ALX Oncology Holdings Inc. (ALXO) Earnings Call Transcript & Summary

Okay. Great. I hope that's good to start. Good afternoon, everyone. My name is Trung Huynh. I'm the large-cap pharma and SMID-cap biotech analyst here at UBS. I'm also joined by Ting Liu, who's our oncology specialist on the team. It's a great pleasure to welcome Jason Lettmann from ALX Oncology. Thanks for coming down, Jason.

Yes. Thanks for having us. Appreciate you and UBS hosting here. It's a great venue. So thank you.

Excellent. So there is one point of admin. [Operator Instructions]

With that said, there's some people here that don't know the story. So perhaps, Jason, why don't you give us a quick introduction about the company, and you've just come off 3Q results, give us a summary of the key highlights that came from that.

Sure. Well, so ALX Oncology is an immuno-oncology focused on the CD47 access. We were founded about 10 years ago, really with a differentiated approach to tackling CD47. Since day 1, we are the only CD47 program with a dead Fc, and it's very important as we'll talk about the space has had some challenges. But there's no question that it's a really important target. And I think what we've demonstrated to date in the clinic is a differentiated both safety and efficacy profile. And I think IO right now certainly needs some new options. So we're excited about bringing forward evorpacept now into late-stage trials.

Excellent. And can we perhaps quickly talk about your R&D priorities that you have in the pipeline today? And how are you prioritizing your assets or your asset in different areas?

Yes, sure. So evorpacept is fundamentally a combination agent. It was designed to work with other drugs. So we're testing it across 3 axes, if you will, one being in combination with anticancer antibodies like Herceptin, 2 in combination with checkpoint inhibitors like pembro, and 3 in combination with ADCs like Enhertu. Those are just 3 examples of ongoing clinical studies that we have. And our drug is meant to utilize each of those in slightly different ways in order to harness the immune system to cancer.

And one of the first questions I always get is around the target CD47. Certainly, we've seen a few high-profile failures within this space, especially the blood cancer setting. So perhaps can you explain why you think yours is different here and talk about the safety side of things, how you [ differentiate ]?

Yes. No, I think what we've known since the early days is that CD47 is a very important checkpoint, if you will, of the immune system. And I think what we've also recognized is it's a challenging one. CD47 is very widely expressed. It is one of the primary ways in which the immune system signals to healthy cells to be left alone. That's why it's known as the don't eat me signal. Historically, the approach has been the conventional one. So let's use CD47 as a tumor-associated antigen. So we do know CD47 is more expressed on tumor. However, the differential expression versus tumor and healthy is relatively narrow. And so what has happened is basically using an active Fc antibody to go after that has been challenging. The therapeutic window has been challenging. It's limited the dose. It's come with serious safety issues. And I think when you have a space like CD47, where Forty Seven was acquired by Gilead for $5 billion and then shortly thereafter, Trillium was acquired by Pfizer for $2 billion, a lot of interest in the space. I think it's a validation of how important this target is. But then when those approaches failed with their active Fc approach, certainly has raised the bar for us. But I'd say from day 1, we recognize those as challenges. That's why we have a dead Fc. That's why we don't have single-agent activity, and that's why we rely on our partner, our combination agent to really drive the EV signal. And by doing those 2 things together, I think what we've now demonstrated in the clinic is that it can work, and it can work in a really safe way, which is different than the others.

Okay. I think it's a good opportunity here to just hand over to Ting, and she's going to pursue the key clinical studies.

Ting has the hard questions.

Okay. Maybe as Trung said, so we'll go through the indications. Maybe start with gastric cancer. Can you give a quick summary of evorpacept in HER2 gastric? What are the needs? And what gives you encouragement so far on the data you were seeing?

Yes, sure. So gastric cancer is where we chose to start really to demonstrate the power of combining evorpacept with an anticancer antibody. In this case, we're combining Evo with trastuzumab or Herceptin, that provides the eat-me signal. And really, the reason to go after that indication was to show how it could work in a very tough tumor type. There's no question, particularly in second-line gastric, there's a significant unmet need. And what we're really excited about is demonstrating in ASPEN-06, the first randomized data in CD47. Those 2 acquisitions I mentioned were off of single-arm data. So to have randomizing this data in this space has been huge. And what we showed and shared at end of July was about a 41% ORR versus 26% on control. So we think it's a very strong signal. And what we're seeing in terms of the durability of response of over 15 months on the Evo arm versus 7.5 months on control is really encouraging, right? It's indicative of an IO agent. And by definition, I think the data takes time to mature, right? I think it takes time for an I-O agent to really show its benefits, and that's why we're eagerly awaiting here the next update.

Yes, definitely. Like the OR data is definitely encouraging in there. The one source of debate is that you must be very familiar with this analysis that sell-side is doing. When dissecting patient pool from those enrolled before and after the interim and looking at the latter half specifically, there seems like nondifferentiated ORs in the investigational arm, which has evorpacept versus the active control arm, which is TRP alone. Have you put more thoughts into that?

Yes. We've been doing a lot of homework there. I think what we've seen by announcing the interim data in the middle of the study, I think we commented on at the time when we shared that is that we are actually enrolling patients. We do think that introduced a bias, which resulted in us enrolling a sicker, more advanced patient population. So I think that was one factor. But what we're really encouraged by is when we look at the HER2 expression data, there's no question that it is a really strong signal. Whether you look at it interim, post-interim, full data, our drug is having a very robust effect. And again, it goes back to the mechanism. We need both to block the don't eat me signal, which we do in a potent way, and we need the eat me signal. And when we don't have HER2 expression, i.e., we don't have the HER2 -- we don't have the eat-me signal, our drug will not work. And for us, I think we've demonstrated that. And it's a powerful biomarker because now we have a very clear sense as to how important this is. And again, we'll get into why we're excited about that, but we know now that what we're doing is also very different than any other HER2-targeted agent alone. So all of these things is building conviction in what we have. Certainly, the Street thinks otherwise given where we're trading, but we have a lot of conviction in our molecule. And we've had our heads down for the last few months just really focused on that.

Okay. And on that HER2 expression, which you just brought about and the prespecified analysis looking at the efficacy in patients with fresh versus archive HER2 biopsy samples, you've shown nice correlation there that in those patients with fresh samples, the efficacy signal is more clear. So that's great. But then another source of debate kind of coming from that data, which could be a stretch analysis for sure, if looking at the TRP arm alone. And then there -- like the OR is actually numerically higher in the archived sample versus those in fresh samples. Yes, what could be happening there?

Yes. I think what's most encouraging for us is we've got the question around HER2 expression, which is, okay, great ALX in the patients that are truly HER2 positive, the ones with fresh, your drug seems to be effective. Well, is that just because you're combining with TRS and that's what's driving the efficacy? And the answer to that, and frankly, what's so exciting is that whether it's a very strong HER2 signal, IHC3+, IHC(2+)/FISH-positive, however you cut HER2 expression, it comes out in about the 20s roughly for ORR. But when you look at ORR data, we're seeing a 55% ORR for those patients that are with fresh biopsy and again, indicative of a strong HER2 signal. So we're working very differently than tras alone. We know that. That's a kinase-driven effect versus what we're doing by driving ADCP against the cancer. It's a different mechanism.

That makes sense. And then to our understanding, you haven't made a decision whether to mandate the fresh HER2 sample in Phase III yet? Was that kind of impacted by maybe there are some executional challenges to run a Phase III trial with a mandate that everyone has to have a fresh HER2 [ biosimilar ] coming into the study?

Yes. I mean, there's been other studies that have run that have required fresh. You look at the Daiichi, I think it was the DESTINY-Gastric02 study in HER2 -- looking at HER2 in second, third-line gastric, same thing. They required fresh. And that's because they know that it's a very heterogeneous tumor type that changes over time. Again, they were able to execute on that study. I think for us, what we're thinking about is, okay, how else do we look at HER2 expression? So circulating DNA, looking at ctDNA, liquid biopsy and understanding how HER2 in the blood is impacting our results are going to be very important. And I think that will give us yet another way to look at the story.

Sounds good. And then also on Phase III, like pending the start -- pending the initiation of that Phase III study, I think you're waiting on the final -- sorry, you're waiting on the mature PFS data. So in that case, do you have a bar of success in there? Do you have like a certain threshold of hazard ratio or PFS data -- delta that will make you comfortable to move forward?

Yes. I think for us, it's similar to what drove the Enhertu approval in second line, which is, call it, 1.5 months, 2-month delta in PFS. There, I think they showed around a 6-month median PFS. So again, that really does reflect where standard of care is. There's a lot of room for improvement. So for us, if we can show a similar gain there, I think that would be success. And again, their bar is around 6 months.

Yes, that makes sense. So maybe my final question regarding to the gastric indication that's -- is it still on the table that there could be an accelerated pathway in gastric based on the Phase II study alone?

Yes. I think to keep in mind what drove the Enhertu approval was, call it, 130 patients roughly OUS. So we certainly have a more robust data set, how our data is stacking up with Enhertu based off what we've already shared, I think, is strong. And again, these patients need new options. There is certainly a lot happening in first-line gastric. There's not a lot coming for these patients that have progressed after first line. So we do think that's an opportunity to approach the FDA. Again, the Phase III will be driven off survival. So we want to wait to make sure we have mature data to share with the agency. But once we have that, I think we have a really compelling story to tell.

Yes, that's great to hear. So maybe move on to something that's more exciting for you in near term. So to our understanding, your partner, Jazz, is running this Phase I/II of evorpacept with their HER2 biparatopic antibody, the zanidatamab. And then there will be some data at [ zanidatamab ] in December this year. So what will be the data there? And what's the patient -- what's the size of the patient population we are looking at from that presentation?

So this is a study that we initiated when zani was with Zyme and now that it's been partnered with Jazz, they've continued to execute this study. So for us, what we're going to be sharing is the data from the breast cancer study we've been running. So this is an advanced breast patient population. Median prior HER2-directed therapies is around 5. So a heavily pretreated population that have seen a lot of HER2-directed therapies. And that gives us the perfect place to run a proof of concept because we know our drug should have relatively little, if not any, single-agent activity. Zani in this setting, again, after that many HER2-directed therapies, it's certainly a high bar. So we're looking to see, call it, a 20% ORR or better -- and that is the bar in that space. If you look at the [ SILVIA ] study, for example, which was a chemo plus rituximab comparator, this is a chemo-free agent, a chemo-free combination. And again, we just demonstrated what we could do with Herceptin in gastric, so to now do -- what we are hoping to do the same in breast with zani, which is a bispecific, also HER2 targeted, as you know, I think it's an important proof of concept.

Okay. You do have a HER2 low cohort in that study. Will you present some -- like data from that cohort as well?

Yes. We'll have about 40 patients across the HER2 high, HER2 low, and then there was a third cohort, which is kind of a basket study of different indications. We're really interested to see does this phenomenon that we're seeing in ASPEN-06 with HER2 expression carry over to breast. So what does the data look like between high and low. And I think that, again, will give us a really great data point. And certainly, gastric, I think, is an interesting commercial opportunity, but the breast opportunity, of course, is much, much more significant. And we think having both the 06 data with Tras plus data is a nice entry for us into the breast opportunity.

Yes, definitely. So within the HER2 low setting, if we could ask one more question. Are you just thinking of having something to replace Enhertu or to compete with Enhertu in that setting? Or are you looking to add additional line of care post-Enhertu?

Yes. I think what's interesting with Enhertu, I mean, of course, it's a great drug with great data. And I think most folks believe it will be first-line breast here soon. And that bar is extraordinarily high. But the question that I think is interesting is what comes next? What -- for those women who advance after Enhertu, what do you have? And I think it's going to be a whole slew of HER2-directed therapies. And what we bring is something different. We bring the ability to use HER2 as more of a handle and drive ADCP against that. And again, we just showed that in 06 in terms of how we can bring something different to the table. So that's where we see the opportunity. For us as AOX, I don't think we would want to compete with Enhertu in the front line, but there's going to be plenty of patients, unfortunately, that progress and to bring something that is new and different than another HER2-targeted agent is pretty exciting.

Can I just quickly jump in? If you fail a HER2-directed therapy in the first-line setting, does that change the expression levels of HER2?

It's a great question. Sometimes it does, but for a large percentage of these patients, it does not. And that's what's interesting about this patient group we'll be looking at, again, roughly a median of 5 prior HER2-directed therapies. We're going to confirm their HER2 expression at study entry. And so those patients are still expressing HER2, but yet they've failed, right? And so that will be really interesting for us. And again, I think that's what's needed in the breast space right now is something that's different.

Sounds great. And within breast cancer, maybe also, within the HER2-positive population, there's also investigational trial going on looking at evorpacept with Enhertu combo, the I-SPY study. So are you equally excited about like evorpacept opportunity with that combo?

Yes. No, we are. I mean, again, it's 3 shots on goal here. It's with antibodies. It's with checkpoint inhibitors and with ADCs like Enhertu. So we are excited about that. Again, fundamentally, Enhertu is an Fc active antibody plus a payload. And so we should be able to be additive there. I think as we've talked about, that study is not one that we're running and controlling. That's part of the I-SPY consortium. So we would like that to speed up a little, but yes, we're definitely optimistic. We just have the Jazz data coming here first.

Okay. Yes, that's great to hear. So the breast cancer is definitely exciting, like data from a whole new brand new indication. So maybe we jump to head and neck. So the ASPEN-03 and 04, everybody is looking into that readout. It's very important. It's a second indication where we'll see data from randomized controlled studies of evorpacept. So what are -- and those 2 parallel studies, they have slightly different combo, evorpacept with KEYTRUDA with or without chemo in slightly different patient population as well in first line. So what are your base assumptions into the readouts of those 2 studies?

So this is testing the CD47 signals, both to the macrophage as well as to the dendritic cell. And so this is really testing the latter, which is can you block that signal to dendritic cells as well as use pembro to block the T cell signal in order to essentially release 2 breaks. We are encouraged by this because of what we saw in the Phase Ib. Again, we are adding 2 IO agents. And so what we saw in the Phase Ib was what I would call a pretty modest improvement on ORR as well as PFS. But then on both of those cohorts in head and neck, we had an over 80% 12-month overall survival. That is very consistent with what pembro saw, right? The KEYNOTE study showed the same. They missed on ORR, missed on PFS and what carried the day was their OS. And again, our mechanism is very similar. So what we're excited about here is we have over 300 patients coming across ASPEN-03 and ASPEN-04. As you mentioned, Ting, those are frontline studies. ASPEN-03 is Evo plus pembro versus pembro alone, randomized 2:1. And then 04 is pembro/Evo chemo versus pembro plus chemo. I think in ASPEN-03, when you look at pembro mono, there's no doubt that, that should be easier, if you will, to move the bar on ORR versus 04, where you have a pretty heavy chemo payload involved. I think the bar will be higher. But we're excited about it. Again, these studies have been running since 2022. It's been a long time in the making. I think with all of the new agents coming in head and neck from [ Meyers ] and Bicara, which also have great drugs in the making. I think it sets the stage well for what we have. Again, those studies are large. They're designed with registrational intent. And given how long we've been at this, I think we have a shot at having something really interesting in terms of survival.

I assume everyone will be looking at the 12-month OS data in there at the top line. One thing is you did push out the top line readout a little bit from late '24, early 2025 to now first half 2025. Was that decision solely based on like the [ event ] collections like a little bit slower on the OS and like you just -- yes, especially on the 12 months, how many events you're collecting there? Or has that also to do with other key secondary points your endpoints you're looking that you thought will be important to present to investors at the top line?

I think, again, going back to what I opened with, this is an IO agent. And combining with pembro mechanistically, we're going to need to see the tail. And if you look at how long it took for the curves, if you will, to separate and KEYNOTE, it was a while. And so for us, it's similar. We are paying close attention to the number of events. We want to make sure we have mature data and don't move too quick. And this is the reality of designing a drug that is immuno-oncology based, right? I mean it just takes a while. So that was the nature of the time line adjustment. It wasn't anything around accrual or patient follow-up. It was just looking at what do we need to see in terms of maturity.

Got you. Okay. Sounds good. And then are you still thinking of potential accelerated path in head and neck as well?

Yes. I think with that size study, it certainly opens the door. And I think one of the reasons for us to wait, again, we know this dynamic in head and neck is very real in terms of the lack of correlation between ORR and survival. And so to be able to approach the agency with randomized data here and have it, frankly, be as mature as we're hoping to have, I think it will make for a really interesting conversation. And we know the agency hasn't seen a data set like this probably since the [ week 10 ] study. So I think it's a good setup for us. It's certainly great that the world is more educated on the need in head and neck, and we're excited about it. It's been a long time in the making. So it will be good.

Yes, that's great to hear. And if we may ask could you provide more granularity of when will you see that data? Would that be about the same time as ASPEN-06? Or would you rather to have everything settled down for ASPEN-06 than present the data from...

Yes. I don't want to get in trouble. So we're saying first half. Again, I think coming off how things were perceived and the reaction in July, we really wanted to just reset the deck a bit and again, give us some flexibility to ensure this data is mature. So we'll see how it plays out. Again, I think we have with the breast data coming, 06 coming ASPEN-03, ASPEN-04, we'll talk about the readout we have with ASPEN-07, but we have at least 5 data sets coming here in the next 7, 8 months. So really excited about that number of catalysts, particularly against our balance sheet, which should take us all the way through all of this.

Okay. Sounds wonderful.

Not a good nonanswer for you.

I know -- yes, I think you did mention at some time point that the data will be presented at a medical conference. So people are thinking there's ASCO GI and then ASCO, which like perfect timing for data at 2 different conferences.

Yes. I think we're always shooting for medical meetings and that's the plan.

Got you. So moving on to urothelial cancer in ASPEN-07. You will also provide some updates from that study next year. What will we see as updates there?

So we shared at ASCO this past year, a really early data look here. We are combining evorpacept in this trial with PADCEV and second-line bladder. Of course, PADCEV has been an amazing drug, and there now is a significant unmet need in second-line bladder. What we shared at ASCO, again, was early data. We were right in the middle of enrolling that study. So at the time, I think it was an early look into some really encouraging ORR, but still had a lot of patients that frankly hadn't even reached first scan. So fast forward to where we are now, I think the goal here is to let that data also mature, get to ORR, ideally get to PFS and be able to come out and share it in a really robust way what we have. Excited about that. That's the first study to ever read out with a CD47 blocker plus ADC. And yes, we're hopeful. I think it's certainly of interest given how many new ADCs are coming and the potential role that those will have in changing care.

Most of the responses we were seeing at ASCO this year, like they were unconfirmed. And when we see the mature response OR data next year, are you still expecting like the same range of OR, like something close to 60%?

Yes. I mean, so when we went out to do those studies, we really wanted to see a 10% delta. I think we tried to give everyone an early look at how things were shaping up. And so thus we had a number of unconfirmed responses. I think the goal for the next data cut here is to get through everyone to confirm. And again, ideally, PFS is what we want to see. I think the other thing we've shared is that we know that these patients were more advanced, heavily metastatic patients, high percentage of liver mets. So we do think that, yes, our patient group is what you'd expect of a study that enrolled right in the middle of PADCEV moving frontline. So that's also in the backdrop. But again, our goal going into next year is to wait, get the mature data and then when we have it, come out and share.

That sounds great. So maybe last question from my side is, do you have any updates from the ScalmiBio's ADC platform?

Yes. So we've been, I think, keeping that relatively quiet, we acquired Scalmi. It's probably been 3.5 years ago now. Scalmi at the time was a really early-stage ADC platform that was working on a couple of things. One was a conditionally activated Shield technology, so to deliver an ADC more effectively on target. And then they had a series of really unique linkers and payloads, et cetera, that we thought was interesting. We've recognized the world wants to hear about evorpacept at this point. So our strategy there is just to be quiet and let things bake. But we are fortunate to have Jaume Pons as our Chief Scientific Officer and President. He ran Pfizer's IO Group. He has a ton of ADC experience. His first ADC was a TROP2 targeted ADC many, many years ago, and the R&D team has been busily working on that. So we're excited about it. I think it gives us another option beyond evorpacept. And again, our strategy there is just to be patient. And I think once we have something firm to say about a time line to the clinic, then we're going to say more.

That's great. Looking forward to hear something...

To close then, it's -- Ting and I were talking before, and it's unusual to see -- when we look at large cap pharma companies, we look at SMID-cap biotech companies. It's unusual to see a company with a lot of shots on goal. You mentioned 5 big shots on goal in 2025. I've been asking companies we've had on stage today. For next year, hopefully, you'll come. So what I want to do is I want to ask you what for 2025, if you're here on stage, do you want to achieve the 3 things that you want to achieve for 2025? And then next year, when you're here again, hopefully, ask you right off the bat, how are you going?

I mean, so this time next year, we're going to have gone through another, I think, step change as a company. This first -- this last year, again, I've been on the job about a year, was really about delivering proof of concept, proof of mechanism. And I believe we've done that. Next year, you look at ASPEN-03 and 04, ASPEN-06, ASPEN-07. If we're sitting here and we've -- and all have worked and we've been successful with all 3 of our shots on goal, well, what does that mean? I think we are now looking at a relatively quick path to registration with 06, potentially talking about an accelerated approval in head and neck, advancing our ADC programs into randomized studies. And I think, yes, I mean, I think it will be really interesting. We are now the only CD47 that's sitting where we are. And really, because of how this has shook out with the field, we have very, very little competition. So if we're sitting here next year and executed on all of that with essentially no one else in the space, it will be exciting. So that's the goal, a lot to do. As you know, we've been really pretty heads down. I think this is the first investor meeting I've been at. And we're excited. I mean we have a lot of conviction in this drug. So excited about next year.

Excellent. Hopefully, we'll talk again this time next year.

Yes. Sounds good. Thanks for having me. Thank you. Appreciate it.