Good afternoon, and welcome to Vir Biotechnology's conference call to discuss the company's VIR-5500 Strategic Collaboration with Astellas and Positive Phase I Data, and 2025 financial results. As a reminder, this conference call is being recorded. [Operator Instructions] I will now turn the call over to Kiki Patel, Head of Investor Relations. You may begin, Kiki.

Thank you, operator, and welcome, everyone. Earlier today, we issued 3 press releases, including a joint release with Astellas, announcing a strategic collaboration with our PSMA-targeted T-cell engager, VIR-5500, a second release reporting the Phase I VIR-5500 data that will be presented at ASCO-GU, and a third release reporting our fourth quarter and year-end earnings. Before we begin, I would like to remind everyone that some of the statements we are making today are forward-looking statements under applicable securities laws. These forward-looking statements involve substantial risks and uncertainties that could cause our clinical development programs, collaboration outcomes, future results, performance or achievements to differ significantly from those expressed or implied in such forward-looking statements. Forward-looking statements include, but are not limited to, statements regarding the potential benefits of our collaboration with Astellas, that the closing of the Astellas collaboration is subject to the expiration or termination of the applicable waiting period under the Hart-Scott-Rodino Antitrust Improvements Act of 1976, the therapeutic potential of VIR-5500, our PRO-XTEN platform, our development plans and time lines, financial terms and milestone payments, and our cash runway and capital allocation priorities. These risks and uncertainties associated with our business are described in the company's reports filed with the Securities and Exchange Commission, including our Forms 10-K, 10-Q and 8-K. Joining me on today's call from Vir Biotechnology are Dr. Marianne De Backer, our Chief Executive Officer; Dr. Mark Eisner, our Chief Medical Officer; and Jason O'Byrne, our Chief Financial Officer. Additionally, Dr. Johann de Bono from the Institute of Cancer Research in the U.K. is joining us for our prepared remarks to provide a clinical and investigator perspective. Let me briefly outline today's agenda. Marianne will start by sharing the high-level overview of the strategic collaboration with Astellas and discuss how VIR-5500 has the potential to be a best-in-class T cell engager to address the significant unmet need in metastatic castration-resistant prostate cancer or mCRPC. Mark will then review the Phase I clinical data for VIR-5500, and he'll invite Dr. de Bono to walk through illustrative case examples. Mark will then summarize the broader data set and outline next steps for the program. Jason will cover the financial terms of the collaboration and provide an update on our 2025 financial results. And finally, Marianne will close the call, and we'll open the line for Q&A. With that, I'll now turn the call over to Marianne.

Thank you, Kiki. Good afternoon, everyone. Today marks a pivotal moment for Vir Biotechnology as we announced a landmark strategic collaboration with Astellas to advance the global development and commercialization of VIR-5500, our PRO-XTEN, dual masked PSMA-targeting T cell engager for prostate cancer. VIR-5500 is our most advanced immuno-oncology asset. And today, we are sharing new Phase I data that will be further presented by Dr. de Bono at ASCO-GU this Thursday, February 26. Together, we believe these milestones position VIR-5500 for rapid advancement and allow us to move forward with both urgency and discipline. The collaboration we've announced with Astellas combines their deep global experience in prostate cancer with our differentiated T cell engager powered by the PRO-XTEN masking technology. Structurally, the collaboration is designed to accelerate the development of VIR-5500 across both earlier and later lines of prostate cancer, unlocking a significant market opportunity while meaningfully derisking our pipeline of cancer immunotherapies more broadly. Importantly, the new Phase I data that we are sharing today show a compelling emerging safety and efficacy profile. While still early, these data increase our confidence that VIR-5500 has the potential to be a best-in-class T cell engager for the treatment of prostate cancer. And finally, today's update is also an important validation for the broader PRO-XTEN platform approach, which we believe can unlock opportunities to develop next-generation T-cell engagers in solid tumors. To understand the significance of this opportunity, it's important to consider the current landscape in prostate cancer. Prostate cancer remains a significant global health burden, representing the most common diagnosed cancer among men with 1 in 8 men being diagnosed in their lifetime. Despite significant progress in treatment, the 5-year survival for patients with mCRPC is only 30% with an estimated 100,000 mCRPC patients in the U.S. and Europe. Across…

Thank you, Marianne, and good afternoon, everyone. I'm pleased to walk you through the latest Phase I data for VIR-5500, which have been accepted for an oral presentation at the ASCO-GU conference taking place later this week. This is the only dual mask T-cell engager under evaluation in prostate cancer, and the emerging signals we are seeing reflect the potential of the PRO-XTEN masking platform to unlock the promise of TCEs for the treatment of solid tumors. As of the January 9, 2026, cutoff, we enrolled 58 patients with advanced metastatic castration-resistant prostate cancer in weekly and every 3-week monotherapy dosing regimens. Importantly, all dose escalation cohorts have cleared the dose-limiting toxicity period. Our dose escalation strategy leveraged insights from our broader TCE platform, enabling us to advance efficiently from very low initial flat doses to step-up dosing with the highest Q3 week maintenance dose of 4,000 micrograms per kilogram. Throughout this escalation, prophylactic steroids or IL-6 blockade were not required and was only explored in 3 patients at the highest 4,000-microgram cohort. Based on the emerging data for VIR-5500, our development focus is now centered on doses at or above 3,000 micrograms per kilogram given once every 3 weeks. These dose levels are where we are seeing the clearest clinical signals. Here, we see the baseline characteristics for patients enrolled in the study. Participants were heavily pretreated with a median of 4 prior lines of therapy and some receiving up to 7. 95% had received prior taxane chemotherapy. These are patients with extensive disease burden, 93% presented with bone metastases, 45% of visceral involvement and 18% of liver metastases. Liver metastases are associated with rapid disease progression and poor response to existing treatment modalities. Approximately half of the study population was RECIST evaluable at baseline, enabling assessment of radiographic…

Thank you, Mark. The 5 case studies I'm going to share with you, many of whom are my patients, demonstrate multiple, impressive, biochemical and radiology responses to this dual masked T cell engager, VIR-5500 in sufferers from metastatic and heavily pretreated prostate cancer. I serve these men in my clinics and have witnessed their experiences and symptomatic improvements on this agent. In advanced metastatic prostate cancer, many subjects experienced significant pain, especially bone pain. VIR-5500 resulted in pain disappearing following treatment in many patients as their disease regressed. Reduction in such pain is incredibly important to these men that we serve. Critically, I believe that the data from this trial show that the dual masking approach works at minimizing cytokine release syndrome, also known as CRS. We are reporting multiple amazing responses with little clinically significant CRS. In fact, circulating interleukin-6 levels remain low and relatively unchanged following treatment with VIR-5500 across these patients, with usually only grade 1 fever being observed, which is really quite remarkable and different to many other T-cell engagers we have studied that have resulted in cytokine release syndrome with hemodynamic instability requiring patient admission, vasopressors and treatment with oxygen, et cetera, for respiratory compromise. In addition to this absence of requirement for prophylactic steroids or tocilizumab in this trial, very few subjects have required treatment with steroids after receiving this drug, VIR-5500. And this is really quite important since steroids are immunosuppressive and can limit immunotherapy with T cell engagers antitumor activity. So this dual masking by limiting CRS has major advantages. Now let's go through these 5 cases in turn. Case study 1 demonstrates complete resolution of multiple, approximately 14, liver metastasis after 9 weeks of therapy with a 99% PSA fall, really very impressive. This was a 63-year-old man who had received…

Thank you, Dr. de Bono. Your clinical perspective on these patients treated with VIR-5500 is invaluable as we continue to advance this program. Turning back to the full study population. The safety profile of VIR-5500 remains favorable. The table on the left displays treatment-emergent adverse events for all patients treated with weekly and every Q3-week dosing. The emerging safety profile supports a wide therapeutic index. We've seen no dose-limiting toxicities to date with grade 3 or higher treatment-related adverse events in only 12% of patients. Most of these are laboratory abnormalities. We had only 2 patients discontinue treatment due to an adverse event. The first patient experienced spinal cord compression that was due to his underlying disease and not attributable to VIR-5500, and the second patient due to treatment-related blurred vision. The bottom half of the table on the left displays treatment-related adverse events at the highest doses of more than 3,000 micrograms per kilogram Q3 week. As you can see, the AEs were mostly grade 1 and 2. The grade 3 and higher events are listed at the bottom, but primarily consist of neutropenia and tumor flare, which are indicative of immune-mediated engagement. We observed 2 events of treatment-related blurred vision with unclear pathophysiology and nonspecific MRI findings that improved toward baseline visual acuity. Overall, limited CRS was observed in high-dose cohorts of 3,000 microgram per kilogram and higher. The bar graph on the right shows cases of CRS by dosing cohort. As you can see, all cases were low grade, either Grade 1 or 2 with no Grade 3 CRS. The Grade 1 events were only fever, treatable with antipyretics. We did not require prophylactic steroids or anti-IL-6 therapy overall. In the highest dose cohort of 4,000 microgram per kilogram, we did evaluate pre-dose steroids in cycle 1.…

Thank you, Jason. Today's announcements mark major progress towards our goal of building a world-class cancer immunotherapy program, the vision we set out for our company just 18 months ago. We believe the data we have shared today for VIR-5500 validates the potential of the PRO-XTEN platform, enabling more rapid advancement of our pipeline of differentiated T cell engagers, and positioning Vir Bio to be a leader in immuno-oncology. We have a lot to look forward to across our pipeline. Our HER2 and EGFR programs use the same dual masking architecture and benefit from shared learnings. We plan to share Phase I dose escalation data from our HER2 program in the second half of this year. The PRO-XTEN's platform plug-and-play design also lets us rapidly engineer new masked T-cell engagers for high-value solid tumor targets, creating a sustainable pipeline of differentiated therapies. We have thus far developed 7 preclinical programs and will progress to development candidate selection by early 2027. As we conclude today's presentation, I want to return to what fuels everything we do at Vir Bio, transforming patients' lives, especially people living with devastating diseases who are vastly underserved by current treatment options. The partnership with Astellas will not only allow for swift advancement of VIR-5500, but also positions us well for more rapid pipeline expansion. All this gives us further flexibility to consider how we develop our pipeline assets and continue to unlock earlier value for patients and our shareholders alike. And importantly, by combining Vir Bio's potential best-in-class T cell engager with Astellas' global capabilities, we will bring complementary strengths to one of the biggest unmet needs in oncology. Together, we can move faster and maximize the potential impact of VIR-5500 for people living with prostate cancer. I would like to close by sincerely thanking the patients, their families and the investigators who have supported the development of this program. With that, I'll turn the call back over to Kiki to begin the Q&A session.

Thank you, Marianne. This concludes our prepared remarks, and we will now start the Q&A section. Joining me for the Q&A are Marianne, Mark and Jason. Please limit questions to 2 per person so we can get through all of our covering analysts. I'll now turn it over to you, operator.

[Operator Instructions] Your first question comes from the line of Paul Choi with Goldman Sachs.

Congratulations on the data as well as the deal. Two questions for us, please. First, either for Marianne or Mark. Can you maybe comment on the range of PSA responses you've seen by prior line of therapies, particularly with regard to prior radiotherapy. Any details there would be helpful, both on the PSA50s and PSA90s. And my second question is, how do you think your data today potentially reflects on the probability of success for your other T-cell engager programs? And just what is your level of confidence that we can make a reasonable inference of higher probabilities of success for your other programs?

Thank you, Paul. Really appreciate it. I'll start with answering your second question and then turn it over to Mark. So we really believe that the data we have showed you today validates our dual masking steric hindrance approach, so really the PRO-XTEN masking. You've seen that the lower systemic immune activation is reflected in limited CRS toxicity, very low incidence of high-grade treatment-related AEs and very limited number of PSMA target-related AEs, again, all very low grade. Also, you saw that we can reach now a wider therapeutic window. So we are able because of the mask to dose higher and less frequently. So we have actually selected a preference for every 3-week dosing. And then thirdly, you have seen that there's great concordance between PSA responses, RECIST responses, PSMA PET responses, which all show great on-tumor engagement. So we think this all bodes really well for our other programs. Of course, every program is unique, but we have, I think, really here shown a validation for the technology. As to your question of range of PSA responses, especially with patients that have been exposed to prior radioligand therapy, Mark, can you...

Sure. Yes. Very good question, Paul. So first of all, we have a very heavily pretreated population with a median of 4 prior lines of therapy. The vast majority of patients have received taxanes. So we -- and we do think we have very strong PSA responses, particularly as we get into the 3,000 microgram per kilogram doses and above. I'd direct you to Case # 4, the one that was presented by Dr. de Bono. This patient had received radioligand treatment, an actinium conjugated agent, PSMA-targeted agent. That patient had a PSA99 response and a complete response in the target lesions. And that patient also had evidence of -- in the lymph node of PSMA decline in terms of expression and T cell abundance in the lymph node 5 weeks after treatment. So we don't have a lot of data yet in the post-RLT setting. We are continuing to look at those patients, of course. But at least in this one patient post-RLT, very, very promising results in this individual. In terms of other prior lines of therapy, it's difficult to say because the patients were just generally very heavily pretreated. So we haven't been able to disambiguate any specific effects of prior line of treatment on PSA responses, but they do appear to be strong across the board, particularly when we get to the higher doses.

Yes. I mean the only thing I would add is we have 2 patients that were exposed prior to the steep TCE, so we have annotated those on the slides, if you would want to go and have a look.

Your next question comes from the line of Cory Kasimov with Evercore ISI.

This is Josh Chazaro on for Cory. Congrats on all the progress. And clearly, you guys have been busy executing here. And question here is, can you give us a little bit more color on what the next steps are before you and Astellas move to Phase III? And are there plans for you and Astellas to explore additional dosing schemes beyond the Q3 week?

Yes. Thank you for that question. Is it Josh? Yes, Josh, thank you for the question. Mark, do you want to take that?

Absolutely. So yes, we're very excited about the partnership with Astellas, and we're also very excited about the next steps of the program. So we do plan to get into -- we've selected a go-forward dose. We plan to get into expansion cohorts in Q2, very shortly in this year. We will be having late-line mCRPC, which is the population here as a monotherapy. We'll have a combination with enzalutamide in the early line taxane-naive setting. And we also will be doing some dose optimization in parallel to satisfy the goals of the Project Optimus to satisfy the FDA's requirements there. So we will be working with Astellas. And I should also mention the combination of metastatic hormone-sensitive prostate cancer and expansion cohort. So taken together with the expansion cohorts, with the dose optimization work, we expect to get into Phase III in 2027 and to be well positioned now.

Yes. This was exactly why we were so excited about entering into the partnership with Astellas. It allows us to accelerate the clinical development and also really broaden the potential and expanding the trials to reach more patients.

Your next question comes from the line of Roanna Ruiz with Leerink Partners.

So 2 questions from me. On the Astellas collaboration, I'm just curious, how are you thinking about unlocking resources for investing into the broader PRO-XTEN platform and thinking about other solid tumor indications? And what sort of calculus will you do in terms of thinking about prioritizing certain programs over accelerating others? And my second question with the larger cohort of patients evaluated on VIR-5500, how does this evolve your thinking about where VIR-5500 could be positioned within the sort of treatment paradigm in terms of line of therapy, combination versus monotherapy and thinking of the future?

Yes. Thank you, Roanna. Yes, as it relates to resourcing, obviously, teaming up with a world-class player in the prostate cancer field and Astellas has entirely internal development capabilities, that will help us tremendously in again, accelerating the VIR-5500 program. From a finance perspective, also it allows us to divert certain expenses to other programs and potentially accelerate those as well. So we think that the collaboration certainly has a lot of benefits beyond just for VIR-5500 alone. As to where to position VIR-5500, do you want to...

Sure, absolutely. So yes, so we are very excited about the progress we've made so far and the data that we've presented to you today. And we plan to really be able to address a broad range of patients with metastatic prostate cancer, including late-line mCRPC and a monotherapy. And these are the data we presented you today, a more early line mCRPC in combination. And as you recall, we've been dose escalating in combination with enzalutamide in the frontline taxane-naive setting already, and that's going well. And then in addition, in metastatic hormone-sensitive prostate cancer in combination. So we really are very excited to continue to progress the program. And these are certainly 3 high unmet need populations within the metastatic prostate cancer setting that we think we can address.

Your next question comes from the line of Mike Ulz with Morgan Stanley.

Congratulations on the data and the deal as well. Maybe just one on VIR-5500 and durability. I guess, maybe just talk about your level of confidence there that some of this very strong early data that you're seeing can be sustained over a longer term.

Yes, absolutely. So we are very encouraged by the RECIST responses that we've seen, particularly those that have occurred up to 27 weeks and the fact that we're able to confirm RECIST responses in patients. We're also seeing a concordance of RECIST responses with PSMA PET responses and deep PSA responses, which we also think is further evidence of the efficacy we can achieve. Also the case studies that Dr. de Bono presented to illustrate patients with up to 1 year of durability, which represents the potential, I think, of VIR-5500. So taken together, we're really pleased with the emerging evidence of durability in the program.

Yes. Makes sense. And maybe just one more question for me. Obviously, you're presenting the data at ASCO-GU later this week. Just curious if the data you shared with us today is the same that will be presented at the meeting? Or are there any additional updates or data points we should be looking out for?

Yes. The oral presentation at ASCO-GU this Thursday by Dr. de Bono, I mean, these oral presentations are rather short. So there won't be additional data, but it will be a subset of this data.

Your next question comes from the line of Phil Nadeau with TD Cowen.

Congratulations on the data and the Astellas collaboration. Two from us. First, in terms of the go-forward dose, could you give us more information on what that dose is? And I guess, in particular, was there a dose response in those doses above 3,000 microgram per kilogram Q3W? Or how did you identify that go-forward dose? And then second, just a clarifying question. It sounds like there's no Grade 3 CRS in doses above 3,000. Was there any below? It didn't seem to be from one of the slides, but we just want to make sure we saw that correctly.

Thanks for the question. So in terms of the go-forward dose, we've done a lot of work on that, integrating safety, the efficacy, PSA, PSMA PET, RECIST responses and so forth. And we have selected a go-forward dose. As you can appreciate, we have now a partner Astellas, which we're thrilled to have on board. And so we're not going to be communicating the exact dose today because that's something that involves both of us in the partnership. But I can tell you, we'll be in the 3,000, 3,500 maintenance dose range. In terms of dose response above -- at or above 3,000, I mean, you can see clearly, we showed you all the data for all the doses tested for PSA responses. I think you can see there's a compelling dose response across all of those doses. Once we get above 3,000, there still is some dose response, but primarily, we're in a range where we're seeing very, very strong efficacy and a very strong therapeutic index. So we feel confident that we've identified a go-forward dose that really optimizes the therapeutic index moving forward. In terms of Grade 3 CRS above 3,000 mg per kilo in the go-forward dose range, we've seen no Grade 3 CRS. We did observe one Grade 3 CRS in an earlier dose cohort in a low-dose patient who had intra-patient dose escalation and had one episode of Grade 3 CRS that recovered rapidly and the patient did very well.

Your next question comes from the line of Etzer Darout with Barclays.

Congrats on this data set. Really nice to see. Just one question I had on the go-forward dose. Just wondering if in the combination study that you've initiated if new patients would be enrolled at these effective doses of greater than 300 micrograms per kilogram. And then also, is there an opportunity with this data in hand to maybe convert to a flat dose versus a weight-based dose in these patients and whether you see this as a potential opportunity moving forward for the molecule?

Yes. Thanks. Good question. So in terms of the go-forward dose in combination with enzalutamide, I mean, we anticipate that the dose should be consistent in the 2 populations. We are doing a -- we're almost complete with the dose escalation in combo with enzalutamide frontline mCRPC, just to confirm that there's no issues there, but we do anticipate should be very similar or the same. In terms of flat dose, right now, we do not have plans for a flat dose. I mean it would be possible in theory, but we're still using the microgram per kilogram dose.

Your next question comes from the line of Joseph Stringer with Needham & Company.

Just a follow-up on the deal with Astellas for VIR-5500. What might this mean for the rest of the oncology pipeline? Is there an opportunity for some of these programs to be stand-alone for Vir? Or do you see the long-term strategy here to seek partners or to partner these programs? And then a question for Dr. de Bono, just based on these updated data, what are the read-throughs in your outlook? And where do you see potential for VIR-5500 in earlier lines of mCRPC therapy?

Thank you, Joey. Yes. So the Astellas deal, again, was a very strategic choice based on the fact that first of all, the unmet need in prostate cancer is incredibly high. The landscape is evolving very quickly. We thought that time to market is most important. So we really were looking for a global partner with scale and with aligned incentives that would help us accelerate the program. And also, as I mentioned earlier, really allow us to really grow the pie, so to speak, and see how much value -- how much more value could we bring to a broader subset of patients. And that's everything that Astellas collaboration really delivers while we can retain a significant portion of the value through the 50-50 profit split, the milestones and the ex U.S. royalties and so on. For the rest of our pipeline, we are going to be very strategic and thoughtful in a similar vein. A lot depends again on the competitive landscape on the size of the commercial opportunity and the indications about the financial need to bring these indications forward. So we will be making very thoughtful choices on what to partner, how to partner, what to keep for ourselves 100%. Also just to remind you that we have 7 preclinical masked T cell engagers. And for sure, on the preclinical programs, this is just too much for us to move forward on our own. We will certainly be looking for partners there. And because of the plug-and-play nature of the platform, again, it allows us to move actually pretty quickly in preclinical research. So you will be seeing that we will be looking for partners in some of those areas. Your second question was related to read-through. Okay. Dr. de Bono is not available here during the Q&A. But Mark, do you want to take that?

Sure. So your question was about the potential in earlier lines of metastatic prostate cancer. So yes, we definitely believe there is potential there. We are planning first-line taxane naive metastatic castration-resistant prostate cancer as an expansion cohort in addition to metastatic hormone-sensitive prostate cancer. So we are really looking across the metastatic prostate cancer landscape to help these patients who really need better treatments.

Your next question comes from the line of Patrick Trucchio with H.C. Wainwright.

Congrats on the data and the deal. I guess just a follow-up on the Astellas deal with Vir having an option in the U.S. to co-promote, what would that look like? And at what point in the development process would you be able to exercise that option?

Okay. Thank you, Patrick. Yes, so we have an option to co-promote alongside Astellas in the U.S. And up to a year before the start of our pivotal trials, we will be able to make that decision.

Great. And then if I could, just a follow-up question for Dr. de Bono. I'm just wondering, just based on the data that you've seen so far, how confident are you that this treatment could potentially move into frontline? And what would you need to see in order to give you that confidence?

Thank you, Patrick. So unfortunately, Dr. de Bono is -- [ given time ] so it's not available for this Q&A. He will be at the ASCO-GU this Thursday. But maybe, Mark, do you want to...

Yes. So yes, I encourage everybody who can to attend this talk or to understand his perspective there. But for sure, we see potential across the metastatic prostate cancer landscape. We have generated, we think, compelling early data in the late-line mCRPC setting. We're currently enrolling patients in dose escalation in the frontline taxane naive mCRPC setting. We would anticipate based on what we've seen to date that we should have an effective drug in that population potentially. They do have lower disease burden overall, and we think our mass TCE approach should work, and we will be generating those data. And as I mentioned before, we are also going to be looking at the metastatic hormone-sensitive prostate cancer setting as well. So that gives you kind of an idea of where we're heading.

Your next question comes from the line of Sean McCutcheon with Raymond James.

Congrats on the strong data. A couple from us. Given the seeming lack of strong dose response on CRS and lack of DLTs, how are you thinking about the limit on the higher end of the range of dose escalation, whether that's saturating on enzymatic activity or otherwise? And second question, how are you guys thinking about the partnership with Astellas and optionality for combining VIR-5500 with other ARPIs beyond enzalutamide?

Thank you, Sean. I will take your second question. Obviously, in partnership with Astellas, we will be determining our future combination strategy, which, of course, could be broader than just the ARPIs, but that will be something that we will need to inform you about at a later time point. Your first question -- what's related to the dose...

Yes. So you're asking about dose response for CRS and efficacy and what's the limit of the dose on the high end. So a couple of points there. I mean our whole goal has been to maximize and optimize the therapeutic window to get the best possible safety with the minimum possible adverse events and CRS. So we've taken a careful look across the data set on all the efficacy parameters and safety parameters, including PSA, including RECIST, PSMA PET, all the safety events, CRS, et cetera. And we think we've gotten to a range in the 3,000 to 3,500 maintenance dose, and we have a specific dose there, which we can communicated in combination with our partner at a later date where we think we really optimize the therapeutic index and can move forward into expansion cohorts in Q2 of this year.

Your next question comes from the line of Alec Stranahan with Bank of America.

Congrats on the really clean update here. Maybe first, just following up on an earlier question regarding durability. I'd be interested to hear your thoughts on how we could correlate PSA declines with -- as maybe a leading indicator for what we could expect on PFS with longer follow-up and when you think you might be in the position to update the markets with that data? And then second, in the 6 patients with the evaluable PSA but not RECIST, assuming these will feed into the overall PFS analysis, could you maybe talk about why we weren't -- those weren't available at baseline and what your prediction might have been in terms of response, given many of them have fairly deep PSA declines, maybe on disease control rate or something else?

So your first question has to do with durability and how do we think PSA declines will track with PFS, right? So in general, I would say that the deeper PSA declines, particularly PSA90s and PSA99s are associated with more durable responses. And we are very encouraged to see that we have some very deep PSA declines, PSA90s and PSA99s. In terms of radiographic PFS, I mean, you're correct, we did not present those data at this update because as you can see, the data, particularly for the high-dose cohorts is still evolving and the patients are still maturing over time. So those data really aren't available yet. And in terms of exactly when we'll present further data, I think that we'll just have to give guidance on that at a subsequent time point. Could you clarify -- I wasn't quite sure I got the second question. Was that why were not some patients not PSA evaluable? Or what was the question?

Well, I think there was 6 patients that had a PSA like, out of the 17, 6 were not evaluable for RECIST. I guess, what was sort of the -- you couldn't get the scans at baseline. Was that kind of the driver there? And then I guess, what would you sort of expect in terms of PFS for those patients?

Yes. No, thanks for clarifying. I appreciate it. So yes, so in the 3,000 or above, we had 22 patients in the cohort. We had 17 patients who were PSA evaluable. Two of those patients are just early at the time of the clinical cutoff. So we will get those subsequent PSA values, they weren't part of the data set. So 2 out of 5 are early, then there's a coming. 3 out of the 5 discontinued early. So we will not have -- they won't be PSA valuable. And that's very typical for prostate cancer trials in the late-line setting. So these patients are quite sick with a very heavy disease burden. We have 11 patients who are RECIST evaluable. And among those, we had 5 responses, 4 were confirmed and 1 is still waiting for the next confirmatory scan between week 9 and week 18. So that one is coming in time.

This concludes the call. Thank you for participating.