirmociclib, Pfizer's investigational selective CDK4 inhibitor, which has just been initiated. We recognize that the treatment landscape is evolving quickly, so running a study with a novel agent like atirmociclib positions us well to stay at the forefront of these emerging new medicines. Consistent with the doublet arm of the study, these arms are also enrolling heavily pretreated patients who have been treated with at least one prior CDK4/6 inhibitor. We look forward to generating data, which demonstrates 2608's broad combinability and ability to address patients across multiple lines of therapy. In addition to the large opportunities for 2608 in breast cancer, we're also excited to explore its potential in PI3K-alpha-driven vascular malformations. Vascular malformations is a blanket term for a variety of conditions that affect the development of lymphatic and/or blood vessels. Here, like in cancer, PI3K-alpha mutations are a major driver in a large portion of these patients, including approximately 170,000 patients in the United States. They are split into different categories. First is PROS or PIK3CA-related overgrowth spectrum, which is where alpelisib has accelerated approval. These are generally patients with a very severe disease. Then there are lymphatic malformations, venous malformations and cerebral cavernous malformations, all of which have no approved systemic therapy. We believe 2608's mutant selective profile is uniquely positioned to inhibit the driver of the disease while sparing patients' toxicities from non-selective agents like alpelisib. As in oncology, lower toxicity should allow for greater target coverage leading to greater efficacy. Additionally, vascular malformations affect children and will likely require chronic treatment, so having a cleaner safety profile will be even more important. Altogether, this is a significant opportunity to address patients in need, and we're excited to start the study with 2608 in the first quarter of 2025. We will start randomized dose finding in adults and adolescents with PROS and other PIK3CA-driven vascular malformations. Here, we will leverage the large existing safety database we've generated in oncology to quickly evaluate 3 dose levels in parallel. One will reflect the oncology RP2D and the other 2 will be at other biologically active doses. As we gain experience at different dose levels, we will open additional brief dose escalation cohorts for children under 12. Given that PROS has a high burden of disease and is the only subtype where data exists to compare to, we will look to enroll PROS patients early in the trial to demonstrate clinical proof-of-concept in vascular malformations. The primary objective of the study will be to identify a safe and biologically active dose of 2608. And the secondary objective will be to evaluate efficacy by radiographic lesion regression, which was the primary endpoint used by alpelisib for its accelerated approval. Critically, we will also evaluate quality of life measures using patients and caregiver reported outcomes. We look forward to providing additional updates once we have the trial up and running next year. I'll now turn it over to Pete to wrap up.

Thanks, Don. As you've heard from Sanjiv and Don, the 2608 data continue to give us great confidence in its ability to address multiple distinct and significant PI3K-alpha opportunities, and we are moving forward to address them as quickly as possible. Our confidence is based on robust clinical data that support the therapeutic hypothesis underlying 2608's design. With more than 9 months of follow-up, we see that 2608 selectivity leads to a differentiated safety profile compared to current therapeutic options in a population representative of the typical western oncology patient population, including those with high BMI and/or elevated baseline HbA1c. 2608 continues to demonstrate a differentiated safety profile across the key AEs that have limited the efficacy of nonselective inhibitors in this class. These low rates of key AEs as well as the absence of unexpected idiosyncratic toxicities leading to DLTs are all indicative of a tolerable and combinable molecule that is well positioned for the breast cancer treatment paradigm of the future. As Don covered, we see that 2608's differentiated selectivity and safety drives to superior outcomes for patients, which underpins our confidence as we move towards the pivotal study next year. Looking across the metastatic breast cancer landscape, we believe 2608 can address significant medical needs for patients across lines of therapy. As we heard from Dr. Sammons earlier, this need is highest in the second line, where despite the introduction of newly approved drugs, physicians lack a durable and efficacious option for patients following CDK4/6 therapy. With more than 13,000 patients in the U.S. annually, the second-line market has the potential to grow significantly with the improved duration of benefit that 2608 could provide, especially given the median PFS we are seeing in the second line, which is greater than double that of the current standards of care. Additionally, there is the potential for meaningful improvement for patients in the frontline setting, which would also translate to significant expansion of the current market opportunity. And our 2 ongoing triplet arms have us well positioned to advance into that setting in the near future. Looking across our pipeline outside of PI3K alpha, earlier this year, we announced 2 first-in-class programs targeting Fabry and NRAS. We are excited to continue advancing these programs toward the clinic. Altogether, these programs make up a focused precision medicine portfolio spanning oncology and genetic disease, which addresses large potential commercial markets. Looking ahead over the next year, we have a number of important catalysts, including the initiation of the pivotal trial in second-line breast cancer, complete Phase I/II doublet data as well as new trial starts in vascular malformations, Fabry and NRAS. Given that we have the significant capital needed to accomplish these goals as well as proven team, we are confident in what the next year will bring. We also remain diligent on how to deploy our team and capital. We will always maintain the balance sheet flexibility to confidently execute the second-line pivotal trial and have multiple strategic levers to extend our cash runway. Thank you again for joining us today, and we'll now open up the call to questions. Operator?

[Operator Instructions] And our first question will come from Salveen Richter from Goldman Sachs.

This is Tommie on for Salveen. Congrats on the data. Just 2 here. So you mentioned that the triplet data was a little bit too early to share, just wondering if you could give any high-level safety commentary about on here? And secondly, with the Phase I/II full data that you plan to show next year, any framing you can provide here, maybe in terms of follow-up would be helpful?

Thanks, Tommie, for the question. I'm going to hand it over to Pete to answer, I think, both of them.

Yes, thanks. On the triple front, just too early to give a robust update today, continue to be happy with the progression of both the ribo and triplet and excited to be able to announce today that we've initiated the atirmociclib triplet with CDK4.

The timing of the Phase I/II full manuscript.

Yes. Well, let's get to full maturity of that dataset, and then, we'll be able to fine-tune the timing of when we could put that out, but we would anticipate that to be a 2025 event for sure.

Our next question will come from Brad Canino from Stifel.

Nice to see the update. 2 quick ones for me. First, the Grade 3 hyperglycemia rate at 3%, is that enough reduction from the rate of the wild-type inhibitors where physicians can now prescribe without needing easy access to an endocrinologist because that was one of the pain points whenever we asked about alpelisib? And then second, how much confidence should we place in the second-line PFS at 11.4 months? It's a relatively short follow-up limited patient number at risk after 9 months. Is there something you see in the granular data that gets you to think that is the right number as the proxy for your planned pivotal trial versus the 9.2 months in all patients?

Thanks, Brad, for the questions. Maybe we'll take the last one first and then -- so maybe, Don, you can take the question around the 11.4 months, how robust it is?

Yes. I think, Brad, we are confident in that 11.4-month number. We do have some patients who are censored before the median. The bulk of those patients who are censored prior to the median are patients who are ongoing in therapy, many of them ongoing in response. So we have confidence as those patients continue to mature that they will continue to support the 11.4-month number.

And then your first question that is around Grade 3 hyperglycemia, the rate of 3%. And obviously, as you know, it's a heavily pretreated patient population that we have. A 1/3 of those patients have a BMI over 30 or high hemoglobin A1c, and so our understanding is that, that is the metric that oncologists are looking for. But given we have an oncologist on the call today, maybe I hand it over to Dr. Sammons.

I can say in the clinic that we're comfortable managing hyperglycemia at low grades. And as long as we can prescribe a low-carb diet into metformin and it's manageable, then that is a drug that we will all be comfortable with, and that's been my experience to date with RLY-2608. With alpelisib, the problem was that they still have hyperglycemia despite the metformin, and then, we had to get them into an endocrinologist, which took months, and that was really cumbersome on us in the clinic. But most of the hyperglycemia that we're seeing is well managed with low-carb diet and metformin.

And our next question will come from Peter Lawson from Barclays.

Great. I guess a key question for us is just thinking about the potential to get an approval in both the kinase domain as well as the helical domain. Or is there route where you would kind of separate the 2?

I mean, as you can see today, the metrics that we show, in 9.2 months' worth of PFS overall, all patients, 11.4 months of PFS in the second line, true in our population. So to answer the kind of higher-level question, we feel very confident against the current standard of care that we are going to go up against capivasertib, where it has a benchmark of 5.5 months that we feel very confident to run that head-to-head trial. Now the difference, as you rightly point out, we are showing slightly better numbers in kinase. In the all-kinase patients, we're showing an 11.4-month PFS across all patients. And so I think that's probably what's prompting the question. I think on the non-kinase, it's not too far from the 9.2 overall and still significantly better than the 5.5 months that we see for capivasertib, which is the current standard of care, so we feel pretty confident.

Our next question will come from Sean McCutcheon from Raymond James.

Just to piggyback on Brad's question a bit. You did see another case of Grade 3 hyperglycemia at 600 mg. Can you provide some context around that additional case and go into how you're thinking about the wide metabolic inclusion criteria you stated on the slide for the Phase III study and any flexibility on a higher A1c?

Yes, you're absolutely right. We saw one extra, so we're at 3% now, which is obviously, just to put it in context, we're seeing Grade 3 hyperglycemia in the 20s for the nonselective agents. But maybe, Don, you could provide the context on...

Yes, it was a patient who's benefiting from 2608, had been on drug for about 10 months and the investigator chose to switch the patient to insulin from metformin, which triggered Grade 3 AE, but again, I think we're very comfortable with the profile we're seeing. We're comfortable with the IE criteria that we have that allows prediabetic patients on study. We're allowing patients with baseline fasting plasma glucose up to 140, baseline hemoglobin A1c up to 7. And as we move forward into Phase III, we anticipate that Phase III inclusion criteria likely will mirror those criteria that we're using in the current study.

And our next question will come from Michael Schmidt from Guggenheim.

Just looking at the planned Phase III study, I guess based on the enrollment criteria, the inclusion/exclusion criteria, how would you expect the patient population compare in the Phase III relative to this Phase Ib study? And would you expect to enroll any PI3-kinase inhibitor pretreated patients? Or will they all be naive based on the current landscape?

Thanks for the question, Michael. I'll hand it over to Don to answer.

Yes. So I think as we are designing the Phase III trial, and I'll caveat this, that we've not finalized the design at this point. But what we'll look to do is use inclusion/exclusion criteria that largely will mirror the ReDiscover patient population, but will be biased towards enrolling a less heavily pretreated patient population. So we're running right now with about 50% of patients who are in the third line or later. We would look to have a patient population that would be more reflective of what we saw in the CAPItello-291 trial that was closer to 80% second-line patients, 20% third line and beyond. With regard to prior PI3K inhibitor therapy, it is our plan to continue to exclude patients who have received prior treatment with the PI3K inhibitor.

And our next question will come from Jason Gerberry from BofA.

Just on Phase III, just curious your evolving thoughts now that you're seeing a more, I guess, almost a doubling, right, of PFS in the second line-only population. How that may be impacting your thinking on Phase III trial design, sizing, powering, et cetera? Just curious if that's something now that you're seeing a greater margin of delta that's impacting your thoughts there? And then for the oncologist on the call, just curious to get your perspectives on 2608 tox profile versus capi and any practical constraints that you see around the capi Grade 3 AE profile that obstruct usage and maybe how you stack the 2 up on toxicity?

Thanks, Jason, for the questions. So maybe, Don, you could take the first one around how does the -- almost, I think, Jason, in your words, doubling of PFS implies just on the trial design?

Yes. I mean I think it's always been our plan, Jason, to design a Phase III trial that will be powered to demonstrate a clinically meaningful improvement in PFS for 2608 relative to the comparator. And I think with the data that we're seeing, that will continue to be the plan. This will be designed as a clear superiority trial. And I think with the data we're seeing today, we continue to have growing confidence in the likelihood of having a positive result in that study.

And then, Jason, on your second question around the toxicity profile of capivasertib in real life versus some of the early experience that Dr. Sammons has on 2608. So I'll hand that over.

Yes. I think it's a really good question. We're using capi in the clinic, and we focused our discussion about talks around hyperglycemia. But with capi, the bigger problem is actually the 80% diarrhea risk, the 30% rash risk, which is actually really substantial. I mean, it's like the patients break out in a full body rash. The mouth sores that are fairly comparable to everolimus actually. And so -- with capi, we're giving all kinds of prophylactic medication, and it's still a bit challenging to manage, and then, it's only a 4 days on, 3-day off dosing schedule. RLY-2608, there is minimal to no GI toxicity. We're not seeing mouth sores. We're not seeing rash. So that's on a day-to-day basis. That is a huge positive for patients because those are the things that are really bothersome. And I think so far, the hyperglycemia, it's their low grade, but it's been manageable.

And our next question comes from Eric Joseph from JPMorgan.

[Technical Difficulty] Just curious to see [Technical Difficulty] variation in PFS to [Technical Difficulty] by baseline measures of what we used to have, whether the composition of the share of [Technical Difficulty].

Eric, this is the operator, you have a lot of static coming from your line. It is very hard to understand.

It's slightly better. I think we understand the question around understanding baseline characteristics versus efficacy metric. So maybe Don you could try and cover it. Is that right -- is that what the right question is, Eric?

I'm wondering if you're seeing much variation in PFS by baseline [Technical Difficulty] and whether the [indiscernible] patients with baseline everybody views that the second line type data similar to the overall population?

I mean I think I will just take a stab at what I think the question was. In general, as we look across different baseline characteristics, whether it's measurable disease versus non-valuable only disease. We're generally seeing the same trends in PFS. And I think across the populations, we would have confidence in having superiority compared to the comparator we're proposing for our Phase III.

And our next question will come from Silvan Tuerkcan from Citizens JMP.

This is Josh Werman on for Silvan. Congrats on the update. So it appears in this update that as the follow-up has increased, there have been several more responses, so in the ORR calculation, there have been maybe 2 more patients that have had a response in your subgroup of interest. Can you speak to the tumor reduction kinetics of 2608 in comparison to inavolisib or capivasertib doublets?

You're absolutely right, the ORR is the -- prior reporting has increased, on the overall gone from 33% to 39% confirmed response rate. And in the kinase domain, it's gone from 53% confirmed response rate now to 67% confirmed. Maybe I'll hand it over to Don to kind of talk through the dynamics there.

Yes. In general, responses come rapidly. We're seeing median time to response of 1.8 months, which is the first scan. So that is generally the case that we see rapid responses. The increase in confirmed response rate that we've seen since the September update also reflects the fact that we had some patients who had unconfirmed responses in the September update that have now confirmed subsequently.

And then in quick follow-up, has there been any changes on the ESR1 front?

With regard to clearance of mutations or with regard to response?

Yes, with regard to response.

No changes. We continue to see benefit in patients with both ESR1 mutations and ESR1 wild-type disease, and we continue to see robust clearance of the mutant ESR1 clone in patients who have ESR1 mutations.

And I am showing no further questions from our phone lines. I'd now like to turn the conference back over to Sanjiv Patel for any closing remarks.

Thank you for joining us on this very early hour. I know there's a lot of data coming out today, and so we appreciate it. We feel very excited about the data that we have, and we are now head down, fully charging towards starting the pivotal trial in the Phase III to get 2608 to patients. So with that, thank you.

This does conclude today's conference call. Thank you for your participation. You may now disconnect. Everyone, have a wonderful day.