Mersana Therapeutics, Inc. (0M40.F) Earnings Call Transcript & Summary

Good morning. Thanks for joining us for another session at the 42nd JPMorgan Healthcare Conference. I'm Brian Cheng, one of the senior biotech analysts here at the firm. I'm joined by my associate, Sean Kim, who's on the -- who's in the audience. On stage, we have Mersana Therapeutics CEO, Marty Huber. I'll pass the mic to Marty for a short presentation, followed by a live audience Q&A. If you're joining us live, you can raise your hand. We have runners in the audience. And if you're joining us virtually, you can also submit questions on the conference portal. Marty, I will turn the stage to you.

Good morning, Brian. I would like to thank JPMorgan for the opportunity to share the Mersana story. Please be aware that this presentation will contain forward-looking statements that are subject to certain risks and uncertainties, including those you see on this slide. For additional details about these risks, please refer to our most recent 10-Q filings available at sec.gov. As we all know, 2023 was a banner year for ADCs, one filled with new product approvals and significant business development activity. In short, ADCs remained at the forefront in oncology and reinforced their role as a foundational therapeutic modality that's benefiting an increasing number of patients. That said, there are significant platform and payload issues that continue to prevent ADCs from realizing their full potential. At Mersana, we have focused on innovating on from a platform and payload perspective to overcome these challenges and generate novel ADC medicines that have the potential to meaningfully improve both efficacy and safety. Let's talk through some of the details. While highly beneficial for select groups of patients, we believe that the first wave of ADCs leaves substantial opportunity for improvement. First-generation ADC anti-tubulin platforms like vcMMAE have well-established toxicities like neutropenia and peripheral neuropathy. These toxicities generally are not driven by ADCs interacting with the target antigen. They are off-target effects driven by the platform and payload. Off-target platform toxicity is typically what prevents ADCs from reaching a maximally efficacious dose and from combinations with other agents, including other ADCs. It is also responsible for patients discontinuing treatment prematurely due to tolerability. As we will discuss, our primary focus at Mersana initially was to develop a next-generation cytotoxic platform that avoided serious off-target platform toxicities so that we can deliver more payload to the tumor. We are now advancing a cytotoxic platform, Dolasynthen, that we believe has the potential to accomplish this mission. The same can be said for issue 2 on this slide. In more recent years, ADCs delivering a topoisomerase inhibitor payload have been an increasing focus for the field. However, similar to anti-tubulin payloads, these ADCs have their own off-target, dose-limiting toxicities, namely myelosuppression and interstitial lung disease. ADCs with topo payloads have another emerging issue as well. With their increasing usage, we are observing the emergence of topo-specific resistance. This has been borne out by multiple real-world data sets presented at ASCO, ESMO, and most recently, at San Antonio Breast Cancer. Mersana is working to provide effective alternatives to topos to overcome these challenges. And finally, we believe there is significant room to innovate from a payload perspective, and our work extends beyond the traditional cytotoxics. We're focusing on establishing an entirely new class of immuno-oncology ADCs, our STING agonist ADC platform, Immunosynthen. We look forward to advancing this vision in 2024. The most important clinical milestone for Mersana will be the initial clinical data set for XMT-1660, which is our Dolasynthen ADC targeting B7-H4. We have made significant progress in dose escalation. We are currently at dose level 6 and still have not established a maximum tolerated dose. We are also enrolling patients in multiple backfill cohorts at clinically relevant doses as we look to optimize both dose and schedule. Once we have defined our recommended Phase II dose, we plan to initiate tumor-specific expansion cohorts in the second quarter of this year. And we plan to present initial clinical data from dose escalation and the backfill cohorts in mid-2024. While we believe that Dolasynthen has the potential to improve upon the limitations of current cytotoxic ADCs, we also believe our first-class Immunosynthen ADC, XMT-2056, has the potential to pioneer an entirely new category of ADCs by delivering an IO payload to stimulate the immune system in the tumor micro environment. We believe preliminary clinical data indicate that XMT-2056 is much more potent before initiating the innate immune system in humans than we've seen in our preclinical data. We recently resolved the clinical hold on our Phase I trial of XMT-2056. We're restarting the trial at a much lower dose and are looking forward to advancing dose escalation in 2024. Underlying all of this is strong balance sheet that provides the capital resources required to support our current operating plan commitments into 2026. Here, you can see Mersana's pipeline of product candidates as well as the collaborations that we have in place with both Janssen and Merck KGaA. We'll be discussing these in a few minutes. This pipeline is being driven by 2 proprietary and very distinct ADC platforms: Dolasynthen and Immunosynthen. As you can see here, we have 2 internal programs. XMT-1660 is our Dolasynthen ADC targeting B7-H4 that is currently in Phase I. XMT-2056 is our Immunosynthen ADC targeting HER2. We have an option deal with GSK for XMT-2056. We received $100 million upfront in this deal, and GSK obtained and maintains the option to exclusively license this candidate in the future. Finally, I would like to note that we have external collaborations with Janssen for the development of novel Dolasynthen ADCs and with Merck KGaA for Immunosynthen ADCs. We believe these 2 ADC platforms will continue to fuel our pipeline and those of our collaborators. Let's provide some details beginning with Dolasynthen. Dolasynthen is our next-generation cytotoxic platform that is designed to produce ADCs that have distinct advantages compared with first-generation ADCs, including our own internal first-generation platform, Dolaflexin. It is clear that anti-tubulin platforms have the potential to benefit patients. As I noted before, many first-generation ADCs are limited by severe neutropenia, neuropathy and ocular toxicity. They also are limited by their narrow range of drug-to-antibody ratios or what we'll refer to as DAR for the rest of the talk. Through deliberate design, we developed a novel or a statin payload to avoid these dose-limiting platform toxicities and we also introduced elements to improve pharmacokinetics. On the left here, you see Dolaflexin, Mersana's first-generation ADC platform. This platform was -- incorporated our novel payload that exhibits a controlled bystander effect, allowing for limited diffusion with the tumor upon release from the antibody until it is enzymatically converted to an even more active metabolite. The active metabolite then becomes trapped in the tumor. In addition, the hydrophilic polymer compensates for the hydrophobicity of the payload, allowing for improved PK while also enabling high DAR. We knew this platform had its limitations because it was inherently heterogeneous. It produced ADCs that had a DAR range 10 at average, but some were much lower and others had much higher DAR. It also attached payload to the antibody in a stochastic or random fashion. Several years ago, we began working on a next-generation ADC platform that was completely homogeneous, one that can allow us to produce ADCs with precise balance and uniform DAR and to customize that DAR for specific targets, matching the ADC to target rather than the typical one-size-fits-all approach. Dolasynthen is a result of that effort. This platform utilizes a proprietary, fully synthetic scaffold linker payload, or SLIP, that precisely balances the payload with both charge and hydrophilicity. It also utilizes precise site-specific bioconjugation and a cleavable payload linker. Our preclinical data suggests that Dolasynthen may not only solve the severe off-target toxicities associated with MMAE, but it also successfully addresses the limitations that we came to see in our clinical result from UpRi, a Dolaflexin ADC that was investigated in platinum-resistant ovarian cancer. Here are a quick snapshots demonstrating how Dolasynthen compares to -- preclinically to Dolaflexin. As we can see on the left, the chromatography for Dolaflexin in orange shows a broad population of ADCs, some with very low DAR and others with very high DAR. Contrast this to Dolasynthen in blue, which has a single uniform peak, demonstrating the homogeneity of the ADC. Also note that this peak is significantly less hydrophobic. This leads to the figure on the upper right. Here, we are looking at the average DAR distribution for time in nonhuman primates. We can clearly see that on administration, the average DAR for Dolaflexin drops rapidly over the first 100 hours. Importantly, this rapid decrease in DAR is not due to deconjugation, rather, it is absorption into -- by lipophilic pathway into non-antigen targeted cells. Free payload levels in the plasma remained very low contrasted to DAR with Dolasynthen in blue, where you don't see this rapid clearance. Why does this matter? This is shown on the bottom right. Here, you can see Dolaflexin ADC compared against the Dolasynthen ADC with an identical payload dose in a lung cancer model. Dolasynthen clearly outperforms the Dolaflexin in orange, and have clearly seen this in many other preclinical models. Importantly, we know that the preclinical differentiation we observed for Dolasynthen may underestimate the impact in the clinic. How do we know that? Well, we have actually brought 2 ADCs into the clinic that utilize the same antibody NaPi2b. Of these, UpRi 1536, was developed using our original Dolaflexin platform. The other XMT-1592 was developed using the Dolasynthen platform with a DAR 6. This afforded us the opportunity to clearly distinguish the clinical capabilities of the 2 platforms. In the first half of 2024, we will share clinical data for both of these discontinued candidates. What you will see is that we saw multiple objective responses in dose escalation with the Dolasynthen XMT-1592. We also saw marked improvement in tolerability issues that we believe were associated with the Dolaflexin platform at its high-DAR subpopulations. Specifically, we saw much lower rates of fatigue, nausea and thrombocytopenia with 1592. On this last point, as some of you may recall, we saw a relatively high incidence of thrombocytopenia and a low rate of severe bleeding in the -- with UpRi in our trial. After researching this extensively, our hypothesis is that the bleeding risk was driven by endothelial cell damage resulting from nonspecific delivery of high-DAR subpopulations within the heterogeneous Dolaflexin mixture. The thrombocytopenia was not sufficient to cause spontaneous bleeding. However, we believe the decrease in platelets was a marker of platelet consumption and a potential indicator of endothelial damage. Importantly, in our dose escalation study with XMT-1592 on the Dolasynthen platform, the rate of thrombocytopenia was significantly reduced, and we did not observe any treatment-related bleeding. While Dolasynthen demonstrates superior to Dolaflexin, one of the benchmarks we are actually working to beat is vcMMAE, which is the most well-established ADC platform of all. You can see some preclinical comparisons here with vcMMAE shown in green compared with Dolasynthen shown in blue. In the first panel, we show significantly improved exposure for Dolasynthen, which suggests that we may be able to dose less frequently than some vcMMAE ADCs. As I mentioned at the outset, one of the key limitations of vcMMAE is that it is known to cause neutropenia or decrease in white blood cell counts. On the lower left, you can see neutrophil counts in mice 8 days following doses. While you see significant reductions in neutrophils for vcMMAE ADCs on the right with the vehicle, there is no impact from the Dolasynthen ADC in blue. And on the right side of this slide are a couple of examples of Dolasynthen's efficacy outperformance across different tumor targets. Let's focus, in particular, on the right -- on the upper right. The black line shows the vehicle control, which progresses rapidly. The gray line is a nontargeted ADC that we include in the experiment to show that the efficacy is -- the model is driven by the antigen-specific interaction. Now note the overlapping green and blue lines. The green line shows the efficacy of vcMMAE ADC with an antibody dose of 6.58 milligrams per kilogram and a payload dose of 0.1 milligrams per kilogram. The overlapping blue line shows the efficacy for a Dolasynthen ADC that has a 10x lower antibody dose as well as a 4x lower payload dose. We can also see the other blue line at the bottom in which Dolasynthen ADC is given at an equivalent payload dose. And in this setting, we flatlined the tumors. XMT-1660 is our lead Dolasynthen ADC that targets B7-H4. We designed this molecule specifically with a DAR of 6 and are currently advancing in a Phase I trial that is enrolling patients with breast, endometrial and ovarian cancers. Now stepping back for a moment and looking at B7-H4 as a target. This is a member of the B7 family of immune co-inhibitory molecules but is quite different from other well-known targets like B7-H1, which is also known as PD-L1; and B7-H3, which is another intriguing ADC target. Interestingly, B7-H4 appears to be rarely co-expressed with B7-H1 or PD-L1 in breast cancer, suggesting a potential role in cold tumors. B7-H4 has low expression in healthy tissue but is broadly expressed on the cell surface of multiple tumor types, including breast, endometrial and ovarian. Furthermore, high expression has been associated with cancer progression and poor prognosis. As can be seen on the left, over 2/3 of patients with TNBC have moderate to high levels of B7-H4 expression. A majority of patients with endometrial, ovarian and hormone receptor-positive breast cancer also have high expression. Finally and importantly, recent clinical data from -- presented at ESMO in 2023, one for a vcMMAE for ADC and the other for a Topo 1 ADC have validated B7-H4 as a compelling target. Objective responses were noted in patients treated with both molecules, most notably in breast cancer. However, both of these presentations also reinforced the dose-limiting nature of platform toxicity associated with the older platforms. And encouragingly, neither showed any obvious B7-H4 target-related toxicities. So let's move on to our development of XMT-1660. Again, one of the unique features with our Dolasynthen platform is our ability to modulate drug-to-antibody ratios for specific targets. On the left is an example of this. Here, we looked at a DAR 2, a DAR 6 as well as a DAR 12 Dolaflexin ADC. And as you can see, clearly, for this specific target, the DAR 6 is outperforming. On the right, we see the data from now, these are PDX tumors in which we take human tumors and put them into mice. And we can see the efficacy against both TNBC and estrogen receptor-positive breast cancer. On the top, you can see that XMT-1660 drives strong tumor responses in both triple-negative and ER-positive breast. It is also -- and you can see below, in the lower panel, that the responses tend to correlate with B7-H4 expression, which is shown in gradients of brown. With regards to the clinical trial, the promising preclinical work led us to initiate a Phase I dose escalation study. We're developing 1660 for the potential treatment of patients with a broad range of solid tumors. And as you can see in the schematic for the dose escalation on the left, our Phase I is designed to evaluate safety and tolerability of XMT-1660 as a single agent in this study. The tumor types that are included in the Phase I dose escalation study include patients with advanced metastatic TNBC hormone receptor-positive breast cancer as well as ovarian and endometrial cancer. These patients are ones -- Phase I population that have exhausted standard of care. Today, single agent is the standard of care for this type of patient, and their prognosis remains exceedingly poor. For patients with these late-stage cancers, overall response rate tend to be in the order of 10% to 15%. For TNBC, it's even worse at approximately 5%. And even then, we do not -- have very little data on what the post and HER2 post-Trodelvy environment looks like for TNBC. In addition to the primary end points of the study of safety and tolerability, we will also be evaluating preliminary antitumor activity, PK and immunogenicity. While we are not selecting patients based on B7-H4 expression, we are exploring the association of expression in tumor and other markers related to cancer or inflammation. We are making very good progress in dose escalation. In fact, as I mentioned earlier, we have now reached dose level 6. We are also continuing to enroll patients in multiple backfill cohorts at clinically relevant doses to better understand and optimize the dosing schedule that we will select for our Phase II dose. XMT-1660 is 1 of 4 B7-H4 ADCs in early clinical development. As I mentioned before, one of the competitive molecules is a vcMMAE ADC, and we have seen its customary platform toxicity issues that limit dose intensity in the ESMO data set. This ADC is dosed biweekly, which may be due to, at least in part, PK limitations. The other 2 ADCs in development utilized topoisomerase 1 inhibitors as payloads. While we have yet to see prospective randomized data as shown -- but as shown on this slide, real-world evidence of topo-based payload resistance is emerging. The data shown here are primarily from patients receiving multiple Topo 1 ADCs. In this figure, we compare the progression-free survival following the first topo in the blue on the left while the second treatment in pink on the right. As you can see, the size of the bars on the left are generally much longer than those on the right. We believe payload resistance could prove to be a challenge for the 2 topo-based B7-H4 ADCs in development, particularly in breast cancer where Enhertu and Trodelvy are increasingly being utilized in early treatment, both of those being built on a topo inhibitor payload. While physicians are using these agents sequentially today given the lack of alternatives, we believe a non-topo payload would be much preferred to address this high unmet medical need. Putting it all together, we're very excited about the potential of XMT-1660. B7-H4 is a clinically validated target that is highly expressed in a number of different cancers. We have demonstrated preclinically XMT-1660's ability to drive deep responses in tumors of interest, and we believe this Dolasynthen candidate has the potential to avoid many of the severe platform toxicities that have hampered the ADC field. 1660 was granted Fast Track designation by the FDA in advanced metastatic triple-negative breast cancer, and we're making great progress in the clinic with plans to initiate our expansion cohorts in Q2 and share initial clinical data in mid-2024. We expect this data set to include meaningful efficacy, safety and tolerability data from the dose escalation and backfill cohorts that can be viewed within the broader context of the B7-H4 ADC landscape. Now turning to our Immunosynthen platform. Dolasynthen has the opportunity, we believe, to improve upon existing ADC platforms. Immunosynthen offers the potential for an entirely new frontier in the development of ADCs, ADCs that are able to harness the power of the immune system in a targeted fashion while avoiding widespread systemic toxicity. Furthermore, by using an IO payload, these ADCs could address resistance to cytotoxic payloads. Checkpoint inhibitors like PD-1 have proven to be effective means of releasing the brakes on the immune system within the immune tumor microenvironment. However, it's been very challenging to date to develop agents that can step on the gas to accelerate the immune response in a tumor-targeted way. STING is a master pathway for the innate immune system that is capable of mounting an efficient immune-mediated tumor response. Preclinical data show that STING activation can induce prolonged antitumor activity, and most importantly, generate immune memory. Unfortunately, systemic STING agonists are generally unable to safely drive tumor activity given their either passive or indiscriminate nature of delivery or their very short half-life, limiting the duration of the immune activation in the tumor. Furthermore, exposure of T cells to STING agonists can result in a proapoptotic activation that is contrary to the goal of enhancing rather than suppressing the immune system. Our Immunosynthen platform is designed to harness the power of STING. We believe that a STING agonist ADC has the potential to overcome the limitations of free STING agonist and intratumoral STING agonist injections by enabling systemic administration of a highly targeted innate immune modulator. With Immunosynthen, we believe we can deliver an impactful one-two punch by activating STING both within the tumor and in the tumor-resident myeloid and dendritic cells while also minimizing the risk of systemic and T cell exposure. Despite recent advantage advancements, most HER2 patients do not have approved immunotherapy treatment options. XMT-2056, being a systemically administered Immunosynthen ADC that targets a novel HER2 epitope, has the potential to change that. Each antibody is equipped with 8 STING agonist payloads, and we have plans to investigate this module in a range of HER2-expressing tumors, including breast, gastric, colorectal and non-small cell lung cancers. Slide 21 shows the sampling of our preclinical work with XMT-2056. On the left, you can see 2056's ability to generate strong monotherapy activity in both HER2-high, and importantly, in HER2-low preclinical models. And on the right, you can see it's used in combinations with different agents, including PD-1 and Enhertu. The Enhertu model is particular interest because this agent also targets HER2 and results in significant tumor regressions even with the lower doses in this -- of Enhertu in this combination setting. This is made possible by the fact that 2056 targets a different HER2 epitope than both pertuzumab and trastuzumab. In Phase I dose escalation, we plan to enroll patients with a range of HER2-positive cancers as shown here. Despite recent advances, most patients with advanced metastatic disease will inevitably progress following treatment, leaving this unmet need. The development of therapeutic agents with unique mechanisms is desperately needed, particularly for those that could be used in combinations with approved agents. Our Phase I dose escalation leverages a rather standard Bayesian optimal interval design. 2056 is dosed every 3 weeks with tumor assessments planned for every other cycle. Similar to our 1660 study, the dose escalation portion of our trial also calls for us to enrich a population with backfill, which in this case, we are calling select enrichment cohorts. Each of these cohorts will include up to 15 patients, and we plan to include at least 1 HER2-high and 1 HER2-low breast cancer cohort. As many of you know, this trial was placed on clinical hold following a Grade 5 adverse event. We dove deeply into the cytokine, pharmacokinetic and other clinical data from the initial patients dosed in the trial. We believe they clearly indicate that XMT-2056 is a much more potent innate immune stimulator in humans than we've seen in our preclinical data. We have accounted for these findings in our revised dose escalation strategy and are now actively restarting a trial at a lower starting dose. In summary, we are excited to be getting this program back underway, and we plan to advance dose escalation in 2024. We believe demonstrating proof of concept with XMT-2056, our lead Immunosynthen molecule, could extend the field well beyond cytotoxics and pave the way for an entirely new class of immuno-oncology ADCs. The final pillar in our strategy is to leverage collaborations to maximize the potential of our platforms. Over the course of the past 2 years, Mersana has aligned itself with multiple strong collaborations as shown here. I will not review the details. We're happy to address any questions. In summary, we are excited about what lies ahead for Mersana in 2024. As we move through the year, our goal is to demonstrate over several scientific presentations how Dolasynthen's preclinical differentiation translates into the clinic. These presentations will include initial dose escalation clinical data from Dolasynthen NaPi2b, the 1592; and as well as our ongoing Phase I clinical trial of XMT-1660, the B7-H4 Dolasynthen molecule. We also plan to advance our goal to enable targeted innate immune stimulation via our Immunosynthen platform by progressing dose escalation with XMT-2056. And finally, we plan to make further progress in our collaborations. I thank you for your attention and look forward to reporting back to you on our progress through the year. This concludes our presentation. We're now happy to take questions. Thank you.

Great. Let's start with the Q&A session. For those of you who are in the audience, if you have any questions, you can raise your hands. We have a runner on the floor. For those of you who are virtually, you can submit your questions on the conference portal. We're joined by Marty, the CEO; and also SVP of IR and Corporate Communications, Jason Fredette. Good to have you guys on.

Thank you, Brian.

Yes. Maybe just to kick off with the -- on 1660, there are a couple of players now in the B7-H4 race. How do you see your 1660s differentiating in this space given that some of the competitors are slightly ahead? And we haven't seen your data yet. So can you just talk about just where you stand in terms of differentiation? And then, ultimately, how do you think about strategically to outcompete -- to outrun your competitors?

Okay. Maybe I'll give Marty a little break since you just went through a lot. And Marty touched on this a little bit. There are 3 other clinical-stage ADCs targeting B7-H4. We think we have a good amount of differentiation. Marty alluded to one of the candidates leveraging a vcMMAE platform. In their initial data set, we saw their customary platform-associated limitations, peripheral neuropathy and neutropenia. That agent also is being dosed once every 2 weeks in their program. So there's opportunity, we believe, to differentiate on both of those fronts with our Dolasynthen platform. With our payload, we have not seen severe neutropenia, peripheral neuropathy, ocular toxicity. Generally, we believe we're able to avoid those toxicities. And I think we've really improved as well versus our Dolaflexin platform, as Marty touched on. The other 2 ADCs are Topo 1 payloads. And Marty alluded to platform-associated tolerability issues with those agents as well. Myelosuppression, ILD are 2 of those. In addition, this resistance to topo has really come to the forefront. And given the fact that breast cancer is a big potential indication for B7-H4 ADCs, we think we potentially could have an advantage there as well.

Any questions in the audience? So turning to your Phase I updates midyear. Can you talk about just the expectations around what we should expect in terms of the sample size that we could get at the next upcoming read? You're doing -- you're up to DL 6 now, and also you're backfilling some of the doses. So how big of that data set could we get? And also just your -- it's also a basket of solid tumors as well. So how much granularity could we get at that time point?

I think it should be sufficient for you to make a judgment kind of how the molecule is fitting in. A couple of things on, one, for the backfills, these are up to a dozen patients at any one dose schedule level. And as we communicated, we started doing the backfills already in Q4. So it is not only the top dose level. We are exploring multiple dose levels in the backfills. The other thing is the overall dose escalation is limited to the 4 tumor types, as I talked about. But within the backfill, we do have the ability to even limit it further to make sure we have enough patients of specific indications. So we think that will be an important element. The other thing, I think, that's important to recognize is the Hansoh data, the Topo 1 inhibitor, that data was all generated in China. And we think most likely, there was a limitation on prior ADCs, which is a key issue. So we don't know how many of those patients have actually seen a prior ADC. Seagen, because it was in the U.S., we suspect had seen prior ADCs, but they didn't share that data. So I think for us, one of the things we do want to make sure is -- given our hypothesis, is getting some data in patients who have seen at least either Enhertu or Trodelvy.

So in the tumor-specific expansion -- I think the time line is also very interesting, right? So the tumor expansion is expected to begin -- the initial expansion is expected to begin second quarter. And your -- the data is in midyear. So can you talk about the expansion strategy here? You have a sense of where your competitors are going. And you're still kind of in the midst of your backfilling work. So how do you -- confident to pick the low-hanging fruit? And what do you think about the low-hanging fruit? Is there a specific -- let's say, the top 2 solid tumor indications that you're going to shoot for?

One, we've not given any details of exactly where we're going and probably are unlikely to for competitive reasons. But what I think -- we have included these expansion cohorts are already in the protocol. So from a protocol execution point of view, they're there. We just have to simply declare a dose and say, go. So we can get a very quick turnaround once we have our recommended Phase II dose. With regards to tumor types, I mean, if you think about triple-negative breast cancer, for example, high -- very high unmet medical need, high use of Enhertu and Trodelvy, which are places where we potentially differentiate; high levels of B7-H4 expression, et cetera. So I mean just looking at the totality, that's kind of an obvious one. As far as everything else, we'll have to see. And -- but I do want to remind you, we will be looking at the backfill data. And there are signals that may come there as well as -- as other data emerge.

Okay. Turning to 2056. The -- your Phase I work is restarting. How soon can we get a sense of the early efficacy and also safety as well?

That's a good question, Brian. I mean we're just getting back into the clinic, right? So I would say we're -- it's premature for us to guide on that front at this stage. What we're planning to do in 2024 is advanced dose escalation. And over time, we'll share some kind of time line for initial data, but we're not ready to go there just yet.

Okay. Maybe just one last one, just to wrap up. During the presentation today, you have painted a picture that Dolasynthen is a much better platform to move forward with over Dolaflexin. And we would definitely have gone through the highs and lows together with UpRi. And so how should we think about just what we have learned from UpRi? And I think you said that we're going to get an analysis of what happened with UpRi. And what are you specifically looking -- what would we be looking for? And what should we be looking for, frankly, in the UpRi data set?

Well, I think, first, while we had -- there's clearly negative outcomes with -- associated with UpRi. The one thing that is clear is we will have clear data showing with our payload the incidence of neutropenia, neuropathy -- well, lack of neutropenia, neuropathy and ocular toxicity. So the issues that we designed the Dolaflexin to solve for, it's solved for. Unfortunately, we realized in the clinical execution of UpRi that there were consequences of these high-DAR species that, as we alluded to, you end up getting nonspecific uptake in injury endothelial cells. So we discovered a constellation of additional toxicities. We saw elevated AST, we saw fatigue and thrombocytopenia. One of the things we are looking forward to sharing is with Dolasynthen, using the same antibody, you see much lower incidence of these phenomena. So that's where we think there is the opportunity to say that we solved -- we addressed, we believe in preclinical and clinical UpRi, that we -- neutropenia, neuropathy, ocular. But we think Dolasynthen helps with us addressing the UpRi concerns.

Maybe -- and I'll just add to tip our hand a little more. We believe Dolasynthen gets payload to target much more effectively. You'll remember with UpRi in its development, we did see on-target pneumonitis, frankly. Type 2 pneumocytes in the lung expressed NaPi2b, we saw that with UpRi. We also saw it with 1592. But as it relates to payload and platform, we saw marked improvements in safety and tolerability.

I know we're out of time, but that UpRi update is going to be in first half?

First half.

First half and 1592.

And 1592, both first half.

And 1592. Okay. Got it.

Thank you, everyone.

Thank you so much. Thanks for coming.

Thank you. Appreciate the time.