Incyte Corporation (INCY) Earnings Call Transcript & Summary

Greetings. Welcome to Incyte investor -- ESMO 2024. [Operator Instructions] As a reminder: This conference is being recorded. I would now like to turn the conference over to Pablo Cagnoni, head of -- I'm sorry.

It's all good. Let's go ahead.

President and Head of Research and Development. Thank you. You may begin.

Thank you. Thank you, everyone, for coming. Thank you, everyone who's joining us online for the webcast. We're very excited about sharing with you data that part of which was shared earlier today on 2 of our programs, retifanlimab being 1 of them and our new CDK2 inhibitor that you will see today data for the first time, so this is a great ESMO for us at Incyte. I think most of you know who I am. And I will be making forward-looking statements. We have a terrific agenda for you today. I'm going to make some short introductory comments. I'm joined by 2 of my colleagues from Incyte, Ekaterine Asatiani, our Head of Early Development, who will talk about the data from our potential first-in-class, potential best-in-class CDK2 inhibitor. We're going to focus on 2 tumor types today, ovarian cancer and endometrial cancer. And then Steven Stein, our Chief Medical Officer, will talk about some of the development plans for our CDK2 inhibitor. We're very excited, however, to have 2 guest speakers invited for this purpose: Dr. Rebecca Kristeleit who will speak about gynecologic cancers and some of the improvements that are needed to manage these patients; and Dr. Sheela Rao, who will review the data with retifanlimab in squamous cell anal cancer that was presented earlier today at the presidential symposium. Before we go to the core of the presentation, I want to remind everyone how the pipeline is being transformed at Incyte as we speak. On the left side of the slide, you have 3, 4 actually, very important near-term milestones for us. We've disclosed obviously pivotal data with retifanlimab in 2 tumor types, 1 of which was presented today. We also recently disclosed pivotal -- a trial of tafasitamab in patients with follicular and marginal zone lymphoma met its primary end point. That could potentially be a really important indication for tafasitamab as well. And as you all know, axatilimab was recently approved by FDA for patients with chronic graft-versus-host disease. And we're in the process of getting ready to launch that medicine. We're also on track to submit Opzelura for a new indication in pediatric patients with atopic dermatitis, so a lot of near-term events that are really starting to progress very, very well. The rest of the pipeline, with a number of potential first-in-class programs that you have on the right side of the slide, also continues to progress very well, with a lot of potential milestones in 2025 that we think are going to continue to transform the pipeline. Axatilimab or Niktimvo is now approved. We're very excited about bringing this new medicine to patients. We think this is a very important advance for chronic graft-versus-host disease and a potentially important opportunity for the business as well. Particularly in patients that have chronic graft-versus-host disease in more fibrotic dominated organs, we think that this data could be very compelling as a new treatment options. And we're delighted by the response rate that we saw in the study and the durability of those responses in patients with chronic graft-versus-host disease. We haven't reviewed this data yet publicly. We will do so later this year. Tafasitamab, which is currently approved for patients with relapsed/refractory DLBCL, now has a positive pivotal trial in follicular lymphoma -- follicular and marginal zone lymphoma. The most common form of indolent lymphoma, as you know, is follicular lymphoma. We really think there's a very important data that could potentially help change the standard of care in patients with follicular lymphoma with addition of tafasitamab to standard of care. And then the data we're going to talk about today, squamous anal cell cancer is an important medical need where very little has happened over the past 10, 15, 20 years. Chemotherapy is widely used, but for patients that recur after initial treatment or have metastatic disease at diagnosis, things haven't really changed much in a long time. And Dr. Rao will walk us through the data that we have generated. And finally, CDK2. This is an important mechanism in patients with a variety of solid tumors. Our focus initially today will be ovarian cancer and endometrial cancer. We have done work in other solid tumors that we will review over time, but today, we're going to focus on those 2 tumor types. And one important catalyst that we are -- milestone that we're adding to 2025 is initiation of pivotal trials in patients with ovarian cancer and our CDK2 inhibitor. And Steven will talk about that a little bit more during his presentation. Let me briefly introduce our 2 guest speakers. And then I'll hand it over to them to start the presentation: Dr. Sheela Rao, who's a medical oncologist who specializes in gastrointestinal cancers at The Royal Marsden hospital in London. She's also a reader at the Institute of Cancer Research; and a member of the International Rare Cancers Initiative, with a focus of anal cancer. She has been, without a doubt, the driver of the data you're going to see today -- is very, very important study for patients with SCAC. Later on, we're going to hear from Dr. Rebecca Kristeleit. Dr. Kristeleit is a clinical senior lecturer and consultant medical oncologist at University College in London. She specializes in the treatment of gynecological cancers, including cervical, endometrial and ovarian cancer. She's a member of the Target Ovarian Cancer Scientific Advisory Board; Oncology and Haematology Expert Advisory Group; and the ESMO Faculty from 2021, 2025; and [ an investigator in the ongoing ] study with our CDK2 inhibitor. And she'll talk a little bit about the current management of patients with ovarian cancer and endometrial cancer. Both Dr. Rao and Dr. Kristeleit will be available for the Q&A. So with that, let me hand it over to Dr. Rao for the presentation on SCAC. [ Dr. Rao ]?

Thank you very much. [ Let me take this ]. [indiscernible]. Okay, can I have my slides? Thank you. So thank you very much, everyone. I'm going to talk to you about some data we presented today. So a bit of background. Anal cancer is a relatively rare disease, but its incidence has been increasing about 3% each year. And this is largely due to human papillomavirus, so HPV. HIV is strongly associated with anal cancer. And there is a greater risk of this disease associated with HIV-positive patients. For many years now, the standard of care in the localized setting has been chemoradiation, and -- but despite this, you'll get about 30% of patients who will progress, so will develop advanced disease, and another 10% to 12% who will develop metastatic disease. So about 40% of these patients will have advanced disease. As previously mentioned, there have been very few developments in this disease. Chemotherapy as a backbone was established a few years ago by our group, with good response rate, durable response but still modest progression-free survival and overall survival at 8 months and 20 months. And HPV-driven malignancies provide a strong and attractive target for immunotherapy due to the depleted T cell microenvironment. And we've seen this both in head and neck and cervical cancer. So just to give you a bit of background. This is about 10,000 new cases per year in the U.S. and the EU. And as I said, about 40% of these will present with advanced disease. So that's the group we're thinking about here. In terms of the current treatment landscape, in the primary setting, it's chemoradiation. And this -- as I said, it's been a standard of care for a long time now. And in the metastatic setting, mainly chemotherapy regimens, so combination chemotherapy regimens that are given up there. In terms of on progression, in the metastatic setting, there are a number of immunotherapy regimens, so nivolumab, pembrolizumab or retifanlimab; as well as other chemotherapy regimens, but of note, none of these immunotherapy regimens in the second-line setting are actually licensed currently. And at the current time, the practice guidelines from the NCCN recommend platinum-based chemotherapy as a first-line option and only recommend immunotherapy on progression. And that is also true for the European guidelines, the ESMO guidelines. So this is just really telling you that again. So both guidelines will basically recommend chemotherapy only at the current time. And no immunotherapy in the first line is approved. And no immunotherapy in the second line is actually licensed, although it is being given. So as I said, there's a really good rationale here for adding immunotherapy. These tumors are HPV -- HPV is tumor initiating and shapes your microenvironment. You get these exhausted T cells. And then clearly, when you add immunotherapy back in, what you can do is reactivate these T cells. And we have seen this in a number of cancers that are HPV-driven. So cervical and head and neck are good proofs of concept. So this is a study we designed. So this is a Phase III study. Basically what we did was add the retifanlimab, an anti-PD-1 monoclonal antibody, to carboplatin and paclitaxel. We took patients who had inoperable locally recurrent disease or metastatic squamous cell anal cancer. And they could not have previous chemotherapy. So they could have had chemoradiation as primary treatment but could not have systemic chemotherapy. Here is the study design. This -- the planned enrollment was 300 patients, and we randomized 1:1. Both arms got chemotherapy. And then the randomization was to the addition of retifanlimab or placebo for a total of 12 months. And patients in the placebo arm could cross over to retifanlimab after confirmed progression. The primary end point was progression-free survival with a hazard ratio of 0.67, with an alpha of 0.25 (sic) [ 0.025 ], with a key secondary end point of overall survival using alpha spending if PFS is positive and a number of other end points as shown here. We stratified for PD-L1 expression, region and extent of disease. So this is the consort diagram. In summary, we screened 376 patients; and enrolled 308, 154 in each arm. The main reason for discontinuing treatment in both arms was progression. And you can see there's a full analysis set of 154 patients in both arms. 69 patients in the placebo arm crossed over to receive single-agent retifanlimab. Demographics, very much as expected and very well balanced between both arms; a higher percentage of female patients, which one would expect; a significant proportion that had radiotherapy previously; and a high proportion of patients with metastatic disease of over 80%, 1/3 of patients with metastatic disease in the liver; a small proportion with HIV positive disease. And PD-L1 expression was greater than 1 combined positive score in over 90% of patients. So that's the primary end point, progression-free survival by blinded independent central review. We met our primary end point. The median PFS for the placebo group is 7.4, on the bottom. And the PFS for retifanlimab is 9.3 months, with a hazard ratio of 0.63 and a significant p-value. And as you can see here, the curves actually separate pretty early on and continue to separate. This is overall survival. This is an interim analysis. These data are not yet mature. We've not yet reached our events, but as one can see here, another similar pattern, the curve is separating relatively early on. We have retifanlimab on the top, the placebo on the bottom. And these curves are separating quite nicely; currently showing a trend towards improved overall survival of about 6 months, which in this disease is certainly meaningful. And obviously we've not yet reached our events. So in terms of crossover, we did rank-preserving structural failure time model just to look at crossover, a fairly standard analysis, because clearly patients in the placebo group will have crossed over to retifanlimab. So the bottom curve is that adjusted survival, overall survival, for patients who've crossed over. And what you do is you recalculate the survival as if they had not crossed over, to give you a true adjusted placebo. And what you can see there is that just enhances the benefit even more, so we're seeing definitely a strong trend towards improved survival compared to the adjusted group. As I say, these data are not yet mature, but the curves are separating early and these data look really promising. In terms of secondary end points, no big surprises. We see an improvement in overall response. We see a really significant improvement in duration of response and a significant improvement in disease control rate. These data, we compared to our original study just to confirm that the control group is basically performing as we would hope and expect. And actually these data are really reassuring. So what we can see is that, compared to the initial study that we conducted with the carboplatin-paclitaxel, the data are very consistent across the board, suggesting that the control group is performing as one would expect. In terms of safety, no concerning readouts here; some adverse events that are greater with the addition of retifanlimab; as one would expect, slightly higher immune-related adverse events, but the data are consistent with data we already have in terms of safety. We have no loss of HIV control. And at the data cutoff, over 58% of patients in the retifanlimab arm still remained on study. When we break that data down further into drilling down on the clinical events, as I said, nothing concerning. The safety events were manageable. There were some increased immune-related adverse events as one would expect, but these were manageable and not out of keeping with previous data. So in conclusion. This is the first and largest Phase III trial of a checkpoint inhibitor in advanced squamous cell cancer. We met our primary end point with a PFS of 9.3 versus 7.4 months with placebo. We have a strong trend towards improved survival, with a meaningful improvement in survival. And we see this also with response, duration of response and disease control. So the treatment was well tolerated and we're really seeing no concerning safety signals. And we didn't compromise the delivery of the chemotherapy with this administration, so essentially this now represents a new standard of care for patients with advanced squamous cell anal cancer. And we had a very good presentation today, followed by a great discussant who also confirmed that this is now a new standard of care.

Thank you. Hello. So just changing tack now, we will move to gynecological cancers. And I'm just going to provide you with a brief background on mainly ovarian cancer. And I think I'm sure you all know that most of these patients present in advance stage. It's very difficult to pick these patients up early even with all of the new technology we have. They're still presenting predominantly Stage III/IV disease, as you can see here in the new cases diagnosed in the lower diagram. The incidence of ovarian cancer is increasing in all these territories, so it's going to become a bigger problem over time, unfortunately. The clinical course of ovarian cancer, this also of advanced ovarian cancer. You have one chance of cure currently as we know it, with a combination of surgery, chemotherapy and targeted agents. And in that first-line setting, we're giving PARP and bev as maintenance, and we continue to do that even with the data presented today. There will be no change to that in terms of PARP use and bev. It's used in most of our patients in the first-line setting. Once patients have relapsed, whilst there might be 1 or 2 that can be surgically cured, we don't have much evidence of that. And the general paradigm is that we keep on treating these patients as they need, mainly according to symptoms and disease burden, with various lines of therapy. And this is where we start talking about platinum-sensitive or -resistant patients, okay, not in the first line. The treatment algorithm. This is ESMO. This is Europe, so I really -- obviously it's very busy, but what I want to draw your attention to is the 2 halves. So that's very much consistent with that treatment paradigm of newly diagnosed ovarian cancer, where we want to cure patients and stop them relapsing, stop them needing any more treatment. And that's where we really need to invest all our efforts to prevent those patients moving into a relapsed setting. However, if we failed with that, the recurrent setting, the greatest needs in these patients -- they all have a need, but the greatest needs is in early platinum resistance or patients that have progressed on their first-line treatment. So these harder-to-treat patients that present with early platinum resistance or even later platinum resistance after 1 or 2 lines of treatment in the recurrent cancers -- recurrent ovarian cancer setting, they really stop responding to most -- well to most of our agents. And this is where most of the development is coming through. So one other kind of truth to ovarian cancer is that all patients end up as platinum resistant. So it's a catch-all phrase, but the ones that begin with platinum resistance have a really, really high need, highlighting here that the focus for Incyte with our CDK2 inhibitor and the needs of the patients is in the HR-proficient or HRD-negative first-line maintenance setting, where [ bev ] is used, possibly in combination with [ bev ]; and in platinum-resistant disease, the 2 yellow boxes. This is a really important slide, really important bits of data mainly generated from within Incyte. What you -- what I really want you to focus on is, in all ovarian cancer, they mainly split 50-50 into HRD negative and HRD positive. So you have expression of cyclin E1, overexpression, in both HRD positive and HRD negative but more of them, there's an increased percentage, within HRD negative, okay? So most HRD-negative patients are CCNE1 overexpressed. The other really important bit of information here is that there's some mutual exclusivity between folate receptor expression and CCNE1 overexpression. This is important with a lot of the new classes of drugs coming through. So you can see that there are many patients that will never be served by a folate receptor-targeting drug, okay? So talking about that maintenance development opportunity in the first-line setting. Most patients receive bevacizumab in advanced ovarian cancer. And the HRD-negative patients are around 52% of that newly diagnosed population. They have a massive need for improving their outcomes. Their outcomes are much less good than the HRD-positive and BRCA patients, and we need new drugs there to improve their survival. So we know that the cyclin E protein expression is enriched in this population. And bevacizumab is used in that population, so there seems a very strong rationale to try to augment bevacizumab and provide a benefit to these patients by the addition of a CDK2 inhibitor, knowing that there's a lot of CCNE1 expression in that population. This is -- the diagram on the left shows you the evidence that tells you that it's around 50% of CCNE1 increased expression in ovarian cancer. So these are the competitive parties, if you like, of other drugs that are coming through. I would highlight mainly the ADCs. They're really pushing forward into earlier lines of treatment, and I wouldn't be surprised if we're seeing some of those approved really quite soon. I would also say that ADCs -- whilst they're a class, they're a class of structure, not a class of targets. So it's the way the drug is put together. Therefore, each individual ADC is going to have a very different use or efficacy. And it's really chemotherapy still. You're targeting the chemotherapy, so you have those side effects potentially. We know we get a lot of specific things like eye toxicity, pneumonitis, so they're not without their issues. And they're also being focused in relatively small biomarker-led groups. And giving these drugs intravenously as maintenance treatment is difficult. So those, I would say, are the main competitors in the ovarian cancer landscape at the moment. And then very briefly, endometrial cancer first-line setting. As you know, we're getting immunotherapy in there with very good benefit in MMR-deficient patients as a single-agent maintenance, but -- and we're getting some benefit in MMR proficient, but that needs to be augmented, so the MMR proficient is a huge area of need in endometrial cancer. And post IO is a huge area of need. So in the second- and third-line settings, historically people have said, "But endometrial cancer patients, they never get there because they're not well enough. They don't do well enough." That's changing. There's more of them getting into second and third line, but they really -- we've got no proper standard of care post IO, so there's huge opportunity in endometrial. So the development here for CDK2 which will be taken -- Ekaterine will move this forward with much more clarity for you, but it's in newly diagnosed ovarian cancer first-line or platinum-sensitive recurrent disease or platinum resistant as monotherapy. So combinations here potentially with bevacizumab, as maintenance, and monotherapy in platinum resistant, many options for combinations as we move through but just getting that activity from monotherapy, first, in endometrial cancer. Really, I mean, pick your line. Thank you.

So thank you, Rebecca, for this overview. And I will remain on the same tack and continue with gynecological cancers and CDK2 data we have, so far. So this slide, you have seen this afternoon. One take-home message I would like to repeat here is that using immunohistochemistry-based tools that we are currently developing as companion diagnostic. Over 50% of patients with ovarian cancer have overexpression of cyclin E. And close to 7% -- 70% of patients with endometrial cancer have overexpression of cyclin E. The tumors that have overexpression of cyclin E have dependence on CDK2 inhibition, which is synthetically lethal. We have a lot of data on that, internal as well as published. And we have selective CDK2 inhibitors that have shown strong preclinical activity, and today, we have disclosed clinical activity as well. So selectivity really matters. I won't go into the details of those slides. I think number speak for themselves, but selectivity against other CDK -- inhibition of other CDKs is important, especially for CDK1 as it's [ pan-cytotoxic ]. In this regard, we believe we might have best-in-class molecule. And also we have most advanced data in ovarian cancer, and therefore, we could have first-in-class molecule as well. Here are listed other CDK2 inhibitors which are in clinic and where we could gather data from published literature based on their biochemical, cellular assays and selectivity. So this is the trial. Again, this was presented from a different podium this afternoon. I won't go into the detail. This trial has been going on for 2 years now since 2022. Today, we presented data, this afternoon, from dose escalation from 84 patients. I will be discussing a lot more expansive and also more mature data with later cutoff from over 200 patients. So we have actually safety database now for over 200 patients. And I will go in a bit more detail also for efficacy in 2 indications that we have treated in expansion cohort, and this is ovarian cancer and endometrial cancer. I must say that we are continuing -- we're kind of finishing up dose escalation expansion part for monotherapy, but we continue with combinations. We just started this combination dose escalation. And it's quite broad spectrum of standards of care used in those indications that we are testing. So this slide, again in different version with less patient numbers, was presented this afternoon. We have 205 patients that have received at least one dose. Majority of those patients are gynecological malignancies that were treated in expansion cohorts. In dose escalation, patients were not preselected. However, we encouraged enrollment of CDK2-overexpressed or -amplified patients; and a lot of them did have amplification as locally detected. In expansion, we select patients with those tumor types for either amplification or overexpression, which we do centrally. So 89 patients with ovarian cancer, 14 with endometrial cancer. Please note that these are heavily pretreated patients, heavily pretreated with chemotherapy, right? So the majority of those patients have multiple lines of chemotherapy, so pretty beaten-up bone marrow, and in this light, we should review the safety data that we have. So this is a busy slide. I won't go through the -- all the numbers, but again, 205 patients in safety database, a lot of them in the doses which we are moving forward in expansions, 88 patients treated with 15 -- 50 milligram BID and 63 treated with 125 milligram QD, so pretty mature data here. Majority of toxicities are on target and as we expect, so a lot of hematological toxicities as we expected, some GI toxicity. Majority are grade 1, 2; very few grade 3. And higher grade are very few grade for just a couple of cases. I should note that majority of patients actually tolerated treatment at a given dose. We had very few dose -- treatment-related adverse event discontinuations. And only 11 patients needed dose reduction. I think it's pretty remarkable in Phase I trial actually. We have some DLTs. So the higher doses -- 75 milligram BID exceeded MTD. 150 milligram also had some toxicities. And we did not pursue that dose level, although it's not declared as above MTD. And these doses, these higher doses, separate very well in terms of expansion from the doses that we move forward in expansion. And this is what I would like to show here, that there's 2 higher doses where we had DLTs. In terms of exposure in terms of PK, they're actually much higher. And they do cross in CDK1 a target into the IC50, for CDK1, but the dose is below that we move forward. And they're boldened here, so 50 milligram BID and 125 milligram BID. They remain well below IC50 for CDK1, but they do cover for majority of treatment period CDK2, so potent and selective. That's basically what I want to say here. Now all dose levels do have reduction in ctDNA, which we think matters. Steep reductions in ctDNA do correlate with tumor shrinkage, and that's what we show on the next slide. So there is a reduction in tumor shrinkage -- there is a reduction in tumor size in those patients. This is both from -- this is from escalation and expansion. And here, denominator is 78, so 78 patients that have post-baseline scan. There were some patients that discontinued without scan, only 3 of them. And there are some patients that just enrolled and we don't have scans on them. So that's where numbers are coming from. Now with the asterisk or star: We have patients which are ongoing. And this is data from ongoing patients basically, so those numbers will obviously change. And on the bottom here, there are prior lines of therapy. And except few patients, most of those patients are pretreated heavily. Now when we looked at the patients treated at expansion doses. And again, we expanded 3 dose levels. There are 37 in denominator. The denominator is selected very conservatively. So those are all patients that discontinued from treatment or had at least one post-treatment scan. And among those patients, we have response rate of 24%. Now 50 milligram BID was expanded a little bit earlier, so data is a bit more mature, and there the response rate is 31%. On top of it, I should say that there are a lot of patients with stable disease, as you saw on the previous scan, and this really matters. I mean, in this setting where patients are heavily pretreated, delaying progression does matter. So disease control rate is 75%. So we just show here a spider plot of the patients that have response. And there are actually 2 which, strangely, overlap. So this -- there were 2 patients that were scanned the same time and have the same tumor reduction, so you -- this is basically the lines do not correspond to the previous slide, but there are PRs and CRs that you can see here. And the reason I would like to show is that there -- these responses are actually durable. And majority of those patients are still ongoing. And I do not have duration of response, but that's because those patients are still ongoing. Here is data in endometrial cancer. So we have not discussed this at all this afternoon, but this is a preliminary data. We have treated less endometrial cancer patients. Only 14 patients were enrolled because we're focusing on ovarian cancer. And in this early set, we have 4 PRs. And majority of patients have tumor shrinkage. I mean this is quite exciting for us as well. Just to summarize. We are working hard to finish up the monotherapy dose escalation, dose selection; and just to bring this drug to patients in the next phase of trial. And this is what Steven is going to discuss in a minute. Thank you.

So thank you, Eka. So just to pause for a second. I think, in 2 years, the team has done an excellent job both in terms of enrolling patients and then getting us adequate data in terms of both dose and schedule, to be in a very comfortable place in terms of meeting regulatory requirements on declaring the dose we'll use in a registration study. Currently, as you saw from Eka, we're expanding both the 50 milligram BID and the 125 milligram QD, just pending regulatory discussion on that, but we're, again, in a comfortable place on those. The second thing just to pause on is we think we've demonstrated very clear proof of concept in platinum-resistant ovarian cancer that is CCNE1 overexpressed; and one could argue, albeit the numbers are much smaller, in endometrial cancer as well. So just to be somewhat repetitive to the prior speakers but just to show you real-world data from the United States in terms of what's actually used. And it's actually largely in keeping with guidelines. So this is the first-line maintenance setting. And you have to be careful of terminology. As Dr. Kristeleit pointed out, the platinum sensitive and resistant definitions only come out later, but in terms of real-world setting in terms of maintenance, if you focus on the left side here, on the BRCA wild-type HRD-negative population, you can see 50% of those patients are getting bev maintenance. Another 28% are getting PARP in the form of niraparib. As Dr. Kristeleit told you, HRD negative represents about half, 52%, of first-line maintenance. That opportunity there is 22,000 patients in the U.S., EU5 and Japan before you do the CCNE1 enrichment part that we do, obviously, with this program. The unmet need part Dr. Kristeleit pointed to, the HRD negatives have the poorest progression-free survival. So a number there for you of 10 to 12 months, in contrast with HRD-positive patients where the PFS is 39 months and BRCA mutants where it could be upwards of 60 months -- or 56 months, but our use of cyclin E protein overexpression -- and we're not declaring our cutoff yet because we need to have the right regulatory discussions. About 60% of those patients who -- HR proficient or HRD negative are cyclin E high, so there's the opportunity. The only PARP approved there, although not used as much as bev maintenance, is niraparib, but the majority of patients receive bev maintenance, so a huge setting of interest to us; a need in terms of trying to improve, potentially, cure rates here; and a very important study that we'd like to do. Just to focus again real-world data, again in the United States from this year. This is the platinum-resistant population. And you'll see it's highly heterogenous, so if you look across the board, those are all single-agent chemotherapies, topotecan, Doxil, gemcitabine, docetaxel's, paclitaxel, et cetera, et cetera. The unmet need remains high in this population of platinum-resistant ovarian cancer, with average of about a 1-year survival. And you saw from Dr. Kristeleit, with each therapy, shorter and shorter progression-free survival. And then chemotherapy single agent use remains the dominant use here, with bev maintenance in settings where it hasn't been used prior, what's not contraindicated. And that may be another population that may be of interest to us. The ADCs have their role. The WEE1 inhibitor remains on clinical hold at the moment. So where are we with regulatory thinking? The major caveat here is we still need to have the regulatory interactions on both sides of the ocean, but let me just walk you through this somewhat carefully. So the study listed first there is single-arm monotherapy study in platinum-resistant ovarian cancer and potentially in endometrial as well given the data we've seen. That would be a regulatory strategy that will be dominant for the United States because it will be a single arm looking at accelerated approval with demonstrating a response rate that's durable to get across the finish line. You saw in one population that Eka highlighted with the most experience thus far that we're hitting a 30% response rate, where they can't even calculate a median duration because most of those responders are ongoing, with long durations thus far. So that's potentially of huge interest and potentially a way we could go forward. In -- sort of in the same theme, below that, you can do that same population with a randomized controlled trial against a physicians' choice or a best chemotherapy choice. There's different single agents in terms of your comparator. Obviously that will give you a global study because again it's a randomized study. The end point here then would be a time-to-event end point, progression-free survival. We could, subject to regulatory discussion, do an interim on response rate there and then use that in the United States to get the accelerated approval, with the study ongoing to suffice for global approval as well as confirmation. If we go that route, you could argue that you don't need the study above, the single-arm study, because you can get both with one study. So that's again an important population, as was highlighted, we want to address. In terms of trying to improve cure is the study below. Again it's maintenance after first-line chemotherapy. It's the HRD-negative population. Bev maintenance is the, in the guidelines as well as real-world data, dominant use here. And it would be a randomization of bev plus/minus CDK2 in this population. It's a progression-free survival end point. We still have to discuss the statistical assumptions here, but it's potentially a larger, bigger study that will take longer to get across the finish line, very important need to address. As Pablo said upfront, we're ready to go in 2025 with this. We'll have the regulatory discussion soon both on dose selection and the populations to go after. These are the ones of huge interests. I'll turn the podium back to Pablo for Q&A. Thanks, yes.

Before we go to Q&A, let me just make a couple of closing comments here. Earlier this year, we highlighted for all of you a number of important milestones that we promised to hit in the second half of 2024 and 2025. And these milestones are critical, as our pipeline continues to evolve, and what medicines become more and more important for patients and for continuing to build the business. Now I think it's important to highlight that 4 of those milestones for the second half of 2024 have been met. Axatilimab is now approved. Retifanlimab has a positive study that we discussed today; tafasitamab, the results that we announced. And just now we unveiled the results of what we consider proof of concept for our CDK2 inhibitor, and we gave you an idea how we're going to develop this new medicine. The submission for pediatric atopic dermatitis for Opzelura is on track for this half. For next year, we're adding milestones just to continue to build value here: the potential approval of retifanlimab in 2 new indications, anal cancer and non-small cell lung cancer, based on the data that we've disclosed; the potential expansion of the label for Monjuvi/Minjuvi for follicular and marginal zone lymphoma; and initiation of pivotal studies or pivotal study, depending on what we decide, for a CDK2 inhibitor in ovarian cancer, as Steven summarized. So the portfolio continues to grow. If you consider current, there's milestones that we just hit. Near term, we promised early next year the Jakafi XR data, with initiation of our CDK2 inhibitor pivotal trials and the first pivotal study for povorcitinib in hidradenitis suppurativa which is coming in early 2025; and then in the future, dominated by a number of novel medicines, that we'll continue to advance and we'll continue to deliver data in the next year and beyond. This will continue to progress to what we announced about a year ago, about 19 months ago, which is that our pipeline is on track to deliver 10 or more potential high-impact launches by 2030. I hope you would agree that, based on the progress we're making over the past 12 months, this is becoming closer and closer to being a reality. So with that, I will stop my prepared remarks. And we'll be happy to take questions. Thank you.

Okay, all right. Okay, we're going to go from the front of the room to the back. There's one, two, three, four, five, six -- I'll try to keep everybody. All right, go ahead.

Yes. Stephen Willey from Stifel. Quick question on retifanlimab. Why was the duration of therapy limited to 12 months in the PODIUM trial? Do you think that you could have extracted additional clinical benefit by using longer-maintenance therapy? And then maybe just another question for the company. How are you thinking about resourcing a retifanlimab commercial launch right now? You obviously have a little bit of GI infrastructure via Pemazyre. I know [ you've one ] in the PODIUM-304 study. You're not there commercially in lung. Is that something that you're aspiring to as well?

So let me have Dr. Rao, first, talk about the design of the study. I can take a first shot at the resources. And Hervé can probably comment additionally. Dr. Rao?

Sure, okay -- I don't think my mic is -- it is working.

It's working, yes.

So the -- so HPV-driven cancer, advance disease that we have a lot of discussion about duration of treatment, but actually if you look at those curves, they're separating very early on, so clearly we're getting that benefit with chemotherapy early on, so at this point, we believe that 12 months is adequate. You can still see the curves separating now. And I don't really think there's evidence at this point to give longer. So that was the basis for our decision for 12 months. I don't think you're necessarily going to get an incremental benefit beyond that given how early the curves are separating.

On the second part of your question. I think you answered it in the question, which is the infrastructure. It's in place because of pemigatinib. And so there's very minimal additional investment that should be required to launch squamous cell anal cancer in that population, based on the infrastructure we have in place from pemigatinib. I don't know if you want to add anything to that, Hervé.

So given -- it's a very targeted group of physicians treating that disease, so I think there will be very -- it will be very easy. And it's not going to add very much to the sales and marketing budget. On the lung question that you asked: Frankly, we don't have the answer yet, so that's a completely separate subject. We don't think we should sort of invest heavily to compete in lung cancer based on the data we have, which is as good as KEYTRUDA, but it's not very differentiated, so we are looking at a very different reason to go in lung cancer. It's to have a reference so that we can combine with our pipeline in the future. And that will be true in the U.S. only. So the lung cancer opportunity commercially, I will put completely on the side; and look at this squamous cell, where we have the team in place. So we can do the launch very quickly in 2025. There is a population waiting and that population is treated by a fairly small number of physicians.

Okay, second. Let's go back. And I apologize if I skip anyone. Over here, please, Ben, lighter jacket -- anyway, I'm trying.

Evan Seigerman from BMO Capital Markets. So I want to touch on your trial design and kind of plan for ovarian cancer, 2 questions. One, are you going to stratify for folate receptor positivity or cyclin E positivity? I think that could be pretty important, based on the slide that you presented. And then as you think kind of bigger picture with a lot of ADCs, bispecifics, how does this asset fit in 5 years from now given how rapidly the space is developing?

Yes, great question. So let's first talk about CCNE1 overexpression. That's by AAC. As Steven mentioned, we need to discuss the cutoff with the agency. We're developing a compound diagnostic. We think, based on the definition that we're using, it's a little bit over 50% of the patients. The overlap with -- the current definition of a folate receptor alpha-positive patient, which is with the only approved one which is mirvetuximab, the overlap there is about 1/3 overlap and 2/3 don't overlap. So when we're starting to think about a study in the current environment with folate receptor -- anti-folate receptor ADCs approved, there's 2 ways to do it. Obviously you can focus on the negatives. Or you can include both negative patients and patients that have progressed on a folate receptor ADC. So that's the current landscape. Now obviously there's a number of ADCs being developed, and I would put them in 2 broad categories: follow-on for folate receptor alpha ADCs. And I think what's been tried to do there is to increase -- or lower the percentage of cells -- the percentage of positivity necessary for eligibility. The way I read that data, so far, is those attempts are, while they've shown efficacy in patients that are lower expressers of folate receptor alpha, the toxicity of those ADCs seems to me prohibitive, particularly [ in further lines ] of therapy. We'll see how those evolves, but I've seen a lot of toxicity, particularly from the Sutro asset. The non-folate receptor alpha ADCs are certainly interesting. I would highlight the [indiscernible] ADC, which has shown very impressive efficacy over 40%. I think, [ though, it's a little bit ] heavily pretreated patients than in our study. And good durability. So the way I see this evolving -- which is why Steven talked about moving to early lines of therapy as quickly as possible. In the maintenance after first-line chemotherapy, which is long duration of therapy, we believe that's the key area where our CDK2 inhibitor needs to move towards as quickly as possible. We will start with a rapid -- hopefully, a rapid approval fast to market, on platinum refractory patients, as we discussed, but moving to maintenance therapy, we think, is the opportunity because of convenience, safety. I mean better balance of safety and efficacy. And you saw the data, the percentage of patients. This is a Phase I study, very heavily pretreated patients. The percentage of patients with discontinuations or dose reductions due to toxicity is extremely low. That tells you that patients are going to tolerate this very well in the maintenance setting. And that's where we want to move to as fast as possible. Let's go here and then we'll go back again for...

Matt Phipps from William Blair. First off, you've -- I think you've kind of hinted at this, but do you think a 30% response rate is almost like a bar to get single-arm approval in these days? The FDA, I think, has been a little bit hit or miss on where they're accepting that. And then of all the registrational paths listed, no chemo combo and no ADC combos, do you think the heme tox that you do see is limiting of going into some chemo or ADC combos?

Yes, so maybe I'll start. And then Eka can address your second question. Obviously it's somewhat tumor histology dependent, given where things have been historically, response rate, but broadly speaking, in this population, we think around -- somewhere between a 20% to 30% response rate that's durable should be sufficient indirect benefit of -- indication of clinical benefit to get it across accelerated approval hurdle. And then you'll have to confirm it with a confirmatory study in this setting. And I think that's the number we're thinking. Obviously it will be subject to discussion. And then in terms of the second question...

Regarding chemo combo, we are testing paclitaxel combination, so we'll see how it's tolerated. And we just saw that in maintenance setting in combination with bevacizumab. We had a better rationale and better way to provide delay in the progression and recurrence for those patients. And that's -- yes, that's what we need to see.

I would complete the thought that Steven gave. So when you think about platinum-resistant ovarian cancer, if you look at the MIRASOL results, the control arm is a contemporary benchmark, right? It's contemporary. 40% of the patients are weekly taxel. There was 3 or 4 different chemotherapy regimens used. The response rate was 16%. PFS was 4 months, so if we think about a single-arm study, something that looks clearly better than that, which is somewhere between 25% and 30%, with some good durability of responses, I think it should be -- potentially it could be an approval package. And that's what we're going to discuss with the agency. Okay, so I'm probably leaving people out. Just pick one, Ben. I don't know...

[ I've got the microphone ]. Andy Berens, Leerink Partners. I guess there's a benefit to sitting at the front of the class. Just a couple on the CDK2: Is there any data that correlates replication stress and repair deficiencies? I can understand why you would certainly want to go into HRD-negative patients, but I'm just wondering how you would expect the CDK2 to work in that setting. And then I noticed -- I didn't get a chance to look at it in too much detail, but there's no dose response, so far. Just wondering about any comments there. And it -- you didn't -- I know you tested intermittent dosing, so just any color on that you could give us would be great.

We can comment on dose response. I -- the first part of the question, I don't think there's data that we're aware of correlating HRD status with CCNE1 overexpression and the importance of one over the other. At least we're not aware of it. Do you want to comment about -- look. It's tricky. It's a Phase I study. Some of the cohorts are a lot smaller than the others -- variability in patient population, including tumor type and prior therapy, so one has to be careful about making -- and as Eka mentioned, the 50 milligram BID cohort has longer follow-up. Those patients were -- entered earlier, so when you compare that with 125 mg QD, there's a different duration of follow-up. Do -- I don't know if you want to comment any further.

So we will perform exposure-response analysis once data matures from all 3 cohorts. 125 milligram QD are probably not separated that much in terms of the dose, but we will look at the exposure and differences in response rates and tumor volumes, et cetera. So we'll put together as comprehensive package as we can for Optimus project and just to defend the dose that we propose. I would not say that there is a very clear correlation of a dose and response right now, but again, once -- as Pablo said, some of those cohorts are very small, so it's difficult to compare.

Michael?

It's Michael Schmidt with Guggenheim. Similar question. Did you see any difference in activity in patients either with CCNE1 amplification versus the overexpression and how finalized is your biomarker strategy? And then the other question was just about how you think about prioritizing other indications. I don't know if I saw breast cancer on your list, but I did notice a new table that selectivity over CDK4 was a little bit lower than the other CDKs, which could actually be a benefit [indiscernible] just curious how you think about that.

Yes. Thank you for noting that...

So why don't you take the first part of the question? And I'll take the second part of the question, on breast cancer.

There is no difference. I mean we have looked and it's not obvious that amplified patients do better than overexpressed. And the way we are using expression kind of correlates with amplification. And I don't want to go through the details of this, but we want to make sure that, patients that we call overexpressed, they correlate with what protein level would be in amplified patients. And there's no, like, difference that you can eyeball. And the second part -- yes.

The second part, I can take it. Look. We have what we believe is potentially a best-in-class CDK2 inhibitor, so for us the priority was to move this as quickly as possible in ovarian cancer, try to get these pivotal trials going. Endometrial cancer was not a surprise, but it was certainly nice to see early responses in endometrial cancers. And we're going to push that fast as well. We are combining with appropriate therapies in breast cancer. We have already dose escalation data. We're combining with hormonal therapy. It is a more challenging development plan for us. We don't have our own CDK4/6 inhibitor. We don't have our own CDK4 inhibitor. We are not abandoning breast cancer. We're going to continue to do work. We'll make that decision probably at some point next year, and we'll communicate at the time. For now the focus is platinum-resistant, platinum-sensitive ovarian cancer and endometrial cancer.

If I may add, Pablo. Those -- the breast cancer will need combinations, probably with CDK4/6 and fulvestrant. And that's what we are just starting now, so this data will come as well.

[ Imogen ] from Cantor and Eric Schmidt's team. So you talked about the CCNE amp and overexpression not necessarily correlating, but did you see a correlation between the level of overexpression of cyclin E and response? We saw a kind of binary high expression or low expression. And then second question, are all of the responses RECIST -- confirmed RECIST responses?

Go ahead...

Yes, whether they are confirmed or not. So the first question: So the way we defined overexpression, it correlates at what you would expect an amplified patient at protein level, so there is a correlation because of that...

Has the level of expression correlated with response? That's the question.

Yes. We have not looked at it. I don't want to answer this question. We have not really looked at level of the -- using the cutoff, yes, you saw the data, but I don't know if, like, higher overexpression would correlate with deeper response, et cetera. Because numbers are small. I mean I would not go [ there probably ].

Do you want to take the second one with -- the answer is yes, which is they're all RECIST confirmed. Go ahead.

It's confirmed based on local assessment. We don't have central radiological assessments of those tumors.

Responses are confirmed and denominators are conservative. I just want to make sure everybody -- there's a lot of noise in this business. Responses are confirmed. The denominators are conservative, okay? The only -- the patients excluded from the denominators are patients that have not had an assessment; or in the waterfall, patients that failed before the first assessment. So you have no measurable disease, but when you calculate the response rate, those patients are included. Kelly?

Alex Kelly, TD Cowen. So I was just curious if you could maybe briefly expand on why you see your CDK2 program as differentiated from maybe some of the other competitor programs, other than the fact that some of those programs may be targeting breast initially versus ovarian.

So we can only speak from public data. And what I see, what we've seen is what -- Eka summarized the preclinical data. I think that, taking data that other competitors present at face value, we think we have the best balance of potency and selectivity. And the selectivity that really matters here is CDK1. They all do, but the CDK1 is the one that lets you push the dose to really be above IC50, IC90 of CDK2, CDK4 as continuously as possible, which is the PK data that Eka showed. That's point number one. The single-agent activity that has been reported by our competitors is pretty modest at best. There are some more interesting combination and data. And there today was a presentation about CDK2 with CDK4, again in breast cancer. The data that has been presented in ovarian cancer, we find it inferior to the data we presented today. And the third point I would make is at least one of our competitors has what I will consider problematic off-target toxicity, which was ocular toxicity which was the -- resulted in a clinical hold. So just to put it all together: I think we have the best balance of potency and selectivity and the best balance of efficacy and safety that I've seen for single-agent CDK2 inhibitor in the clinic.

Kelly Shi from Jefferies. I have 3 questions. I hope you don't mind. They're like too many.

No.

The first one is you've got a -- 2 complete response out of 37 patients. And it's not [ very easy ] in the PROC setting. I was curious. Any particular patient characteristics you noticed from the 2 CRs? And the second one is we also noticed a very promising disease control rate, so curious whether you have the spider plot for those patients [ who achieved SD ]. Are they, like, durable? And lastly, do you have information on the mutual exclusivity between CDH6 and CCNE1 as you have shown for FR alpha?

Great. So let me take the last 2. And Eka can talk about those 2 CRs and whether any patient characteristics. I commented not that long ago that I thought disease control rate was important in this setting, particularly for a well-tolerated drug in the maintenance setting, because it should lead to prolonged progression-free survival. I think that's really important. I mean, 75% disease control rate in a population of patients with 3, 4 prior lines of therapy, I think it's a substantial number. And I think we need to, you need to keep an eye on that. We don't know: What are the parameters of the CDH6 test? So we can't tell. We're trying to generate some in-house data, but none of that has been made public. Allegedly, they can cover 65% to 85% of the patients. We recognize that that's the way it's been communicated. For now we're starting to do some work to understand the potential overlap, but I would go back to the way I answered your question earlier, which is this is why we want to move earlier. We want to move to early lines. We think maintenance is the place for this medicine. It's a pill, easy to take, very limited toxicity. If we can drive the efficacy in those patients, that's where the most patient benefit will be and the longer duration of therapy. Eka, I don't know if you want to talk about the 2 CRs and whether they have any...

Yes. So the CRs. I mean I didn't want to go into the details of this, but responses deepen with time, which is a good thing because we might have more responses in stable-disease patients as well. 1 of the CRs is patient that is longest on treatment. And it was actually PR and converted into CR -- should I say this or not? We have another...

You shouldn't, but you did, so go ahead.

Okay, I didn't say anything, but basically responses deepen. So the 1 of the 2 CRs is -- has converted as a CR gradually. The other one is patient with lymph nodes. And that -- I mean that's all I can say about that. So that patient had a lymph node. Like majority of disease was in a lymph node. And that was also CR.

We're going to go to the phones now to see if there are any questions. Greg? 5, okay.

[Operator Instructions] Our first question is from Kripa Delavkronda (sic) [ Devarakonda ] with Truist Securities.

Congrats on all the details. I have a question for Dr. Kristeleit, but I wasn't -- I want to make sure she's online still.

I'm here, yes.

Okay. So the retifanlimab data looks really strong. And you gave us a little bit of an idea of the market opportunity. I was just wondering. First, are you already using the drug in recurrent SCAC? And if approved, where do you expect it to be most used? You talked about 40% of advance patients. And if the drug were to be approved, what percent of your patients would you prescribe the drug to?

Thank you for that question. Let me just paraphrase it just to summarize. So the question here is -- for Dr. Rao (sic) [ Dr. Kristeleit ] is what do you see as the biggest opportunity for retifanlimab in squamous cell anal cancer. And if approved, what percentage of your patients would you offer it to? That's the gist of it.

Okay, thank you. So in terms of what this offers, well, clearly, for any patient with advance squamous cell anal cancer, this would be the treatment of choice. So patients with inoperable disease or advanced metastatic disease, this would certainly be a standard of care. And this will be a standard of care globally. In terms of the proportion of patients, it's usually about 40% of patients with anal cancer. So you saw that instance earlier about 10,000 new cases a year in the EU and America. So that, it's the ballpark there is about 40%. Most of those patients would be eligible, so near to 40%.

Thank you -- yes? Sorry. Go ahead.

I was just going to ask. I was going to ask about the strategy. So I know you haven't talked about the other indications where CDK2 might be relevant. If you can't talk today, can you give us an idea of when you can talk about the potential broader opportunity for CDK2?

Yes, yes. So beyond ovarian cancer and endometrial cancer, I think my comments were, next year, we'll provide an update on our approach to other tumor types. At some point in the next 6 to 9 months, we'll refine that guidance, but for now we're not going to give any more details.

Our next question is from Jessica Fye with JPMorgan.

This is Nick on for Jess. I was hoping -- I know you mentioned that you don't have duration response yet in ovarian, but any -- based on the data you have on hand, is there any color you can provide around that and what the duration of response could shake out to be as well as any insights into PFS?

Yes. So thank you. So obviously it's going to be a key point to use in potentially regulatory discussions and very important based on our accelerated approval discussions. The problem is -- one is limited numbers. And two is most of the responders in the curves Eka showed you are ongoing. So you can't really calculate, but the point she was making is many of those are already beyond 24 weeks. So we think we're in a very good place in terms of duration of response and then potentially obviously PFS as well, but it's hard to put precision on the number. But it's looking healthy is the best way I can describe it, yes.

Stay there for now.

Yes.

Yes. We need more follow-up in a lot of these patients.

I was hoping to ask another one, on the safety profile of CDK2. Just looking at that, I know there's discussion of it being consistent with the CDK2-targeting agent and hitting it on target, but it looks so far -- looks good so far. But at any of the higher doses or even at the 50 milligram BID or the 125 milligram QD, are you seeing any evidence of off-target or potentially any CDK1-related tox or any potential ocular tox in some of those TREAs (sic) [ TRAEs ]?

Yes. So let me take a shot. And you can complete it if you think I'm off base. So when you look at the PK data, at 50 milligram BID and at 125 milligram, maybe there's a little bit of CDK1 effect based on what we know about the selectivity of this molecule. Now to tell you for sure that some of the heme tox is CDK1, CDK2, I mean, that's impossible. What we know is that what we've seen in terms of heme toxicity has been limited to grade 3. I don't even know if there's an occasional grade 4. Some of the grade 4s are nonheme and they're rare. They're below 10% of the 2 dose levels that we're talking about. And they -- when you look at patient discontinuation and dose reductions, those were very, very rare events. So for me that's usually the measure. I mean either one of our doctors can comment on how they use these medicines. If you can keep patients on drug without reducing the dose, that tells you that the drug is tolerated. Dr. Kristeleit, you've used the drug. Do you want to comment?

I mean it's a really easy drug to give. I mean I've been involved in a lot of DDR inhibitor Phase Is. This is a really easy drug to give in the Phase I setting, to both endometrial and ovarian cancer.

Thank you.

Our next question is from Brian Abrahams with RBC.

Really appreciate the very detailed presentation. Just a couple for me. First, can you confirm that you haven't observed any ocular tox across the arms? Secondly, what's the right way to think about your expression threshold and internal diagnostic? Any potential FDA back-and-forth or possible pushback around it? And then lastly, just curious on your dose selection plan for the registrational studies; and whether, 50 milligram BID, we should think about as the most likely go-forward dose; or whether it might actually depend on whether you do monotherapy or combo.

Okay, let me -- we have not seen any ocular toxicity. We are not ready to discuss expression threshold. We think -- we've spent a lot of time in the past year working on companion diagnostic for this program. This is not something that started last week, so we think we're going to be in very good shape when we go talk to FDA. We'll have that conversation. We'll explain why the cutoff that we selected and the methodology we selected, so is there going to be a back-and-forth? We certainly hope there won't be. And we believe we're going to have a very productive and very constructive conversation with the FDA. When it comes to dose selection, I'm not going to start picking favorites. We have 50 milligram BID and 125 milligram QD. You saw the data. The follow-up for 50 milligram BID is longer, so I would caution about not putting your money on that dose yet. Once we have longer follow-up and a few more patients, then we're going to sit down -- first off, we're going to do an exposure-response analysis, which the team is already working on. And that will be followed again with, first, an internal conversation; and then a conversation with FDA once we have the dose selected. All this should lead, will lead to a pivotal study start next year.

Our next question is from Gavin Clark-Gartner with Evercore ISI.

So I have 2. First, on the CDK2 in ovarian, what was the response rate at the 50 mg BID dose if you pool the expansion and escalation portions?

We'll look at the number and we'll let you know. I mean they're small numbers. We can add them up, but we never looked at them that way.

Okay, got it. Secondly, what was the average time on drug until a lot of the responses have been seen? And what's your expectation for how much the QD expansion arms, the response rate may improve over time?

Yes, I'm not going to speculate how much better the data is going to get. We will do the experiment and we'll communicate it externally when we're ready. Do you want to talk in general terms about the speed of response, Eka?

[ Patients with ] sequential scans, majority of -- I was actually looking at it now. So majority of responses happened on a first or a second scan but then tumor shrinkage continues, basically.

Our last question is from Salveen Richter with Goldman Sachs.

This is Matt on for Salveen. Congrats on the updates. Just 3 quick ones, if I may. I know it's still early, but do you think the 50 mg BID might be best based on just what you know about the PK profile? Second, will you share PFS data from the study in the near term? And then lastly, on the companion diagnostic for cyclin E, will this be something that's straightforward to incorporate into a potential launch? Or would it be more of a lift?

So the companion diagnostics should be straightforward. I think it was mentioned today, I think, by one of the speakers IHC is a pretty straightforward test. It's widely available, no special equipment. I don't think that's going to be a problem. And I think, as I mentioned, we are on track to have that ready at the right time. I can't give a time line for PFS. We'll disclose PFS data when we consider data mature enough to be disclosed. That's what we try to do. That's why we waited until we had 200-patients-plus worth of safety data and the efficacy data you saw to disclose this for the first time. Once we have mature PFS, we'll disclose that. And the 50 milligram BID, I'll answer the same way. The 50 milligram BID cohort right now has longer follow-up. Does it look better because it's better; or because, as Eka explained, the responses deepen over time and the 50 milligram BID has had a little bit longer follow-up? It's unclear at this point. That's all the questions online. Let me come back to the room, see if there are any burning questions. We have time for 1, maybe 2, more. One, two, three. Thank you, everyone, for attending. I hope this was helpful. It was terrific for us to update you both on the results of an amazing pivotal trial as well as what we believe is now a program that puts us on the lead for CDK2 inhibitors, at a minimum in ovarian and endometrial cancer. Thank you, everybody, for attending. And enjoy Barcelona.

Thank you. This will conclude today's conference. You may disconnect your lines at this time. And thank you for your participation.