Hi, everyone. Welcome to Day 2 of our Cantor Healthcare Conference. My name is Li Watsek. I am a biotech analyst here at Cantor. I am very pleased to welcome our next presenting company, ALX Oncology for our fireside chat. With me today is CEO, Jason, and CFO, Harish. Maybe to kick us off, I would turn it over to Jason to walk us through the story.
Sure. Well, thanks for having us. Appreciate it. Always a good meeting. Harish is our CFO today, but given we do not have Barb, our CMO, he is also here for any super hard
Top pressure.
clinical questions. No, pleasure to be here. It is an exciting time for the company. Appreciate all the support from Cantor over the years. As a quick background, we are developing two programs, both we think have the potential to really change care in oncology. Our lead program is in CD47, evorpacept or EVO. We have been pursuing that program since the company's founding. I think the story with CD47, like a lot of things in biotech, you could probably write a couple of novels on. Certainly, a lot of ups and downs and changes in the space and failures. CD47 historically has been a tough target.
The reason it is so hard is because it is how our immune system signals to essentially be left alone. It is how our healthy cells communicate. Cancer hijacks this, and because of that dynamic, it has been a really tough target to pursue. There's no question that it's an incredibly important target, and for us, what has been unchanged is our approach to that target. Dating back to the company's founding, we took the approach that we really had to separate the "don't eat me" signal, which CD47 is known as, from the "eat me" signal.
That approach of using EVO with a dead Fc to block the "don't eat me" signal, and then combined with an Fc active antibody to provide the activation and the targeting, is how the company was created and what we're still pursuing today, and what is really, we think, built into a really promising program. We can talk about what's changed, but I think what's really important for this program, and frankly for IO, is our ability to predict what patients will respond. That's been a real shift in our approach, really. We're now focused on bringing a targeted therapeutic to patients.
Our second program, which we'll also get into, is ALX2004. This was a homegrown, internally developed effort out of our labs in California, really led by a strong group of protein engineers and chemists who have put a lot of drugs on market. ALX2004 is an EGFR-targeted ADC. Certainly, EGFR is a very well-validated target, and it's also been a tough nut to crack for ADCs. Our approach is very different. We're the only ADC that we're aware of using matuzumab as an epitope. We think that's going to give us some unique advantages. We also spent a lot of time optimizing the linker and payload architecture, utilizing a topo-based payload to really put an ADC together that we think has the potential to drive strong potency as well as a wide therapeutic window.
That's where we're at right now. Again, exciting time for the company, and happy to go through some questions.
Yeah, great. Maybe let's start with CD47. Jason, as you mentioned, we've seen some ups and downs in this space. But right now, it seems like you guys have found a pretty compelling use case in breast cancer, and you also have some biomarker insights. Tell us how you got here.
Sure. If you look at where we've combined EVO with an Fc active antibody, we have five different clinical data sets to support that mechanism. We've combined with Herceptin, we've combined with a HER2 bispecific zanidatamab, we've combined with matuzumab. That whole body of data, including a randomized study with gastric, has really supported the drug's activity. As I mentioned, I think what we've learned from our global ASPEN-06 study, our gastric study, as well as the study that we did with Jazz, with Zani, is that CD47 is also a very strong and predictive biomarker. It's been known for a long time and well-published in the literature that CD47 high is a negative prognostic.
Patients that overexpress CD47 typically do poorly. What hasn't been really elucidated is can you use it as a biomarker? Both in our gastric study, our HER2 positive gastric study, as well as the HER2 positive breast study that we did showed that the patients that had the best response to our drug were CD47 high. Again, I think it's really important, as a former investor, I think one of the things that always scared us about IO is there's been a lot of failures. Whether it is IDO, STING, you could put a long list up, there is really only one PD-1 story.
Yeah.
To me, it is not that those pathways were not active and the targets were not valid. It was how do you identify the patients that can respond best? That has been the shift to our story. I think that insight is really important for our development going forward.
Jason, maybe you can talk a little bit about your ongoing phase II ASPEN-09 study, then you will be sharing data next year.
Yep.
I guess in the CD47 high patient population, what would be a good outcome in this HER2 breast cancer population?
Sure. I think some of the data we have shared over the past year really is built off our ASPEN-06 data. That data in gastric, again, showed a really compelling response in the patients that were HER2 positive and CD47 high. We had an almost 40% delta in terms of ORR. PFS hazard ratio of 0.39 in a randomized setting. That then carried through to what we saw in our HER2 positive breast study. There, in combination with zanidatamab, again, small data set, but of the five responders, all five overexpressed CD47, including a CR. That segment is what we are targeting with this study.
If you look at the landscape in HER2 positive breast, it is changing, and changing pretty dramatically. HER2 is now moving to the frontline. HER2, as everyone knows, is an incredible drug, and is now firmly will be, and is, standard of care in the frontline. What happens next for those patients after progression is still really TBD. What is known is that those patients are not doing well. The real-world studies that we have seen would suggest a 15% ORR and maybe 3-4 months in terms of median PFS. Just at ESMO, there was a real-world study that supported that of over 600 patients. Again, that showed around a 14% ORR and a few months of PFS. That sets the bar for us.
Again, our goal is to focus on CD47 high patients that have progressed on HER2. We know that it is a sizable population. It is probably around 20,000 patients in the major markets, if you will. That is our focus with this study.
For the phase II study, you guys will have both high and low patients.
Exactly
Included. Okay.
I think that is really an important point. This study that we are running is a single-arm study.
The advantage of what we're doing and the targeted approach is we're going to have patients that are both high and low. So although it's single arm, we should have subgroups in our study that perform differently.
For us, a really important comparison will be the CD47 high patients versus the low, and I think that delta, if you will help inform what we do going forward.
Jason, you mentioned it's a single-arm study, and you guys are testing a triplet regimen. How should we tease out contribution of components here?
Yeah, I think, again, the benchmarks are really pretty well-established. For patients here, they've seen in HER2, a lot of them will have seen another HER2-directed therapy, maybe a TKI, and then enter our study. It will be a mix. But what we know and what we've, I think, gathered from a lot of the real-world evidence is that it's a low bar. When you're talking about 15% ORR and a few months of PFS, I think there's a lot of room for improvement. We certainly aren't expecting to go five for five and put up 100% ORR, but I think somewhere between 15% and 100%, we have a lot of room to benefit.
That's our expectation.
You have 100%.
100%.
Good. That's good.
Expectation.
Is 100% POS on our phase III, too, or not?
Yeah, that would be a home run scenario.
Right.
How should we think about the cutoff for CD47?
Yeah, I think the good news, we shared this at the ESMO Breast meeting in the spring. We had a lot of investor questions around, "That data looks great in gastric. Did you just cherry-pick a cutoff that looks good, and the rest of them don't? We shared a range of cutoffs at that meeting. It's IHC scoring of 2-3, IHC 3. In terms of CD47, we shared four different cutoffs. Regardless of what we picked, we showed a very strong delta on ORR and PFS and OS hazard ratios that look good.
That, to me, just speaks to the magnitude of benefit that we're seeing in these patients that are CD47 high. That again, was very similar to what we saw in our breast study, where the patients that responded were CD47 high. There, we used a pretty simple cutoff. It's almost on/off. Is it there or not? That worked as well. Of course, we wouldn't want to base a phase III selection criteria off of that. One of the key drivers of doing this study before phase III is making sure we have that right.
If I understand you correctly, you're going to be maybe looking at CD47 expression levels across these patients and look at the response rate and other metrics to maybe finalize a threshold?
Right
in terms of the cutoff.
Right. Exactly.
Okay. Where are you in terms of developing a companion diagnostic?
Yeah, we're well down that path. We initiated efforts on this front with Ventana probably at least a year ago. A great partner for ours. There's no magic in the test itself. We're looking at CD47 expression on tumor. We're looking at samples that are gathered at diagnosis, so it's something that every patient will do, every clinician is comfortable with. There's no, like I said, some big research effort to develop the right test. We just need to operationalize that with Ventana and put that in place for the phase III.
You have to get that done prior to your phase III initiation?
Yeah. That's the goal.
Okay. Can you talk a little bit about enrollment and testing for CD47 to require fresh biopsy, or archival, or blood test? I assume it's not a blood test. Then just based on your experience so far, what's the percentage of patients that have CD47 high expression?
Yeah. We don't require it. We're strongly encouraging fresh, I think it's an important question.
I think to the extent that we can drive more fresh biopsies, it allows us to know with 100% accuracy where these patients sit. There shouldn't be change in terms of CD47 expression. If anything, the literature would suggest it should go up over time. But again, I do think where possible we are trying to get fresh. Again, I think that then allows us to be very clear as to what bucket, if you will, each of these patients is in.
For HER2 testing, is that fresh biopsy?
HER2 testing, it's similar. Again, to the extent we can get fresh and we know HER2 status at time of study entry, I think that is important. We do know, and I think it's an evolving question in the field, but I think what is becoming clear is that over time, patients will lose HER2 expression, particularly after they see a powerful ADC like an HER2. Understanding that is really important for us because our drug works when we have HER2 expression and CD47 expression. I think the really compelling thing about what we're seeing with this biomarker is that it's right on mechanism, meaning when we have those two things present, that's when our drug is working best.
Can you use the same tissue biopsy for both HER2 and CD47?
Yeah.
That's what you guys are doing.
Right. Exactly.
Okay. In terms of the regulatory path forward, you guys talked in the past that the base case scenario is perhaps a phase III randomized control study. Can you tell us about maybe the patient inclusion criteria, the size of the trial, and is there an upside case here where you might be able to get a single-arm accelerated approval if you see something truly compelling?
Yeah. I think the base case for us is we execute this study, call it 80 to 120 patients. Again, this is really meant to inform the phase III design. The idea that from here we're going forward with the same inclusion, exclusion, really the same treatment arm that we're testing right now, EVO plus TRAS plus chemo. The control, I think would be TRAS chemo, which is standard of care right now. Of course, the data here will inform the powering. I think the hope when you're running a targeted playbook is that you're able to get to a smaller pivotal study, and can run a relatively tight study to get to market. I think that is what we're trying to do.
Again, if we see a 40% delta in ORR and a PFS of more than double like we did in the gastric study, I think that should then result in a relatively modest phase III. I think that's TBD. I think on the accelerated path, of course, dare to dream. I think that's possible. I don't know how anyone could promise that as a base case. But if you look at what it's taken in the past for companies oncology, they've needed a screaming unmet need, which we have. They usually need some sort of targeted approach, and they need efficacy that jumps off the page. Is it possible? Sure. But I don't think it's our base case, but I'd love to be wrong about that, of course.
Can you touch on the commercial opportunity here for you? We're in second line HER2 positive, but also CD47-
Sure
positive setting.
Yeah. We estimate about 20,000 patients fall in this criteria in the core market, basically looking at patients who we think retain HER2 after a post HER2 setting, as well as have a high CD47 expression, which we estimate are on the 50% mark. So it is a sizable opportunity, so to speak, right? Clearly more important than the number, what is also I think important to recognize is, I think Jason alluded to, is the very poor outcome in this setting that we've seen so far. The unmet need is huge there. Clearly this is the reason why we're focusing on this indication now and in this setting with EVO.
In terms of competitive landscape, we know Jazz is running a phase III study of trastuzumab versus their own zanidatamab, also in this sort of post on HER2 setting. If that trial reads out positive, how do you think that's going to impact you guys? Or do you think it's going to perhaps raise the bar?
Yeah, I think for us, we need to compete effectively in the CD47 high group. I think that's what we're focused on. Certainly, a lot of enthusiasm, and deserved enthusiasm, for zanidatamab and the potential there. I think what we're trying to enable is a decision out of phase III to do either a combination with zanidatamab, if that ends up being successful, or a combination with trastuzumab. We definitely want to combine with standard of care. I think the way that we've set this up should allow that optionality, so certainly are watching that closely. I think in general, because of the way our drug works, we should work with a naked antibody, with a bispecific.
I think there's a lot of interesting strategic angles behind the ability to combine with bispecifics. Because as these bispecifics come along, whether it's zanidatamab or pertuzumab or amivantamab, they're going to need to find combinations, and ideally combinations that are unique. Because of how this all played out in CD47, there's really no other CD47s out there. Again, if we're successful, we're staring at a relatively large market, as Harish just mentioned, and we're essentially the only CD47 available for these patients.
I want to maybe move on to your EGFR ADC program that is also very exciting. Jason, you mentioned in the beginning that you guys picked a pretty unique epitope for EGFR. Tell us why, and then how confident that you can avoid some of the GI scheme toxicities associated with prior EGFR ADCs.
Well, I think there has been efforts in the past to develop a monospecific ADC against EGFR. AbbVie and Amgen, for example, both had programs that were terminated. EGFR certainly is a challenging target, and I think if you look at cetuximab and panitumumab, both of them have on-target toxicities. Skin, derm-related issues, blistering, et cetera, is the primary issues. I think the insight from our research team was we need to take a different approach because if you take that approach, you put a payload on it, you are going to have on-target issues. So our team identified milatuzumab as an interesting antibody.
Quick history there, it was developed by EMD Serono after a very large antibody screening effort to specifically find a unique binding pocket that would not create these skin issues. A binding pocket and modality that was specific to tumor. They then took it into clinic. In fact, it showed it does not cause skin-related issues, but also continues to drive activity. That's why we picked it, and that's what we're hoping to see translated. Certainly translated into the NHP work. We didn't see derm or skin-related issues there, and I think our hope is that will translate in the clinic as well here as we move forward.
Where are you in terms of dose escalation?
We started dosing patients about a year ago. Importantly, this dose escalation study is 100% done in the U.S. We are enrolling only at the top academic centers, so we are enrolling at Moffitt, MD Anderson, I think really the top centers in the country. We are doing that intentionally. I think that is a tough place to go, and if we can show activity in that setting as well as we think deliver on the safety promise is what we are trying to do. We started at 1 mg/ kg. We announced we were able to double to 2 mg/ kg , we then went to 4 mg/ kg , and then over the summer, we shared that we have escalated beyond 4 mg/ kg .
I think going into this, if you look at certainly what we saw in our preclinical work as well as the history of topoisomerase ADCs, in and around 4 mg/ kg should be the start of the therapeutic window. That is what we thought going in, and we continue to feel that way. Our goal, of course, in dose escalation is to continue to escalate until we see DLTs, with the idea that then you have between 4 and X to play with. The study has been going quite well. We are looking forward to sharing an update here soon. We have guided towards back half of the year. We are now in the back half of the year, so we are expecting to provide an update on the program here soon.
How many patients should we expect from this initial data exclusion?
Yeah, I think it'd be call it 20 or more.
20 or more.
We intentionally enrolled four tumor types only here that we think are EGFR relevant, so lung, head and neck, colorectal, and esophageal. I think that should give us both a good sense of safety, and again, that's the primary focus of this update, just given the stage of the study, but also ideally activity because we know in those settings that, again, an EGFR ADC should work.
You have not shared the split of patients between these histologies?
No, we haven't shared that.
Okay.
No, not to date.
What would be a good outcome just from a safety and early efficacy perspective?
I think from a safety perspective, again, it's to show that we have a therapeutic window that is wide enough and feasible enough to take forward. I think for us, it's really also answering the big safety questions. There's no question in terms of topoisomerase ADCs, ILD has been a concern. We want to be able to ensure that we're not seeing ILD. Ocular tox can be an issue with some of these ADCs.
I think if you look at skin blistering, et cetera, going back to what I just said in terms of the sales pitch as to why this should be different, we want to see that translate. So I think with this update, ideally, we're able to show that we are safe, and I think that's important for an ADC in this class. Of course, if we could show activity beyond that, I think that would be great. As I mentioned, we're treating four tumor types that we should see activity. It's just going to be a question of sort of maturity of our data at that point.
So I assume for the initial data disclosure, we might see some sort of response rate from this data set.
I think the right time to measure response rate is probably going to come more into next year. Ideally, I think, although it can work for some companies to say we're 4 for 5 and it's an 80% ORR, everybody should be happy.
Yeah
Buyer beware, right? Hopefully, that works out, but I think that can set a tough expectation. For us on efficacy, we want to be able to report that when we're at a place where we have a good sense of what we're seeing. I think this update is important, but as we think about going into next year, we want to deliver proof of concept on this drug, at some point, call it mid-next year. I think for us, that means one or two doses in the right patients, call it second and third line, head and neck and lung, for example. Clearly enough follow-up on those patients to be able to say something about confirmed responses. I think at that point, those are typically pretty big inflection points.
If we can deliver both the potency that we think we can as well as a therapeutic window, I think that for our perspective, is the right time to start thinking about what the right ORR percentage is.
How do you rank these four tumor types?
I think it depends on the day. I think for us, all four are interesting. We see a lot of opportunity in lung. I think if you look at both mutant and wild type, there is opportunity for a topoisomerase ADC. There is no approved topoisomerase ADC, for example, in the wild type lung setting. I think for us, mutational status shouldn't matter. I think there has been a lot of interest and enthusiasm for an EGFR ADC in those spaces, and those are very sizeable opportunities. Head and neck is also interesting.
Certainly a lot going on with PFS and amivantamab. I think some promising options there. This epitope advantage we have is very interesting, right? I think there is no question that in the future, new drugs are going to need to compete in combination with an EGFR or after an EGFR, and that is very challenging when everyone is using the same epitope. For us, this is where the milatuzumab advantage comes in, and I think could be particularly relevant in head and neck. We think esophageal is interesting as well, and CRC, I think historically, has just been very tough. We'll see. Again, it's an important area to study, but certainly, I think it's been a harder area to move the needle.
I think particularly for head and neck cancer, we don't have a ton of ADCs here, so clearly an open field. On the other hand, we have bispecifics moving to the front line. So how are you thinking about maybe in the second line setting, what would be the bar for success?
I think if you look at the recent amivantamab data, that was 41%, 42% ORR. I think the question is how is all of this going to play out, how will the PFS first-line data look? Amivantamab certainly is a promising drug. Where will that play? And then, of course, there's some of the NME-based ADCs coming along as well. Again, I think our advantage with the milatuzumab construct is in using topoisomerase, frankly, which is easily the most validated payload, is we think we have an opportunity, certainly in the second line
because we'll be able to treat patients who have progressed, is also potentially in the first line, depending on how this plays out. Because again, there's no question EGFR is a very well-validated target in head and neck.
Okay, great. Looks like we're out of time. Jason, Harish, thank you very much.
Great. Thank you so much. Really appreciate it.