Hi, everyone. Welcome to our next session with Pyxis Oncology. My name is Li Watsek, a biotech analyst here at Cantor Fitzgerald, and it's my pleasure to introduce Lara, CEO of Pyxis. Great to have you with us. And I think you know you guys have a very unique ADC story. We've got some exciting data coming up, so I would love to turn it over to you to give us an overview.
Excellent. Thank you, Li, for having us here, and to all the Cantor colleagues. It's a great start to the fall, so we're happy to be here. As Li mentioned, Pyxis Oncology is an ADC-focused company. We have a lead asset called MICVO, which is an extracellular targeting ADC. It targets extracellular domain B, which is a splice variant of fibronectin, and the target is expressed ubiquitously across a variety of solid tumor types and negligibly in normal tissue, so it's a highly specific expressing target. This ADC was originally designed by Pfizer. It was optimized in this construct for potency, permeability, stability, and we took it on at development candidate stage, brought it into the clinic in the first quarter of 2023, had our part one data readout last November, and are now in clinical trials for both monotherapy and combo therapy with Pembro, focused in head and neck.
Hey, great. And I know you guys sort of picked the indication of head and neck, and many investors are very familiar with the landscape right now with some bispecific showing some great data. So I wonder if you can just talk about where you can potentially fit in and how you look at the landscape of head and neck, obviously, is evolving quite rapidly.
Yes. So our phase I part one data that we read out last November was out of a basket trial that tested nine different tumor types, of which we saw significant tumor regression in six, of which head and neck was the strongest signal with a confirmed ORR of 50% and 100% disease control rate. That was in six patients at the therapeutic 3.6-5.4 dose range. As you mentioned, Li, the head and neck space has intensified with a number of entrants over the last couple of years, which has been terrific for patients. 20 years with not much. I think Cetuximab was approved in 2000 or 2001, so they've been waiting a long time. It's been really terrific, actually, to see these new bispecifics coming in, targeting EGFR sort of plus with Merus and Bicara leading the way, shaping the space, creating engagement.
We actually think it's a terrific place to be with the work that they're doing to raise the profile, given that we have seen such promising data in our own phase I part one with that 50% confirmed ORR compared to the Merus 37% confirmed ORR in their own phase I and compared to Bicara's 0% ORR monotherapy in their own phase I. So we think it's a competitive place for our agent. We think that there's certainly a place for a novel ADC. And we saw in our phase I data last year that we had responses in patients who had been resistant to prior treatment with Cetuximab, as well as prior platinum, PD-1, other ADCs. So clearly, there's still a need, even as they're making good inroads and demonstrating great data.
The other thing I would just say specifically about those bispecific competitors, we are seeing in the data sets that are out there some differential in their performance or response rates for HPV positive versus HPV negative patients. We have not seen that in our data set as of yet. We saw responses, including a CR, last year in both HPV positive and HPV negative patients. And there's nothing in our mechanism or our target that suggests there should be any reason to see a differential in response based on HPV status. So we also think that as this market takes shape, there may be segments that Merus and Bicara don't go after that are perfectly appropriate for MICVO. So lots of opportunity, I think, for patients.
I guess going back to the phase I data from last fall, obviously, Lara, like you said, small patient numbers and sites across a dose range 3.6-5.4. So how many doses are you guys testing right now, and when would you be able to pick a dose?
Yeah, great question. So we, as you noted, had tested a number of dose levels in dose escalation last year. I think it was nine or 10, and then there were nine different tumor types. So it gets very segmented from a data perspective pretty quickly. But the signal in head and neck at 3.6 through 5.4 was very strong. We had dosed a total of 77 patients as of the data readout last November, of which 39 were at 5.4, which was the highest clear dose. That was across those tumor types. So we are dosing at 5.4 in this cohort expansion. And we are interested in, obviously, pursuing, as part of our development plan, the dose optimization approaches that are required from Project Optimus.
How exactly we'll do that, in what sequence, will be part of our next data disclosure as we guide on the next stage of the development plan, but we do expect to have a robust data set in head and neck out of this 5.4 cohort expansion.
I guess just from that data set, do we have any sense of the durability of response?
Yeah. So it's tough always in phase I trials, not just for MICVO, as you know, right? If you really want to get durability data, you've got to give a year or two, and we just don't do that in the setting of phase I. But what we can do is identify any early enrollers, particularly early enrollers that are early responders, and at least tell some vignettes and some qualitative insights into durability. What we did see in the phase I part one data set that was released last November was a median time on therapy of our head and neck patients that was around 16 weeks or so, whereas the average that you tend to see in phase I studies is closer to 65 or 70 days. So that gives a bit of a hint.
In one of our responders in the head and neck patient population, we actually have a swimmer plot that we had disclosed last year. I think that particular patient, when we last updated on that data set last February, had been on therapy 16 or 17 months. We also had a confirmed complete response, which that patient ended up, after achieving the complete response, moving to a different region and it was complicated giving that patient continued dosing so they ended up going off treatment, but held that CR for six months while they were off treatment so I raise these as anecdotes since I can't give you statistically significant durability but I think in our next data disclosure, we hope to have a rich set of anecdotes that can at least, by proxy, give some insights into potential durability that could come ultimately with a pivotal trial.
Okay. As you move from dose escalation to the expansion cohort, I understand the phase I heavily pretreated patients, and now you're perhaps enrolling patients less pretreated. So maybe talk to us about any differences in terms of patient characteristics that we should keep in mind so that we can maybe extrapolate what we would expect.
Yeah. I think the biggest delta is what you put your finger on, which is the pretreatment status. When we were giving our data for the data disclosure last November, the median prior lines was four. So we were essentially median fifth line for these patients. And now the enrollment criteria for our expansion is second and third line. So that's a material difference. When you're recruiting into a part one with no restrictions on number of lines, you can get patients with 8, 9, 10 prior lines, which we had, who are essentially end-of-life patients. And if you also just think about what's going on in their bodies after that many series of treatments, it can be tough to rev up a response. But even in that setting, we had that 50% confirmed ORR.
So we're pretty optimistic that going into second and third line, where we're moving into less heavily pretreated patients, that that 50% ORR will hold. That's our hypothesis, and we'll see how the data validates that or not. I do think that when we look at the bispecifics data in phase I , Merus had a 37% ORR in their own phase I , and that was a median second line. Bicara had zero. That was a median second line. We had 50% median fifth line. So we need to kind of keep that context in mind as we continue to advance.
Okay, so MICVO is targeting EDB, so that's tumor microenvironments. But you don't seem to see strong correlation between EDB expression levels and the response rate. Why do you think that's the case?
Yeah. So it's really interesting. I think a couple of observations. One, with our agent, with EDB being an extracellular target and the ADC cleaving extracellularly, that requires the proteases that cleave the ADC to be in the extracellular environment. So we think that protease concentration may be one of the things that influences for a particular patient, whether they respond or not. Things like the acidic or basic nature of the tumor microenvironment can influence it. So the target, the way we think about it is the target is necessary, but may not be sufficient to fully, for any particular patient or tumor type, predict a response. So we're doing a lot of biological work, proof of mechanism work with our translational team. Some of that will be shared at ESMO.
We have posters coming out there for proof of mechanism, and we have other translational work going on that we'll share throughout the course of the year. We think that will help inform the thinking on the clinical data. So that's one piece of it. And then I think the second piece of it is that we've learned this year from Enhertu and some of the work done by a team at Duke that the bystander effect is maybe more pronounced with ADCs than had been previously anticipated. And so we think that's a key component of our mechanism. And again, as long as that target, the EDB target, is enabling the ADC to unlock and dislodge its payload, that part of the mechanism should kick in just like we're seeing elsewhere.
So I think there's a lot of canonical beliefs that people had held previously about ADCs regarding internalization and direct tumor killing that are just actually not proving to be as critical as maybe we had thought. And an ADC like ours that has those other components, those non-canonical approaches like bystander effect or extracellular targeting, is sort of well-positioned to benefit from that evolving philosophy.
Okay. And I think another thing that's interesting coming out of the data set is if you look at payload-related toxicities, it seems like it's lower when you compare to other MMAE payload ADCs. So I think that's interesting. And also, you have one Grade five event. I mean, how confident are you guys about this is not related to the drug?
Yeah. So whenever there's a new target or a new mechanism coming into humans for the first time, those phase I are incredible learning experiences. We're not improving on a different target where you have a prediction around what's going to happen. You have to learn by experience. And so one of the things that, as we went through the dose escalation, we were paying particular attention to is the categories of AEs that are well-known for ADCs, right? Would we see things that look a lot like our pure ADCs that came out ahead of us, like ocular tox, neutropenia, neuropathy, et cetera? Or do we see other things? So there's a page in our disclosure that looks at our safety profile from the phase I part one compared to the tolerability profile for the approved ADCs.
To your point, we've seen in each of those dimensions, neuropathy, neutropenia, ocular tox. We looked at pneumonitis and cutaneous, namely rashes. We looked either as good as, equivalent to, or better, depending on the dimension and depending on the comparator. I think we're in the field we need to be in. In terms of the actual safety data set itself from last November, we did observe in our patient population a grade five non-treatment-related. This was an extremely sick patient that was dosed at the 8.0 dose level. Probably in retrospect, that patient probably should not have been on a clinical trial at all. The patient had a number of underlying comorbidities in addition to the cancer diagnosis. So we're quite confident in the scrutiny around that, both at the site and by our team, that it was not attributable to drug.
So I appreciate the opportunity to clarify that because in this age of quick internet headlines, sometimes those things can get lost in translation.
Yeah, I appreciate the color. And then you're going to have data coming up, I think, in the second half of this year. So we're going to see some pretty nice update from the expansion cohort very shortly. So I understand you're going to maybe share around 20 patients of data. Maybe walk us through types of data, the expectations for that.
Yeah, so we have two programs going on now. One is monotherapy, second- and third-line, as we mentioned earlier, and there's actually two arms to that. Arm one is looking at the PD-1- and platinum-resistant patients, about 20 patients enrolled with data expected second half of this year. The other arm is EGFR- and PD-1-resistant patients, and that data is expected first half of next year. That's also 20 patients. The reason we chose to do those two arms is we wanted to really suss out how our agent performs against today's standard of care, which is platinum and PD-1, and what we think tomorrow's standard of care will be, which is EGFR-targeting agents.
So this way, we have data sets that show how we can compete with the emerging landscape, but also follow right behind the emerging landscape and go right behind Merus and Bicara by having that data set in EGFR resistant populations. So it's a really nice place to be from a drug development perspective because there's a number of ways we can actually participate in this market, and they're actually unique segments. So from an evaluation perspective, you can look at different market size for competing against Merus and Bicara and alternative if we're coming right behind them. There's also nothing for patients right behind Merus and Bicara now. So in that scenario, we are adding to the treatment opportunities that are out there. So we think that the data will inform the next stage of our development plan in mono.
And then the second program is in combo with Keytruda. We have a supply agreement with Merck. We're very grateful to get that for free, the supply of Pembro. And we are looking at preliminary data out of that combo study in the second half of this year. That is in dose escalation. It's a part one dose escalation study. So we have multiple tumor types in the escalation phase, and then we're backfilling in head and neck. So in general, I think 20 is usually what you want to shoot for for any kind of data set. I can't promise exactly 20 head and neck patients because other tumor types are included in the escalation, but we'll have some preliminary view on the combo data second half of this year.
I guess for the cohort one monotherapy data, what do you think you need to achieve in terms of response rate and how much durability data will we get later this year?
Yeah. So I think Merus and Bicara are setting the standard on that, and in particular, Merus in the monotherapy setting because they had that 37% confirmed ORR out of their phase I-II study. So we look at that as what we should shoot for and shoot to clear is that 37%. And that would apply to the data catalyst that's coming up, the platinum- and PD-1-resistant population. The second arm, the EGFR-resistant arm, since there's nothing really there as a backup from PIs and KOLs and experts in the field, they're guiding us that a 20%-30% response rate for that population that is resistant to Merus or Bicara would be a win. So we see that as two differential expectations in terms of the hurdle of what you'd want to see from a data set for each of those monotherapies.
And then, in terms of durability, yeah, we just can't deliver beyond the anecdotes, but we will comment on maybe some ways to interpret some of those anecdotes and our ORR data on how we might be able to, or investors may be able to, infer durability until we can actually deliver it in a longer-term trial.
Would you be sharing maybe swimmer plot?
We haven't committed to any particular views for our disclosure as of yet, but we did include the swimmer plots in our data disclosure last year. So I think they do tend to be pretty standard views, but we haven't worked out specifically yet what we'll share.
Okay. I guess for the second cohort, maybe just clarifying that. So it's supposed to be EGFR and PD-1. So that includes the Cetuximab, and also patients can be treated with Merus and Bicara's bispecific.
Exactly. So we allowed ourselves a little extra time to recruit that cohort because we're trying to actually get Merus and Bicara clinical trial participants into that data set. The Cetuximab patients are easy to get because they're global, but we are targeting very specific sites that have Merus and Bicara trials where we are also present, but also we're targeting them for referrals to other sites if they have Merus patients or Bicara patients who have started to progress. So it'd be great if we can get a handful of those into that second cohort so we can explicitly share how our drug is doing post those new EGFR agents.
Okay. Have you said, I guess, how many patients will be treated with Merus and Bicara drugs in cohort two? I understand total patient number is around 20, but is it 10 patients?
We haven't tried to legislate that split as much just because it's so hard to predict. If you have even just one or two PIs who are pretty excited to send their resistant patients over, it could change it. So we're hopeful we'll get at least a handful.
I guess, what can you say on the safety front that you observed so far, maybe just at the high level?
Yeah. So in the phase I part one, we had focusing on the 5.4. I think that we hadn't observed grade three, grade four neuropathies and ocular tox. We had some neutropenia. We had some cutaneous lesions. Again, we looked at all of this in the context of how does that compare to other approved ADCs. And as we continue to dose patients in this expansion, we monitor all those ADC-specific types of AEs very closely. So I think that's a helpful way for us to evaluate the potential of this agent relative to its pure ADC set, but also, of course, relative to where the emerging profile is going in the head and neck space by the bispecifics, et cetera.
Okay, so we talk about the bar for the monotherapy. What about for the combination?
Yeah. Again, I think there, Merus and Bicara is not far behind, has set the bar in combo with Pembro. It's about 65%. So it's a pretty high bar. And again, we're thrilled to see that for patients. If we step back and think about our mechanism, our biology, that level, that target doesn't faze us because our mechanism is three-pronged. We have direct tumor killing. We have bystander effect, and we have immunogenic cell death locally in the tumor microenvironment. If we look at combination with Pembro, you can imagine that third part of our mechanism synergizing really nicely with Pembro. We've seen that story before in Padcev plus Keytruda in bladder. So if you step back and think about what's similar to Padcev to our ADC, it kind of starts to look like a similar setup.
So we're very keen to see how that combo will perform when we get through the dose escalation to the identified dose level in combo compared to what's out there.
Okay. So it sounds like later this year, we're going to get the data from the combo cohort, maybe response rate, some sort of durability.
The combo data later this year will be preliminary out of dose escalation, so it'll be primarily safety and dose level, but some insights into efficacy and potentially a mixture of tumor types because we're also recruiting sarcoma, breast, et cetera, into the escalation because that allows us to speed through escalation with more tumor types, but then focus the backfill on head and neck, so we'll try to get as many head and neck patients in there as we can, but it may not be exactly 20.
Okay. And then in terms of the bar for PFS and OS and Merus and Bicara also set a pretty high bar. So is that sort of the benchmark that you guys have to go against?
Yeah. We're looking at them as setting the bar there too, and unfortunately we just being as early as we are don't have the clear kind of comparator to them just yet in terms of how we shape up on that dimension, but that will come in time. We do have, again, some of those anecdotes like the patient that was on for 17 months, the patient that had the CR that was off for six months and the CR held. Those kinds of things, I think, can help us infer, but we won't have that full data until we probably are in a pivotal stage.
I wonder if we can just maybe comment on patient enrollment. It sounds like some sites may have to choose between Merus and Bicara, but for you guys, you don't really have to compete with an ADC.
Yeah. Yeah. It's a great question. So we definitely have seen sites either fall to Merus or to Bicara and that has not precluded them from having Pyxis there. So we are at sites that have each of those studies. And of course, they're at many, many more sites than we are because they're enrolling many more patients than we are right now. Our strategy with enrollment is to really try to have a mixture of subject matter experts like head and neck experts in our case and some general phase I sites where we can source multiple tumor types to help speed through enrollment. We used the global footprint that we had established for our sites last year, which was both U.S. and Europe based. For the monotherapy cohort expansion, we've built upon that footprint and expanded it.
And then in terms of combo, a number of those sites that were in mono are also in our combo study. And then, of course, there's some new sites there as well. So it's a good mixture of experts and generalists.
Okay. So I guess right now, Merus and Bicara are really competing in the front line. They're enrolling phase III trials. And obviously, it seems like you guys want to do maybe the second line, but you also have Pembro combination potentially can go to front line as well. So as you think about potentially enrolling second line versus front line, do you see maybe one setting is more challenging versus the other or maybe it's a time sequencing issue?
Yeah. Yeah. So I think it's probably a little bit of all of that, meaning I think there's definitely a sort of a fight for first-line patients between Merus and Bicara, but we are also seeing that sites are sending first-line patients to us as well in combo. So we're not getting boxed out of enrolling first-line patients in combo because of Merus and Bicara in second and third line for mono. We're also not really getting boxed out there either. Partly, I think those trials are more mature. So they're farther along. The sites that have enrolled us into the monotherapy, again, there's a lot of sites that were already known to us. So we had an execution advantage there. So that really hasn't. Competition for sites hasn't really been a large issue for us.
Okay. So, Lara, I think in the beginning, you mentioned about the HPV status and it sounds like so far you guys haven't seen any sort of difference in terms of response based on the HPV status. So I guess if Merus or Bicara they just maybe got approved in certain subset of patients, does that change sort of the comparator arm if you guys were to go head-to-head with them?
Yeah. Yeah. No. It's a fascinating study, this competitive landscape, right, because there's a lot of decision points that are going on simultaneously with a lot of players. So I think that when we put our data set out or both of them, right, what we put out in the second half of this year and then next year, we'll include more explicit guidance around trial design and how we're thinking about some of those topics. There's a little bit of sort of projection, right, on what we think is going to happen with Merus and Bicara into this space.
But I'd say sort of unequivocally, if we or any competitor can get a registrational study started ahead of their approval, that gives a lot more flexibility in terms of what that design will be than if you're getting out there after they've already sort of carved their niche. So there is a window, whether it's us or whether it's somebody else in the near term to advance. And I think maybe that was one of the reasons Pfizer jumped in with their PD-L1 announcing registrational trial plans. They announced that earlier this spring. So people are moving swiftly.
Okay. We're almost out of time. So maybe just last question. What are the top two or three priorities for you guys?
Yeah. Execution, execution, and execution. It has been the year of execution, getting sites up, patients enrolled, data generated, data cleaned, quality analytics, all of that. We're just continuing to do more and more of that. We don't get too caught up in what's going on at the FDA or this or that. We can't control that. We still hold an empirical belief that a good drug will declare itself. A drug that declares itself is developable. It's commercializable. It has value. And regardless of what's going on, those things will be true. So our job is to help the drug reveal itself. So we spend a lot of time on those core drug development fundamentals.
Thank you so much, Lara.
Thank you.