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

Good morning, and welcome to Mersana Therapeutics' Emi-Le data conference call. [Operator Instructions] Please note this event is being recorded. I would now like to turn the conference call over to Jason Fredette, senior Vice President, Investor Relations and Corporate Communications. Please proceed.

Thank you, operator, and good morning, everyone. Just a couple of quick notes before we begin the presentation. First off, we are using slides for this conference call. If you have dialed in, please visit the Investors & Events section of our website to access the webcast and view the slides. Those of you who are already viewing will see on our title slide here that we now have a formal approval of XMT-1660's international nonproprietary name, which is emiltatug ledadotin. Going forward, we will refer to this product candidate by the abbreviation Emi-Le. Now please note that this call will contain forward-looking statements within the meaning of federal securities laws. These statements may include, but are not limited to, those related to the potential clinical benefits of our product candidates and platforms, clinical trial progress and designs, dosing and patient management strategies, addressable market opportunities and anticipated future milestones, including with respect to future data disclosures. More detail on these statements is reflected in this slide. Each of the forward-looking statements is subject to risks and uncertainties that could cause actual results to differ materially from those projected in such statements. These risks and uncertainties are discussed on the slide as well as in our quarterly report on Form 10-Q for the period ended September 30, 2024, filed with the Securities and Exchange Commission on November 13, 2024, and in subsequent SEC filings. Our filings are available at sec.gov and on our website, mersana.com. Except as required by law, we assume no obligation to update forward-looking statements publicly even if new information becomes available in the future. On today's call, we have Mersana's Chief Executive Officer, Dr. Marty Huber; our Chief Development Officer, Mohan Bala; and our Chief Operating Officer and Chief Financial Officer, Brian DeSchuytner. With that, let me turn the call over to Marty to begin the discussion.

Thank you, Jason. We are going to keep the upfront portion of the presentation brief, but it is important to reiterate our corporate focus at the outset. We believe there are significant platform and payload issues that continue to prevent ADCs from realizing their full potential. Here at Mersana, we have been innovating for more than a decade to overcome these challenges and generate novel ADCs designed to meaningfully improve both efficacy and safety. Our Emi-Le data suggests we are on the right path. You may have seen Slide 4 from us before, which frames up some specific challenges the ADC field has been facing. For today, we will focus on the upper 2 challenges specifically. Current anti-tubulin ADCs are effective but have been limited by well-established toxicities like neutropenia, peripheral neuropathy and ocular toxicity. Meanwhile, ADCs using topoisomerase-1 inhibitors are generally limited by severe myelosuppression and sometimes also by ILD. These are platform and payload related toxicities that not only limit efficacy but also effectively prevent the ADCs from being used in combination with other standards of care. Additionally, there is an emerging understanding about topo payload resistance. This has become a significant issue in breast cancer, where topo-1 ADCs are rapidly becoming the standard of care in early lines of treatment. We believe the Emi-Le clinical data we will share today demonstrate for the first time within the broader B7-H4 landscape, clinical activity, specifically in patients who were previously treated with topo-1 ADCs. On Slide 5, you can see that B7-H4 express in a range of solid tumors. As you can see on the left, the highest expression as reported in literature is seen in TNBC, but it also reveals increased expression in patients with hormone receptor-positive breast cancer, ovarian and endometrial cancers. B7-H4 has been clinically validated as a target by competing B7-H4 ADCs. Importantly, these data sets also lack evidence of target-mediated toxicities. Emi-Le, shown on the right, is a B7-H4 ADC developed using our next-generation Dolasynthen platform with site-specific bioconjugation, Mersana's proprietary auristatin payload and a drug-to-antibody ratio of 6. Emi-Le previously received fast track designation from the FDA for patients with advanced metastatic TNBC. And earlier this morning, we announced that Emi-Le also has received fast track designation from the FDA for patients with HER2 low and HER2-negative breast cancer, inclusive of both TNBC and certain hormone receptor-positive breast cancer who have received a prior topo-1 ADC. These are areas of very high unmet need. In fact, the control arm in ASCENT, which you'll recall was the Phase III trial for Trodelvy in late-line TNBC, showed a PFS of only about 7 weeks and an objective response rate of only about 5%. And in terms of market size, global revenues for topo-1 ADCs in relapsed and refractory TNBC are projected to eclipse $1 billion annually starting in 2025. We believe the data we are sharing with you today, which is summarized on Slide 6, delivers on the potential for our Dolasynthen platform and shows that Emi-Le has the potential to be a best-in-class B7-H4 ADC. Emi-Le appears to have a differentiated safety profile with the most common TRAEs reported being transient increases in AST, low-grade nausea and fatigue and generally asymptomatic and reversible proteinuria. Importantly, unlike many other ADCs, we have not seen dose-limiting neutropenia, neuropathy, ocular toxicity, ILD or thrombocytopenia. In fact, as of this data cutoff, no grade 4 or 5 TRAEs were reported. We believe this profile not only could be attractive as a monotherapy, but also could enable combinations with standards of care like platinum chemotherapy and other ADCs that our competitors would be hard-pressed to pursue. We'll get into the different dose ranges that we investigated in just a moment, but we have seen encouraging clinical activity at intermediate doses with Emi-Le. Specifically, we observed a 23% confirmed response rate in very heavily pretreated Phase I population among evaluable patients with high B7-H4 expression across all tumor types. We also have seen this 23% ORR in the subset of evaluable patients with TNBC. Notably, all of these TNBC patients had previously been treated with topo-1 ADCs. These are the first public data we are aware of, showing confirmed responses in TNBC patients who have received prior topo-1 ADCs. These data, combined with a generally well-tolerated safety profile, led us to initiate expansion in patients with post topo TNBC at the upper end of the intermediate dose range at 67.4 milligrams per meter squared every 4 weeks. We also continue to investigate higher doses in our ongoing escalation of backfill cohorts. As we'll soon discuss, while our current confirmed ORR at both intermediate and high doses is similar, the depth of clinical activity at high doses appears to be even greater. In fact, 7 of 9 patients with high B7-H4 tumor expression achieved a reduction in target lesions of at least 30%. That said, we observed treatment delays in this dose range that we believe impacted our ability to confirm responses. We are now working actively to mitigate these delays. So let's move on to the study design and demographics. Key elements of the Emi-Le Phase I trial design are shown on Slide 8. The population includes patients with TNBC, HR-positive breast cancer, endometrial and ovarian cancer, as well as a rare tumor type with a very high unmet need known as ACC-1. We enrolled patients in dose escalation using a Bayesian Optimal Interval design, and a safety review committee that includes 2 external investigators oversaw both the dose escalation steps and the enrollment of backfill cohorts. The primary end points for Phase I, of course, include safety and tolerability and efficacy measures are included among the secondary endpoints. Tumor imaging was conducted every 6 weeks. In escalation and backfill, we have been assessing B7-H4 expression levels retrospectively using tumor proportions scores via an immunohistochemistry assay. On the left side of Slide 9, you can see the methodical approach that we took in dose escalation. We investigated a broad range of doses from about 7 milligrams per meter squared to 115 milligrams per meter squared. We have also investigated 3 different schedules, dosing every 3 weeks shown in light blue, every 4 weeks in dark blue and dosing on day 1 and 8 of a 4-week cycle in purple. This approach has allowed us to gain substantial insights into the relationships between dose, schedule, safety and tolerability and efficacy. To orient you in the slides that follow, we will be showing safety and efficacy in 3 dose ranges that you see identified on the far left here, a sub-therapeutic range below 29-milligram meter squared or less than 1 mg per kg per cycle, an intermediate dose range from 38 to 67 milligrams per meter squared or from about 1 to 2 milligrams per kilogram per cycle and a high dose range above 76 milligrams per meter squared or 2 milligrams per kilogram per cycle. We had enrolled 130 patients in the dose escalation of backfill cohorts as of our data cutoff of December 13. And among them, 5 patients experienced dose-limiting toxicities. Three of these DLTs were transient grade 3 AST elevations, one of which occurred at a dose of 59 milligrams per meter squared every 4 weeks and the other 2 of which occurred at 115 milligrams per meter squared every 4 weeks. These latter DLTs led us to deem 115 milligrams per meter squared as a non-tolerated dose. Additionally, 1 patient experienced reversible Grade 3 proteinuria, accompanied by peripheral edema at 80 milligrams per meter squared every 4 weeks. This event was confounded by a concurrent gout flare. And finally, 1 patient experienced Grade 3 pyrexia, which was self-reported and rapidly resolved on a high dose day 1 and 8 schedule. As noted before, we have identified an initial dose to take into expansion in patients with TNBC, which is 67.4 milligrams per meter squared every 4 weeks. Consistent with the FDA's Project Optimus, we are now working to identify a second higher dose for expansion. The key characteristics of patients enrolled in the trial are shown on Slide 10. The most common tumor types in the dataset are TNBC with 63 patients and hormone receptor-positive breast cancer with 34 patients. The remaining 33 patients had ovarian cancer, endometrial cancer and ACC-1. Across all tumor types, the median age of patients was 55. Overall, this is a very heavily pretreated population. For instance, in TNBC, we enrolled patients with up to 9 prior lines and a median of 4 prior lines. And in hormone receptor-positive breast cancer, we enroll patients with up to 15 prior lines and a median of 7 prior lines. As we noted at the outset, there is a large unmet need, particularly among breast cancer patients who have previously been treated with topo-1 ADCs. We enrolled many of those patients in this trial. As can be seen here, approximately 92% of patients with TNBC had previously been treated with at least 1 topo-1 ADC and 27% had received 2 prior topo-1 ADCs. In HR-positive breast cancer, those figures are approximately 59% and 29%, respectively. Based on our preclinical data, we can hypothesize that B7-H4 expression would influence clinical response. As of the data cutoff, we have determined the B7-H4 status for 103 patients or approximately 79% of the patients in the trial, and we continue to analyze the samples from the rest of the patients who are most recently enrolled. For the purposes of this dataset, we used a tumor proportion score or TPS of 70 as the cutoff for high versus low expression. Based on our TPS data, nearly 45% of enrolled patients were above on this cutoff. It is important to note that this may not be our final TPS cutoff for high expression going forward as we are still gathering data on patients at all levels of expression in both backfill cohorts and in expansion. Additionally, a commercial B7-H4 assay does not currently exist, so we caution against comparing our TPS data to those from other companies using B7-H4 research assays. Now let me turn things over to Mohan, our Chief Development Officer, to talk through the safety and efficacy data.

Thanks, Marty. To begin on Slide 12, we would like to emphasize that Emi-Le was observed to be generally well tolerated with a safety profile that appears differentiated from other approved and clinical stage ADCs. In fact, no grade 4 or grade 5 treatment-related adverse events or TRAEs were reported among the 130 patients as of the data cutoff. We have split the data here into the 3 dose ranges that Marty spoke about earlier: subtherapeutic, intermediate and high. And in the far right column of this table, you can see data for the overall population. Focusing on the far right column, less than 5% of patients experienced a treatment-related serious adverse event, or SAE. Only 3 patients or about 2% of the population had TRAEs leading to dose discontinuation and approximately 9% of patients had TRAEs, leading to dose reduction. Those delays occurred in approximately 12% of all patients. Here, there was a difference across dose ranges, with about 21% of patients in the high dose range experiencing a delay. The most frequent cause of dose delays at high doses was proteinuria. Finally, as noted below the table on the slide, there were only 2 treatment emergent deaths reported. Both of these were deemed by investigators to be unrelated to Emi-Le. One was a case of non-neutropenic sepsis in a patient with a history of inflammatory bowel disease and the other was a respiratory failure related to progressive disease that occurred more than 30 days after the patient's last dose of Emi-Le. Slide 13 contains a summary of the treatment-related adverse events. On the far left, you can see the TRAEs among all 130 patients. In the middle, you see those specifically in the intermediate dose range and on the right are those in the high dose range. We are including on this slide the TRAEs seen in 10% or more of patients in any of these 3 dose groups. The most common TRAE overall was increased AST. The AST increases were transient and reversible and generally, they did not require dose modification. Also, among the most common TRAEs were proteinuria, which was generally asymptomatic and reversible; and nausea and fatigue, virtually all of which were low grade. Notably, you do not see treatment-related alopecia on this slide. This is an AE that can meaningfully impact patient's quality of life in TNBC and other cancers. Before moving on, even if you were to focus only on the AEs on the right side of the slide for the high-dose group, we believe the safety profile compares favorably not only to our clinical stage B7-H4 competitors but also to many of today's approved ADCs. On Slide 15, we show the waterfall plot for all evaluable patients. Patients were defined as evaluable for the efficacy analysis. If they had measurable disease at baseline and had at least one post baseline scan. The figure is broken down into our 3 dose ranges, subtherapeutic on the left, intermediate in the middle and the high dose group on the right. And the bars are colored for B7-H4 expression, with gray being not yet determined, light blue being B7-H4 low and dark blue being B7-H4 high. Visually, you can quickly see here that Emi-Le's observed clinical activity appears to be correlated with both dose and B7-H4 expression. While there are isolated patients with low B7-H4 expression that show tumor reductions, even some that extend below the 30% cutoff for partial responses or PRs, none had led to confirmed responses. That said, we are continuing to evaluate patients regardless of B7-H4 expression. Here on Slide 16 is a summary of the waterfall on the last slide, focusing on the patients with known B7-H4 expression. On the Y axis on the left, you can see the 3 dose ranges. And on the X axis on the bottom, you see the patients who are either below or above the TPS 70 cutoff. There were, of course, no confirmed responses at subtherapeutic doses in the bottom row. And as previously noted, there were no confirmed responses in patients with low B7-H4 expression on the left column. In patients with high B7-H4 expression in the intermediate and high dose ranges, we have observed a confirmed ORR of 22% to 23%. However, you can also see that the percentage of patients who achieved at least a 30% reduction in their target lesions more than doubled from 35% in the intermediate dose range to 78% in the high dose range, suggesting a potential for increased efficacy at higher doses. On Slide 17, we see the waterfall and spider plot for patients with B7-H4 high tumors in the intermediate dose range with the bars in the waterfall and the lines in the spider plot color coded by tumor type. Patients who remained on treatment as of the data cutoff are highlighted with an asterisk in the figure on the left and then arrow on the right. We also have labeled the waterfall for confirmed objective responses. In the spider plot on the right side, you can see TNBC, shown in light blue, appear particularly noteworthy in light of a PFS of approximately 7 weeks for standard of care chemotherapy in this indication, which leads us to Slide 18. Here, you can see the waterfall for all evaluable patients with B7-H4 high TNBC in the intermediate dose range. Importantly, as you can see marked below the waterfall, this is a heavily pretreated population. Also, every single patient had previously been treated with a topo-1 ADC and some of these patients had received more than one prior topo. The B7-H4 high TNBC ORR was 23% in the intermediate dose range. These figures compare with an ORR of only about 5% for the standard of care single-agent chemotherapy. Given the encouraging data we have seen in the intermediate dose range in TNBC, we have initiated expansion in this indication. Importantly, we believe the opportunity exists to demonstrate enhanced efficacy. As we move into expansion, given that we are now investigating the top dose from this range, 67.4 milligram per meter square every 4 weeks, and we are now limiting enrollment in this TNBC cohort to patients with 1 to 4 prior lines. The 67.4 milligram per meter square dose was generally well tolerated across the enrolled patients in all tumor types. And notably, all 4 of the evaluable patients across B7-H4 high tumors who received this dose achieved target lesion reductions and all remained on treatment for durations of approximately 16 weeks or more as of the data cutoff. Now let me turn it back over to Marty to discuss our next development steps.

Thanks, Mohan. Slide 19 shows the design of our first expansion cohort. Again, we are initially limiting enrollment to TNBC patients with up to 4 prior lines of treatment, including at least 1 prior topo-1 ADC. These TNBC patients may have originally been diagnosed with hormone receptor-positive breast cancer, and we are including patients with HER2 low and HER2-negative tumors. Unlike the approach we took in escalation, we are stratifying patients and expansion for B7-H4 expression. We expect to eventually advance at least 2 doses into expansion with the first one being 67.4 milligrams per meter squared. And leveraging the data from this portion of our Phase I, we would expect to select our dose and establish a TPS cutoff for B7-H4 expression for a potential monotherapy pivotal trial. Finally, while we are starting our expansion work in post topo-1 TNBC, we have seen confirmed responses with Emi-Le in all of our enrolled tumor types. So we have ensured that our protocol includes the option to open multiple expansion cohorts. And finally, in terms of data, Slide 20 illustrates why we believe we have the potential for even greater efficacy in the high dose range. You can see from the waterfall on the left that all but 2 patients receiving high doses had reductions greater than 30% in their target lesions. We highlight here 4 patients, including 2 that were unconfirmed partial responses at the time of our data cutoff, who experienced dose delays due to proteinuria and then showed progression by the time of their confirmatory scan. Per protocol, treatment is paused if urine protein is 2 grams per 24 hours or higher, which is within the grade 2 range. The proteinuria we observed was generally asymptomatic and reversible with dose delays. We believe this AE is payload related and it is also seen with some other auristatin ADCs and we are now actively implementing mitigation strategies to reduce proteinuria-associated dose delays as we continue dose exploration and seek to bring a second higher dose into expansion. That concludes our data presentation. We are excited by this initial clinical data snapshot and are pleased to be actively enrolling patients in our first expansion in cohort. Before opening the call to questions, let's just recap Mersana's overall areas of focus for 2025 on Slide 21. In terms of our plans with Emi-Le, we expect to continue enrollment in our first expansion cohort and initiate expansion enrollment at a second higher dose in post topo-1 TNBC. Also, we plan to present additional Phase I dose escalation and backfill cohort clinical data for Emi-Le in 2025. As many of you know, we also are actively dose escalating in a Phase I clinical trial of another product candidate, XMT-2056. This is a STING agonist ADC candidate targeting a novel epitope of HER2 that was developed utilizing our Immunosynthem platform. In 2025, we plan to present initial pharmacodynamic STING activation data from this clinical trial. And finally, we plan to continue our pipeline development work for ourselves and to support our collaborators. The team here at Mersana would like to thank the patients and investigators who have been involved in our Phase I clinical trial of Emi-Le. And with that, let's open the call to your questions.

[Operator Instructions]. The first question comes from Tara Bancroft with TD Cowen.

So I was hoping maybe you could provide more detail on the proteinuria lab requirement and what that entails or proteinuria as an AE in general, since we haven't really seen it with other ADCs? Like I guess what I'm looking for is it cumulative? And if so, what would the longer-term implications be for this? And then second, if you could better define what you mean by dose delay, even in terms of length of time, and what impact that has on relative dose intensity, please?

Thank you, Tara. First of all, I would like to remind you that the dose we're taking forward, 67 Q4, is in the intermediate dose range. And at that range, as you saw in our data, while we did observe proteinuria of any grade in 1/4 of patients but it's much lower, and we didn't see the grade 3 to the same extent as we did the higher. So when we talk about proteinuria, the primary focus of the conversation is for that ability to go to that second higher dose. Now to go into the more direct answer to your question is what we see with this proteinuria is, it is -- we now have results of biopsies. We have urine protein electrophoresis from patients. We have our preclinical data. And what we're seeing is this is primarily an albumin urea. This actually is seen with other ADCs. For example, [ Glenmark ] has proteinuria in their label. One of the reasons it doesn't become an issue or a challenge with some of these others is remember, for most ADCs, you're getting dose-limiting neutropenia or neuropathy with auristatin payloads before you get to the doses. So this is a little bit of a byproduct of the design of our molecule to eliminate these kind of traditional auristatin payloads. With regards to the delay is it does reverse, but essentially it is a matter of while we're not getting into specifics, we're talking weeks is the general concept for reversal and the protein comes back down and then we can restart. What we are looking to do about proteinuria is when we consulted with nephrologists is they identify the way you mitigate this in patients with similar -- with other conditions is you give them ACE inhibitors, you give them ARB inhibitors -- I mean ARBs, not ARB inhibitors. You can look at SGLT2 treatment. And the fundamental goal is to make sure you have tight control of blood pressure, tight control blood sugar and then we will be applying that in our high dose exploratory part going forward.

The next question comes from Charles Zhu with LifeSci Capital.

I guess, also maybe just a follow-up on the proteinuria. How confident are you in your mitigation strategy in reducing proteinuria and maintaining drug pressure? And have you historically implemented these mitigation strategies with some of your historical pipeline drugs like UpRi have also seen some similar signals of this type? And could you provide any incremental color on your experience there, if any?

Well, first of all, with regards to UpRi, while we did observe proteinuria, as you recall, our dose was around 1 milligram per kilogram. So we -- because we had the dose-limiting toxicity associated with the NaPi2b and the ILD, which was target mediated. So similar to like other ADCs, we didn't -- you didn't get to this point because we had another toxicity that basically prevented us from getting there. With regards to the efficacy of these mitigations, therefore, we didn't really test these in the prior regimens. We've been putting these in place very recently. The data you're seeing today, and so all this data is not reflecting any of those interventions. So that will be in a future dataset where we'll be able to measure the effects. As with regards to our confidence in this, the nephrologists are pretty straightforward of what you see in this setting that there is literature for when you have this type of podocyte injury that you can put in place ACE inhibitors, ARBs, SGLT2s are well-established interventions to manage this.

The next question comes from Jonathan Chang with Leerink Partners.

How do you see the Emi-Le program positioned in the B7-H4 competitive landscape? And how do you see these data comparing to other datasets we've seen in the space?

Yes. Jonathan, maybe I'll take that one. We actually view this as potentially best in class. It's a little bit difficult to make an apples-to-apples cross-trial comparison for all the usual reasons, but I'll just note a couple of things. Many of the competitors in clinical development today have topo-based payloads. And we actually believe that those topo ADCs are at a competitive disadvantage in breast cancer. Since Trodelvy and HER2 are already entrenched in that space, and there's increasing evidence of resistance to topo-based payloads for patients who have already had that mechanism. Just case in point, one of the competitors active in this space just reported out data where they had a 0% response rate in patients who had previously seen a topo. So for the clinical competitor that's got a non-topo payload, the data that they've shared with the molecule is quite limited. And so I'll just point to a couple of things. Pfizer's initial data were in patients with a median of 3 prior lines of therapy, we have a median of 4.5 prior lines of therapy. We've shown, I think for the first time, responses in post topo-treated patients, and we actually don't know the prior therapies for the Pfizer ADC because they didn't report that. We also don't know what doses Pfizer is using in the most recent data they presented. And if you recall, they had to abandon the more intensive dose regimen following some Grade 5 events. So we actually believe our tolerability profile is highly differentiated. As of the data cutoff, we don't really see dose-limiting treatment-related neuropathy, neutropenia, ocular toxicity, ILD, thrombocytopenia that might enable us to explore combinations that others might be hard pressed to pursue.

Understood. And maybe just as a follow-up to that, how does the TPS of 70% or higher compare to how other companies are looking at B7-H4 as a biomarker?

Yes. I want to be -- we want to be a little cautious on direct comparisons because everybody is still in the development phase for their assay. And I think what is important though is at our TPS 70 cut point, that's capturing 45% of TNBC patients. So if you -- I think what's probably more relevant is if you look across studies, what percentage of their patients are in a given bucket? That's more likely to be consistent than the exact TPS cut point.

The next question comes from Michael Schmidt with Guggenheim.

Just a few follow-ups. Obviously, great to see association of response with biomarker expression and dose. And so as you think about additional optimization that you're doing, I mean, first of all, I was wondering did you break out response rate at just the 67.4 mgs per square meter dose, which is the expansion dose that was selected? I think there were 15 patients at that particular dose. And could you comment on the 3 responses and TNBC, at what dose levels where these patients treated? And then lastly, any learnings from your evaluation of the different dosing schedules. I know you've looked at every 3 weeks, every 4 weeks and then an alternative twice per, I think, month dosing schedule.

Yes. So I can maybe give a little bit more color to the data. So as we have mentioned, we started seeing responses at the 38-milligram per meter square dose and we've seen activity in the whole range from 38 to 67. We did not break it out by dose because the number of patients at each of the doses is limited and the dose increments were quite small. Having said that, we are very encouraged by the data we are seeing at the 67.4 milligram per meter square dose. As I'd mentioned, all 4 patients who are TPS high who were treated at this dose had a tumor reduction and had a very meaningful duration of treatment with patients staying on treatment for approximately 16 weeks or more as of the data cutoff. So that is the encouraging data that really prompted us to move forward at this dose into expansion.

And then, Michael, with regards to your question on learnings on dose and schedule, at this point in time, as we continue to explore up to 95 milligrams per meter squared for the higher dose, we are still looking at the optimal way to deliver those high dose by a split dose or by an every 4-week or 3-week schedule. So that is still an open question. We will -- but for now, 67 Q4 is our -- is the dose that we are taking forward.

The next question comes from Colleen Kusy with Baird.

Can you talk a little bit more about how you determined the 70% cutoff for the biomarker? And if there's any other data to kind of support why that would be the best cutoff? And then on -- for the dose response, I know we had the proteinuria at the higher doses that kind of impacted the high dose, but do you see a sort of dose response within that intermediate dose level?

I'm going to answer your question on TPS cut, and then I'm going to turn it over to my colleague, Mohan, to handle the dose response question. First of all, for TPS cut point, I would like to remind you and this is from the literature as well before we start editing our data, the B7-H4 expression is somewhat bimodal. So you tend to have higher expressors and you tend to have lower expressors. So to be frank, whether you're at 50 or 70, that is kind of a middle ground that may not be that relevant one way or the other. So what we did for this data cut is there tended to be a natural breakpoint. And I think as you saw our data below 70%, while we did see some activity, we didn't see confirmed responses. So that's probably one of the things we're going to be very careful on, the TPS 70 is the cut that we identified for this dataset. The final TPS score that we would use for a diagnostic, however, will be based on the totality of data, which not only includes this dataset, but we are enrolling patients regardless the biomarker expression in the initial part of our expansion and in the ongoing escalation backfill as we study higher doses. So once we get that full dataset in, then we'll do kind of the traditional receiver operating characteristic curve with our diagnostic partner to define the final cut point.

And to address the dose response question. As we showed in that intermediate dose range, there were very limited numbers of dose interruptions, dose delays, dose reductions, right? So overall, that dose range was very well tolerated. As I mentioned before, it is a little bit difficult to do a formal kind of dose response analysis because of the limited number of patients at each dose. Having said that, the impact of proteinuria in that dose range has been very limited in its ability to confirm, right? And as I've mentioned before, with 67, particularly at the high TPS we've seeing kind of meaningful duration of activity, which has not been limited.

The next question comes from Asthika Goonewardene with Truist Securities.

Just going back to the proteinuria. Is it -- just a quick question here on this. Is it possible to dose down a patient to help manage the proteinuria instead of just pausing therapy? And then even with managing the patient with ACE inhibitors, ARBs, et cetera, do you still anticipate taking a meaningful pause to their treatment with Emi-Le? That's one. And then second, can you talk to us a little bit about next updates from this and the expansion cohort up to 67 mgs per meter square. When should we expect that?

I'm going to let Jason answer the second question first, then I'll take your first question.

Rather not answer that. So we've guided to additional clinical data from dose escalation and backfill this year. We have not yet guided to expansion data. So that guidance likely will come over time, but not today.

And actually thank you for bringing up the comment about the dose reduction. That is, in fact, one of the strategies that we're evaluating is instead of delaying do you maintain dosing but with a dose reduction. So that is part of the overall strategy we are evaluating currently.

[Operator Instructions] The next question comes from Yigal Nochomovitz with Citi.

I guess I want to come back to sort of the basic premise around the technology platform and the Dolasynthen given that you're seeing appreciably better tox profile. So is the explanation that it's not a topo-1? Is the explanation that it's a better linker? Do you have more specific targeting to B7-H4, so obviously less off-target? Is it all of the above? What is the fundamental reason why you're seeing what appears to be a very significant departure from the tox profiles we're all very familiar with in the ADC space?

Part of it was this was the design intent. And in fact, we had shown that with our first-generation molecule already. Unfortunately, we had target-mediated toxicity with NaPi2b. So I think to kind of tease out a little bit, we said, the core tenet is one, we have a much more stable linker. So we do not see just loose systemic exposure. If you think about it with Trodelvy and the topo payload there, you have very rapid release of the topo payload systemically. Ours is what an ADC was intended to be is the payload is delivered to the cell that expresses the antigen. And so if you think about it, that's kind of what we were all trying to achieve with ADCs, where we've kind of gone astray from that and using this kind of pseudo targeted where then you have free payload everywhere. A second element of that as for a statin payload is our controlled bystander effect that once it's internalized, it activates in the cell. But importantly, there is a period when the linker is cleaved and it becomes free payload in the cell. There is a period for which the permeability. So in that tumor cell, it can leak out to adjacent cells before it gets locked in. But I think probably the most important tenet is our goal and our strategy overall is to ensure that the payload is being delivered to cells that have the target and minimize exposure of payloads to nontarget mediator.

Okay. And I may have missed this, but did you make any comments with respect to durability or time on therapy at this point or that's coming later?

We -- I mean, we've been -- as we've previously stated, this initial dataset, especially at the 67 and higher doses has very limited duration of follow-up. So we are not trying to prematurely -- we are not prematurely commenting on durability, PFS and those types of endpoints, that will be at a future update.

The next question comes from Brian Cheng with JPMorgan.

Maybe just one from us. Can you comment on the correlation of the timing of the proteinuria cases and the payload exposure or any details of the payload in circulation?

At this point in time, we're not going into PK data from this dataset. I do think one of the things that we have seen previously, we presented with our molecules, is the exposure to free payload with our platform is actually very, very limited. But as far as a direct answer with this molecule and is there any correlation with payload exposure, we don't have that data in this presentation.

The next question comes from Justin Zelin with BTIG.

I wanted to ask about the ALT/AST elevations. Were these purely biochemical adverse events or did you see any evidence of clinical liver injury and do you believe that this AE is also payload related?

Yes. So maybe I'll start and then Marty can comment. So to date, we've not seen any [indiscernible] cases, so no cases of drug-induced liver injury, right? So at this point, these are primarily isolated lab abnormalities.

And to build on that, while the scaffold linker is different, the payload is the same payload and in UpRi with over 600 patients, we had seen a grade 3 AST elevations in up to almost half of the patients. We're seeing a much lower incidence of that here. And despite that much more frequent AST elevation in the UpRi experience, once again, 600 patients, we didn't see -- we don't see hepatic failure. So I do think it's a transient phenomena, the exact pathophysiology we don't know today, but we don't see evidence that this is leading into meaningful liver injury.

Understood. And maybe just as a follow-up, the focus here has obviously been on breast cancer. Can you talk to your plans with Emi-Le in endometrial and ovarian?

Sure. We have not given any specific details other than, one, we would like to note, we did see responses in both of those indications in the datasets. And if you play with adding up the bars, you can kind of guess that for endometrial. You'll be pleased with that. I do think we are looking at TNBC post topo as a focus today is because that is the highest unmet medical need. We're clearly differentiated space and we can go very fast there. When we think about endometrial and ovarian, our key point of differentiation, because the topo after topo was less of an issue there, will be our ability, because of our differentiated safety profile, to combine with other agents such as platinum chemotherapy. So what we want to think about and we haven't made a decision, but if you kind of step back and go, well, there's a whole bunch of people fighting with each other now in platinum-resistant ovarian cancer or how many people can actually go into platinum-sensitive ovarian cancer and combine with carbo. And of all the B7-H4s, we're the only one that has a profile that would allow us to do something like that. So one of the things we're going through the discussion thinking on is you -- do you go to the traditional work your way up to the bottom or you just skip to the head of the class?

The next question comes from David Nierengarten with Wedbush Securities. Please go ahead.

Just a quick one on the higher dose patients and the unconfirmed responses and you note that 2 of the patients progressed on their confirmatory scans. So I assume the treatment delay was, was it at least 1 cycle, was it more than 1 cycle? And I just wanted to check on that. And then on the intermediate-dose patients, how many are with the tumor reductions, are they all still on study? Is that correct?

Say that second question again, they're all?

Still on study?

On the first question about treatment delays in the high-dose patients. As Marty had mentioned, these patients had meaningful dose delays, right, in the range of weeks, after their kind of initial scan before their confirmatory scan. So the delays were kind of, of a meaningful amount. In terms of the ongoing patients, I think they are marked on the slide with an asterisk if the patient starts to -- on treatment.

Maybe my screen is too small. I can't see that.

Well, there's also a PDF that's up on the website, maybe the webcast portal viewer may hide those a little more. But you should be able to see them in the PDF on our website.

And you'll also -- when you see for the intermediate dose, the spiders will have an arrow on the ongoing events.

This concludes our question-and-answer session. I would like to turn the conference back over to the CEO, Marty Huber, for any closing remarks.

We'd like to thank everybody for joining us today, and we will be heading off now to San Francisco. So we look forward to seeing many of you at JPM next week. So thank you.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.