Good afternoon, everyone. My name is Farzin Haque. I'm one of the biotech analysts at Jefferies. I'm happy to introduce and welcome Lara Sullivan, CEO of Pyxis Oncology. This is a fireside chat format. Thank you for joining us today.
Thank you for having us. We appreciate it.
For those who are new to the story, can you provide a short overview of your program?
Yes. Pyxis Oncology is an ADC-focused company. Our lead program and the sole focus of the company these days is our ADC that we had in-licensed from Pfizer at the development candidate stage. It is an extracellular targeting ADC called micvotabart pelidotin, otherwise known as MICVO. It is an asset that is now in later stages of phase I development. We are in cohort expansion in monotherapy. We have just launched a dose escalation study in combination with KEYTRUDA, funded by Merck. The KEYTRUDA is funded by Merck through a supply agreement. We are very excited about this particular agent. It is targeting EDB, which is extracellular domain B. It is a splice variant of fibronectin, which is a key structural component of the tumor and a key target in the tumor extracellular matrix.
Cool. Maybe talk more about your differentiated approach in targeting the stroma, and then basically suggest a payload cleavage in the extracellular matrix that has enabled you to achieve the clinical translation.
Yeah. I think there's been a lot of activity in ADCs over the last few years, as everybody's aware. The canonical thinking in the ADC modality has been that the ADC needs to be internalized by targeting a cell surface target, engulfed into the tumor cell, cleaved intracellularly. That's how the payload activates the killing machinery. What we have actually found, what Pfizer did in its early days in some of the exploratory biology work, in which our team and scientists have continued to augment, and now that we've seen through clinical responses, is that, in fact, that canonical belief is not true.
The ability to cleave the ADC in the extracellular environment with the presence of proteases in that ECM is allowing that ECM to essentially dismantle or alter the tumor microenvironment, not only release the payload for direct tumor cell killing and the bystander effect, but also to alter that local environment and potentially sort of create inhospitable conditions for tumor growth. As I said, we saw some of the early work that Pfizer had done demonstrating the presence of these proteases in the ECM. We're seeing, I think, even with some recent work that some team out of Duke did in the context of in HER2 with low HER2 expressing patients, that the bystander effect and the extracellular activity of ADCs and payloads in that environment is actually becoming sort of well known now as a key mechanistic component.
Here we are with the first in concept extracellular targeting ADC that's really relying on that mechanistic component.
Got it. Let's talk about the clinical data that you have shown last fall, I believe. Yes. The data was an early cut across doses and different tumor types. You show that the drug is active and has good safety. What are the expectations basically from that data set from the investors and what you're thinking about it?
Yeah. We were very pleased with the data that we put out last November out of the dose escalation study. It was a phase I study, basket design. Ten tumor types were eligible, nine tumor types enrolled. We dosed 80 patients. We dosed across, I think, nine or ten different dose levels. And we identified 3.6 mg-5.4 mg as the effective dose range. We found tumor regression in at least six of the nine tumor types that were enrolled, and a very strong signal that emerged, particularly in head and neck, where we had enrolled nine patients at 3.6mg and above, six of which were evaluable. The data set clearly demonstrated that the drug has activity. From a safety perspective, we had no grade 5s, and we had zero grade 3 and above neuropathies and ocular tox. We only had, I think, a less than 10%.
It might have been 8% grade 3 rate in neutropenia and in cutaneous superficial rashes. The agent was extremely well tolerated. I think this speaks to two things about this program. One is the highly differential expression of our target. We see this target expressed in the extracellular environment of multiple tumor types, and it is negligibly expressed in the companion normal tissue. It is a high differential expression gradient, which bodes really well to attracting the ADC right to the tumor as opposed to normal tissues. The construct, the ADC construct itself that Pfizer had engineered through really empirical testing and optimization, built a really nice, stable molecule that does not fall apart in the blood. The ADC goes right to where we want it to in the target tumor tissue. The safety profile, I think, reflects those components.
No evidence of an antigen sink either. We saw that same stability and potency of the ADC across dose levels during dose escalation. As we took a look at this data, as we stepped back, as we interpreted it with signals across multiple tumor types, in this climate, we decided to really focus our development efforts going forward on head and neck, where the signal was extremely loud and strong, and to leverage the learnings from the other tumor types for potential lifecycle management strategies in the future. Really, to make sure right now we're clear, we're focused on our execution in elucidating that signal going forward in head and neck.
Makes sense. Maybe for context, what was the response rate that you saw in the head and neck cohort, and particularly in the background, how many median lines of therapy they had?
Yeah. Our phase I study was addressing a very sick population. The population was very heavily pretreated, median of four prior lines of therapy. We were essentially fifth line. In fact, we had some patients enrolled who had 9 or 10 prior lines, very sick end-of-life patients. The confirmed ORR was 50%, 50. That included a confirmed complete response and two confirmed partial responses. The other three patients had stable disease, so 100% disease control rate. It was a pretty compelling data set. We had dosed another three patients that were not efficacy evaluable, two at 6.6, which was a dose level that we had not fully evaluated during dose escalation. It was not a clear dose level. Those two patients also showed tumor regression of about 10% and 25%.
One additional 5.4 mg patient had a complicated medical history unrelated to oncology and was only able to take one dose, but still, after only one dose, had a 15% tumor regression. I always sort of give the context of those three additional patients, even though they're not evaluable, because all three showed meaningful tumor regression in pretty extenuating medical situations. In the case of the 6.6 mg patients, they were both end-of-life patients. We still showed tumor regression when these patients were close to end of life from disease progression.
Makes sense. Safety has been good so far. Is there any reason to think that the AE profile would be different in early line patients now that you're exploring that, especially with respect to neuropathy, neutropenia, pneumonitis, ocular, and skin tox?
Yeah. You named the big things that we all worry about in ADCs that we keep a very sharp eye on. I think we're very pleased with the starting safety profile that we saw in this very sick population. I think as we go into earlier lines, there's some things that will actually continue to support our expectations around this robust safety profile. For example, by the time we got our patients in phase I in the dose escalation with the heavy pretreatment regimens they received, many of them had already been exposed to agents that induce neuropathy.
As we move into earlier lines, we can be more selective about patients that may have those predispositions and potentially not just sort of hold the line with where we're at with a strong profile where neuropathy has not been an issue, but perhaps even be even stronger than what we saw. We expect the safety profile that we generated to be validated as we increase the n through the cohort expansion. There is no reason to think in the earlier lines that anything will jump out. If anything, we think some of those profiles might sharpen up even a bit more.
Got it. You have two data updates coming up in the second half. The first line combo and then second, third line monotherapy. How are you setting expectations for that readout? Let's start off with the number of patients and duration of follow-up you expect to report.
Yes. In head and neck, as you mentioned, we have two programs that are underway. One is a monotherapy study, and the other is a combo therapy study in combination with KEYTRUDA. In the case of the monotherapy study, we have two arms. Arm one, which is targeting the current PD-1 plus platinum treated patients. These are the current standard of care or the patients that are resistant to today's treatments. The second arm is EGFR plus PD-1 resistant patients. This is where we think standard of care is going. We have all seen really nice data coming from Merus and Bicara. Cetuximab is already out there. We decided to get this arm off the ground in parallel with arm one so that we can generate data in the case of arm one that helps us see where we fit relative to the current competitors that are out there.
In the case of arm two, where we can fit as the line right behind the competitors that are out there, this gives us maximum strategic optionality so that we are essentially agnostic to which of the two EGFRs wins the race and at what line they settle in, because we can either coexist with them or compete with them either way, depending on the data set. We expect in each of those arms to enroll about 20 patients. In the case of arm one, which is the PD-1 plus platinum arm, we're looking to have that data in the second half of this year. In the case of arm two, which is the EGFR plus PD-1, we're looking for that data in the first half of next year.
We've built in a little extra time for arm two because ideally we can enroll some Merus and Bicara patients who have not responded to those therapies. We want to make sure we have enough time to capture some of those patients. We're already able to capture the cetuximabs because those are out there. In the case of the combo study with pembro, that is at the dose escalation phase. We are starting at 3.6 mg per kg plus 200 mg fixed dose of KEYTRUDA. We're anticipating 3.6 mg, 4.4 mg, 5.4 mg as the dose escalation levels. This aligns with the effective dose range we saw in monotherapy. Depending on how swiftly the dose escalation goes, and in the dose escalation for combo, we are allowing other tumor types in there as well because that will speed our ability to get through the dose escalation.
Breast and sarcoma, gastric, and I believe it's cervical are all eligible for dose escalation. The ultimate data readout for that, or the initial data readout for that, I should say preliminary, that we plan to do second half of this year, it may be focused on the dose, or it may have some early signs of data related to head and neck. It really depends on how the recruitment pattern goes. The plan there with dose escalation, as I said, is to leverage multiple tumor types to speed through the escalation itself. We will be backfilling immediately as the dose level clears with head and neck. It's a little hard to predict exactly how many end will come. Ultimately, we're looking to release data in the combo setting of about 20 head and neck patients, as well as obviously a recommended dose level.
More to come as we get closer on whether those two things will be together or split into two different data readouts.
Got it. It seems like the uncertainty with enrolling and dosing patients, so the update will be at a medical conference or a company-sponsored event.
Yeah. We have not guided to a specific format just as of yet because, as you know, with early stage trials where you are wanting to capture as much data in real time before the data release, the medical conference timelines can sometimes be challenging since they like to know far enough ahead. That being said, we always are looking at every vehicle, whether it is a medical conference or whether it is a company event or press release. As we get closer, we will guide more explicitly around the format.
Got it. Your data seems agnostic to the HPV status. Is this something that you can leverage in a development strategy to enable differentiation related to competitors?
Yeah. We were really thrilled to see in our dose escalation that we had confirmed responders in both HPV positive and HPV negative. It's a small n. We recognize that. As we're recruiting into these next stages of development for head and neck, we're looking to balance the recruitment across HPV positive and HPV negative so that we can confirm the agnostic nature of what we think our therapy is delivering here. I do think that can be a key competitive advantage. It goes back to even why we're designing arm one, which could be competitive to today's emerging leaders who may, in fact, ultimately have differentiated market positions depending on HPV status. We could have a potential privileged place with demonstrating that we're agnostic. We're looking forward to that.
Got it. You mentioned about the refractory, the Merus and Bicara and the cetuximab patients. So are these patients easily identifiable to enroll in your study, or is there something that could be potential competition for those patients?
Yeah. It's actually a great question. We spend a lot of time with PIs and KOLs at ASCO. We've also seen where the Merus and Bicara clinical trial sites are. Those are published out there on clinicaltrials.gov. What we're finding is that a number of PIs and KOLs, while excited about the EGFRs, remain equally excited about a mechanism like ours. We are absolutely able to sign up sites that are already signed up with Merus and Bicara. What we're seeing is one site typically won't sign up both Merus and Bicara itself. They have to pick between EGFRs, but they're very open to our mechanism. In many cases, since we're recruiting, I neglected to say this earlier, but in the monotherapy arm, we're recruiting for second and third line. In the combo, we're recruiting for first and second line.
At the sites that are also enrolling Merus and Bicara, we are finding that we are getting patients that could be eligible for their trials as well. It is great because we want to be able to, again, take a look at the earlier line or the line coming right behind them. We are able in the recruiting to source patients in both of those situations.
Got it. And then the key question, what do you view as the ORR and the bar for success for each of the cohorts that you're reporting?
While we haven't explicitly set guidance around the expectations to come out of our study, what we do see is that the landscape is aggregating around the benchmarks that Merus and Bicara are setting. I think in the setting of combo, we're in the world that you're at sort of an expectation for new competitors is around 65% ORR or more. They've reset that benchmark. I think in the case of mono, Merus has set that benchmark at 37%. We feel confident because in our November data release, people who looked at that may recall that we looked at the Merus Bi cara data out of their own early dose escalation phase and looked at their data in comparison to ours that we had put out. Our confirmed ORR was higher than each of theirs. Bicara's was 0 in monotherapy, and Merus' was 37%. Ours was 50%.
Both Merus and Bicara in their data sets at the same time point that our dose escalation came out, their median prior lines of therapy was one. Our median prior lines was four. So we were already showing a higher confirmed ORR in later line treatment. That is why we remain convicted and enthusiastic about what we're going to see as we increase the end here. I think with an end of 20 in each arm and ultimately an end of 20 out of the combo, it should be pretty persuasive at that point, the data set that we put out.
Got it. Your data set is not designed to address durability. How do you view the competitor updates at ASCO from Merus and Bicara with regards to PFS and OS? How does that shape your strategy for MICVO's clinical development?
Yeah. I think, first of all, we hear a lot from the PIs and the KOLs around durability is the thing. Response rate is easy for us to all sort of calculate and measure quickly. It is certainly not the name of the game. It is all about durability responses. We have been pleased to see the quality of data that they have both been putting out and really extending the time frame there for patients. I think they are also setting those new benchmarks there. With the cohort expansion, just as with the dose escalation, neither of those studies are designed for durability. We are not going to be able to put up data that is comparable to theirs.
We will be able to share some anecdotes around some of the very early enrollers because we'll get some number of months of data around durability for some of those early enrollers. I can tell you that for the phase I dose escalation of the six responders, the six patients that were efficacy evaluable of the three responders, we have at least one of them who's been on therapy for 17 plus months. The other, the complete responder who unfortunately had to end treatment shortly after the CR for personal logistical reasons involving household move and other things that we couldn't problem solve around, maintained their CR for four months off therapy with no further treatment. These are two anecdotes. Those are two separate patients that give some hints around durability that we think we're going to get additional anecdotes in this cohort expansion.
Ultimately, the next phase of development, which could be designed for those kinds of endpoints in the study after this.
The one patient that was at 17 plus months, that was the CR or a PR patient?
That was a PR. In fact, that patient ended up, we actually learned a lot from that patient because that patient started out at 3.6 mg, got a PR at about -50%. We had dose escalated that patient to 5.4 mg. That patient had some tolerability challenges as well as just some other medical issues. Basically took a dose holiday. This was after about 14 months. Came off, came back on at the 4.4 mg dose, which we had gotten approved, and then dropped tumor burden down to basically -90%. What is really interesting about that patient is they had a strong response. They went off. When they went back on, they immediately responded again. There was no evidence of any kind of resistance to the therapy that they had already had.
We actually also saw when we went back and looked across tumor types, significant responses in some of our patients who'd been on prior microtubule inhibitors. We think there's something going on here mechanistically with our extracellular approach that may allow us to generate responses in areas where patients have typically had some signs of resistance with current therapies that are out there. These are a handful of anecdotes. In early stage drug development, those are the kinds of nuggets that really help you in designing the next phase and foreshadowing, hopefully, what could be strong clinical data.
Just to clarify on the CR patient, how long was he on therapy before he left?
I think he was on about five or six months. We have to double-check that. If I'm wrong, I'll send you a note to clarify.
In this way. Now, I get this question a lot. How are you thinking about the market opportunity in the different lines of therapy where MICVO could potentially be used in clinical practice?
Yeah. Head and neck cancer remains the sixth largest tumor type in oncology. Having been fairly underserved for the last 20 years, now it's great for these patients that they're starting to see a variety of options here as Merus and Bicara continues to develop, we and others coming in with ADC approaches, et cetera. I think you're going to see lots of different segments start to sort out. I think the HPV positive negative vis-à-vis the Merus and Bicara is a natural one. We see a lot of folks that are slugging it out for first line between Merus and Bicara. I think Pfizer now with their PD1LV is looking to attack the first line setting. What we hear from PIs and KOLs is that there's still a huge amount of unmet need in the second, third line setting.
For us, I think the fact that we keep going back to the design we have put in place for monotherapy where we are testing two hypotheses. One is our ability to compete in the front line, basically the earliest line therapy with Merus and Bicara, and the second being to come right behind them. Either of those segments is actually pretty large and attractive. We have not guided on what we think those estimates or what those market sizes look like. Either of those provide us, I think, a material market opportunity just in head and neck. We may, through the dose escalation where I mentioned earlier in combo where there are other tumor types, those other tumor types might give us early signals of where we could have potential additional life cycle opportunities downstream.
In the case of the monotherapy, we already know there are other tumor types as well. Depending on what happens in the macro, what happens in the funding landscape, our head and neck data, which is where we are 100% focused right now, could also open up further franchise opportunities later on. In the meantime, I think head and neck, whether it is first, second line, or second, third line, both of those are pretty attractive opportunities for us.
Got it. I believe you have free drug supply agreement with Merck, which Merus and Bicara does not have. Curious if Merck has any input on your overarching strategy across multiple tumor types and if they're seeing the data since it's open label.
Yeah. We were very fortunate to get that supply agreement with Merck. They have a very high bar for providing the supply for free for collaborators. We were thrilled. They have been a wonderful partner to work with so far. We enjoy a really trusted relationship with them. Typically, around these agreements, the infrastructure is set up. There is kind of a joint steering committee where the data is viewable by both parties. They are able to see the data as it unfolds. I think without sort of—I cannot speak for them as I was not in their decision meetings, obviously, where they decided to approve the agreement with us.
My hypothesis is that they were intrigued by the biology, intrigued by the mechanism and the potential there, especially after they saw an MMAE-based ADC in the case of PADCEV combining with KEYTRUDA and the outsized response rate there. I think there were a number of sort of pattern recognition clues when they looked at the PADCEV plus KEYTRUDA experience that we were maybe signaling could be something that we could see in the setting of head and neck or some of the other tumor types as well. We take their input and their ideas seriously and vice versa.
Got it. I believe you're doing some exploratory work to better understand responder versus non-responder patients. Can you talk more about the strategy and how you're using it for recruiting patients into the study?
Yeah. So it's interesting because of the nature of our target, a biomarker is not necessarily a natural thing that we will find. I mean, we're doing work with our translational team to understand what are the conditions that lend themselves to responses versus non-responsives, and how can we leverage that information to select patients going forward. We ultimately think that the IHC, the target expression, is an important component to that, but unlikely to be the only component. That better characterizing the nature of the stroma, the extracellular environment, including protease concentration, stromal maturity, there's several different stages of stromal maturity. We're also looking at AI-enabled digital path that's helping us envision at the sort of hypercellular level the morphology of what's going on in the stroma. All of those things taken together are giving us significant insights into how this agent is working.
Whether we can ultimately translate that to patient selection or not, I think, is going to be an important insight that comes out of this next batch of clinical data as well as this translational work that I just mentioned, which we are going to be publishing more of over the course of the year. What you can expect from us over the course of the year is deeper insights on the biology and the mechanism in advance of the clinical data, and then the clinical data coming along the timelines I mentioned earlier. I think that is going to give us a pretty fulsome view of what is going on here.
Got it. Maybe to wrap up, what is the cash position on the runway and then the key events that investors should pay attention to in the next 6 to 12 months?
Yeah. I think our last cash balance was about $105 million, if I remember correctly. Yep. The runway is through the second half of 2026. It safely clears these three catalysts. We have the two catalysts from monotherapy and the one from combo that are coming up. I would guide investors to really keep an eye out for this translational work, the mechanism, and the tumor biology that will come ahead of the clinical data. I think it is going to be very informative, very helpful in really understanding this agent. When we started out, we had novel target, novel mechanism. Double sources of novelty can sometimes be tough to quote de-risk. We are working our way through all of that. It will be a pretty fulsome several months ahead.
Great. Looking forward to the updates. Thank you, Lara.
Thank you.