Zymeworks Inc. (ZYME) Earnings Call Transcript & Summary

Good day, and thank you for standing by. Welcome to the Zymeworks AACR Webcast and Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, Ken Galbraith, Chair and CEO. Please go ahead.

Thank you. Good afternoon, everyone. Thank you all for dialing in. I'm Ken Galbraith, the Chair and CEO of Zymeworks. I'd like to welcome you to our AACR Conference Call and Webcast to discuss the exciting events presented and discussed at AACR over the past few days. We sincerely appreciate you taking the time out of your busy schedule to join us virtually. Before I hand the call over to Paul Moore, our Chief Scientific Officer, and his team, I want to highlight some housekeeping matters. In today's presentation, we will be making forward-looking statements. We rely upon standard safe harbor disclosures, common in companies in our sector and state of development. For more information on risks and forward-looking statements, we encourage you to read statements made on this slide, along with our SEC filings and disclosures available on our website. Additionally, for anyone listening to the webcast live today, if you have questions on today's material, you may e-mail them to [email protected], and we'll try to answer your questions during the Q&A at the end of the prepared presentation. So with that -- I'm really excited to speak to all the programs you see on this slide. While there will be a few posters not discussed specifically today, we're happy to answer questions on these at the end of the call. Further, I'd like to note the progress we've made since AACR of last year. Last year, we didn't have any posters to present at the 2022 edition of AACR. This year, we had 11 abstracts presented from 11 abstracts submitted. We've been working diligently with focus on both our ADC and MSAT program to be where we are today. I really want to thank Paul and his team for the effort that has gone into developing these exciting presentations showcasing our progress over the past year on both products and platforms. Today's presentation follows on from our Early research and Development Day in October of last year in New York City, where we unveiled our name programs and targeted research areas. While we continue to make significant progress across all of these programs, I want to highlight the team's work in bringing together all the data presented over the past few weeks. We've been incredibly focused equally on both the antibody drug conjugate and multi-specific antibody therapeutic areas of our R&D strategy. These teams have been incredibly focused on delivering high-quality data that will help advance these assets toward IND quickly and efficiently. So thank you goes to all the team, both those present in the room today and back in Vancouver and Seattle that worked so hard to get all of this pulled together. With that, I'll spend a minute here on Slide 4 to briefly talk about the strategy we laid out in October. There are 2 important sides for our R&D program, ADCs or antidrug conjugates and our multi-specific antibody therapeutics or MSAT. Together, both -- with these 2 modalities, we're able to look at and hopefully address treating and improving patient outcomes, indications with the poor prognosis as measured by 5-year overall survival rates. Through this strategy and our ability to integrate these 2 sides of the business, we intend to target indications with first-line and second-line opportunity, where we have the opportunity to achieve timely full or accelerated approvals globally, and do all this an indications of significant global peak sales potential both globally and within the U.S., where we plan to 1 day commercialize our own products. Broadly speaking, we also have more towards Zymeworks than our preclinical programs that we're speaking about today. Our strategy as a company is to continually create value by progressing each of the 5 areas of focus for Zymeworks. These 5 areas include our most advanced asset, Zanidatamab, our HER2 bispecific antibody currently in pivotal studies, which we partnered with Jazz and BeiGene; second, our HER2 x HER2 bispecific ADC, zanidatamab zovodotin or zani zo for short, currently progressing to Phase II clinical studies, which we'll speak to briefly today. Also, our portfolio of legacy technology licenses with global pharmaceutical companies, which provide potential financial benefits in the form of future milestones and royalties, and of course, our portfolio of early-stage ADC and multi-specific programs that's been the focus of our AACR presentations over the past several days. We are looking to progress each of these 5 areas of our business this year, and in doing so, we believe we can continue to improve our enterprise value compute on a per share basis and continue to create novel and differentiated biologics that have the potential to help patients around the world impacted by difficult-to-treat cancers. With that introduction, I'd like to hand the call over to Paul Moore, our Chief Scientific Officer. Paul?

Thank you, Ken. As Ken shared and we've shared previously and also reflected in our AACR presentations, Zymeworks is uniquely placed as an antibody-based therapeutics company to have expertise and proprietary technologies in both protein engineering and antibody drug conjugates. In practice, this allows us to leverage the most effective therapeutic modality to treat a particular cancer type, balancing efficacy and safety. As shown on the left, the cancer types we intend to focus on are where there is greatest unmet patient need. These are the more challenging cancers biologically to treat with existing therapies requiring these types of improved therapies that we're building to overcome these limitations that are not adequately addressed by existing therapeutics. As shown on the right of this slide, with ADCs, we are able to customize the drug to the molecule -- the ADC through selection and engineering of the antibody itself to enable optimal targeting internalization, using a choice of monospecific biparatopic or bispecific formats through selection of the payload, conjugate and the drug-to-antibody ratio. Multi-specifics allow us to incorporate multiple mechanisms of action into a single molecule, achieving synergistic biology not possible with monospecific targeting agents. On the next slide, what we're showing here is our pipeline of early-stage programs, including zanidatamab zovodotin, which I'll shorten to zani zo during the presentation. That's our biparatopic ADC targeting HER2. As we presented previously, we've established a recommended Phase II dose for this molecule at 2 mg -- 2.5 mg per kg every 3 weeks, and we plan to move that into Phase II testing later this year in both non-small cell lung cancer in combination with anti-PD1 [indiscernible] cancer, where we believe zani zo's ability to be used in combination with existing standards of care combined with the differentiated safety profile will allow us to become the HER2 ADC of choice following TDXd where TDXd may have challenges due to tolerability concerns. We then have 2 programs on track for IND filing in 2024. ZW191, which leverages our proprietary TOPO1 inhibitor ADC platform, targeting folate receptor alpha and then our first in-house T cell engager targeting mesothelin, deploying a unique anti-CD3 binder and an optimized 2+1 geometry that we'll hear more about later. While we've had partners progress T cell engagers into the clinic using our Azymetric platform, this will be our first in-house program to advance to the clinic. We took learnings from these partner programs and have further optimized ZW171 with additional features we think are appropriate for targeting mesothelin, which is a broadly expressed cancer target. Behind these 2 IND programs, we then have 2 additional publicly announced preclinical stage ADC programs ZW251 and ZW220 that we continue to progress. As we first stated at our R&D Day in October of last year, our goal is to have 5 new molecules in the clinic over the next 5 years, leveraging both platforms, investing where we can to accelerate development while continuing to integrate an increased level of novelty and diversity. So before elaborating more on our early-stage preclinical pipeline molecules, which will be the focus of today's webcast, I did want to share one piece of zani zo preclinical data that was presented yesterday in which we demonstrate the ability of zani zo to induce expression of DAMP molecules, hallmarks of immunogenic cell death. As shown here, a result of zani zo treatment on various HER2 cell lines is the induction of calreticulin, HMGB1 and levels of extracellular ATP. This is important as it supports the rationale for a combination of zani zo with checkpoint inhibitors such as anti-PD1 sustained immune response and anticancer activity even beyond that achievable [indiscernible]. The concept of combination is at the heart of the design of ZW49 or zani zo. And interestingly, this ability to induce immunogenic cell death appears that at least based on levels of accumulated extracellular ATP greater than that achievable with trastuzumab-based ADCs comprising either vedotin or deruxtecan. Before handing over the podium to the ADC and MSAT, multi-specific research leadership, I wanted to share a few high-level slides demonstrating how we are leveraging platform technologies, both with from conjugates as shown in the top and is a metric-based multi-specific as shown in the bottom. Building on a target -- building on target specificity of a single or multiple antibodies to build a robust and differentiated pipeline. What is decorated on this slide are the various programs and molecules presented at AACR. As you can see, we have a balance of both ADC-based programs shown on top with presentations on our TOPO platform pipeline molecules 191, 220 and 251. And then with our multivalent multi-specific platforms with presentations on 171 and our next-generation trispecific T-cell engagers where we incorporate co-stimulation or checkpoint inhibition directly in the same molecule. We also show focus on our ProTECT program, which incorporates a unique approach to conditional molecule activation. So a little further on the design of these molecules shown here are representative of structures for each flavor of molecule or product opportunity. Starting on the left-hand side with the multispecifics, you can see the relationship between the different structure and how they all build on the Azymetric platform. At the top is our lead ZW171, a 2+1 structure incorporating [indiscernible] and involved this structure to support trispecifics, which incorporate either co-stimulation or TriTCE Co-Stim where we're including CD28 or checkpoint inhibition or TriTCE through incorporation of PD-1. From the TriTCE PD-1 or the CPI, we could then incorporate a mask over PD-1 using ProTECT that helps address targets with normal tissue liability. The theme of masking is also incorporated into our engineered affinity modulated IL-12, an example of using Azymetric to build an engineered cytokine. How much of the ProTECT molecule we use the same proprietary previously but a custom mask for IL-12. On the right-hand side, you then see our ADC portfolio zani zo a biparatopic targeting antibody incorporating our proprietary Auristatin payload and our focus on TOPO-based ADCs with the ZW191, 220 and 251, all employing a proprietary TOPO payload and design criteria for ADCs, all of which you're going to hear about next. With that, I will now hand over to Stuart Barnscher, Director of Preclinical Programs and our ADC Therapeutics Development Department. Stuart?

Thank you, Paul. Before we dive into more of the exciting data we presented at AACR this week, I'd like to take some time to outline our antibody drug conjugate strategy. As we've seen over the last few years, developing targeted chemotherapy through antibody drug conjugates has emerged as an extremely effective therapeutic strategy for multiple types of cancers. At Zymeworks, we are committed to this exciting therapeutic modality through the development of differentiated fit-for-purpose ADC product candidates and novel ADC technologies with which to build our therapeutic pipeline. At the foundation of our strategy is the commitment to an integrated design philosophy that considers all components of the ADC, including the antibody, linker conjugation, payload and target. We use empirical evidence from our own data and the broader scientific community to think deeply about all 4 of these components and how they combine when we design each of our ADCs. This thinking is demonstrated in our AACR poster, Abstract 1538 entitled, Revisiting the dogma of antibody drug conjugates, Emerging data challenge the benefit of linker stability and the primacy of payload delivery. We also have a cancer cell paper entitled, The therapeutic window of antibody drug conjugates: A dogma in need of revision and 2 articles, 1 in bioprocess online and 1 in clinical leader, which further demonstrate our evidence-based integrated design philosophy for ADCs. We are driving multiple topoisomerase-1 inhibitor ADCs towards the clinic, while simultaneously investing in our novel ADC technologies to support future program development. With our novel topoisomerase-1 inhibitor technology, we are initially focused on validated targets as this provides an opportunity for benchmarking a preclinical development and expected clinical differentiation. The novelty of the targets we pursue will increase over time, both as our topoisomerase-1 inhibitor technology gains clinical validation and as we address difficult-to-treat cancers in need of new approaches to derive clinical benefit. Further, alternate payload mechanisms of action to topoisomerase-1 are under investigation for long-term development opportunities and to provide more options when matching disease and target biology with payload mechanism. We are currently leveraging validated peptide-cleavable linkers and stochastic conjugation for the design of our topoisomerase-1 inhibitor ADCs. However, new chemistries are under development to complement our novel payloads as we believe that linker design should be dictated by payload properties, including potency, solubility, metabolism and mechanism. We continue to discover antibodies with properties well suited for the ADC mechanism and we are also exploring both biparatopic and bispecific ADC formats that may provide opportunities to develop differentiated therapeutics. With that, I'd like to get into the data presented at AACR for 3 of our ADC assets, ZW191, ZW251 and ZW220. Our AACR poster on ZW191, Abstract 2641, entitled ZW191, a novel folate receptor alpha targeting antibody drug conjugate bearing topoisomerase-1 inhibitor payload, outlines our most advanced topoisomerase-1 inhibitor ADC that targets folate receptor alpha. The expression of folate receptor alpha is compelling in indications such as ovarian cancer, other gynecological cancers and multiple other solid tumors like non-small cell lung cancer and triple-negative breast cancer. We paired our internally discovered antibody with our novel topoisomerase-1 inhibitor platform at DAR8 to generate ZW191. The antibody was selected from a large panel to find a unique antibody with characteristics we chose to best suit the ADC format, including optimal internalization, payload delivery and tumor penetration. Since ZW191 was introduced last October, we've made tremendous progress, and we are on track to file the IND in 2024. In our poster, we presented data detailing the antitumor activity of ZW191 in 9 patient-derived xenograft models of ovarian cancer. We analyzed the expression of folate receptor alpha in these models using an in-house immunohistochemistry assay and signed an H-score to each model. These 9 models represent a range of folate receptor alpha expression from high expression with an H-score of 210, down to a very low expression with an H-score of 20. After a single IV dose of 6 milligrams per kilogram of ZW191, we observed antitumor activity in 8 of the 9 models. Conversely, the clinical benchmark of mirvetuximab soravtansine dosed at 6 milligrams per kilogram was only active in 5 of the 9 models. When we stratify the models by H-score and look at response, we see that in models with high folate receptor expression, each score is above 150. ZW191 and mirvetuximab soravtansine have similar antitumor activity. But in models with low levels of folate receptor alpha, H-score is below 150, ZW191 has improved antitumor activity and is active in 83% of the models tested compared to the clinical benchmark with activity in only 33% of the models expressing low levels of folate receptor alpha. The strong antitumor response we observed with ZW191 in folate receptor alpha low-expressing PDX models set a precedent for potential activity not only in ovarian cancer expressing lower levels of folate receptor alpha, but also in other indications with lower levels of folate receptor alpha, including endometrial cancer, non-small cell lung cancer and triple negative breast cancer. We believe our ability to target lower levels of folate receptor alpha compared to the clinical benchmark is driven by our unique antibody, which was selected for optimal internalization and payload delivery as well as our novel bystander-active topoisomerase-1 inhibitor payload, which constitutes a differentiated mechanism of action compared to mirvetuximab soravtansine. We also present the data outlining the tolerability and pharmacokinetic profile of ZW191. The tolerability was assessed in a 2-dose every 3-week non-GLP, nonhuman primate toxicology study. Nonhuman primates represent a relevant antigen binding species as ZW191 cross reacts with the nonhuman primate version of folate receptor alpha. And in this study, we found that ZW191 was tolerated up to 30 milligrams per kilogram. Histopathology findings at 30 milligrams per kilogram included changes to the thymus and stomach and these were considered as background and low severity and were not considered adverse. Clinical chemistry and hematology findings at 30 milligrams per kilogram were considered mild and nondose responsive. And clinical observations were limited to fecal abnormalities with no effect on body weight. From this study, we also see that ZW191 displays a linear and dose-proportional toxicokinetic profile at exposure levels above those projected to be efficacious. Our differentiated antibody designed specifically for the ADC mechanism, paired with our bystander-active topoisomerase-1 inhibitor payload, constitutes an exciting opportunity to make a potential impact in folate receptor alpha expressing cancers, including both folate receptor alpha high and folate receptor alpha low expressing ovarian cancers and in endometrial cancer, non-small cell lung cancer and triple negative breast. Our GMP process is well underway. And as I mentioned earlier, we are on track to file the IND in 2024. Additionally, we are planning aggressive time lines for Phase I study execution. I will now hand the call over to Jamie Rich, Director of Technology and the ADC Therapeutics Development Department. Jamie?

Thank you, Stuart. Another topoisomerase-1 inhibitor ADC program that Zymeworks is advancing is ZW251. This ADC targets 3 glypican-3 or GPC3, a proteoglycan that is expressed in close to 80% of hepatocellular carcinoma or HCC, with high expression in over 50% of these liver cancers. We provided an update on this program at AACR and our poster, Abstract #2658, entitled ZW251, a novel glypican-3 targeting antibody drug conjugate bearing a topoisomerase-1 inhibitor payload. ZW251 comprises a novel GPC3 targeting monoclonal antibody conjugated interchain disulfide cystine residues to our topoisomerase-1 inhibitor drug linker, the same drug linker that we were applying in the ZW191 program. We're evaluating both DAR4 and DAR8 ADCs in this program. The antibody was selected for its characteristics that support ADC activity, strong binding to the target, rapid internalization and potent cytotoxicity when conjugated to our bystander-active topoisomerase-1 inhibitor payload. Thus far, we've seen robust activity from both the DAR4 and DAR8 ADCx in preclinical patient-derived xenograft models of hepatocellular carcinoma. And we've completed the in-life portion of our repeat dose nonhuman primate toxicology study. Encouragingly, there was no mortality at doses of 60 milligrams per kilogram for the DAR8 ADC and 120 milligrams per kilogram for the DAR4 ADC. Our hepatocellular carcinoma cell line and patient-derived xenograft models revealed ZW251's broad antitumor activity after a single dose of 8 milligrams per kilogram, either at DAR4 or DAR8. We ran a total of 6 CDX and 9 PDX efficacy models with tumor volume data shown on this slide. The tumor volume data for the CDX model is grouped in the top row where we can see the ZW251 at both DAR8 in blue and DAR4 in purple were active in all models. We also included an isotype control ADC, shown in orange and its minimal activity across the models highlights the target specific nature of the antitumor activity. Robust antitumor activity at both drug-to-antibody ratios is also evident in the data from the 11 PDX models shown in the bottom row. Overall, the ZW251 was active in 82% of the high-expressing models that either DAR and in 50% and 75% of the models with lower GPC3 expression for DAR4 and DAR8, respectively. The in-life portion of our nonhuman primate toxicology study examining repeat dosing of ZW251 is complete. We dosed the ADC -- the DAR8 ADC at 10, 30, and 60 milligrams per kilogram and the DAR4 ADC at 20, 60, and 120 milligrams per kilogram. No mortality was observed in any of the treatment groups prior to necropsy and only minimal changes in body weight, hematology parameters and clinical chemistry parameters were observed. While we're still waiting for the toxicokinetics and pathology reports, overall, we are very encouraged that these high doses appear tolerable. We're pleased to note that the pharmacokinetics of ZW251 in Tg32 human FcRn mouse model appear to be unaffected by conjugation, and that the exposure of the ZW251 ADC in this model is comparable to a clinical-stage antibody comparator, codrituzumab. Our third topoisomerase-1 inhibitor ADC program is ZW220. We're really excited about the therapeutic potential of this program, and we're very close to finalizing our candidate. This program was just presented as a poster at AACR, abstract #1533, entitled ZW220, a novel NaPi2b-targeting antibody drug conjugate bearing a topoisomerase 1 inhibitor payload. NaPi2b is a sodium-dependent phosphate transporter that is highly overexpressed in both high-grade serous ovarian cancers and lung adenocarcinomas. NaPi2b is a clinically validated ADC target that is seen in 2 [ auristatin-based ] ADCs reached the clinic. ZW220 ADC uses an internally discovered and ADC tailored antibody that exhibits strong target-specific binding, rapid internalization and good PK in preclinical models. It's conjugated to our novel topoisomerase-1 inhibitor payload and utilizes the same drug linker and stochastic, cysteine conjugation that we have described for ZW251 and ZW191. So far, we've generated a really strong data set in ovarian cancer patient-derived xenograft models, where ZW220 has proven active in almost every model irrespective of NaPi2b expression level. We've also studied this ADC at a repeat dose, nonhuman primate toxicology study and have found that both DAR4 and DAR8 ADCs are tolerated at very high doses. Here, we show some of the antitumor activity efficacy data that we've generated around ZW220 using ovarian cancer patient-derived xenograft. The expression of NaPi2b ranges from very high at the left to relatively low on the right. We've also included immunohistochemistry from the study, which was scored by a pathologist. Both DAR4 and DAR8 versions of ZW220 were assessed in these models using a single dose of 6 milligrams per kilogram of the ADC. The tumor volume data plotted as blue open circles and closed circles, respectively. We've also included the lifastuzumab vedotin benchmark at the same dose shown in red in these studies. ZW220 was active and superior to the benchmark in 5 of the 6 models. Breaking it down a little further. It's apparent that ZW220 is active in models with high target expression in almost every case. But notably, it does not tend to -- it does tend to retain its activity at H-scores that fall at 150 and below. The DAR8 ADC version of ZW220 is more active than the corresponding DAR4 ADC, but keep in mind that the amount of cytotoxin available to be delivered in the case of the DAR4 ADC is low. ZW220 has also been assessed in a 3-dose non-GLP toxicology study in male nonhuman primates, where we dosed at 15, 30 and 45 milligrams per kilogram for the DAR8 ADC and 30, 60 and 90 milligrams per kilogram for the DAR4 ADC. Remarkably, all of the doses were well tolerated with no mortalities, adverse clinical or microscopic observations or impact on body or organ weight. Furthermore, there were no clinical or anatomic pathology findings related to ZW220 at any dose. A stable dose-proportional pharmacokinetic profile was observed in the nonhuman primate study with no evidence of any liabilities. With this data, we're confident in our assessment of 45 and 90 mg per kilogram as the maximum tolerable doses for DAR8 and DAR4 ZW220, respectively. In summary, we recently presented posters on 3 different preclinical pipeline programs at AACR this week ZW191, ZW251 ZW220. ZW191 is on track for an IND in 2024. We plan to nominate candidates for both ZW251 and ZW220 programs in the near future. In addition to our focus on these exciting topoisomerase-1 inhibitor ADC pipeline programs and the advancement of multiple topoisomerase ADCs into the clinic, we remain committed to further diversifying our ADC portfolio. We are currently pursuing new drug linkers featuring payloads with alternative mechanisms of action, and at the same time, we're shifting our focus to more novel targets, and we're exploring biparatopic and bispecific ADCs, where this makes sense. So now I'll pass the call back over to Paul. Paul?

Okay. Thanks, Jamie. And before I hand it over to Nina, who's going to discuss the posters presented on our multi-specific technology and programs, I wanted to provide some high-level perspective on where we are presently focused on leveraging multispecifics biologically. I shared in the introduction, all molecules are built on our Azymetric platform. And you look on your -- and if you look at this slide left to right, you can see the design evolution that I discussed earlier. Biologically, our main focus is on generating a suite of molecules, the build on the promise of T cell engagers to mobilize immune cells, in this case, T cells to eradicate tumor cells. While both T cell engagers and CAR-Ts have seen much progress in this application. This has primarily been observed successfully in liquid tumors. Success in solid tumors has been limited for various reasons. But each of our T cell engager-based molecules potentially provide the ability to overcome the limitations encountered by prior T-cell engagers, whether in the context of improving therapeutic window using a low affinity CD3 to reduce systemic cytokine release and a 2 plus 1 geometry in the case of 171 or trispecific T cell engagers that include either co-stimulation or checkpoint blockade to overcome tumor microenvironment obstacles for TCEs to work on solid tumors. With our ProTECT platform, we have a proprietary masking strategy to potentially target tumor antigens expressed in tumor cells, but also with normal tissue liability, while also incorporating the feature of our TriTCE CPI molecule. Our collective approach to TCEs or T-cell engagers provide capability not presently feasible with existing T-cell engagers. In addition, we can also apply our Azymetric technology to other biology, and we share an example of this with our masked IL-12 program, which illustrates our ability to incorporate design -- various design features to overcome challenges impended by others in design of engineered cytokines. So with that, I'll hand over to Dr. Nina Weisser, Director of multi-specific Antibody Therapeutics, who will talk through each of these programs.

Thank you, Paul, and hello, everyone. I'm very happy to present an overview of the multi-specific data that the team recently presented at AACR. For all of the programs I will discuss with you today, we have harnessed the flexibility of our Azymetric platform to identify multi-specific therapeutics with enhanced and novel functionality to address challenges in the treatment of solid tumors. I will start off with our most advanced multi-specific programs, ZW171. ZW171 is a bispecific T cell engagers that was designed as a 2 plus 1 antibody format, where there are 2 paratopes targeting mesothelin shown in green and a single paratope targeting CD3 shown in blue. Heterodimeric heavy chain assembly is facilitated by the Azymetric platform mutations in the Fc regions. ZW171 facilitates this antitumor activity by binding CD3 on T cells and redirecting T cell cytotoxicity in mesothelin expressing cancer cells. The tumor target mesothelin is a glycoprotein that is highly expressed in several cancer indications, including pancreatic, mesothelioma, ovarian and many others, for which there is a high unmet medical need. ZW171 has a novel anti-CD3 paratope that was designed to widen the therapeutic window by having low T cell binding and cytokine release and potent tumor cell lysis. We applied our protein engineering expertise and the flexibility of the Azymetric platform to modify and perform extensive assessment of various aspects, including valency, geometry and affinity. Progress since our last presentation includes the work we have done for the pilot NHP as well as PK data showing that ZW171 is well tolerated and with an extended pathway. ZW171 is on track for an anticipated IND filing in 2024. ZW171 was engineered to widen the therapeutic window of T cell engagers in solid tumors with enhanced antitumor activity and improved safety profile for the treatment of mesothelin expressing tumors. Engineering strategies that we use to improve the safety profile will build up the learnings from the field. In particular, with generation 1 anti-CD3 paratope based on OKT3 and SP34 paratope are associated with high affinity CD3 binding and dose-limiting toxicity related to cytokine release syndrome or CRS. Thus, we engineered the antibody to have low affinity CD3 binding that results in low cytokine release yet maintained potent tumor cell lysis in an in vitro assay to potentially avoid CRS. In the data presented, we have compared the [ ZW-CD31 ] paratope in blue to a generation 1 high affinity and generation 2 low-affinity paratope in purple and orange respectively. As illustrated [ ZW-CD31 ] has reduced binding to T cells, stimulates reduced cytokine release in the presence of tumor cells yet equivalent and potent T cell lysis of mesothelin-expressing tumor cells when compared to the same Generation 1 and Generation 2 CD3 paratope. To improve the antitumor activity, we screened a panel of antibody geometries, formats and affinities to identify an antibody that mediates optimal T-cell synapse formation and antitumor activity. The ZW171 2 plus 1 format has 2 anti-mesothelin paratopes that facilitate avidity-driven binding that directs the binding and cytotoxicity to cells with high mesothelin expression, but not to those with low mesothelin expression such as healthy tissue. As illustrated in the graph up bottom, you can see that ZW171 mediates T cell cytotoxicity and high-mesothelin expressing tumor cells but little to no activity in low mesothelin-expressing tumor cells, which acts as a surrogate for low-mesothelin expressing healthy tissue. Thus mitigating the risk of on-target off-tumor toxicities. We believe the profile of reduced CD3 binding and avidity driven mesothelin binding is ideal to maintain antitumor activity to potentially avoid dose-limiting toxicity related to cytokine release syndrome or on-target off-tumor toxicity in patients. To further evaluate antitumor activity, we compared ZW171 to clinical benchmark or HPN TriTAC or HPN-536 that has an anti-mesothelin single-domain antibody, single anti-CD3 scFv paratope with an anti-albumin single-domain antibody in the center for half-life extension as shown in the cartoon that way. In vitro cytotoxicity at low effect-to-target cell ratio or E:T of 1 to 10 shows that the ZW171 is more potent compared to HPN-536. This observation translated in vivo when we compared the 2 molecules in a PBMC engrafted established [ old CAR ] tumor model where ZW171 also showed superior antitumor activity. A single dose tolerability study in nonhuman primate showed that ZW171 is well tolerated up to 30 milligrams per kg. All observations were mild -- minimal-to-mild and associated with the 171 mechanism of action. Additionally, ZW171 showed prolonged half-life in the NHP study. The GLP toxicology study is scheduled for later this year, and GMP process manufacturing is underway. In summary, ZW171 has been engineered for optimal format paratope affinity and stability with an improved safety profile to wider therapeutic index, and we are excited about the opportunity of ZW171 in treating mesothelin-expressing cancer. Moving on, I will present data highlighting our co-stimulatory trispecifics program. The activity of bispecific T-cell engagers in solid tumors have been limited by low numbers of intratumoral T cells and T cell anergy. Conventional T cell activation and sustained proliferation requires signaling via CD3 and Signal 1 and co-stimulatory molecules Signal 2, such as CD28. The balance of signaling between Signal 1 and Signal 2 is critical for optimal T cell activation. Signal 1 in the absence of Signal 2 results in T cell anergy, while overactivation via Signal 1 and Signal 2 can make to T cell dysfunction and cytokine release as observed with toxicities associated with anti-CD28 superagonist antibodies. Treatments that rely on only T cell activation via Signal 1, such as traditional bispecific T cell engagers in solid tumors are further limited by treatment mediated T-cell anergy and death to lack of sufficient co-stimulation. Optimal Signal 2 co-stimulation via CD28 results in improved T cell fitness, activation and proliferation. At Zymeworks, our aim is to develop co-stimulatory trispecifics with the balance of engaging Signal 1 through CD3 and Signal 2 through CD28 and a tumor-associated antigen or TAA or to be able to provide optimal T cell activation and improved T cell responses in solid tumors in 1 molecule. To address the limitations of traditional bispecific T cell engagers, we have engineered trispecific co-stimulatory antibody that we refer to as TriTCE Co-Stim. TriTCE Co-Stim has the potential to provide more durable responses and reinvigorate cold tumors with lower T cell infiltration via tumor-dependent T cell activation, CD3 Signal 1 and co-stimulation via CD28 Signal 2. Novel engineering and screening solutions were employed to optimize signal strength for T-cell activation and antitumor activity, including modifications compared to affinities and the antibody format geometry. The antibody [indiscernible] show some examples of the antibody formats and geometries that were evaluated. The graphs on the bottom left illustrate the range of affinities in our CD3 and CD28 paratope library that were also evaluated. Our in vitro screening identified TriTCE Co-Stim molecules with enhanced tumor target-dependence or TAA dependent antitumor activity compared to a bispecific T-cell engager. As described in more detail in our poster and as hypothesized earlier of screenings that allow for fine-tuning of TAA dependent enhanced potency at low effector-to-target cell ratios, while not activating T cells in the absence of the tumor target, and eliciting cytokine release and our approach has been transferable across other targets. One other tumor targets we have evaluated for our TriTCE Co-Stim program is claudin18.2. Claudin18.2 is a clinically validated target that is highly expressed in gastric -- gastroesophageal junction cancers and has restricted normal tissue expression. Claudin18.2 TriTCE molecules show enhanced TAA dependent antitumor activity and T cell functionality compared to bispecific T-cell engagers. Specifically, our claudin18.2 TriTCE shows enhanced cytotoxicity at low effector-to-target cell ratios, which reflects the conditions in solid tumors with low T cell numbers and where CD3 bispecifics show low activity. Claudin18.2 showed TAA-dependent cytotoxicity and cytokine release. As shown in the upper right, TriTCE Co-Stim cytokine release in the presence of tumor cells but not in the absence of tumor cells. claudin18.2 TriTCE also show enhanced piece of functionality compared to bispecific T cell engagers as demonstrated by increased T cell proliferation and expression of the anti-apoptotic protein BCL-XL. In an established PBMC engrafted tumor model, our lead claudin18.2 TriTCE Co-Stim molecule showed superior antitumor activity compared to bispecific T cell engager including a format matched bispecific control and an AMG 910, claudin18.2 TriTCE bispecific to the clinical benchmark. Taken together, our screening approach allows for the selection of a TriTCE format that exhibits increased cytotoxicity and activation of T cells through CD3 and CD28, while not finding an activating T cells in the absence of a tumor-associated antigen. In summary, bispecific T cell engagers have limited efficacy in solid tumors due to low numbers of intratumoral T cells and treatment-related T cell anergy. TriTCE Co-Stim molecules are designed to address these challenges by providing increased T cell fitness, activation and proliferation via tumor-dependent T cell co-simulation in 1 molecule. TriTCE Co-Stim may provide more durable responses in solid tumors and show superior activity in cold tumors with low T cell count. We are currently evaluating multiple tumor targets, including claudin18.2 as we continue to make progress with this program. Next, I will present an update on our trispecific checkpoint inhibition program or TriTCE CPI. Treatment of solid tumors is hindered by immunosuppressed microenvironment that can be refractory to traditional CD3-bispecific T cell engager. Immunosuppression in the tumor microenvironment limits treatment responses due in part to the expression of inhibitory immune checkpoints such PD-1 on exhausted T cells and PD-L1 on tumor cells. To improve T cell responses and antitumor activity in immunosuppressed solid tumors, we generated trispecific T-cell engagers that target a tumor-associated antigen CD3 and PD-L1 via PD-1 moiety to stimulate concurrent tumor-directed T cell killing and checkpoint blockade at the tumor site. The unique and differentiated activity of the TriTCE CPI was mediated by multiple mechanisms of action as illustrated in the top panel. Number one, by blocking of PD-1/PD-L1 to restore T cell activation at the T cell tumor synapse. Two, avidity-driven binding to PD-L1 positive tumor cells through engagement of the tumor-associated antigen and PD-L1 positive tumor cells. And three, additional potentiation of T-cell activation and proliferation via cross-linking of PD-L1 positive dendritic cells within the tumor environment. We employed now engineering and screening solutions to optimize TAA-dependent cytotoxicity and checkpoint inhibition that includes modifications to antibody formats, geometries, paratope and PD-1 domain affinities. And -- our in vitro screening identified TriTCE CPID molecules with enhanced TAA-dependent antitumor activity and checkpoint inhibition compared to a bispecific T cell engager as illustrated in the table at bottom. Looking further to the enhanced functionality of the TriTCE CPI, the data shown at top presents 2 examples of TriTCE CPI molecules that induce enhanced TAA-dependent T-cell cytotoxicity when compared to a bispecific T cell engager and enhanced PD-1/PD-L1 checkpoint blockade compared to a bispecific and bispecific plus the checkpoint combination. In PBMC engrafted xenograft model, the 2 TriTCE CPI molecules display superior in vivo antitumor activity compared to bispecific T cell engager as shown at bottom. In summary, T cell engagers have limited efficacy in solid tumors due to immunosuppressive tumor microenvironment, checkpoint upregulation and the emergence of resistance mechanisms over time. We generated multiple TriTCE CPI antibodies that combine tumor-dependent T-cell cytotoxicity with checkpoint blockade, which may translate to improved T cell responses in immunosuppressed solid tumors. The evaluation of multiple antibody formats, geometries and paratope affinities allowed for the identification of antibodies with enhanced TAA-dependent anti-tumor activity and superior site-specific checkpoint inhibition. Key factors that may contribute to a wide therapeutic index and improve clinical outcomes. Additional mechanistic assessment and TA target evaluations are underway. Moving to the next program. I will present updates to our ProTECT platform that apply a novel masking solutions to enhance the therapeutic window of T cell engagers. The critical limitation of bispecific T-cell engagers is the potential on-target off-tumor toxicity due to normal tissue expression of the targeted tumor associated antigens. These potential toxicities limit the number of applicable tumor targets that can be targeted by a bispecific T-cell engager. Masked, protease activated T cell engagers are promising new modality to limit potential on target off-tumor toxicity and have the potential to widen the therapeutic index of T cell engager targeting TAAs with normal tumor expressions, which are not applicable for bispecific T cell engagers. With ProTECT, we are developing a novel, differentiated T cell engager masking approach that builds on a TriTCE CPI platform and combines masking and linker technology with TriTCE CPI immune modulation to enhance the therapeutic window and efficacy of T cell engagers. As shown in the figure on the top, the ProTECT mask consists of an engineered PD-1/PD-L1 protein pair that sterically hinders the anti-CD3 paratope and CD3 binding in the periphery. The PD-L1 moiety is fused to the anti-CD3 antibody via proprietary linker sequence containing a protease cleavage site. Once cleaved, the resulting molecule is the trispecific TriTCE CPI antibody, providing T cell engager activity, checkpoint inhibition and additional differentiated functionality. In our initial proof of concept design, we evaluated the activity of a HER2xCD3 ProTECT molecule. And as shown in the middle panel, the masked ProTECT has greater than 1,000-fold reduced potency in vitro compared to the unmasked control and when cleaved, fully recovers the activity of the unmasked trispecific. Illustrated in the bottom panel, in vivo treatment of established tumors with the unmasked ProTECT results in complete and durable antitumor response. Furthermore, as shown in more detail in our poster, the unmasked ProTECT showed enhanced in vivo activity compared to the combination of the bispecific T cell engager and anti-PD-L1 antibody tislelizumab, combination, suggesting potentially an advantage of ProTECT over combination therapies. We next evaluated ProTECT on a relevant tumor target that has normal tissue expression and broad expression and application across multiple tumor types. As shown on the top left, this TAA x CD3 ProTECT molecule showed greater than 500-fold masking with efficient cleavage and full recovery of cytotoxic activity after cleavage, demonstrating the plug-and-play nature of the ProTECT platform. In addition, the protease-activated ProTECT displays the unique activity of the TriTCE CPI, including increased checkpoint inhibition shown at bottom left. As illustrated in the top panel, after protease activation, ProTECT show increased binding to exhaust the PD-L1-positive T cells. This might provide enhanced selectivity within the suppressive PD-L1-positive tumor microenvironment and potentially further enhance T cell activities by avidity driven T cell engagement via CD3 and PD-L1 to exhaust the T cells within the tumor. Additionally, we have shown previously that protease-activated ProTECT is highly active in vitro cytotoxicity assays with exhausted T cell. A novel mechanism of action in the protease-activated ProTECT is the ability to cross-link PD-L1-positive dendritic cells, or DC cells, and T cells independent of expressions of the tumor target. As shown on the bottom right, this DC/T cell cross-link can lead to the potent increase in T cell activation and proliferation but is protected in a masked ProTECT molecule. ProTECT masking and protease-dependent activation might allow this cross-linking to be specific to the tumor and may have the ability to induce and potentiate T cell activation via PD-L1 on antigen-presenting cells in the tumor microenvironment. Selective cross-linking of dendritic cells and T cells in the tumor microenvironment is a novel and potentially very attractive mechanism of action that we are currently further evaluating in relevant preclinical models. Overall, as illustrated here and described in more detail in our poster, ProTECT incorporates several mechanisms to improve tolerability of T cell engagers: an anti-CD3 masking, low affinity CD3 binding and enhanced selectivity for PD-L1-positive tumor -- within the PD-L1-positive tumor microenvironment with a protease-activated molecule. Additionally, protease-activated ProTECT displays the unique mechanisms of action of the TriTCE CPI to increase efficacy, including increased avidity and checkpoint inhibition. In summary, T cell engagers have limited efficacy in the clinic due to a narrow therapeutic window, checkpoint upregulation and the emergence of resistance mechanisms over time. The ProTECT platform is designed to tackle these challenges by combining a masking domain that when cleaved, provides additional immunomodulatory properties. In addition, ProTECT's therapeutic provides multiple unique mechanisms of actions that cannot be achieved through combination therapy. And finally, I will present updates to our masked IL-12 fusion program, ZW270. IL-12 is arguably one of the most powerful immunotherapy agents ever discovered, but its translation to the clinic has proven to be very difficult due to [ interim data ] associated toxicity. The ability of IL-12 to potentiate anticancer immunity and potentially remodel noninflamed tumor microenvironment makes it a very attractive therapeutic modality for multiple indications and combination treatment. Due to these unique immune-modulating properties and potential wide therapeutic application, IL-12 holds renewed interest as a cancer therapeutic. We have previously presented early data on IL-12 Fc engineering, and we are happy to provide an update on the evaluation and selection of ZW270, which uses a unique engineering approach to widen therapeutic index and might have the potential to overcome current limitations of IL-12-targeting therapies -- or IL-12 therapies. Wild-type IL-12 fusions have so far not provided significant therapeutic benefit in the clinic, and that multiple different approaches of systemically delivered IL-12 are currently pursued with the aim of widening the therapeutic index. These include potency attenuated IL-12 Fc fusion and protease-dependent activation of masked IL-12 to localize activity to the tumor. But achieving sufficient and sustained exposure of active IL-12 in the tumor microenvironment remains a challenge. To overcome those challenges and design a potentially superior IL-12 Fc fusion that can be systemically delivered but has an improved therapeutic index, we have been building on both the learnings of the field and our internal learnings on the cytokine engineering. ZW270 is a masked, potency attenuated IL-12 Fc fusion that is activated via extended-release, gradual protease cleavage. ZW270 is comprised of a single IL-12 molecule attached to the C-terminus of the Fc and an anti-IL-12 Fc scFv antibody mask with high affinity to p40 that blocks IL-12 activity when in the periphery. The IL-12 p40 subunit was, in addition, engineered to reduce potency without affecting scFv masking, providing a highly effective mask and an attenuated IL-12 Fc when activated by protease cleavage of the scFv mask. ZW270 also employs a proprietary cleavable linker technology that is the technology used in ProTECT and combines slow extended release, gradual unmasking in the periphery and highly efficient cleavage in the tumor. We hypothesize that combining these engineering strategies of potency attenuation, plus masking with an extended release protease cleavage has the potential to widen the therapeutic index of IL-12 therapeutics and be superior to comparator IL-12 engineering strategies. To evaluate our hypothesis, we compared ZW270 to other IL-12 comparator molecules in a pancreatic adenocarcinoma humanized mouse model. The comparator molecules included wild-type IL-12 Fc, an attenuated IL-12 Fc and a masked, cleavable IL-12 Fc with wild-type IL-12 potency, as shown in the main figure. ZW270 and the comparator IL-12 molecules were all dosed to the maximum tolerated level. And in the table and graphs below, only the treatment group [ and time points ] with less than 20% loss of [ mice ] are shown. As you can see, at the tolerated dose levels in this model, ZW270 was able to control tumor growth, while none of the comparators was equivalently effective, suggesting ZW270 might be able to achieve a higher exposure of active cytokine in the tumor while not exceeding the maximum tolerated exposure in the periphery. To further investigate the potentially superior exposure response relationship of ZW270, we performed quantitative systems pharmacology modeling based on our experimental data and literature data. As described in more detail in our poster, in this model, the projected therapeutic index of ZW270 was superior to the comparator Fc fusion. And the data suggests that the enhanced therapeutic index of ZW270 is mediated by a combination of Fc half-life extension, IL-12 attenuation and the extended release gradual protease unmasking. In a single-dose, nonhuman primate study, ZW270 was well tolerated at 10 mg per kg and at 31.8 mg per kg, while the wild-type IL-12 Fc had a maximum tolerated dose of 0.2 mg per kg. In summary, our activity attenuated and masked IL-12 Fc lead ZW270 has potent and superior antitumor activity to comparator IL-12 Fc fusions and is well tolerated in nonhuman primates greater than -- up to greater than 30 milligrams per kg. ZW270 utilizes a novel differentiated engineering approach, and our data suggests ZW270 might have the potential to overcome current limitations of IL-12 therapies. In addition, the unique geometry of ZW270 allows design of targeted IL-12 Fc as a next-generation molecule to increase tumor retention and potentially further enhance the therapeutic index. We continue to evaluate this program path forward as we maintain our focus on our portfolio of multi-specifics and ADC candidates. However, we believe this program has the potential to address shortcomings of previous IL-12 cytokine programs, and we'd look to continue development in the future with a partner. Shown here is a summary of the multi-specific poster that was presented at AACR. In terms of next steps for the 5 programs presented to date, ZW171 is anticipating IND submission in 2024. Both the TriTCE co-stim and TriTCE CPI programs are performing lead format selection and pilot toxicology studies. The ProTECT T cell engager program is also active in lead format selection. And with ZW270, we are targeting potential partnerships and evaluating opportunities for next-generation molecules. And with that, I will hand the call back over to Paul.

Thank you, Nina. So I just wanted to conclude by placing the effort described today and shared at AACR into the context of our R&D strategy moving forward. As we introduced at our R&D Day back in October, our 5x5 strategy entails the goal of bringing 5 new molecules into the clinic in the next 5 years. While previously, Zymeworks is focused on the development of a more select pipeline with broad investment in platform technologies and tools, moving forward, our goal is to accelerate our in-house product pipeline, taking advantage of our advancements in technology with both the ADCs and multi-specifics that you've heard today. [ We'll then augment ] this with select [ product ] partnerships. This slide then summarizes our early-stage pipeline, demonstrating an integrated pipeline of ADCs and multi-specifics. On the top of the slide, we list the programs discussed today, the tumor types they are designed to treat and the stages where we sit in our preclinical pipeline. The bottom presents our thinking moving forward to fuel the pipeline, which will come from both our ADC and multi-specific platforms, expanding our technology to additional targets and cancer types. Some of these programs will progress through internally developed programs, while others we anticipate progressing through partnerships. On this slide summarizes just some key bullets that we want you to take away from our focused R&D strategy. We plan to build a diverse pipeline of both best-in-class ADCs and multi-specifics, with the 5x5 goals by 2027. We'll do this through accelerating the speed of early clinical development through global diversification of early clinical studies ex-U.S. from our hubs in Dublin and Singapore. We have ongoing efforts to reduce the time from preclinical development candidate selection through IND filing. We plan to supplement internal portfolio with advancements of additional product candidates through collaboration and partnerships. We will retain U.S. commercial rights wherever possible with an ex-U.S. partnering strategy prior to registration studies. And we plan to do some selected expansion of R&D efforts in adjacent areas for continued innovation in ADCs and in multi-specifics with expansion of core capabilities where appropriate to ensure long-term competitiveness. I'll now hand over the talk to Ken to finish off with the concluding slide.

Thanks, Paul. While today's focus has been on the early R&D efforts, as you can see, we have an exciting slate of upcoming catalysts from now through 2024. Just to highlight a few noteworthy items. Later this year, you'll see our full data set from our pivotal study of zanidatamab in previously treated, HER2-amplified biliary tract cancer. And we'll also be moving forward with our Phase II studies of zani zo in non-small cell lung cancer in combination with PD-1 and in breast cancer looking at both HER2-low patients and patients who progressed on [indiscernible]. Further, we also plan to present additional data from our Phase I study of zani zo or present additional monotherapy data on patients since the cutoff from the 2022 ESMO presentation, including data from our weekly dosing cohort. We also intend to nominate our next preclinical candidate for an expected 2025 IND in the second half of this year. For our partner programs, later this year, we expect to complete and announce additional partnerships on our preclinical programs as well as announce progress and milestones from our portfolio of legacy platform licenses. 2024 will be a continuation of our focus on early R&D with 2 INDs planned for our lead preclinical programs ZW191 and ZW171, as well as the nomination of another preclinical program for a scheduled IND in 2026 to keep us on track for the Zyme 5x5 program objectives by 2027. These events highlight the continued progress we've made since I took over as CEO in January 2022. This year, 2023, is a year of focus and execution on the plan that we laid out last year when I started as CEO. As we continue to make progress across all elements of our business, I believe we're on the right track to continue driving value for the benefit of our shareholders and continue looking to making a meaningful impact to the lives of patients with difficult-to-treat cancers. I'm incredibly proud of the progress we've made, exemplified by an amazing group of our scientists and drug developers here at AACR this week. And we look forward to providing continued meaningful updates to the focus plan we've laid out during the remainder of 2023 and 2024. With that, I'd like to thank you all for listening and open up the line for questions. Operator?

[Operator Instructions] Our first question comes from the line of Charles Zhu from Guggenheim Partners.

Congrats on all the progress across your pipeline. Maybe a couple of specific ones for me, maybe a general one, but specific one is regarding the driving of immunogenic cell death with ZW49. It looks like a good chunk of that could be driven by the antibody rather than the payload. So would you expect to see potential synergy not only between zani zo and checkpoint, but also potentially between zanidatamab and PD-1 that may not be present with trastuzumab and PD-1.

That's an interesting thought, Charles. Thank you for that. I think our feeling is, is that -- you could be right. It could be completely driven by zanidatamab itself. I think a lot of the release of those [ damps ], those markers is more driven by the sort of cell death you get driven by the payload. And you're going to sort of -- so I think there's definitely a contribution of the payload there, and I think that's where we feel that the combination is really more appropriate with zanidatamab zovodotin. It's possible there could be some combination there with zani. But I think really the level that we see of those with the doses that we're treating is really more driven by the payload.

Got it. Great. And maybe one other somewhat specific question this time regarding GPC3 in hepatocellular carcinoma. Looking at the current landscape for systemic therapies in liver cancer, it looks like liver function remains a key eligibility criteria that potentially precludes a considerable portion of these patients for never receiving therapy. How much of an opportunity do you think potentially more targeted approaches can broaden the potential patient pool, particularly those with poor liver function?

Yes, that's a great question. That's something that we are considering as we'll be talking to our -- when thinking about our clinical strategy. What we do feel is, is that there is a clear unmet medical need in liver cancer, these clearly unserved patients, and strategies targeting molecules like Glypican-3 are being welcomed and encouraged from the community and from docs. So we feel that there is definitely a path there. And we definitely, like from our profile of our molecule like we've seen from preclinical data, the safety data that we've seen in nonhuman primates, that we're really encouraged by the therapeutic window that we achieved with the potency and as Jamie shared with you, on the potency in the xenograft molecules -- in the xenograft models. So I think our [ balance in the ] therapeutic window we see and the encouragement that we see with the safety profile, we're very continuing on that path. But we certainly have to bear in mind what is discussed, and we'll be discussing that with KOLs in the design of our clinical studies. But so far, the feedback we've got is there's a lot of encouragement for additional therapies, including the types of strategies that we are deploying.

Got it. Great. And perhaps if you can humor one last general question for me and I promise I'll hop off. Just one on capital allocation. You have so many different emerging assets. To what extent might your expected cash flows from zanidatamab commercialization be able to fund those programs? And how much additional potential partnership activity might you need to pursue to support further development of any one asset or the pipeline?

Yes, thanks for the question, Charles. Obviously, when we selected the strategy of the 5x5 by 2027, capital allocation and availability of capital stemming mainly from our partnerships with BeiGene and Jazz on zanidatamab were extremely important. And it's not that we only have 5 good ideas. We have many more than that. But we felt comfortable with the cadence of 2 INDs next year and one every year thereafter for our own account to get to 5 unencumbered assets we can work on ourselves and move forward in clinical studies, at least to the registration study states, that we can do that in an unencumbered way, relying mainly on sources of capital that are either already come in from the Jazz and BeiGene collaborations or expected to come in through milestones and eventually, commercial revenues from those sources. So not relying upon other future capital sources like equity capital, et cetera. So we feel comfortable that we can handle that 5x5 with the capacity of the team we have and the capital allocation we can put towards that. I think for all the other good ideas that we have beyond those 5, and you've seen some of those talked about today and presented at AACR and you'll see more of them in future conferences, we're going to be looking for potential partnerships and collaborations to provide the capital to move those forward in the clinical studies and additional resources from a partner to manage those clinical studies. I think we have a very good predetermined idea of how to have the right amount of capital allocation to our internal programs, partnership capital to move other things forward and make sure we can do that and recognize the fact that our current runway guidance is consistent with that capital allocation strategy, which takes us through at least 2026. And obviously, additional capital comes in from some of those partnership arrangements, will provide additional runway as we move forward with that strategy. So I think we feel very comfortable from a scientific perspective and a development perspective that we can get to the 5x5 by 2027, and we feel comfortable with the way we've thought about capital allocation to get there and continue to retain the strong balance sheet and runway guidance that will continue to be important to fund those programs long term and also be somewhat opportunistic on maybe something new that may be presented to us beyond that, as well as potentially have the ability to opt in to some of these other programs beyond the 5x5 where we're going to rely on partnership capital first, maybe be able to opt in later to retain additional U.S. co-promotion rights or joint development rights at least in the U.S. So I think it's something we thought a lot about because it's extremely important. It's even more important today than it was a year ago, and I think it's going to continue to be important to find a way to run a business which builds a broad portfolio of really great assets, [ put enough ] financial responsible ways to continue to provide the funding for those opportunities to continue to move forward.

Our next question comes from the line of Stephen Willey from Stifel.

Congrats on a pretty significant presence at AACR. Just a question on ZW171. So I know that dose optimization and selection of TCEs, even within the confines of, I guess, more homogenous key malignancies, has been kind of a bit of a challenging endeavor. So I guess I'm just trying to understand how you guys are going to try to control for some of that variability within the confines of a Phase I dose-escalation study where presumably, you have different patients, different tumor types, different mesothelium expression levels, different E to T effector ratios. And so how do you think about dealing with that variability in the context of a Phase I and then ultimately, being able to generate data that points you to a dose that you think that you can take forward?

Sure. Sure. I mean this is a good question, Steve. I mean you'd be aware, I've worked on quite a lot of T cell engagers in the past before I came to Zymeworks, really paid attention to the field. I think that really comes down to the nuances of that particular clinical trial. What we'll be focusing on is the dosing strategy and the escalation to try and get to a dose within the Phase I that will allow us to then expand into more broader patient set at a dose that we think can be efficacious and safe. So that will -- that's just typical Phase I development. I don't think -- I don't anticipate us doing any sort of unique patient-by-patient dosing. It will be just standard expansion. But I think by the design of our molecule, what we're really encouraged about and what I've seen is that the potency of our molecule and the balance of -- with the balance of safety, it really does look like we've kind of come across a molecule that really seems to have very functional activity in tumor models where when we compare that to other molecules, either 1 by 1 or even different 2 by 1 structures, we don't see that potency. So we think we're going to gain on having efficacy based on the preclinical models that we hope will overcome the challenges that people had before because they haven't had that balance of efficacy while having and maintaining a safety profile that we've kind of exemplified by going into cynomolgus monkeys at such a high dose. So I don't know if that answered it, but I think it comes down to the design of the molecule that we think does overcome limitations with prior T cell engagers that will allow us to just broaden the patients and hopefully have an improved profile.

And do you think...

Yes, and just to add to that -- sorry, just to add to that, again, we've obviously developed both 171 and 191 specifically to address a breadth of indications. And so we've thought a lot about the design selection and format of both of those molecules with that in mind. I think for mesothelium itself, I think the other thing that you'll see us do is, again, not try to take on all of the potential indications or target indications for mesothelium in a Phase I study. You'll see us likely select a number, but not all of them. So I doubt it will be an all-comer strategy. So I think that will be, in addition, one way that we can control some of those things that you're talking about. And I think if we're able to do that in several indications, then we can think about potentially expanding beyond that. So I think we've thought about that going forward with the breadth of indications we're trying to pursue at least for these first 2 programs, if that's helpful.

Okay. Yes, no, that's helpful. And then maybe just a quick question on, I guess, data that was presented at AACR. We saw the KEYNOTE-966 data, which, I guess, kind of looked very TOPAZ-like. But where are you guys just with respect to the generation of data with zani and GemCis in frontline biliary? I know that you're running, I think, what, a 20- or 30-patient cohort there. And I guess, is there a chance that we could see some data within the next kind of 12 months or so?

Yes, no, good question. We've obviously looked at that data set and also the detailed subpopulation data from TOPAZ-1. And I think what we're hearing back from KOLs in the space is exactly what you're maybe alluding to, which is zanidatamab as monotherapy with the top line data, and you'll see the detailed data coming up pretty soon, seems to be very effective and consistent and durable for a good group of patients. And we just haven't seen that with a more broad therapy involving a PD-1 plus GemCis. So obviously, that means our first-line data becomes more relevant to maybe inform physicians on the ability to use zanidatamab potentially with GemCis in an earlier setting. So we're continuing to recruit on our Phase II study, which is looking at first-line patients with zani plus GemCis, exactly that combination. And I think that's just -- hopefully will illustrate a little bit of our hypothesis that a HER2-targeted therapy might be the best thing for that population of patients who has a HER2 alteration/aberration that you'd like treat. We haven't given any guidance yet on when that study might fully recruit or when you might see data. We're obviously interested in working with BeiGene and Jazz to recruit that as quickly as we can because I think that will inform physicians of not only the use of zanidatamab with GemCis, because right now we just have monotherapy data, but also maybe understand what's possible in earlier line setting with a higher-quality patient in a first-line setting. So we're anxious about that. But as soon as we can guide on when that might be enrolled and when we might see that data, we'll do that. We're just not ready to do that today.

Our next question comes from the line of Derek Archila from Wells Fargo.

This is [indiscernible] on for Derek. Congrats on all the progress. So regarding ZW191, can you discuss some of the differential properties for the molecule relative to other folate receptor alpha targeting ADCs? And how much juice is left to squeeze in ovarian cancer in your view? Like do you view this as a real market or something you want to use as a benchmark for your novel platform?

Yes, well, I think -- thanks. I mean I can speak, first of all, to the differentiation. I think that as we shared, what we've done is spent a lot of time thinking about this antibody drug conjugate itself, both the antibody and the drug conjugate and the payload. So even on the front end of the molecule which binds the folate receptor, we have -- we selected a somewhat unique antibody based on its comparisons to other antibodies that target folate receptor that really is very efficient internalizing, and a lot of that data we shared in the poster presentation. And so that, we think, gives a lot of horsepower on the front end to get to the target and get internalized. And then on the back end of the molecule with the conjugate, obviously, as you know -- as we discussed, we have our own proprietary [indiscernible] payload that we think then, combined with the antibody and some of the other features and the way that we've attached the drug conjugate to the antibody, differentiates it from the comparator molecule. And that's been reflected in the data that we show and that Stuart discussed that what we see is that we see efficacy with those -- with our molecule in patient-derived xenografts and models that you don't achieve with the comparator molecule. So we can target the lower-expressing folate receptor. So I think functionally, both by design and by proof of data, we see differentiation. And then I think to your latter point, I think you're on the track of what we are thinking. We do see the higher-expressing ovarian cancer is a way to have some sort of benchmarking, but that's not really our design and our goal. Our goal is to go deeper into the lower-expressing ovarian cancer, and then go into other tumor types that express folate receptor also at more modest levels but we think we can potentially treat with our molecule.

That's very helpful. And as a more general question, as you look across the state of technologies you can leverage, can you walk us through the process on deciding which one to utilize [ for existing targets and time to development ] candidates?

So you're asking the question how do we pick one technology over the other technology for developing?

Yes, basically, for like specific targets, how do you choose one technology over the other?

Right. Yes. I mean I think some of that does come down to the target biology itself, the expression of the target, the level, its ability to support internalization. And then that's also then combined with then if we make a molecule and design it, how well does it actually function and support the profile that we would want to have to develop it. So it's really somewhat based on the biology, and then some of it is based on empirical testing because we do have the ability to make both molecules and then we'll take forward. The other factor is that in -- with some targets like, for instance, folate receptor, it's already clinically validated [ to simply support ] an ADC. So there, because it was our first target that we're developing, that was attractive because it did give us benchmarking and sort of derisked a little bit of taking a more novel target. But as we move forward, we'll also use the ADCs against novel targets as well.

Our next question comes from the line of Yigal Nochomovitz from Citi.

This is Ashiq Mubarack on for Yigal. Really appreciate all the detail today. I had just a couple on the ADC programs. For your folate receptor alpha ADC, I guess building on the last question, do you think the benchmark in the clinic is to be better than mirvetuximab? Or do you think potentially being able to target patients with lower levels of expression is sufficient as a differentiator, especially given, I think, Sutro is in the clinic with a sort of similar strategy? And then the second question on the same topic, how do you view the Phase III MIRASOL data, which is expected very shortly, as potentially impacting your future development strategy?

Yes, thanks for that. I think for the second part of it, obviously, that's -- it's not our drug and not our clinical study. So it really doesn't have an impact on how we might function. We root for all studies that might have the potential to help patients, whether they're ours or someone else's. So hopefully, for patients, we find something beyond the subset of patient populations that we've been able to seem to address so far, and we're looking to go well beyond that. So we specifically designed our folate receptor alpha ADC with 2 things in mind. One is there's a lot of interest in looking at TOPO1 as a payload for that patient population. So we're really excited to put forward a product format that allow us to do that. I think our specific interest in the way that we designed and selected that product was to find a way to find something that might be encouraging for the breadth of the indications that folate receptor alpha might be. So well beyond the subset of high-risk patients in ovarian cancer that -- or high-expressing patients in ovarian cancer that seem to be suited to [ at least mirv ]. And so we're looking to go well beyond that, and that's the way we design engineered the drug. That's the way we're going to approach the clinical studies is to look for evidence of efficacy and safety in our broad class of patients at multiple indications and at multiple levels of expression of that target. So that's our goal. That's what we're striving to do, and we're trying to do it with what we think is a really interesting payload that -- through the combination of payload and the antibody and the entire ADC structure that we have with the clinical strategy might be pretty encouraging for a much broader group of patients than you're currently referring to.

Okay. Got it. That's very helpful. And then maybe asking one on the NaPi2b program. I guess looking at some of the preclinical data you shared, why did you compare ZW220 against -- with trastuzumab versus the NaPi2b agent currently in the clinic, the one called UpRi?

Yes, I'll answer that, and then Stuart or Jamie may add to that. I think there, part of that is some of those molecules, to actually make them accurately requires some technology and some approaches that are not as straightforward as some of the other benchmarks. And I think that particular benchmark, if I recall correctly, it's got some additional complexity. And so it's a benchmark against you have to be more careful to generate it correctly. So you're really genuinely benchmarking against something. So I think that's something that the team is looking into to also test against that. But right now, we just didn't feel comfortable sharing that data at this point.

Okay. I understand. And then last quick -- the last question for me...

Yes, let -- actually, let -- maybe Jamie can expand on that as well.

That is, in fact, correct, Paul. Yes, the drug linker component of UpRi is somewhat unusual in the sense that it has a polymer backbone. It's got some features that we're not as familiar with in terms of generating representative chemical structures. And so we view lifastuzumab vedotin and UpRi as having roughly equivalent clinical outcomes. And so benchmarking against LIFA seems appropriate.

Okay. One last quick question. One thing I haven't -- I didn't hear you talk too much about with your new ADCs is potential future combinability with other agents or maybe standard of care depending on the setting. Is that something you're currently thinking about? And are you generating potential combo data in the preclinical setting?

Definitely, that will be considered a strategy moving forward. And certainly, we factor that in. I mean I think, of course, right now, it's going to be monotherapy, getting the data for that molecule. So our focus really is driving that, but we do see great opportunity for combination treatments. I think that's kind of -- just to kind of [ go off the times ] a little bit, if we -- what we talked about with zani zo, in that case, that's where we're thinking of a combination. So a little bit different with the topo platform and moving forward. But again, we may employ that. And in addition [ to behind ] the topo platform, the team is also thinking about other payloads as well that we can incorporate in advances of the technology. So that's the other thinking on supplementing what we're doing with topo.

And our next question will come from the line of Brian Cheng from JPMorgan.

Maybe just -- I want to start off with a general question first. So Ken, you mentioned that you see your assets with predefined path in a sense that you're clear on which asset will be partnered and which asset is to keep. So what are the factors that you will need to consider to sort out which asset will be predefined to be partnered? And just one more on top of that as to how do these newer platforms, such as masking technology, the TriTCE and cytokine platform, [ those on ] what you are leading with 171 and 191, how are you using those platforms to build in your priority?

Yes, I'll address the first one. I think -- as I said, we -- I think right now inside the company, we've got an abundance of what we think could be really interesting products. And so our goal is sorting through those and determining which are the ones will be in the 5x5, which obviously will be things that we'll keep which we'll have a future commercial interest in versus the things that we'd like to work with partners on and maybe retain some future opt-in rights or future co-promotion rights within that commercial interest. So I think there is some strategy around both of those sets of products that we look at we think are attractive to work on from a development standpoint and competitive standpoint and eventually, commercial size of the market, which is important for us and one of the criteria we've talked about. I think there is some logic around what those 5x5 programs will look at, the potential end markets, the specialties that will be using those. And so there'll be some logic around how those will all fit together in addition to the potential for zani zo to be something that we would retain future U.S. rights on an ex-U.S. strategy. So I think we're starting to think about how that will work together so we can do this in a very methodical and deliberate effort. And so I think as we start to develop that portfolio of assets that are in the 5x5 and as you start to see our first partnership and collaboration move forward, I think that will start to be more evident as you see those assets fall into one group or the other. And so I think as we play this out, I think it will become evident how we're thinking about doing that in a way that [ it leaves us with not just a ] cohesive R&D strategy in terms of what we develop, but something that ends up being a cohesive commercial strategy at the end of this, which is something we're at least interested in the longer term.

Yes, so I think the back end of your question was just talking about how do we integrate these different T cell engager platform technologies that we discussed. And I mean I think there, part of this is what the Azymetric platform allows us to do is to actually design molecules that on paper maybe other -- that really address certain limitations of other molecules, okay? By having the co-stimulation in there or the checkpoint inhibition in a single molecule, we can get synergistic biology not achievable. We still have to move these programs more through additional clinic -- preclinical testing and select which one will really make it and satisfy our go/no-go criteria. But it does allow us to plug in different targets for different tumor types that are not addressed at the moment with existing therapies and existing T cell engagers. And that, I think -- we, of course, want to do what's best for the patients, providing new treatment options. And I think these strategies can get us to that possibility. But of course, we need to think also about the target that we pair. We shared one of the targets [ called out in '18 ], but we're also thinking about other targets we can pair there. And there may be certain targets and certain biologies that really need either co-stimulation or overcoming checkpoint inhibition. Depending on that target, we'll apply the appropriate technology.

I'm not showing any further questions in the queue at this moment. I'll turn it back to management for any closing remarks.

Well, that's great. Thank you very much, operator. And again, I sincerely appreciate everyone dialing in and asking the questions. Thanks for being patient with the amount of material that we wanted to download with you. Obviously, we're extremely excited about being able to present a pretty broad spectrum of work across the portfolio, all of which, we think, is extremely high quality. And so it did take us a little while to do that given the number of posters that we're presenting and wanting to make sure we did justice to every part of that. So appreciate that. If there's questions we haven't gotten to during the course of this call, [indiscernible], we're always available. So please reach out to us, and we're happy to follow on as you dive through all the abstracts and posters that we've presented over the past 2 days. Again, I want to thank the R&D team internally, those that were here in Orlando and those who are back in Vancouver and Seattle for working so hard over the past year to make progress with a range of programs in the company, which I think is going to lead to a development of a pretty exciting portfolio of both ADCs and MSATs that we think could really make a difference in the treatment of patients with some poor prognosis of indications that we're pursuing. And we're really excited to continue to make that progress and continue to report it to you and have you follow along with us. So thank you very much and look forward to reporting some more progress in the not-too-distant future. Thank you.

This concludes today's conference call. Thank you for participating. You may now disconnect. Everyone, have a great day.