For our next Fireside, we have Pyxis represented by Lara, their CEO. Thank you for being here.
Thank you for having us. We're always excited to be here.
Great. Just to start off, presented data, which we'll get into, but maybe just for a second, a zoom-out, bigger picture. Can you remind us what's special about MICVO, maybe the target, the unique, optimized, payload linker structure?
Yeah. MICVO is our lead asset at Pyxis Oncology. It's a first-in-concept ADC, which is drugging extracellular domain B. That's the target, which is a spliced variant of fibronectin. The construct was actually developed by Pfizer, who had originated the molecule prior to when we licensed it in a development candidate stage. Pfizer spent a lot of time and resources optimizing the construct, the ADC construct itself, for better potency, better stability, better permeability. We have a site-specific conjugation chemistry, which, for those who follow the ADC space, know that was a key part of the thesis that Ambrx held when it was acquired by J&J in terms of the quality of the data set and how it linked to the construct of the ADC. We have that same site-specific conjugation chemistry. We've got val-cit linker, and we've got a highly specific target.
EDB is expressed quite high across a variety of solid tumors and negligibly on normal tissue. The combination of the differential and the specificity of the target, along with a well developed or created ADC construct that surrounds it, coupled with a novel mechanism that's targeting the extracellular domain, essentially the extracellular matrix where the target sits, kind of those three things together have us very enthusiastic about the clinical potential for this asset as we continue to advance it in clinical development. While the ADC field has evolved quite a bit over the last several years, the approach we're taking with the novelty of our target really built upon sound technical construct, I think bodes really well for the data that's coming up.
Yeah, really helpful overview. Thank you. Last year, you guys presented the first look at efficacy data for MICVO. Obviously, there were some mixed reactions to the data, but I guess what's the key takeaway from that data for you guys with respect to the responses you've seen and sort of how it validates a path forward?
Yeah. We were really excited to put the data out last year and are even more excited about the clinical data that's coming up this year. The reason for that enthusiasm is when you're launching new targets and new mechanisms into the marketplace, you're building a case. You're building a wall of evidence from both a mechanistic, a biologic, and a clinical perspective. Last year was just the beginning of that story, the very, very beginning. The first time we put out clinical data, it was a dose escalation study across 80 patients, and there were 10 tumor types eligible to participate, nine of which were dosed.
We knew with a dose escalation of about 8 dose-10 dose levels across nine different tumor types that any one field of data, meaning any one combination of tumor type by dose level, was going to be a small N. We knew that this was signal finding, establishing safety data, and giving us the tools of what we need to design the next stage of development, which is where we're at now in cohort expansion to prove without a shadow of a doubt the strength of the signals that we already observed. I think last year, well, this year too, the market conditions continue to be challenging for all of us. We launched our dose escalation data out into a market that was, I think, looking for a higher N out of a dose escalation than you can necessarily deliver in a dose escalation study.
Here we are now with the next set of data that will be coming up out of our cohort expansion, able to deliver the N that the market wants to see, to have that kind of full faith and confidence in the signals. The reason we are so excited is we saw tumor regression in six out of the nine tumor types that were dosed. We saw significant disease control in other tumor types as well. We saw clear evidence of activity, certainly in head and neck, where we had a 50% confirmed response rate out of six evaluable patients, including a confirmed complete response all by RECIST. These were all in the context of a highly heavily pretreated population. We were median fifth line in this study. We had some patients who had 8, 9, 10 prior lines of therapy.
I mean, these are very, very sick, very end-of-life patients. Now, in this stage of development, we are working, we have both a monotherapy program and a combo therapy program, which we'll talk more about in a moment. In both of those programs, we are in much earlier lines of therapy. We're going into second, third line in mono and first/second line in combo. We expect to see much more fit patients, and we expect a higher number of patients. I think those two things will help alleviate sort of any lingering concerns around how to interpret the data that we had initially put out. Because it's very hard to interpret when it's a small N.
You think N of 80, you have a lot of data to work with, but when you break it down, at any dose level, it's actually a smaller N, and it's a smaller N when you segment it by tumor type. We recognize that can be hard to interpret.
Yeah, that makes sense. I think one of the most frequent questions we heard post the data was actually on safety. There was a patient death, but can you talk about maybe the overall tolerability of the ADC candidate, what you saw coming out of that trial, and maybe sort of as we look forward to the next cuts of data, if there's any specific adverse events that we should be looking for?
Sure. Patients die on phase I oncology trials, not just ours. The patient that passed away passed away from progressive disease. The patient did not pass away on anything to do with the study drug. That was an end-of-life patient who passed at week five. That was a Hail Mary pass from a PI trying to give a patient an experimental therapy that could potentially prolong their life. Interestingly, in that patient, we had tumor regression even while the patient was clinically decaying and beginning the preparation to pass away. There was not a safety issue at all associated with that patient or with the drug. Unfortunately, in this age, when the word death is mentioned or appears on a slide, all of a sudden, people attribute something to it. With the internet, all of a sudden, it's out there.
I appreciate the opportunity to clarify that because the safety profile, when you actually look at the safety tables, shows no grade fives that are drug-related, zero. In fact, the safety tables show quite a well-tolerated drug. We had only one patient out of the full 77 discontinue. We had a very low dose reduction or dose holiday rate, well within the norms of what you tend to see with ADCs. We also, in our data set in the fall, compared our safety table to the FDA-approved dose levels of the launched ADCs that are out there. We looked at that from the lens of neuropathy, neutropenia, pneumonitis, ocular, and skin tox. We were as good as or better in each of those dimensions on every approved ADC that was out there. Remember, those approved ADCs, we looked at the label, those are earlier line patients.
We were comparing our median fifth line against approved agents' earlier lines. We fully expect that with this next data set coming into earlier lines, that any lingering questions around interpretability of safety from the dose escalation will be fully assuaged and put to rest.
Got it. On the efficacy side, I mean, you mentioned the 50% response rate. How does that compare to what's currently in the treatment landscape for head and neck?
Yeah. It's been very exciting, I think, seeing the entrants that are out there in the head and neck space, and clearly, Merus and Bicara with their EGFR mechanisms are bringing some much-needed potential therapies for patients. We've clearly been watching, and we're seeing contributions to the potentially emerging standard of care that raises the level of expectation around ORR. Of course, durability is quite important, overall survival, progression-free survival, those types of things. Within the stage of development we're at right now, we're obviously focused on validating the signal, showing durability wherever we can, recognizing, again, the stage of development we're at is not designed to demonstrate durability. We're not a two-year study. We are in a cohort expansion where we'll be putting out data as soon as we can when we reach sort of an N of 20-ish patients.
We may have some durability capture in there, but it'll primarily be ORR-driven. When we look at Merus and Bicara's performance in their own phase I or phase I-B studies, we saw a 0% ORR for Bicara and a 37% for Merus. We look at that Merus level as what we need to beat or exceed. Merus' ORR in that study was in median second line patients. Again, we're median fifth line where we showed our 50%. While our N was lower, we do have confidence that as we expand the N, we're going to meet or exceed the threshold that Merus had put out there. Similarly, when we look at combo, it's interesting to us that you see Bicara generating ORR in combo with pembro when they had a 0% in mono. Clearly, there's something interesting going on there.
Same with Merus, having an improved ORR in combo with pembro, and we're looking there in the 60%-65% ranges. We see that as the level that we need to get to to demonstrate the efficacy of our agent. Again, we go back to the six patients that were evaluable in head and neck, nine total patients that we dosed in head and neck where we saw tumor regression in seven out of nine of them, clear tumor regression in seven out of nine of them. We feel very good about the opportunity to meet or beat the expectations of what's being set in the market as the standard of care is evolving.
Yeah. Maybe one last question on the data you presented last year. It was obviously, as you mentioned, a broader basket study. Head and neck was not the only thing you guys looked at.
Correct.
Among the other tumor types, I guess, are there other signals you guys think are interesting that are worth pursuing maybe after head and neck?
Yeah. The way our mechanism works, our target works, we do have a broad multi-tumor potential agent here. We're a small biotech. We're constrained by resources. We're focusing our execution right now on head and neck. That being said, to your point, Leo, the signals that we saw in breast, we saw signals in sarcoma, we saw signals in ovarian, we saw signals in lung, were very intriguing to us. The way we are handling that in a resource-constrained environment is twofold. In monotherapy, we're still doing, in preparation for Project Optimus, some additional work and fleshing out our data set across doses. We're continuing to do some scholarship with the 4.4 dose that we added at the end of the dose escalation last year. We only had four patients back then, so we want to round that out.
We're using that as an opportunity to put some additional tumor types in at the 4.4 dose. It's primarily around gathering PK and FO to support, again, Project Optimus downstream, but it gives us another opportunity to test some signals in other tumor types. That's a great place to do that in monotherapy. In combo with pembro, we are in the dose escalation phase. We're starting at 3.6+ the pembro. The dose escalation is eligible for multiple tumor types. We've got head and neck, we've got breast, we've got sarcoma, we've got gastric in there. That will allow us to better understand and characterize those signals. Once we clear a dose level in combo, we're going to immediately backfill with head and neck.
This is how we're going to amplify the head and neck signal while we are dose escalating. Also at the same time, getting some additional signal-finding benefit for other tumor types that could form the basis of more lifecycle management opportunities downstream or a more holistic development plan for the asset, of course, depending on the funding climate that we find ourselves in.
Yeah. I wanted to touch on some preclinical data that you presented at AACR, including on tumor signatures. Can you talk about sort of how it informs the mechanism more broadly and also what you've learned from that data and potentially being able to identify who might be most responsive or what tumors might be most responsive?
Yeah. It is always really exciting with a new molecule or with a new mechanism to better understand and characterize the biology. Our hypothesis, and those investors who have been with us since our Series B and IPO days will recognize that our hypothesis around the mechanism has always centered around three specific components, one being the direct tumor-killing effect, the second being bystander effect related, and the third being around sort of local immunostimulatory types of effects. The degree of influence of each of those three components to our mechanism has been something we have been spending a lot of time with our translational team to better characterize. To your point, our AACR posters recently have given some insights into these mechanisms.
We have a lot more work going on, and you're going to hear a lot more from us around this over the next several months in advance of our clinical data. Sticking with AACR, some of what we put forward, and the posters are available on our website for those who are interested, has demonstrated the impact that adding MICVO to the tumors is having on local immune cell activation. In one of our images on one of our slides, we actually have a picture that shows T cells that are sort of stuck outside the tumor in untreated tumor types. We show with MICVO treatment, the T cells coming right into the heart of the tumor.
That gives us a huge amount of, A, validation around the mechanism, but, B, enthusiasm around what we're going to see as we add pembro and combo, and, C, just the kind of rationale for some of the responses we've been seeing in monotherapy. We are also doing some work around gene signatures, understanding what that looks like with respect to patient-responding population or non-responding population, better characterizing our understanding of the stromal environment, how do some of the features of the stroma, whether it's protease concentration or specific proteases or oxygenation status or architecture, all of those things can contribute to the likelihood of a patient potentially responding. We want to understand those factors, and we want to understand how they correlate to clinical responses.
That additional evidence, that wall of evidence that we're building around the mechanism, around the tumor biology is coming over the next several months. I think the AACR posters are a really nice clue in terms of the direction we're going. I think having kind of the totality of those insights in hand from a translational perspective in advance of this next tranche of clinical data will really equip investors to be able to make much more informed judgments about our data versus last year where it was a first cut of data that was harder to interpret for the reasons we discussed.
I want to spend a few minutes talking about the next set of trials that you're running. With respect to the monotherapy data in second and third line, I guess, how many patients' worth of data should we be thinking about? Do you have a sense of what the timing might be for that data? I know you sort of alluded to it, but what you'd like to see in terms of response rate?
The next stage of development for us in head and neck is focused in two programs, the monotherapy, to your point, and then the combo with Keytruda. In the monotherapy development program, we have two arms. We're looking to generate data and share data on 20 patients, 20 head and neck patients in each arm. That will ultimately be a total of 40. Arm one is the PD-1 plus platinum resistant patients. This is today's current standard of care, patients who aren't responding to today's current standard of care. The second arm is the EGFR plus PD-1 resistant patients. These are patients who don't respond to cetuximab today, but also patients who will not respond to Merus and Bicara tomorrow. Many of our sites are actually also enrolled in the Merus and Bicara trials.
We did that very deliberately so that we could try to enroll into our trial, into our monotherapy trial, patients who did not respond to Merus and Bicara. What this does is this enables us with these two arms. Arm A, the first one, against the current standard of care, allows us to, quote, compete with Merus and Bicara, right, because we're all looking at that same population. Arm B, where we are picking up the EGFR-resistant patient population, allows us to exist with Merus and/or Bicara because assuming the standard of care goes in that direction, we'll have the data showing how our patients perform who've been exposed to EGFR therapy. Each of those arms, as I said, 20 patients, the current standard of care arm, the PD-1 plus platinum arm, we expect that data the second half of this year.
The EGFR plus PD-1 arm, we expect that data first half of next year. We are building in a little extra cushion of time for the EGFR arm because we want to try to get as many Merus and Bicara patients in there as we can versus just cetuximab patients. We want to give a little more time to enable that to happen. That is second, third line for us. In the combo setting with pembro, those are first and second line potential patients. Again, Merus and Bicara have their combo studies ongoing. We may see, while we are recruiting patients eligible for both first and second line, we may see a little geographic difference in terms of where we can pick patients up.
We've got a lot more potential opportunity to pick up first line patients in Europe just given there's hundreds and hundreds of patients already in Merus and Bicara enrolled in the U.S. It allows us to get the second line patients right behind them so we can coexist with them as well as compete with them. That data for the combo is expected preliminary data because remember, it's dose escalation. As I said earlier, we are backfilling as we're escalating. We expect to be able to share some preliminary data, hopefully about 20 patients second half of this year. That 20 patients may be mixed with other tumor types because it just depends who walks in the door and dose escalation.
You always have to have that caveat in managing expectations around dose escalation when you've got multiple tumor types that you don't have full control over the end by tumor type.
Got it. On the combination study, how do you think about teasing out the response rate of Keytruda?
Yeah. This is, I think, the beauty of the way we've designed the development program because we've got the monotherapy progressing right in parallel to the combo. Contribution of parts is always the question on everybody's mind, right? To your point, pembro is a powerful agent. It's demonstrated itself over and over and over again. You want to be able to understand how much is coming synergistically with your own agent versus just pembro. I think the fact that we're going to have 20 patients in each arm from the monotherapy coming around a similar time, at least the first arm, right, in parallel to the combo, will give us and will give investors lots of confidence in how to interpret our combo data.
Whereas if it's, for example, Bicara had 0% in monotherapy and now they have a nice response rate in combo with pembro, those kinds of situations, it's much harder to interpret. In our situation, we'll have that data in hand to be able to triangulate.
Got it. You mentioned that the PD-1 combo study is a dose-ranging study as well. How are you thinking about that 3.6 mg-5.4 mg per kilogram target dose for monotherapy as it translates into the combo? I mean, do you think that's still going to be the right range? Is it possible to push dose higher just based on what you saw in the monotherapy? On the other hand, is there any potential in overlapping toxicity that you're watching for?
Right. No, it's a great question. We felt very comfortable by the effective dose range that we identified of 3.6 mg-5.4 mg in mono. We had dosed in the mono dose escalation 42 patients at 5.4 mg. It's a really large amount of safety data in that setting. We feel very confident in the range of 3.6 mg-5.4 mg in pembro and, of course, in combo with pembro. Pembro is very well characterized. It's a fixed dose of pembro. Our plan is a 3.6 mg cohort, a 4.4 mg cohort, and a 5.4 mg cohort. In taking a look at the pembro toxicities in ours, we're not that concerned about sort of overlapping or synergistic toxicities. We also can look to the Padcev plus Keytruda bladder study. Padcev is an MMAE payload. Ours is a modified orostat and payload.
We have some ability to have some predictability from a pattern recognition perspective of what to expect here. We are very keen just given from a mechanistic perspective and the early clues that we put out in AACR and the translational work we are doing with our team that the combo with pembro sort of right on top of the immune priming we are doing in the tumor through MICVO is going to be pretty powerful for these patients. Whether we ultimately need to dose all the way up to 5.4 mg or the 3.6 mg or 4.4 mg dose will be enough remains to be seen. To your point, the dose escalations are independent. We could still dose above 5.4 mg in combo with pembro. We are not limited by what we have done in mono other than, of course, the learnings, right? The insights are always helpful.
From a procedural perspective, we could choose to go higher. I have a feeling that the 3.6 mg-5.4 mg range, since that was such a nice sweet spot in mono, will probably be a similar nice sweet spot in combo.
Got it. We only have 10 seconds left, but wanted to ask what the reception to the data has been in the physician community, sort of how excited the sites are to keep enrolling in the study, add new patients, just as a contrast to maybe what we're seeing elsewhere.
Yeah. Our dose escalation, we had waiting lists throughout the entire study. We are continuing to see that level of enthusiasm with the investigators in the mono and the combo. We've got a number of the top sites in the U.S. as well as ex-U.S. that are enrolled. There's a real appetite because I think the investigators, the KOLs, the PIs, they are really getting a granular understanding of the mechanism and seeing how that can benefit where there's been sources of resistance for their patients in the past. We didn't get a chance to talk about it in great depth, but we saw in the dose escalation study responses in patients who failed prior taxanes, responses in patients who had not responded to prior ADCs, responses in patients who had not responded to prior EGFR.
I think they're seeing us as a potential all-utility agent, meaning there's a lot of different patient populations that could be or prior treatment populations that could be benefiting from our therapy.
That's great to hear. I think that's all the time we have, but thank you so much.
Thank you.