Pyxis Oncology, Inc. (PYXS) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by. Welcome to Pyxis Oncology MICVO Clinical Update Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to turn the conference over to Alex Kane, Senior Vice President of Investor Relations and Capital Markets. Please go ahead.

Hi, and good morning, everyone. My name is Alex Kane, and I'm the Senior Vice President of Investor Relations and Capital Markets at Pyxis Oncology. Thank you all for joining us on the phone and webcast today to discuss the MICVO recurrent metastatic head and neck squamous cell carcinoma clinical update. Our President, CEO and CMO, Lara Sullivan, will review prepared remarks, followed by a Q&A session that will include Lara and additional members of the team. We will be making certain forward-looking statements throughout today's discussion. I would recommend that you read the SEC filings, including our forward-looking statement disclosure and related risk factors in our recent Form 10-Q. With that, I will pass the call over to Lara to review today's presentation and data. Lara?

Great. And thank you, Alex. As Alex noted, we appreciate each of you taking the time today to be here with us virtually to hear more about the MICVO clinical update for recurrent/metastatic head and neck squamous cell carcinoma. We are excited to share with you the preliminary data out of both the monotherapy and combination studies that we have been working hard on throughout the year. Please note, on Slide 2, we will be making certain forward-looking statements today. We refer you to our SEC filings for a discussion of the risks that may cause actual results to differ from the forward-looking statements. We undertake no obligation to update these forward-looking statements, except as required by law. Without further ado, if you turn to Slide 3, you will see it's been an exciting year for Pyxis Oncology. I'm pleased to share this preliminary data update and the progress for MICVO in recurrent/metastatic head and neck squamous cell carcinoma, both in monotherapy and in combination with KEYTRUDA or pembro. Starting with the monotherapy data, we delivered a 46% confirmed objective response rate and 92% disease control rate in an n of 13 at the 5.4 mg per kg dose across our ongoing Phase I study. This data is very exciting as it shows how powerful MICVO is right from the very first dose, generating rapid and deep tumor regression with responders showing initial durability to date. The picture of broad tumor control, either through response or stable disease, is an important one that may be a positive predictor for durability metrics, such as overall survival and progression-free survival as we proceed in clinical development. Moving on to our study in combination with pembro, which is in the dose escalation phase, we are showing data today in 7 first and second-line plus patients at 3.6 and 4.4 mg per kg. This ongoing combination study has cleared the 3.6 and 4.4 mg per kg doses, and we are currently testing the 5.4 mg per kg dose right now. To date, we have observed a 71% confirmed ORR, and we see a 100% disease control rate, which we believe shows best-in-class potential for combination therapies in head and neck. In addition, we have FDA alignment on the trial design for a planned monotherapy pivotal study in second-line plus head and neck. Given the exciting signal in monotherapy and combination therapy, respectively, we are currently evaluating the path forward for potential pivotal studies for each program in head and neck, and we'll provide additional detail next year as the data continues to mature. Before we dive into these exciting clinical updates in depth, we want to provide a bit of background on MICVO, a first-in concept extracellular targeting ADC. On Slide 4, you can see that MICVO targets EDB+FNs or extracellular domain B, a novel noncellular ADC target. Fibronectin is a noncellular component of the extracellular matrix that provides structure and shape to the tumor and acts as a scaffold to keep the tumor functioning while supporting the tumor's ability to grow. EDB+FN is a splice variant of fibronectin associated with tumor growth and metastasis that is seen in tumors, but not in normal tissue. EDB+FN is expressed across a variety of tumor types. And as you can see, there are high levels of expression in head and neck, in particular with negligible expression in the companion normal tissue. Through recent translational learnings, we have observed that target expression is necessary but not sufficient to engender clinical response, meaning there are other factors at play in the extracellular environment that influence the degree that a tumor will be sensitive to MICVO. For example, the pH level in the tumor microenvironment and the presence of extracellular proteases, the cathepsin family in particular, is very important in the tumor matrix, and this helps influence tumor sensitivity or responsiveness to MICVO. Additionally, the actual architecture of the strong environment, meaning the geometry of the noncellular elements in the extracellular matrix, influences a tumor's ability to respond to therapies. We have observed with AI-enabled digital pathology tools at hyperresolution differences in the noncellular elements and how they are organized in different tumor types. For example, in head and neck, we have seen the fibronectin sheet in the tumor cells organized very neatly, regularly and linearly, which allows the payload to traverse without an impediment through that extracellular matrix. In the case of other tumor types such as pancreatic, we have seen those fibronectin elements tangled, jumbled, almost acting like a mechanical barrier, preventing the payload or the drug from traversing. So these are other components besides just the target expression that ultimately can influence a tumor's susceptibility to a therapeutic agent, such as MICVO. On Slide 6, in terms of the ADC construct itself, MICVO was purpose-designed by Pfizer, and we licensed it at the development candidate stage. MICVO has site-specific extracellular protease-cleavable Val-Cit linkers and the predictable DAR of 4. These components both lend themselves to the stability of the ADC as well as the cleavability of the ADC. The payloads have been optimized for potency and permeability and are important in helping support the efficacy potential of this ADC for killing tumors both directly and through the bystander effect. Putting it all together on Slide 7. The mechanism of action that we see here is focused around three key prongs. First, when the payload cleaves in the tumor extracellular matrix, it diffuses directly into the tumor cells and begins killing them. And then as those tumors light, that triggers the bystander effect, which creates a repetitive sequence of events where the killing continues. Finally, we have the immunogenic cell death mediated component of our mechanism noted here as the third prong. This means that the payload is activating the T cells in the tumor microenvironment and helping them to infiltrate the tumor. These components of the mechanism that we just described are essential elements that allow this ADC to uniquely combine with a variety of agents, in particular, checkpoint inhibitors, where we can amp up this third prong of the mechanism and really use the immune systems to help augment the potency of the killing that the payload has mediated. We've got this three-pronged mechanism at work in the dose escalation study last year. Slide 8 is a reminder of how strong the signal was. This was the data in monotherapy in head and neck shared last November and updated this past February across all doses. We saw responses at 3.6 and 5.4 mg per kg, and we saw a 50% confirmed ORR in this efficacious dose range in the head and neck squamous cell population. We took the 5.4 mg per kg dose into expansion due to its strong efficacy signal and tolerability profile. Notably, our dose escalation study was a basket study, in which we saw responses in head and neck and multiple other tumor types. So strategically, we decided to layer on both market and competitive considerations as well. This market analysis, combined with the clinical data, you're seeing in front of you, is what got us really excited about the opportunity to support patients in head and neck cancer. The head and neck market is a large growing and relatively uncrowded market that is ripe for new therapies. As you can see on Slide 9, it is the sixth largest market in oncology, and there are only a handful of companies with drugs in later-stage clinical development, and those competitors have primarily come into the space in the past 2 years. So innovation is happening right now in this space after years of inactivity, and we're also seeing enhanced business development activity. There are a lot of exciting things going on for patients in head and neck as well as ample space for new entrants and new mechanisms to come in. As these new entrants continue through later-stage development, we expect that the market will reshape, but it's important to ground ourselves first in what the market looks like today as a starting point. Slide 10 is a simplified depiction of the current head and neck treatment paradigm. It is a pretty simple treatment algorithm. Depending on CPS status, a patient will either get a chemotherapeutic agent, a checkpoint inhibitor or Cetuximab, and their CPS status will dictate the order in which they tend to receive these therapies. This is a U.S.-based depiction that we have represented here. Practices globally are pretty similar, with the exception that certain regions in the world bringing Cetuximab into earlier lines of therapy, but otherwise, these are the choices that patients have today, and there are not many alternatives for them. With the expected introduction of next-gen EGFR therapies into the head and neck market in the coming years, we anticipate more nuanced segments emerging based upon CPS status and HPV status. What you're looking at on Slide 11 is the projection of the U.S. head and neck market emerging in 2029, where we see next-gen EGFRs in combination with KEYTRUDA, primarily focused in the HPV-negative or HPV-not-applicable population with CPS greater than or equal to 1. This means that in the frontline setting, the patients on the top left that are HPV-positive or with CPS less than 1 will primarily remain unserved or underserved. Importantly, this shift in the market dynamic also leaves a massive vacuum for patients that progress into the second and third-line settings as the EGFRs will not be used serially or in sequence, meaning prescribers will need to select one of their preferred next-gen EGFRs and then move on to other mechanisms. As there will be no established alternatives in the post next-gen EGFR setting, we see the second and third-line setting as a huge opportunity for new mechanisms and treatments such as MICVO. And in fact, as you can see on Slide 12, the second and third line market is nearly as large as it is for the next-gen EGFRs in the front line. So there is ample space for new competitors to create value for patients. Additionally, in the frontline setting, the HPV-positive patients and the low CPS patients scoring less than 1% also represents an exciting frontline opportunity that is unlikely to be served by the next-gen EGFRs based upon the clinical data that they have put out to date, which has been focused in the HPV-negative or HPV status not applicable setting. The combination of our clinical data from the dose escalation study last November as well as our nuanced understanding of the evolution of the market in head and neck has informed the plans that we have in place in our program this year, which you can see on Slide 13. We have two studies underway, our monotherapy study and our combo study with pembro. In the case of our monotherapy study, we have focused on the 5.4 mg per kg dose. We've taken second-line-plus patients and we've assigned them into two arms. Arm 1 contains the platinum and PD-1 experienced patients and Arm 2 contains the EGFR and PD-1 experienced patients. Arm 1 is the data set that is most comparable to Merus, J&J, Corbus and other competitors who've entered the market. Arm 2 is a new data set that has not been shown by other companies in clinical development and is being shown for the first time by Pyxis today. The patients Pyxis has enrolled in both of these arms have been a bit more heavily pretreated than the patients enrolled in the competitors' data set, and Arm 2 is more heavily pretreated than Arm 1. This is because today's EGFR therapies, typically Cetuximab, are used in later line settings. In the case of the combination study, we are still in the dose escalation phase. Our combo study is a basket study design across multiple tumor types for the escalation component, but we are focused on recruiting head and neck patients for backfill. The 3.6 and the 4.4 mg per kg dose levels have cleared, and that's the data we are showing you today in the head and neck population. We are currently enrolling at the 5.4 mg per kg dose level and dose escalation. I want to point out that this time line in the middle of the slide indicates that just last November, we had put out our monotherapy dose escalation data across multiple tumor types, and here we are, 12 months later, with 3 preliminary data sets in head and neck across both the two arms of the monotherapy study and the emerging preliminary escalation data in combination with pembro. Additionally, we have accomplished a tremendous amount operationally, moving from a generalist Phase I clinical trial site population to a head and neck specialist clinical trial site population. As a result, there's a hockey stick effect of activating these sites over this time frame, and you will see that reflected in the fully enrolled data sets to be shared next year. We thought it was important to show today's preliminary data set to give you an idea of why we are so excited about MICVO's potential in head and neck. The data we are about to review provide validation for both the mechanism of action and the initial clinical data that we had shared last November. Slide 14 displays a summary of the monotherapy data in Phase I. In short, we observed a confirmed response rate of 46%, or the 13 patients dosed at 5.4 mg per kg and an impressive disease control rate of 92%. In the case of Arm 1, which is the post-PD-1, post-platinum population, we see a 60% confirmed ORR. As a reminder, this is the population that is most directly comparable to Merus, J&J and other competitors, with the Pyxis data set showing an even more diverse patient population across HPV status and a more heavily pretreated patient population overall compared to those competitors. This is also the data that recapitulates our dose escalation data from last November and gives us a high degree of confidence in the competitive efficacy profile of MICVO. In the case of Arm 2, this is the post-PD-1, post-EGFR population. As noted previously, these patients in this preliminary data set are post-Cetuximab and have been more heavily pretreated on average than the Arm 1 patients or the competitive data sets that are out there. We are recruiting, post-Merus, post-Bicara and post-J&J next-gen EGFR patients into the Arm 2 study and expect to have some representation of these next-gen EGFR patients included in our fully enrolled data set to be shared next year. We felt it was important to bring forward the data set in this Arm 2 to highlight MICVO's ability to demonstrate response across any form of prior therapy and any level of prior treatment status. We do recognize that our mechanism and our tumor biology would not suggest any difference in response rate depending upon specific prior treatments. Efficacy expectations in the heavily pretreated late-line post-Cetuximab population are driven by PI benchmarks, who are looking for a 20% minimum efficacy bar to establish an opportunity for an additional line of later-line therapy. It's early days with the low N of 4 patients in our Arm 2, but our confirmed ORR of 25% in this late-line, heavily treated post-Cetuximab patients population clears that PI-driven benchmark. The patient demographics on Slide 15 reflects a few trends. The initial demographics of patients in our preliminary data set shared today are more heavily weighted towards HPV-positive patients. This is because we are in the midst of migrating from generalist Phase I clinical trial sites to more expert head and neck clinical trial sites. At these more expert head and neck clinical trial sites, we will have access to a more broad, diverse array of HPV status patients. Given the recent operational progress and push towards global expansion of our clinical trial sites, we expect to see a more balanced demographic profile going forward. For now, what you see in the upper right is amongst the patients with oropharyngeal tumors, which is the only location for which HPV status is relevant, 7 of the 9 are HPV positive. The other 3 anatomical locations, oral cavity, larynx and hypopharynx, do not measure or evaluate HPV status and therefore are considered HPV status not applicable. Of note, in terms of prior treatment regimens, 2/3 of patients across our Phase I monotherapy study have had prior taxane therapy. This is really important because we are not seeing resistance conferred from prior taxane use in our patient population. Additionally, 100% of our patients have also been on prior platinum and prior checkpoint inhibitors. Slide 16 shows the responses of the efficacy evaluable patients at the 5.4 mg per kg dose level across our Phase I study, including 4 patients from dose escalation and 9 patients from dose expansion. As a reminder, we have observed a confirmed response rate of 46% amongst these 13 patients and a 92% disease control rate. You can see across the bottom of the slide, the identification of each patient by Arm or escalation phase as well as anatomical location and HPV status. We are pleased to see responses regardless of HPV status and regardless of number of prior therapies. I would also draw your attention to the fact that we observed responders across a variety of tumor sizes, including tumors greater than 100 millimeters. At the bottom, we have identified the number of prior therapies that these patients have been exposed to and it presents a very interesting and important data point. If you were to draw a line directly down the middle of the waterfall, bisecting it into responders on the right and disease control patients on the left, you can see that there are significantly fewer prior treatments on average among responders relative to the patients with disease control, roughly 2.2 prior treatments for the responders on the right and roughly 4 prior treatments for the disease control patients on the left. While we continue to see responses to MICVO regardless of the nature or specific prior treatment, including responses in patients previously exposed to taxanes and EGFRs, we do see a picture more of disease control when we enroll sicker patients with more prior rounds of therapies. The profile of the one patient at the very far left is an anomaly in that this is the only head and neck patient we have treated with MICVO who had early disease progression. Said differently, we have had 100% disease control across all head and neck patients treated with the exception of this one patient. So it was clear to us that there was something unique going on worth looking into. As we dug into this patient's case, we discovered that this patient has a rare subtype of squamous cell carcinoma called verrucous subtype. These patients tend to be surgically treated and are typically not enrolled in clinical studies as they tend to be chemotherapy resistant. This patient enrolled because they qualified for the study, but the standard of practice is typically not to put them in a trial like this. And this is something that we will take into account going forward as we continue to evaluate our inclusion and exclusion criteria. On Slide 17, we can see the swimmer and the spider plots. We look forward to sharing initial durability data in 2026, but in the meantime, we are encouraged with the early signals seen on these two plots. Starting with the swimmers plot, responses have generally occurred early during treatment and the very low rate of early progressive disease as a function of the impressive disease control rate we've seen in head and neck cancer bodes well for the potential of our upcoming durability data. We're in the middle of the hockey stick of enrollment that I referenced earlier. You can see this on the plot with the 4 patients who are dosed and ongoing at the bottom, who have not yet received a post-baseline scan as of the November 3 data cutoff. Of note, on the tail end of the swimmers lanes, we have seen what appears to be a unique profile for MICVO in patients who have discontinued treatment and either maintained response or stable disease following discontinuation for an extended period of time. This could be a result of the mechanism of action and, in particular, the third prong of the mechanism of action highlighted earlier, where we drive immunogenic cell death through the activation of the immune system. Turning to the spider plot. We can see a very clear demarcation between the patients who are responding in green with rapid and sustained tumor responses and the stable disease population in yellow. You can see the outlier patient with the verrucous subtype in the gray and the complete responder patients from dose escalation noted in the blue. Of note, the complete response patient had to drop off therapy for logistical reasons related to their life circumstances, but they continue to hold their complete response for 6 months with no therapy. This is another anecdote that provides us with confidence in the emerging durability profile of MICVO, which we will validate as we continue to move the program forward in 2026. On Slide 18, we've laid out the key aspects of the safety profile of MICVO in head and neck at 5.4 mg per kg. Our philosophy as a company is to dose right up to the maximum tolerated dose in order to maximize potential efficacy and better understand the tolerability profile of MICVO. The second part of our philosophy is to then tune the AE profile of the agent. As we sit here today with a much more thorough understanding of the benefit/risk profile of MICVO, we have a clear sense of how to risk mitigate and manage these AEs moving forward. The data on Slide 18 does not yet incorporate these risk mitigation and patient management practices. MICVO is a very potent ADC. And when we dose a patient with 5.4 mg per kg for 1 or 2 cycles, we are triggering an incredible tumor response very, very quickly. Patients don't necessarily need the same level of payload exposure for months and months at a time to continue to benefit from the potency. Therefore, dose reductions and dose delays can be used strategically to reduce the overall drug exposure for patients. So if we can create an approach to managing the side effect profile, where we reduced the overall exposure of the drug to patients who are more susceptible for developing these AEs, we are confident we will be able to land the optimized benefit-risk profile. We can do that through potential dose reductions and dose delays as well as other alternative dosing schema, which I'll talk about in a moment. On the table at the bottom of the slide, you can see the ADC payload treatment-related AEs of interest. We have no Grade 4s and no Grade 5 events. We have seen some cases of Grade 3 neuropathy and neutropenia as well as one case of ocular tox and one case of pneumonitis. In the case of pneumonitis, this was a patient that was particularly overweight. The point around weight is a very important one, and I want to spend a minute here. What we have learned as a field is that ADC safety profiles have a lot of sensitivity to patient body weight relative to ideal body weight. This is because ADCs have historically been dosed as milligram per kilogram. So a heavier body weight patient of the same height compared to a normal body weight patient will receive more absolute drug as part of their dosing regimen. We have seen in our data set -- our preliminary data set presented today a higher number of heavier body weight patients. As such, we are migrating our dosing approach from a milligram per kilogram way of dosing to an adjusted ideal body weight means of dosing. This will allow our patients to get the optimized dose for their body weight and will reduce their overall risk of AEs that are correlated with higher body weights. The adjusted ideal body weight approach is another lever that we are pulling to mitigate the AE profile while maintaining MICVO activity moving forward. This is in addition to the opportunities to provide dosing reductions or dosing holidays or delays, as already noted. We have provided additional information that gives us confidence in this approach in the appendix slide. Also for the individual AEs of interest, there are prophylactic measures we can consider employing moving forward for each of them. For example, we could consider prophylactic eye drops for ocular AE. All of these are things that we are actively working on implementing right now, and we are confident that while MICVO has a manageable safety profile, we believe it can be improved upon through the risk mitigation and risk management measures. Moving on to the combination data. As a reminder, on Slide 19, our mechanism of action has an immune-driven component in addition to the direct payload-mediated effect. This is what has us so excited about the opportunity to combine MICVO with pembro. As noted, MICVO initiates a strong and powerful effect very early on in its initial cycle, while pembro has a profile of tumor control that is longer-term in nature. So the one-two punch of MICVO and pembro may represent a very powerful combination therapy for patients. You can see the patient demographics in the combination study here on Slide 20. Just as with monotherapy, we have had a focus on U.S.-based sites early in this clinical trial and are now expanding our global clinical trial site footprint. All 7 combination patients in the data we are showing today are HPV positive. We expect a more balanced enrollment of HPV patients moving forward as these additional global sites come online. All of the patients in this preliminary data set have had prior exposure to platinum, several have had prior exposure to checkpoint inhibitors, and 2 out of the 7 have had prior exposure to taxane. What I'm showing you on Slide 21 in our preliminary combination data set is an emerging best-in-class profile for our agents combined with pembro in the frontline-plus head and neck setting. These are early days, but we are excited about this data with a 71% confirmed response rate and 100% absolute disease control. We see these responses in patients across CPS status and a mixture of first and second-line-plus patients. Five out of seven of these patients are still ongoing. The two that are no longer ongoing did not discontinue for safety or efficacy reasons. One of them was found to have an exclusionary criteria after they were dosed and the other patient withdrew consent for unknown reasons. What we're seeing here is a picture of MICVO's agility in being able to support patients at different lines of therapy with different prior treatment regimens and different CPS status. On Slide 22, you can see we have included the swimmer and spider plots for combination therapy. Again, these are early days with preliminary data, but the swimmers are continuing to swim with no evidence of disease progression. And on the right side, the spider plot is going down and sustaining exactly in the direction we want to see. It's a very exciting profile that is continuing to mature, and we look forward to providing more robust data next year. The combination safety table on Slide 23 presents a rather exquisite picture. This is at the 3.6 and 4.4 mg per kg doses in combination with 200 milligrams of pembro. As a reminder, in our monotherapy dose escalation, we had previously identified an efficacious dose range for MICVO from 3.6 through 5.4 mg per kg, and we were very deliberate in dosing in our monotherapy expansion at the top range of that efficacious dose range. Here, we're starting at the bottom of that range, and we're seeing in combo with pembro a beautiful efficacy and safety profile. We have no Grade 3, Grade 4 or Grade 5 events. Rather, all of the AEs that we have seen to date are Grade 1 and 2. It is a very manageable safety profile to date in combination with pembro. On Slide 24, we've summarized the points coming out of the combo data. We see a best-in-class response rate and a disease control rate profile of 100%. Similar to what we saw in monotherapy, this efficacy profile should bode well for durability, longer-term survival and progression-free survival. There is considerable potential in the frontline opportunity for these underserved patient populations, and we're very excited to get the HPV-negative and HPV status not applicable patients enrolled going forward. Putting this all together on Slide 25. As we reflect on the marketplace and as represented in the upper right of the frontline setting, we expect the standard of care to migrate to the next-gen EGFRs in combo with KEYTRUDA in the setting of HPV negative or HPV not applicable patients with CPS status greater than or equal to 1. This movement leaves a number of underserved patient segments in the evolving standard of care landscape. The combo data we just reviewed for MICVO demonstrated clear efficacy across CPS status and in an HPV-positive population, which gives us a clear opportunity to support patients in the upper left segment of the frontline setting. Keeping in mind, we do not expect the next-gen EGFRs to ultimately play a meaningful role in that segment. In the case of our monotherapy MICVO data, we see a significant opportunity in the second and third line settings where there is a vacuum of treatment options that leaves patients with very few alternatives. MICVO as a non-internalizing ADC with an extracellular target can play behind these next-gen EGFR agents can ultimately compete with these agents or even combine with them besides combining just with pembro. In short, there is near-term opportunity for MICVO in the upper left in the front-line setting and in the second and third-line settings. We may ultimately see opportunity across the entirety of this funnel as we continue to generate data in the HPV negative or HPV not applicable segments over time. On the final slide, you will see that we have a very busy year ahead in terms of clinical updates. We plan to provide mature data from the monotherapy dose expansion study in second-line-plus in the middle of next year. In the second half of '26, you should expect additional data from the combo dose escalation study, including a more robust head and neck data set as well as data including the escalation from other tumor types that have participated. Thank you for listening, and we will now take any questions or comments.

[Operator Instructions] And our first question comes from Brad Canino with Guggenheim.

Thank you for providing all the detail. It's great to see the validation of that head and neck signal. First question for me on the post-EGFR efficacy, it looks like 2 of the 3 second-line-plus KEYTRUDA combo patients have higher EGFR, but I don't see that noted on that on the water. Do you know what the response profile was in those patients?

Thanks for the question, Brad. Just taking a quick look at the waterfall. As -- you're right, we didn't disclose which ones were the prior EGFR-specific exposed patients. Yes, we just haven't disclosed that additional detail. We haven't seen a meaningful appreciable difference in the case of the combo program based on the specific prior treatment regimens that patients have received. We decided to call out the prior IO therapy specifically because sometimes people could assume a concern of resistance to prior IO might mean that they're unlikely to respond here, and we actually didn't see that being the case. So that was why we chose to spotlight the prior IO.

Okay. I think with the small end for the monotherapy post-EGFR, it might be helpful. So if we could get that detail over time, that would be great. Second question on the mid-'26 data update, I think we're all keen to know if the adjusted ideal body weight dosing can reduce the discontinuation rate due to AEs. Do you expect to have a handful of responders at this new dose schema, at that update with sufficient follow-up to start to show that?

Yes. Thank you for the question. We're very excited and enthusiastic to get this body weight dosing adjustment in place and to see the impact in the data set because we've seen through observation of other ADC programs in the field who have implemented this a meaningful reduction in their more severe AEs. So we're very optimistic on that. It takes a little bit of time because, obviously, those kinds of amendments to protocols have a time frame associated with them as well as IRB approvals. So we are handling this in two kind of parallel processes. One is going through the actual protocol amendments and IRB approvals. And in the meantime, we're working at the patient level, patient by patient with each individual PI, to assess the body weight profile of the individual patient and prospectively implement either reductions, delays or dosing caps. And that will essentially achieve the same effect because we can do the math of what the ideal body -- adjusted body weight would be, since that's a standard formula in the field. And having that patient-level information, we can do that calculation and then we can get at that same drug exposure through these other tools. So ultimately, everything will be kind of tied up from a regulatory perspective at sort of the pace that those things evolve at, but from an operational pace, we're tackling it right now. So we do expect that the data that would come out middle of next year should reflect the implementation of these measures, whether it fully reflected across the entirety of the data set or is the dominant component, we'll see depending on the actual numbers of patients, for whom this needs to be applied. The one other comment I would make, the adjusted ideal body weight has an additional benefit because it helps also improve the drug exposure of the underweight patients. So we've been talking about the AE profile of the overweight patients. But when you dose with a straight mg per kg and you have patients with cachexia or significant underweight, we may not have been providing them enough drug. So this will also help -- this type of calculation will also help ensure that our lower body weight patients are sufficiently treated as well.

And our next question comes from Farzin Haque with Jefferies.

We noted tumor control after treatment discontinuation, implying like immunogenic cell death on the payload release. So how many patients -- you mentioned one, they had sustained responses post-dose hold or discontinuation. And what's the time frame from the last dose of the best response in those?

Farzin, it was a little difficult to hear. But just to confirm, the question is about patients where we have disclosed response post-treatment discontinuation. Is that accurate?

Yes.

I'm sorry, I couldn't hear what the specific question was.

The question is more on the discontinuation. Like how many patients basically have sustained responses post the dose hold or discontinuation?

Okay. Got it. So thank you for the question. We have not annotated the swimmer plot for every particular patient case. We have shared some particular patient anecdotes regarding the continued tumor control or tumor regression that we've seen in particular patient cases. We will be providing as part of the more comprehensive data update in the middle of next year, a window into both the more emergent durability profile that the more emergent data set will show as well as more explicit commentary patient by patient around the post-treatment tumor control that we've observed continuing. The one patient -- yes. The one patient anecdote that we had shared previously in escalation was related to that confirmed response patient who had the logistics of moving to another continent and needed to stop treatment for that reason, where we were able to very clearly and easily keep track of that patient's change in logistical circumstances. And the tumor control of 6 months was clearly documented. We are working on all of that across the patient sets, and we'll share the updates on that in the next data disclosure.

Got it. And then for the combo cohort was really great with no Grade 3 AEs. So can you provide any context for impact of body weight there? And are you planning to use the same AIBW in the combo setting as well?

Yes. So I think it's really helpful, I think, to look at both the combo safety table and the mono safety table in conjunction with one another. So if we step back for a moment and think about the dosing levels, the 3.6 mg per kg dose level is 50%, 5-0, lower than the 5.4 mg per kg dose level. So the Grade 3 AEs in monotherapy that we saw at the overweight patients were patients who were overweight by about 10% to 20% above their ideal body weight. So at the 3.6% level, where the overall drug exposure is already 50% lower, even when we have overweight patients, they're at the beginning of the efficacious dose curve and the tolerability right range. So we're not seeing as pronounced an impact of the over body weight at the lower dose levels of 3.6 and 4.4. So I think that's why we didn't see any Grade 3s at 3.6 or 4.4. So somewhere between kind of 4.4 to 5.4, we really start to see the sensitivity to every basis point of increased body weight. We are, as a practice, planning to implement adjusted ideal body weight globally for any patient who received MICVO in any clinical trial. We think it's important to do that for a number of reasons. One, we think that is actually the best way to be dosing ADCs. And again, we have looked at many of the metaanalysis of other ADC programs who've done this. Elahere is a great example. They migrated in their Phase I dose escalation from mg-per-kg dosing to adjusted ideal body weight, and their Grade 3 ocular toxins went down essentially to 0, and now their label has an adjusted ideal body weight. So we know that, that works while holding efficacy. We also think, as I mentioned earlier, that it's important that any light patients or lower body weight patients get their full drug exposure. We are testing the 5.4 mg per kg dose in combo with pembro. So we've cleared the 3.6 and 4.4 doses, and that's the data we shared today. We're in progress right now of dosing the 5.4. So we are applying the adjusted ideal body weight, the same regulatory and IRB processes that I mentioned for mono are happening in combo as well as the same proactive dose reduction, dose delay, dosing holidays that we could use at the patient-by-patient level. So we are going to apply that at all 3 dose levels. I do think that the 3.6 and 4.4 doses in combo are very, very interesting because it's showing the potency of the agent and the importance of the mechanism, particularly the immunogenic cell death component, how our mechanism synergizes with pembro even at lower doses to have such a pronounced efficacy effect with minimal safety issues. So we really think that the adjusted ideal body weight is a good practice overall, and we'll only serve patients going forward regardless of dose level.

And our next question will come from Sam Slutsky with LifeSci Capital.

Just kind of two quick ones. You mentioned that you're entering the hockey stick phase of enrollment. Any expectations you can provide though for how many patients we might get in the 2026 updates? And then just on the potential second-line-plus registrational study, what does that look like as you think about the control arm? And then when are you hoping that might start?

Yes. So our intent is to share data in the middle of next year of the fully enrolled monotherapy study, both arms. Our target enrollment in each arm is approximately 20 patients. So you could expect an update of approximately 40 patients in that time frame. And the hockey stick enrollment and site activation is a real phenomenon, right? Because for us, when we went last year coming out of our data -- our dose escalation study in mono, we had been looking at 9 different tumor types. So we had to have generalist Phase I sites. So this year has really been a big operational push in migrating to head and neck expert sites, and it takes 6 to 12 months to open up those sites. So we're right now sitting in the sweet spot of those sites being activated. So those enrollment targets are well within reach and should inform the more mature data set that we'll put out in the middle of next year. Similarly, for combo, that is in dose escalation phase right now. That also is a multi-tumor basket escalation with a head and neck-focused backfill. So right out of the gate in combo, we have started with a mixture of both head and neck expert sites and more broad generalist sites. So we're confident that we'll be able to put up about 20 patients in the combo study readout in the second half of next year. So in terms of the registrational study design, we have a slide that outlines the key design elements. I want to start by saying we were to have a very productive and collaborative conversations with the FDA about this design earlier this quarter. There's a lot of concern in the news these days about time lines of responses and interactions with the agency. We've been very fortunate our same FDA team has been in place that we've dealt with over the last few years. They have institutional knowledge. They're quick on. We have access to them very quickly. We had a really, really [indiscernible]. So that's reflected in the design, where we have their strong endorsement. Essentially, the discussion we had with them was supporting a randomized approach off of a control arm is [indiscernible] rituximab, methotrexate or docetaxel. We wanted -- it's actually very important. We're committed in getting the design and plan these stage trials, because we see the field moving extremely quickly. And we wanted to make sure that the design that we are able to put forward reflects today standard of care so that the data that could ultimately come out of this pivotal trial could be very comparable to the ongoing late-stage development studies that are being run right now by the next-gen EGFRs. So that we're drawing on a similar comparative population, and the adjusted ideal body weight approach will be included in that design as well.

And our next question will come from Jeffrey LaRosa with Leerink Partners.

Also on the dosing and safety data, it seems that the -- it is striking how much better the tolerability is at the combination than the monotherapy, and you implicated the lower dosing is maybe a big driver of that. Are you considering also testing going back and testing the 4.4 mg per kg maybe as a monotherapy? If -- would that have a similar effect to using the adjusted ideal body weight strategy? And I guess another follow-up specific question on the example that you gave on the adjusted body weight dosing in the neuropathy patients, were these patients that had -- were those that did have the prior taxane treatment or were these taxane-naive?

Yes. Thanks, Jeff. I appreciate your question. As you know, as anybody who follows this space knows, Phase I is a learning time, right? And especially when you're dealing with a novel target and a novel mechanism. So we have been very, very appreciative of the hard work of our PIs, our team and our patients as we are collectively working to optimize the benefit-risk profile here. And I think the benefit side of the equation is clear as day. We saw that signal very strongly in the dose escalation data put out last year, and it's clearly validated here in the expansion study and in the combo. So I think your question hits at the heart of the matter of what Phase I is designed to do, which is to figure out what is that optimal balance. As a company, our philosophy has been to dose up to the maximum tolerated dose and figure out how to titrate and optimize around that. And so that was reflected in the choice that we made to take the 5.4 mg per kg dose into expansion. As you note, we've had sort of another crack at understanding even more about the 3.6 and 4.4 dosing levels in the context of the combo study. And it tells us, right, since we know that 3.6 through 5.4 dose range is the efficacious dose range, it tells us a bit of information about where we are on that sort of S curve, right, of efficacy and safety. And that's very steep in that kind of 4.4 to 5.4 range. So we haven't committed to additional dose expansion cohorts at this time. We are evaluating and taking a look very closely at the benefit-risk profile at each dose level in combo, and what we're learning in mono as well as how the implementation of some of these sort of dose softening, for lack of a better term, techniques are working for each patient, what those maybe ultimate drug exposures could look like. And then if that ends up being something that may influence our thinking around the ultimate dose, we can incorporate that. Of course, Project Optimus really is the best way to do all this, right, so where we take x number of doses and compare them with each other. So we will be providing further insight and guidance on how we're thinking about ultimately the approach to the Project Optimus component of the development plan, but I think your question is a very good one and goes to the heart of what we try to do in Phase I in figuring out where to land the plane.

Just a quick follow-up. Has the FDA weighed in on the adjusted body weight? Do they fare your perspective that this is probably something that all ADCs would benefit from, and in addition to the new [indiscernible].

Yes. Yes and yes. We've had very good dialogue with the FDA on the adjusted ideal body weight approach. We also benefit from a number of our PIs being ADC experts, who actually have done this in other studies. There's a few additional sources in the literature, I can point you to on this topic that are publicly available. Dr. Tony Tolcher, who is one of our KOLs at our data disclosure last year, had put out perspective on this topic as well as his experience with specific programs that have implemented it through either an AACR or an ESMO presentation. Similarly, Dr. Emiliano Calvo has done the same. You can see some literature from Elena Fontana. All -- these -- all three of them have put these things out in the ESMO or AACR domain, in addition to the work that Elahere did, which was published by Dr. Katie Moore. And interestingly, Pfizer with its Integrin B6 program, has gone right into dose [indiscernible]. So I do think we're at a juncture in the field now where people are recognizing that in a lot of ways, we need to think about dosing these things more like chemos where you have the sensitivity to the drug exposure versus a more fixed dose antibody-like dosing. And that may have been really a part of the scholarship over the last decade that is starting to fall into place now, now that we have critical mass of these ADC programs. Does that answer your question? And I realize, I think you had another question about the taxane use. Okay.

And our next question will come from Stephen Willey with Stifel.

So it looks like neuropathy was driving the majority of Grade 3 adverse events and really discontinuations. And I guess, in the monotherapy cohort, these are all platinum-experienced, I guess, in many cases, also taxane-experienced. So just curious if you were allowing patients with preexisting neuropathy to enroll.

Thanks, Steve. I will also pick up the tail end of Jeff's question since it's a similar topic. So the short answer is yes. We did -- we have allowed patients with prior neuropathy to enroll in the study, low-grade neuropathy. We actually started the study with an attempt to be more restrictive because we wanted to really reduce the risk of neuropathy happening in our study, but because we need to study later line populations, that proves to restrict too many patients from enrolling in the trial. So we had to become a little more open to including patients with prior neuropathies. And we knew that would be a risk that would likely materialize into our risk-benefit profile here. And in fact, we've seen it. So the way we think about these Grade 3 neuropathies are, they're cumulative tox, right? They reflect cumulative tox. And how does that occur? Well, that could occur from prior exposure to neuropathic-inducing agents. It can occur right in the setting of patients who have prior histories of neuropathy, and it can occur from the dosing that is happening in the MICVO study. Now if you almost imagine that a patient has a threshold of where they can tolerate that exposure and then they get tipped over, if a patient walks in with a significant history of prior taxane use or prior neuropathy, we have very little room to work with before our agent might trigger them to the threshold versus a taxane-naive patient or a patient who never had neuropathy before. So this is why we think the adjusted ideal body weight piece will actually be very helpful, particularly in those patients who have prior taxanes, prior neuropathy experiences because it's going to dampen down the amount of potential exposure to neuropathic-inducing agents that are contributed from our agents. Bringing in Jeff's part of the question, we did not explicitly disclose at the patient level, which patients had the prior taxane and which ones had the prior history of neuropathy. But to the point that was made, 2/3 of patients have -- that are enrolled, have had prior taxane use. And we see on the appendix slide where we've disclosed the Grade 3 correlation with body weight, we see 5 patients with Grade 3 neuropathy, 4 of whom were overweight, 1 of whom was underweight and the underweight patients could have been a patient that had excess taxane use or a prior history of neuropathy or a long-duration patient, right, that on top of all that since we haven't disclosed those particulars, but I think the key point is all of those things contribute. And as we adjust the ideal body weight to dampen down the contribution of our agent to an individual patient's overall exposure to neuropathic agents, that should improve. As we move into earlier lines, where we have fewer patients walking in with prior taxane use or prior neuropathy history, that should improve as well. So I think that the Elahere example, the Phase I paper that Dr. Katie Moore put out, I think it's a terrific reference because it showed the power of this approach to reducing the exposure to the ADC and how quickly that sharpened up the AE profile while maintaining efficacy. And then as we continue to prove out the use case for this agent and as we continue to develop it and continue to earn the right to compete in earlier lines of therapy, that should also help sharpen the safety profile. So this profile we showed today is very much one that's in flux. It's early days. And as I said earlier, Phase I is a powerful learning experience. It gives us valuable information on how to make this more optimized for broader use going forward. So did I address your question fully?

You did, you did. And that's helpful. And maybe just as a follow-up, actually.

Stephen, you can go ahead, please.

Okay. So maybe just as a follow-up to a prior question as well. Does your preliminary Phase III trial design, is that including kind of a seamless Phase II/III where you build in the dose optimization portion required by Project Optimus into that registrational study? Or do you think that because some of the nuanced work you need to do around dose optimization will be done kind of as a separate carve-out?

Okay. Excellent. Thanks, Stephen. We will be spending more time next year providing greater insights into the plans around -- the details under the hood of the design principles that we disclosed here. Obviously, dosing is a critical issue, or the critical issue, right, to trigger the enrollment of hundreds of patients into a pivotal study. So it's something we are actively working through. And we'll be ready to comment on next year.

And our next question will come from Sudan Loganathan with Stephens.

Thank you for this robust data presentation, and congrats on achieving this milestone. The efficacy outcomes seem very promising. Do you still anticipate to hit that 5.4 milligram per kg MICVO dose in the combination with pembro even with the adjusted ideal body weight strategy, although it seems that the efficacy is really there with the 3.6 and 4.4 doses and trying to manage the adverse events is probably the most important thing. And then, can you give us any details on the weight range and estimated percentage of patients enrolled that would be candidates for the 3.6, 4.4 and 5.4 doses at this new adjusted body weight strategy?

All right. Yes. So great question. So we are actively enrolling at the 5.4 mg per kg dose in combo with pembro right now. To your point, we were extremely pleased and excited to see the strong efficacy signals come out at the lower doses in the combo setting, and the absence of any progressive disease has been very notable to us as well. And I think it really speaks very powerfully to this mechanism of this extracellular ADC, which is doing something to the tumor microenvironment. I think that's absolutely coming through loud and clear. And just a little bit of context on that. Our translational team has done a beautiful, beautiful job over the past year of putting out an extremely high number of posters across ESMO, AACR triple meeting. I think we have a dozen that are essentially proving out this mechanism. We've shown that we activate T cells. We've shown that we are supporting the T cell infiltration into the tumor. And I think that was one of the compelling elements of the mechanism that got Merck excited to provide the pembro to us to do this experiment because they have seen prior example, and we've all seen those prior examples of powerful ADC plus checkpoint combos, right, in the case of bladder, for example. So here we are in head and neck, I think, similarly seeing something as exciting. To your point, seeing that at such a lower dose level in the escalation is notable. And we are treating the two studies somewhat independently in how we are thinking about the relative benefit-risk profile. I think it's important to complete the experiment in the combo setting with the 5.4 dose level. We may ultimately decide that the dosing regimen looks different in combo than it does in mono. I think it's still very early days here in both studies. The fully enrolled dataset that we'll put out next year, I think, will provide more definitive answers to these questions. But to your point, seeing this kind of efficacy at lower dose levels and really understanding how the mechanism is supporting that is a great -- gives us great degrees of freedom in not needing to think about higher dose levels as needing to be consistent, right, across those programs. So that maintains optionality for us. In terms of adjusted ideal body weight, I want to be clear, that is an approach that is going to be applied to every single patient in the study. So adjusted ideal body weight is a standard formula. It's not something Pyxis is making up. It's a well-known formula, which, I think, is in some of those sources that I had cited earlier. And it takes into account things like gender, current weight, height. And it basically says you put in those numbers and it fits out what your optimal dosing should be if you were at the ideal body weight. We know -- we all know the demographics, especially in the developed markets tilt towards overweight or obese patients. So I think this is a very, very important thing in dosing agents like ours with chemotherapeutic components that have known correlation with weight-based effects. So a normal weight patient would still have the formula applied and the outcome of that formula would essentially be the same answer as what the straight mg per kg dosing would be. An underweight patient would have the formula applied. That answer would put out a little higher dose because that current underweight patient has probably been underdosed by the current mgs per kg. So it should bring up the lighter patients to where they need to be. It will really essentially just validate the dosing for the patients in the middle, who are at the normal weight. And it will soften the excess drug exposure for the patients who are higher weight. And since we're at such a -- in that 4.4 to 5.4 dose range, for example, where a couple of percentage points of excess drug could be the difference between a Grade 3 or Grade 1 or none, we think that's going to be very impactful.

Great. Just a quick follow-up, also. You mentioned some challenges with enrolling HPV negative and nonapplicable patients. Would you also attribute that challenge to trial competition with the other ongoing competing trials for the head and neck squamous cell carcinoma?

Yes, yes. Thank you for that question. So yes, I'd say it's two factors, the first factor being the reflection of the demographics of our earliest opened sites. Those have tended to be in more affluent areas within the U.S. where the U.S. is seeing more of an influence of HPV-positive patients. Of course, there are still HPV-negative patients around, but there are also studies that are in competition for patients who are HPV negative or HPV not applicable only in their recruiting. And so if you're a PI with a portfolio of 3 or 4 studies and 2 or 3 of them are HPV negative or not applicable only, and there's another study that says, HPV all comers, you're going to rationalize your own patient population to send your positives to the all comers and your negatives to the dedicated studies. So we have had some impact of that. The way we are mitigating against that is opening up more sites globally where there are more HPV negative patients, for example, ex-U.S. And we also think as the next-gen EGFRs continued to progress through their later-stage development, the competition for patients with those studies will ease in certain sites and in certain geographies. So we think this is very much a solvable thing, and you will see a more diverse HPV status population reflected in the fully enrolled study demographics next year in those studies. That being said, I do want to point out in the monotherapy study, and I think this is important, that when we look at the demographic slide, and we think about where HPV status is relevant, it is measured in the oropharynx. It is not measured or relevant in oral cavity larynx and hypopharynx. So typically, those HPV status not applicable patients have been kind of lumped in with the HPV negative. So even in monotherapy right now, we have 50% of our enrolled monotherapy patients are in the not applicable segment. So even though we only had 2 outright HPV negatives in this monotherapy study, we had 9 not applicable. And then those tend to travel with the HPV negative.

And the next question will come from Swayampakula Ramakanth with HCW.

This is RK from H.C. Wainwright. Congratulations on all the data sets that you talked about this morning, Lara. A lot of my questions have been answered, but just a couple of quick questions. So the combination cohort showed a 71% ORR. Of course, it's 7 patients. Just trying to think through the next stages for this, for the combination. Do you think you're not, I mean, initiated any conversation with the FDA because you still need to get some additional data into this study? If so, at what point or how many patients do you need to initiate a conversation with the FDA regarding a pivotal trial? And also, a second quick question is, obviously, your -- the mechanism of action for MICVO is quite different from what other drugs in the clinic are at this point. Does that allow you to use MICVO as a combination with either the Merus drug or the Bicara drug? Is that a possibility at all? And also, does this also allow you to look at beyond head and neck cancer, especially in the EGFR-negative tumors?

Great. Excellent. Thanks, RK. So sort of hitting each of those one by one. In terms of the combination data and our enthusiasm there, I mean, we're thrilled. We're thrilled with the confirmed ORR and the disease control rate. And I just want to spend a moment on the disease control rate. It's 100% in combo, first, second, third line, all here, all CPS statuses. It was 100% all doses, all in head and neck in that escalation, right, from 3.6 through 5.4. It's essentially 100% in the expansion right now, if you remove that one patient that should not have been treated in a clinical trial. So the all durations we're seeing in the tumor microenvironment are absolutely real and fundamental and impacting the tumor biology. And so I think that bodes extremely well for the long-term potential durability of this agent kind of in any setting because we're not seeing that any particular specific prior therapy is impeding our mechanism for working. We see responses in prior taxane-treated patients. We see them in prior platinums. We see them in prior chemo, in prior checkpoints, in prior EGFR. What we're seeing is the needing to obviously land the benefit-risk profile and moving earlier into lines of therapy where we trigger more of a frank response than a tumor control one. So all of that is terrific. And then you layer on this combo data with waterfalls all going south and no progressive disease, and it's a very compelling picture. We have not commented on any FDA dialogues yet with the combo. I'd say stay tuned next year for further updates around our regulatory strategy for both programs, and we will kind of pick up insights around our thoughts on further development with this program here very specifically. In terms of your second question around can we combine with other agents, specifically the next-gen EGFRs? Absolutely. And again, I think going to the point of the mechanism of action, we -- our mechanism is orthogonal to what the other agents in the space are doing, meaning we are targeting a noncellular element of the tumor microenvironment. Whether it's another ADC or whether it's another mechanism, those other programs are addressing cell surface-based targets. And as we've seen through our own work mechanistically in this space, we believe this has multi-combinatorial potential. We started with the checkpoints, because we saw things like the power of KEYTRUDA plus PADCEV in bladder. And then as we began to prove out our mechanism and the impact we have in the tumor microenvironment, we got even more enthusiastic about the checkpoint inhibitor combination options here. And sure enough, now we've got the clinical data that's proving it. If you actually look back at what we've said over the last 4 years since we went public, everything that we predicted mechanistically has been proven out. It's taken us a couple of years to do it because that's early stage drug development, but we've done it. And so we have a strong belief that this can combine with the next-gen EGFRs, that this can combine with other agents, that this could take the place of chemo essentially, right, in some of those other regimens, those other combination regimens. So that's absolutely something that we think is interesting. And I think if you look at the landscape, the competitive landscape with the funnel of the different segments that we have in here, the summary on Slide 25, we think we have a unique position here in being able to provide benefit to patients in segments that the next-gen EGFRs won't be tackling. We have the ability to provide benefit to patients behind them because we have not seen resistance from any specific prior therapy used ahead of us. And our mechanism gives us the ability to combine with them. So it gives us a lot of agility in how to play. And of course, that shifts the responsibility to us to make the right operational and strategic decisions to prove that out, right? So that's what we're doing here. So I'd say great question and something we're very excited to continue to explore, the combination potential overall. And I think then your third question is because of this mechanism and this approach that we're taking, how do we think about opportunities beyond head and neck? Well, we saw in our dose escalation study that we read out last November strong tumor regression signal and disease control across a variety of other tumor types. In particular, we saw strong signals in breast, sarcoma, lung, ovarian. Each of those markets has their own particular nuance from a commercial perspective in terms of the degree of unmet need, the particular competitors, pricing dynamics, et cetera, et cetera. So those evaluations need to be taken as well as that of the clinical data, but we believe our job right now is to demonstrate focus in head and neck, prove the optimal benefit-risk profile for this agent and the optimal positioning. And as we do that, we would welcome the opportunity to do that in other tumor types that follow as well. But we want to be extremely deliberate and focused to make sure that our position in head and neck is extremely clear. And as the asset continues to develop, we will continue to evaluate the opportunities to apply what we're learning in head and neck to other tumor types. One additional point I'll make on that from a translational perspective, and we encourage people to look at our posters, our translational posters that we put out this year that I referenced. I mean, I believe there's clues hidden in plain sight for everybody on what this drug's power and potential is. And one of those is the work that came out of our transitional team identifying the factors in the tumor microenvironment that influence responsiveness to MICVO. It's not just whether the target is there. It's things like pH levels, things like the family of cathepsins and their concentration in the tumor microenvironment. It's things like AI digital path enabled tools showing us that the geometry of the tumor microenvironment matters. How those noncellular elements are organized? Are they linear or are they chaotic influences, whether the payload will traverse, how far and whether it will end up delivering a frank response or tumor control. All that stuff matters. Well, what does that mean? We see a right set of conditions in head and neck in all those elements that line up behind the clinical experience we're seeing here. We can use that information and say, well, what are the tumor types look like head and neck at the highly visualized cellular level? What do they look like? Are there other tumor types that have similar stromal architecture, that has similar pH range, and that could also influence where we go from there. So this is ultimately a pipeline in a product is what we see, but our job today is to deliver its positioning without a shadow of the doubt in head and neck.

Perfect. If there is a possibility for me to ask a quick follow-up? When I look at your Slide 16 and look at the baseline tumor size. As you said, even you had varied sizes within your cohorts. Going into the next stage of studies, are you planning to have any criterion around the tumor size, the baseline tumor size in the inclusion/exclusion criteria?

You mean like whether we would restrict it to particular sizes?

Yes, yes.

Yes. So that's an interesting question. We wanted to include this baseline tumor status info because, like I just said, I think geometry matters. I think spatial-like orientation of how the tumor microenvironment is organized matters. And we were thrilled to see these incredibly large tumors, 113 millimeters and 133 millimeters, responding. Like the 133-millimeter patient is sitting there at minus 70%. I mean, when you think about what the impact is in the reduction of the tumor of that size, and it came on fast. This agent demonstrates deep responses, cycle 1, cycle 2, like these patients are responding. So I don't know that there's a lot of benefit, per se, for us to restrict tumor sizes because I don't think we need to do that to find a sweet spot of efficacy because I think the messages we keep seeing whichever lens we look at this from. If we look at it through anatomical location like we've laid out here, if we look at the lens of HPV status, if we look at the lens of tumor size, if we look at the lens of which prior therapy patients had before getting MICVO, we're not seeing any factor that tells us don't go there. It's telling us, oh, okay, it works in any of these scenarios. So I don't know that we need to have those kinds of restrictions in place. Obviously, we spend a lot of time thinking very carefully around inclusion/exclusion criteria, like we talked about earlier with the neuropathy. We knew as an MMAE-based ADC that there's a risk profile for neuropathy. And so we tried at the beginning of the study to take that risk down to 0 as best we could by excluding patients with prior neuropathy or prior taxane use and exposure. But then you can't recruit patients. And so at the end of the day, I think our mission is to serve patients. And so I think we're better served being armed with an understanding of what we're doing, and how it works, and where it can benefit and then working to shape the profile. And very fortunately, in the case of size, we're seeing large tumors respond just as well as small tumors, which is extraordinary.

And our next question will come from Andy Hsieh with William Blair.

So to help us better understand MICVO's efficacy and safety compensation with the adjusted ideal body weight dosing, do you have response data on hand for Slide 28? Obviously, it will be very interesting to see if overdose patients are having enriched response and that patient who's underweight do not respond. So that's question number one. Question number two, for the combination studies, are you enriching patients with CPS of at least 1%? And going forward, are you looking at kind of the entire spectrum of CPS expression level? I'm asking just because it will be kind of interesting to use that CPS low or nonpopulation to validate some of the mechanism of action that you proposed.

All right. Thanks, Andy. So we didn't disclose the particular demographics of the individual patient's weight or responses. We do, at some level, especially in the setting of a public disclosure, need to be cognizant of HIPAA and identifying patient characteristics and that sort of thing. And weight is one of those factors that's thought of that way. So that's not something that we're able to include and comment on in a broad setting like this. But rest assured, those are the kinds of analyses we're doing as a team. You can somewhat intuit or infer from the swimmers plot to the slide in the appendix on Slide 28, right? When you see some patients that have strong PRs, strong responses or stable disease and then who are no longer on treatment, and we know there's a discontinuation for treatment-related reasons, and we have a slide that shows a set of patients that have treatment-related discontinuations. So we're not doing that mapping at the patient level, but you might be able to narrow down and connect those points. So that's kind of the best I can offer you in terms of that question. And I will say, again, these are the things that we look at extremely closely and feel very, very confident that when you look at an S curve, a dose response curve for patients, the efficacy dose response curve for a drug is not necessarily the same dose response curve that the safety S curve is, right? Those things are not always superimposed, which means there's deltas in between those curves that you can optimize around. And we know from prior ADC experience with other programs that they have been able to land those optimized benefit-risk profile, where they've been able to tune down the safety without sacrificing the efficacy. And here, I can give you, again by inference, a bit of comfort around that because if you look at our Slide 28, you see our overweight patients are somewhere 10% to 20%. And if you go to our combo safety table on Slide 23, we know the 4.4 dose is about 25% lower than the 5.4 dose, just in raw escalation, right, 3.6% is 50% lower. We have no Grade 3s at 4.4 in combo. By the way, the risk of that is even higher because you've got pembro contributing to it, but we still had a goose egg. So you can see by tuning somewhere within plus or minus 25%, there is a space that you can occupy where you tune down those side effects and maintain your efficacy. And I think this 4.4, no Grade 3 safety table on Slide 23 kind of gives you some guidance around potentially what you could expect with the AIBW reductions of 10% to 20%, right? Every percentage point matters when you get to the top of those S curves. So we want to land this very precisely. And so I guess the second -- can I move on to your second question about the CPS status then?

Yes, absolutely. That's helpful.

So we did -- we have not restricted or enriched for specific CPS status to date within the combo study. And I think we feel very validated by that decision by seeing what we see here on Slide 21. And I think, in particular, for some of these lower CPS status patients, it's particularly exciting. We -- so we feel very comfortable with the net of -- we were casting for these patients in the first line, second-line-plus setting. One thing we haven't covered yet, which maybe I'll just sort of throw out there in conjunction with this question, circulating tumor DNA can be also an exciting measure here of response and so forth. And while we don't have a sort of mandated circulating tumor DNA assay in place, many of our sites do have their own measures or assays in place to capture that. And we have continued to see strong evidence reflected in circulating tumor DNA of the clinical picture that we're showing here on Slide 21. So that's also something that might be interesting for us to think about providing more insights into in the fully enrolled data sets next year.

I am showing no further questions in the queue at this time. I would like to turn the call back over to Lara for closing remarks.

All right. Terrific. Well, first, I want to thank everyone for your time today. We very much appreciate the investment of your attention in thinking about our program and your really important questions. You hit on the key things that we think about, and we work on every single day. We are extremely, extremely excited about what we have seen to date and are very much looking forward to sharing the fully enrolled data, as it becomes available in the middle of next year for mono and second half of next year for combo. The enthusiasm for this program remains extremely high amongst our PI community. We've had several notable PIs who've been with us since the very early days, even preclinically, as we were planning this study. And I think their enthusiasm is also reflected in our ability to both recruit and enroll patients and to figure out to put these pieces of the puzzle together around optimizing the benefit/risk profile. So we look forward to continuing to serve our patients and working collaboratively with our PIs, the institutions. And we appreciate the time and care and attention from the Capital Markets participants, such as yourself today. So thank you.

This concludes today's conference call. Thank you for your participation, and you may now disconnect.