Thank you, operator. Good afternoon, everyone. Thank you for joining our third quarter 2025 results conference call. As usual, before we begin, I would like to remind you that we'll be making a number of forward-looking statements during this call, including, without limitation, those forward-looking statements identified in our slides and the accompanying oral commentary. Forward-looking statements are based upon our current expectations and various assumptions and are subject to the usual risks and uncertainties associated with companies in our industry and at our stage of development. For a discussion of these risks and uncertainties, we refer you to our latest SEC filings as found on our website and as filed with the SEC. In a moment, I will hand over to Leon Patterson, our Executive Vice President and Chief Business and Financial Officer, who will provide an overview of our recent business and partnership updates, along with financial results for our third quarter 2025. Following this, Dr. Sabine Mikan, our Senior Vice President of Clinical Development, will provide progress updates on our Phase I programs ZW191 and ZW251. We will then pass the call over to Dr. Paul Moore, our Chief Scientific Officer, who will provide a brief overview of recent R&D developments. At the end of the call, Leoni, Sabeen, Paul and Ken Galbraith, our Chair and CEO, will be available for Q&A. As a reminder, the audio and slides from this call will also be available on the Zymeworks website later today. I will now turn the call over to Leon.

Thank you, Shrinal, and good afternoon, everyone. I'd like to start the call by walking you through recent progress on both clinical and preclinical programs within our wholly owned product pipeline. As you know, our team was pleased to present initial clinical data from the Phase I trial of ZW191, an antibody drug conjugate targeting folate receptor alpha at the ENA conference in October. Sabeen will provide a recap of the data we presented during our poster presentation later on today's call. We are encouraged by the preliminary Phase I data for ZW191, which provides early clinical validation of our ADC approach. And we are pleased to announce that we have dosed the first patient in the Phase I clinical trial of ZW251, a DAR4 ADC targeting GPC3 in hepatocellular carcinoma. Again, Sabeen will talk more about the trial design later on today's call. We also continue to present preclinical data of ZW1528, a bispecific inhibitor of IL-4 and IL-31 to address respiratory inflammation at the European Respiratory Society Annual Congress. Additional information can be found on the ERS Congress website, and a copy of the poster is available on the Publications page of Zymeworks website. Meanwhile, our partnered programs also continue to provide encouraging data at ESMO. Our partner, Jazz, presented a trial in progress poster on the DiscovHER PAN-206 Phase II study of zanidatamab in HER2 overexpressing solid tumors as well as a 2-year follow-up in first-line metastatic colorectal cancer showing durable responses and a favorable safety profile. In addition, yesterday, Jazz announced that the ITT population for the primary PFS and interim OS analysis of the HERIZON-GEA-01 trial will include the full patient population enrolled in the study of 920 patients. Also at ESMO, J&J presented translational findings from the first-in-human study of pasritamig in metastatic prostate…

Thank you, Leone, and good afternoon, everyone. I'd like to start off by providing a recap of the initial Phase I data for ZW191 as presented at the AACR-NCI-EORTC conference last month. As it pertains to the safety, we are encouraged by the tolerability profile that we've seen. The safety profile for ZW191 allowed us to escalate dose up to 11.2 milligram per kilogram, which is quite high for topoisomerase payload of this potency similar to deruxtecan. Across all treated patients, there was a low incidence of grade 3 or higher treatment-related adverse events and adverse events leading to dose interruptions or reductions were infrequent. The most commonly reported events were nausea, fatigue and anemia, which are generally consistent with our expectations for an ADC. Importantly, there were no serious treatment-related adverse events, no discontinuations due to adverse events and no deaths observed in this study. These findings support a favorable safety profile, particularly in a population that has been heavily pretreated. Overall, these data gave us confidence that this drug is well tolerated at clinically active doses, providing a solid foundation for ongoing and future studies. Moving now to the efficacy results. This slide shows the waterfall plot summarizing the best change in tumor size across dose levels. What we see here is also very encouraging. There are meaningful reductions in tumor size across multiple dose levels with objective responses observed at doses as low as 3.2 milligram per kilogram and the majority of patients continuing on treatment at data cutoff. Importantly, these responses were seen across the spectrum of folate receptor alpha expression, an important observation as we think about future development and patient selection. In participants with gynecological cancers dosed between clinically relevant doses of 6.4 and 9.6 milligram per kilogram, we observed an objective response rate…

Thank you, Sabeen. I'd like to just add a few final thoughts on the developments disclosed this quarter for both ZW191 and ZW171. Firstly, the initial data presented on ZW191 provides important translational insights that could help accelerate and reduce risk in the future development of ZW251 and other pipeline ADCs using our ZB06519 payload. As you can see on this slide, behind 191 and 251, we also have preclinical stage candidates targeting more novel antigens such as Ly6E and PTK7. Also, our NaPi2b program remains IND ready, and we continue to explore next-generation ADCs. Importantly, each of our ADCs has been tailored to factor in target biology by toggling drug-to-antibody ratio and the Fc modifications. Furthermore, we also ensure to utilize the most optimal antibody to deliver an internalized payload, whether this be a superior monoclonal antibody to benchmark as ZW191 or LE or a biparatopic antibody such as in the case of PTK7. Our approach of tailoring these parameters to target biology, patient population needs and preclinical safety efficacy data aims to ensure optimal therapeutic windows while minimizing off-target toxicities. Secondly, I wanted to touch briefly on our decision to discontinue the development of ZW171 and importantly, the valuable insights, both scientifically and operationally that we took from this experience. Internally, we hold ourselves to very high standards when it comes to our target product profiles. That discipline is important because we have a broad and productive pipeline, and we want to ensure our capital and our focus go to programs with the clearest path to meaningful patient benefit. Based on the totality of the dose escalation data, we concluded that as a monotherapy, this program did not fully meet our internal threshold to advance further within our portfolio as it was unlikely to support a benefit risk…

Thanks, Paul. Over the last 2 years, we've redefined what this company can achieve by combining R&D innovation, smart partnerships and disciplined capital allocation to help deliver potential best-in-class therapies while helping to grow shareholder value. Our partnership-based model continues to generate value today while also providing opportunities for growing potential cash flows. We plan to continue leveraging partnerships across our wholly owned pipeline to bring in external capital and accelerate development. We believe this approach allows us to main control of our R&D innovation while helping to derisk clinical development and to help ensure that every investment we make has the potential to contribute meaningfully to durable value creation. As we look beyond important near-term events for our pipeline and partner programs, our long-term focus is on compounding returns from Ziihera and protecting and enhancing future cash flows that can be reinvested to drive the next wave of innovation. With this in mind, this quarter, we announced some changes to our Board of Directors to align governance and leadership with the next phase of our strategy. We welcomed 2 new directors in August and 3 members transition off the Board effective today. We'd like to thank those 3 directors for their service to Zymeworks. In October, we appointed Dr. Adam Schayowitz as acting Chief Development Officer to help advance our portfolio and strengthen our partnership-driven strategy. With this refreshed leadership, we believe we're well positioned to transit our scientific innovation into a scalable model that builds durable royalty streams and deliver sustainable long-term value for our shareholders. To close, I want to emphasize that our capital allocation decisions, whether investing in R&D, advancing partnerships or returning capital through share repurchases, all serve one purpose to help build sustainable long-term value. Our R&D priorities remain focused on programs with clear differentiation and strong scientific rationale, and we'll continue to fund those using partnerships to extend our reach and offset development risk. Those collaborations also aim to provide a meaningful revenue floor through milestones and royalties, giving us the flexibility to invest with conviction and discipline. This is how we plan to sustain momentum through focus, partnership and the power of compounding. I want to thank you for your continued support. I'd like to turn the call back over to the operator for the question-and-answer session. Operator?[ id="-1" name="Operator" /> The first question comes from the line of [Technical Difficulty].

Can you hear me?

Yes, we can hear you.

Perfect. Congrats on the progress. Two from me, if you don't mind. First one, we heard it from Jazz yesterday and I think earlier today as well. But wanted to get your thoughts perhaps on the update in the PFS analysis for HERIZON-GEA to include the ITT rather than the PITT population, your thoughts here and perhaps what drove that change?

Yes. Thanks, Charles. I think Jazz provided some guidance on that yesterday in their earnings call in the prepared remarks, I think in question-and-answer session. We don't have anything to add beyond what Jazz has shared other than that we're aligned with the regulatory strategy that they laid out for the readout of HERZON-01 and how to analyze that data. So I really can't add anything beyond that.

Got it. Fully understood. And perhaps for my second question, I want to say congrats on the folate receptor alpha data at Triple meeting. That was quite impressive. Kind of also wanted to get your thoughts on what does this mean for GPC3, especially when we're thinking about a DAR4 construct in the liver cancer population. And similarly, if we see anything that comes close or is similar or even exceeds what we saw with 191, what would your thoughts be on potential development in-house versus partnership versus out-licensing of this asset in liver cancer?

Yes. No, good question, Charles. Yes, we're intrigued as your question suggests as well and looking forward to continued recruitment of ZW191 in dose escalation, moving to dose optimization, which provide a larger, more mature data set. At the same time, as we announced, we're recruiting patients now in the ZW251 study. So far, our clinical execution is as good as it has been to date with our prior programs. We're looking forward to that. I think in terms of what we think about that, I think maybe I'll give Sabeen and then Paul both a chance to add their flavor to that because it's a really, really interesting intriguing question for us as well. So I don't know, Sabeen, if you want to go first and I'll ask Paul to follow up.

Yes. I can go first. So as you know, hepatocellular carcinoma is a population with very high unmet medical need, particularly post first-line setting. There are not many treatment options for those patients. And that's why we think we should be able to create a difference given the construct of our ADC and what we've observed in ZW191 based upon the clinical data that we've observed. One of the key concerns with the hepatocellular population is concern for safety because this patient population often is very fragile and they have underlying liver disease. So the concern for safety is very important. And it is for this reason that we have selected DAR4 for this ADC molecule. And given the safety profile that we've observed with ZW191, we're fairly confident that we should be able to have a good safety profiles and to be able to have a therapeutic window in terms of treatment for hepatocellular carcinoma patients. I'll pass over to Paul.

Yes. No, I think Sabeen really captured the key points. I think the tolerability -- I mean, from the 191 study, it was both the tolerability was really what we were hoping for, but we also got the efficacy. And we've gone with the DAR4, we know from preclinical studies that we can maintain the same -- we can get to the same activity level with that. So we were really being careful on just making sure we had the most tolerable molecule to develop in such a challenging cancer indication. And I think the data from the Phase I sort of supports that we're in the right direction with the way that we selected the payload. We were very careful in how we pick that payload in the sort of the space of the topisomerase inhibitors that it would support a tolerable profile while maintaining our ability to get good dose into patients, and you could see that from our data. I think ultimately, we want the molecules to be combinable with other modalities as well so that we can go up in line. But obviously, first, we want to establish the profile as a monotherapy, and this really energizes us now after seeing the 191 data to really chase after the 251. [ id="-1" name="Operator" /> The next question comes from the line of Yaron Werber of TD Cowen.

Congrats as well on the folate receptor alpha. I got maybe a couple of questions actually on the pipeline. Maybe the first one, while we're staying on GPC3, B1 today on their call said that with their bispecific GPC3 4-1BB, they actually established proof of concept. So they're moving forward. That's definitely very encouraging. In terms of the payload that you're using is irinotecan and typically -- I'm sorry, a TOPO1 kind of based payload. That's not -- are TOPO1 typically used in liver cancer? And kind of maybe give us a little bit of a sense from the preclinical data, what are you expecting in terms of showing efficacy? And then secondly, for the next IND in the first half of next year, the DLL3 CD3, CD28 trispecific, we know DLL3 is a great target, and we've seen a lot of activity with both the bispecific on the market and the ADC. CD28 has not worked out so far in most other cases. So maybe what makes you more optimistic this time around?

Yes. I'll let Paul talk about the DLL3 and then maybe let Sabine and Paul both comment about CDC3, that's okay.

Yes. So yes, maybe I'll -- yes, I'll take the second question first. And then -- so great question, Yaron, on why do we think we can make CD28 work where others have had challenges, right? And so I think we do take precedents from the CAR-T space where adding in co-stimulation has shown benefits. So something like CD28 or 4-1BB that you refer to in the context of the B1 molecule. But -- and a lot of people have chased after that because of the attraction of getting that CD28 costimulatory signal to the T-cell to maintain or enable activity that you don't achieve just by having signal 1 through CD3. And I think what the challenge has been is actually getting that timing, that simultaneous engagement of CD3 and CD28 in the kinetics and the timing that you need to get that benefit. So that is what we took the challenge on when we developed a trispecific so that we knew that when we engage CD3, that T-cell could be then engaged with CD28. And no one's really developed a solution until what we think we have the solution for that. And so that's where we feel we can make an impact based on our preclinical data that gives us encouragement that we'll see that impact in the clinical setting. So it's a little bit to do with just the way we design it. Others have tried doing CD3 bispecific plus the CD28 bispecific. And in some cases, that may work. But we feel a more precise way is to hit the same T-cell with the primary and the secondary signal in a concert in a manner that can be done with a single molecule. So that's the DLL3. And certainly, we've presented data, we'll show a little…

Sabeen, anything you want to add from a medical perspective with this patient population and the idea of chemo versus a payload delivery with an ADC construct?

Yes. So I would like to say that chemotherapy has been tried in hepatocellular carcinoma with limited success, but there has been some incidence of success there, especially trying to localize chemotherapy that's been effective. And that actually makes us believe that giving cytotoxic in an ADC format, particularly with our higher internalizing antibodies and the fact that hepatocellular carcinoma has very high expression of GPC3 gives us confidence that we should have the therapeutic window that is needed in this patient population to be successful. [ id="-1" name="Operator" /> The next question comes from the line of Andrew Berens from Leerink Partners.

Congrats on the progress. Just a question. I know Jazz is controlling the trial, but I was wondering, would the increase in the -- to the intent-to-treat analysis today also increase the number of PFS events that are necessary to trigger the analysis? Just try to put this announcement in context.

Yes. No, thanks for the question, Andy. I think I going to answer the same way before. I think Jazz provided all the guidance appropriate around that decision of the patient population that will be utilized for the ITT patient population, both from a PFS perspective and OS. And I don't want to go further than the guidance they've provided. Obviously, we've been working on the study for 4 years from a Zymeworks perspective and proximity data is very close. And so I'll just let Jazz provide that guidance, and we'll just have to wait for a future announcement and presentation to understand anything further beyond that. [ id="-1" name="Operator" /> The next question comes from the line of Stefan Wiley of Stifel.

Just curious how we should be thinking about the starting dose levels of 251 relative to 191. I know obviously, different DARs, different target organs. But is there anything you can say qualitatively or maybe even quantitatively about how you're thinking about pushing dose here? And I guess, did that dose escalation schema for 251 change at all as some of the 191 data started to come in? And I just have a follow-up.

Yes. Good question, Steve. Sabeen, do you want to -- obviously, we haven't disclosed the starting dose yet. We'll obviously look to do that probably a similar way we did with 191. But Sabeen, is there anything you want to add about the dose schema for dose escalation for 251 as it relates to the 191 schema that now people have seen?

So I would say that the schema for 251 is very similar to 191, although as you rightfully pointed out, this is a DAR4 as opposed to 191, which was a DAR8. So there are differences. And also with 191 was our first ADC into the clinic. So we were very conservative with our initial starting dose. And now that we've gained some clinical experience, particularly with regards to safety, I can say that we have more confidence in our starting dose, but we are not disclosing that yet. We will be disclosing that later similar to what we did with 191.

Okay. That's helpful. And then -- maybe just a question for Paul. Just curious how big the universe of target antigens you think is for a trispecific format beyond DLL3. I know that target has a pretty exquisite expression profile between tumor and healthy tissue that obviously mitigates some of the concerns about amplifying off-tumor tox with a signal 2. But just curious where and how you might be able to leverage this format to other targets of interest.

Yes. No, thanks, Steve. That's definitely very much in our mind. And actually, I sort of alluded to we have actually a presentation this weekend at SITC and what we're going to show there is application of the technology to different targets and the way that we designed the molecule for the target. So the base molecule on the context of the CD3/CD28, we know sort of the positions of those molecules. We're not telling people really the secret sauce there and how they're in the geometry of the molecule. But what we also can think about is how do you then target the antigen and design the targeting of the tumor antigen in such a way to get that maximum window. So we are looking at that. We're looking at targets both in solid tumor and in hematological cancers. We can deploy against sort of the 2 plus 1 strategies. We can think about logic gated strategies as well. So there are ways with the Azymetric so versatile and flexible that we can put in multi binding sites to help us get more selectivity and targeting. We can share more of that. But that we're very much thinking about how do we tailor that so that we can have that therapeutic window. We don't rule out the use of masking. We do have masking technology. We're actually applying that to the IL-12 molecule, and that can also be adapted to our T-cell engagers. So we have that toggle if we feel we need it as well. But we just run it through, we test all the different permutations of the molecules, let the data drive and then we have the preclinical models that then allow us to understand the toxicity profile and the therapeutic window. So we're very excited about the application of that. And again, looking forward to pushing forward with the DLL3 program, but we do have other molecules coming behind as well. [ id="-1" name="Operator" /> The next question comes from the line of Brian Cheng of JP Morgan.

Just 2 quick ones from us. So in a trial design for GPC3, we noticed that you're recruiting patients actively through the patients who have been through standard of care. So just one is, Paul, I'm curious what you saw in the preclinical setting that gives you confidence that GPC3 will be active in the post-IL setting, given that NIVO IPI got approved in the first-line HCC not too long ago. And then just on the biomarker side, I'm curious if you perceive a potential need for -- to develop a biomarker assay near term, is there a need for it today? Just curious what you think about that front, too.

I'll let Paul start on that. I think Sabine may have something to add also on those, but I'll go ahead, Paul.

Yes. No, thanks. So I think from the preclinical setting, your question was how do we -- what's our confidence that we can go behind other standard of care, right? So I think the expression level of GPC3, we've looked at that. And that doesn't -- we don't anticipate or any -- there's no proof that, that would be modulated by IL treatment. So I think the complementary mechanisms and how they work wouldn't really preclude us going with a targeted medicine that's going after GPC3. And we've got preclinical data in different PDX models. Some of those will be collected potentially after treatment, right? But I think just mechanically, we don't see that as a barrier. And I think one of the Yaron mentioned there's encouraging data with other GPC3 modalities that are also going behind -- they would be tested behind standard of care from those clinical trials. So we take encouragement from what others have seen with GPC3 targeted therapies using different modalities. We just feel the ADC modality could just give us an additional mechanism that -- and the power of that approach can just give us an opportunity for more meaningful differences in benefit there. So that was the thinking there. And then on the biomarker side, there, just like we did with folate receptor, we will look at GPC3 levels and make a decision on whether that would be something that we would need as we move forward in clinical development, and we'll collect that data as we go on that. And Sabeen can reiterate that or elaborate on that.

So I'm answering a question regarding NIVO IPI being approved not too long ago. I don't think that changes our development plan. If you look at the treatment landscape for first-line hepatocellular carcinoma, the treatment is currently includes checkpoint inhibitors and VEGF and also checkpoint inhibitors plus CTLA. So prior to approval of NIVO IPI durvatremi has been approved as well. So it's the same mechanism of action, and it really doesn't have an impact on how we think an ADC, particularly a topoisomerase ADC would perform in this setting. So I think we remain confident with regards to that. And I think Paul answered all your questions regarding the biomarker. We are enrolling similar strategy to our 191 enrolling patients regardless of expression and be able to ultimately do a correlation of how the expression level relates to clinical activity. [ id="-1" name="Operator" /> The next question comes from the line of Mayank Mamtani of B. Riley Securities.

Congrats on a productive quarter. Can you talk a little bit more about your expectations on durability for 191, just given what you've seen at the dose levels you're at? And at what point you'd also be able to explore combination, obviously, important in PROC, but also in other solid tumor types that you may want to explore there? And just kind of put it together, when you think you have a sort of partnership enabling package here, just your latest thoughts on that. And then I have a follow-up.

No. I'll just answer the partnership question quickly, and then I'll turn it over to Sabine, I think can answer part 1 A, B and C of your first question. But obviously, we found the data from 191, although early in initial clinical data, very interesting. I think there are others who are interested in other ADCs that are differentiated. We think ours clearly are. And so we'll continue to talk to parties who might have an interest in joining us and moving that forward that might allow us to accelerate development, might allow us to find a better ability to compete even on a time basis and explore the full potential of ZW191. So we'll continue to have those discussions. And as I think you've seen that data was very intriguing to KOLs and obviously, people on this call and especially to us. And I think there are potential partners where that data was also very intriguing. And so we'll continue to let the data mature, continue to collect more data and have ongoing discussions at the same time. And I'll let Sabeen answer the subparts of your first part of your question, if that's okay.

So with regards to durability of responses, I think some of the key things when you look at efficacy is the number of patients who responded. So our overall response rate looks pretty encouraging, particularly in the doses we would like to take forward that is 6.4 to 9.6 milligram per kilogram. Key things that we noted were we have a pretty wide therapeutic index with responses starting at 3.2 milligrams per kilogram, that actually gives us a lot of confidence. And if you look at our swimmers plot that we shared in our poster, a few things that give us encouragement is that most of the responses, particularly at higher doses occurred early. And looking at the waterfall plot, the depth of responses, we had a pretty good depth of responses in terms of reduction in tumor size for the target lesions, even though our follow-up is relatively short. And vast majority of patients are staying on treatment. So combined with our safety profile, which would hopefully allow patients to stay on treatment for a long period of time, I think that would help us give the durability that we're going to need to achieve the PFS and OS that we -- that would be important in these indications.

And then on the learnings from 171 to 209, were there any step-up dosing learnings you're looking to apply here as the DLL3 program gets into the clinic? I know you're not saying what dose levels you may start at, but I was just curious, the therapeutic window should be very different consideration given the target differences in mesothelin and DLL3 from an off-target toxicity standpoint. Any thoughts there would be great.

Yes. I'll let Paul talk about just learnings from our 171 program and how they're going to apply to our thoughts around 209.

Yes. And I think one of your questions was just thinking about the dosing and how we go about thinking about the step-up. And what we did for 171 was we used QSP modeling and we sort of leaned on prior clinical precedents to allow us to really nail what we thought was a good starting dose and then how we could accelerate through the dose escalation. And that approach we will use a similar approach for projecting the starting dose and the step-ups for 209. And what I would say is that those projections when we looked at the exposure levels in the PK, they seem to really fit nicely with what we had projected. So we're anticipating that we can use that again. Obviously, the target toxicity profile, the safety profile is a little bit different for DLL3 than it is for mesothelin. But we still think there's relevant learnings from the design of them -- from the clinical design. But then also there was also some design features in 171 that we're also carrying over into 209. I think that also gives us confidence then that we have that human experience with that approach that it sets us up well for 29. [ id="-1" name="Operator" /> The next question comes from the line of Robert Burns of H.C. Wainwright.

Just one, if I may. So one of the things that I noticed in the presentation for ZW191 was that you used an score categorization, low negative 0 to 74 intermediate 75 to 199 and high 200 to 300 versus the majority of the competitors are using a PS2+ method to define high versus low. So I was just curious about the correlation between those 2 different scoring methodologies so we could sort of assess them in a more apples-to-apples comparison.

Yes. Thanks for the question, Robert. I think I'll let Sabeen start addressing that question and then see if Paul has something to add to that response as well. But excellent question.

So I would say that H-score is pretty well-known and very well-validated research method in evaluating expression levels of different targets, and it combines both intensity, which is usually measured in IC treatments for 1, 2 plus intensity as well as the number of patients with positive -- number of cells with positive scores. So it's a pretty well integrated method for evaluating the number of pay cells that express the target. And it has had a pretty good correlation with both TPS score, which is often used in certain assays that are ultimately commercialized as well as IHC scores. So that is why we use the H score. It's a composite, and it's got a pretty wide range from 0 to 300, and that gives us a pretty good evaluation across the range for the expression level. So one of the things that we did also do in our poster is categorize the H-score into 3 different categories, high, intermediate and low. And within those categories, the high category that we defined is correlates with high expression that of folate receptor that is used for treatment with ELAHERE. And that is a measure where we can evaluate how many of our patients responded who would have been candidates for ELAHERE versus patients who were low or negative and were not candidates for that treatment. Paul, if you want to add something. Please go ahead. Yes.

Yes. No, I think that's good. Yes. No, great, Sabine, you covered it. I think with the H-score, it just gives us a little bit more granularity across that than using the PS2+ score. But by having the H-score, the score the way that, that specimen sample is scored with the test, we can -- as Sabeen alluded to, we can also calculate the PS2+. That's doable. So you can -- we can take that data and analyze it whichever way we want. But we felt for this analysis, this was the appropriate way to show the H-score because it just gives people more breadth and understanding of the profile of the patients that we're seeing.

Yes. No, I completely understand that, and I appreciate the granularity. So just if you don't mind, like would it be an accurate assumption to say the patients that you defined as high expression per the H-score of 200 to 300 would fit the category of PS2 plus greater than or equal to 75? Or would there be some discrepancy between them?

It should be a very high correlation.

I guess last question for me. Given the data that we've seen from RENA-S as well as the Eli Lilly compound, obviously, they're using a PS2+ scoring system. In those non-high patients, how do you think that ZW191 stacks up against those 2 compounds in the lower expressing or intermediate expressing folate receptor alpha patients?

So as you saw from our data, we showed pretty transparently across a spectrum of H scores across low and negative that we observed clinical activity across folate receptor expression levels. We're looking at data from -- which we are pretty confident about, and we're showing pretty good activity. Given in our sample size, we had roughly around 2/3 of patients who were low negative roughly. which correlates very well to the number of patients who are not candidates for ELAHERE. And comparing our data to the competitors you talked about GES and Lilly, I think we feel pretty confident about our activity in the low negative patient population from what we've observed so far. Obviously, we're going to continue to follow our patients. We are enrolling very actively in our study with more patients in dose escalation and longer follow-up, and we are initiating our Part 2 dose optimization, which will provide us more data at the doses that we would like to move on. I think that would give us a lot of confidence in our activity across the spectrum for expression levels, including low and negative.

I can't wait to see the additional data for ZW191. [ id="-1" name="Operator" /> The next question comes from the line of Akash Tewari of Jefferies.

This is Phoebe on for Akash. On ZW191, it looks like a key differentiator between this and other next-gen folate receptor alpha ADCs is safety, specifically on Grade 3 cytopenia. Can you talk about the importance of this difference in terms of potential combinations maybe in earlier treatment lines?

Good question. I think we see a number of potential differences, differentiating factors between ZW191 and data we've seen from others. But I'll let maybe Sabeen talk specifically about the tolerability profile we've seen so far in our data set.

The tolerability profile, we're very pleased, particularly with our safety event rate. Most of the safety events that we saw were pretty expected, as we mentioned, which were mostly not vomiting cytopenias. Our cytopenia rate is -- actually, we're very pleased with that rate because this is something that you would expect very typically from a topoisomerase ADC. And the rates that we observed for anemia, neutropenia and thrombocytopenia are well within expected for an ADC, particularly with the fact that we're going at relatively high doses compared to other ADCs with similar payload. So with that, we are confident that, that would help us drive efficacy. And at the same time, the safety profile with cytopenias helps us combine with treatments in earlier lines of therapy. As you know, in ovarian cancer, earlier lines of treatment consists of a combination of platinum, taxanes and bevacizumab. So that gives us a lot of confidence to combine with all of these treatments going in earlier lines of treatment, particularly some of the pitfalls that we've seen with other ADCs, cytofolate with combinations in earlier lines is neutropenia, and that often leads to reduction in doses to the point that it affects efficacy. And we're hoping with our safety profile and particularly the lower rates of neutropenia that we're observing, ZW191 should be able to be combinable with the platinum agents at a much more efficacious dose. So that's one of the key areas. The other thing that we're thinking about in earlier lines of therapy from a tolerability perspective, obviously, is the ability to treat patients for much longer, particularly in the maintenance setting. And I think these are all areas where we can differentiate. [ id="-1" name="Operator" /> The next question comes from the line of Jon Miller of Evercore.

I'll follow on a question on the DLL3. I guess we've seen some really great data from other T-cell engagers there even pretty recently. So I'd love what do you think are the key places where you'd hope to differentiate? What would make your molecule a best-in-class molecule in your opinion? And where do you think you can target that? And then I've got a follow-up.

Okay. Paul, do you want to...

Yes. No, I think absolutely. I mean, DLL3, a lot of excitement. It's a attractable target in solid tumors. We're getting encouraging response rates. For sure, we think there may be patients that could still -- that could benefit from -- that don't respond that don't have the T-cells that can really mount a response or a prolonged response. And that's really what we're trying to do here with our molecule. So really change the game and really get the next level of response and durability of response is really what we're hoping for. And we think by having the CD28 co-stimulation, it gives us opportunity to do that. So there may also be some benefits in the mechanism that can lead to thinking about the duration of response and the way that we dose the molecule. They could also be intrinsic in the design and the fact that we have that extra T-cell response, but that will away further analysis. But it's really more patients responding and longer responses. Again, that's the goal, Jonathan, and our data suggests that we can from a preclinical, we have a chance to achieve that.

Fair enough. I guess since you were talking earlier about being almost finished with your repo, I am curious, given the expected upcoming milestones from Jazz would be one on the GEA readouts, what is your expected use for future milestones? Should we be expecting that repo will make a return when you have cash inflows in that response? Or is that money spoken for, for your internal programs?

Really good question. I think as we started this last year, we do want to have the ability to always have an authorized stock repurchase program that then gives us the ability to allocate capital to reducing share count at what we think is attractive prices to boost TSR. So we always want to have that optionality. So I think you should expect that we will always have an authorized stock purchase plan in place. We get to decide when and how we use that for shareholder benefit. So I think you should just expect as we're getting to the end that we should always have one in place to be able to do that. It's not the only place we've been allocating capital over the past period of time. We have been allocating capital to R&D programs that make sense for us and when the data justifies it, continuing to move forward with additional investment as we are with ZW191. We've obviously talked a little bit about our strategy of maybe creating another area to allocate capital as capital comes in from milestones and royalties from Ziihera and hopefully eventually pasritamig. Having the ability to then decide to allocate that capital back into a royalty portfolio, given that those royalty portfolio we have right now, it earns very attractive annualized rates of return, we think, from holding it from development through commercialization and having the ability as some of those gains are realized through payments from our partners to put that back into an attractive royalty portfolio that could generate really interesting rates of return is something another piece of the puzzle and strategy that we've talked about putting in place, and we'll talk more about this in the weeks and months ahead. So I think you should see…

This is [ Shuan ] on for Yigal. Congrats on the progress. Maybe just a quick one from us. You spoke on it a bit already, but just wondering if you could provide additional color on potential time lines of third-party milestones beyond what might be expected from Jazz.

Yes. We haven't, as a practice, provided much guidance in that regard. Obviously, we've tended to wait until we've earned or receive milestone payments as we did this quarter with the $25 million that we earned from Johnson & Johnson with respect to pasritamig moving into Phase III studies. So for right now, I think we'll keep that guidance. I think as we move forward, especially with Ziihera into commercialization, we might provide some additional guidance around milestones from both Jazz and B1 as they become closer, more approximate and more probable just so people understand a little bit more about cash flows that might be realized in those licensed products and then obviously, then where that capital might be allocated to. So until then, you just have to wait and see, but not too long, I think. [ id="-1" name="Operator" /> This does conclude the question-and-answer section. I would now like to hand the call back over to Chair and CEO, Ken Gabre. Ken, please go ahead.

That's great. So thanks, everyone, for your time and attention and questions on today's call. Obviously, back in 2021, we designed and initiated a really important clinical study with zanidatamab, the HERIZON-G01 study. And we're really pleased that Jazz continues to be optimistic and confident of reporting out the top line data in this quarter. And we're as interested as anyone in understanding that data set and what the potential is for zanidatamab to be practice-changing in this patient population. And we're very pleased that we won't have to wait that long to understand that. And so please stay tuned and look forward to talking about that further with our partners, Jazz and B1 as appropriate. So thank you very much for your time, and we'll talk to you all very soon. [ id="-1" name="Operator" /> This concludes today's presentation. You may now disconnect.