Hey.
Good morning, everyone. This is Michael Yee, Managing Director and Senior Biotechnology Analyst at Jefferies. I am very pleased to introduce to a standing room audience ALX Oncology. Up here on the stage with us, we have the CEO, Jason Lettmann, and a new member of management, new Chief Medical Officer, Alan Sandler. Many of you know him from his prior company, so maybe it would be a good opportunity, Jason, for you to make some introductory comments and obviously give an opportunity for Alan to introduce himself and tell us about his background and what he saw here at ALX since he just joined. Obviously that should speak to something.
Yeah, well, thanks for having us. Always a great meeting. Appreciate all of you coming to join. I'm Jason. I'm the CEO of ALX. We are pioneering a novel approach to drugging CD47, which is a very, very important pathway and very important target. We've been at it since 2015. We've raised over $400 million to date. We have nine ongoing clinical studies. And as you alluded to, Mike, we're in the midst of an exciting period with a number of important readouts. Alan Sandler is our new Chief Medical Officer. He's now, what, four days on the job? Five?
Maybe five. Depending on the travel and language, coming from the U.S.
But yeah, so he's an ALX grizzled veteran at this point. But thrilled to have him. And Alan, I'll let you go.
Great. Thanks, Jason, so thanks again for the invite. It's great to be here, so brief background, a couple of decades as a medical oncologist in academics, and then specialty was thoracic oncology, and then left to join Genentech in 2013. Spent about seven and a half years at Genentech, rose to head up late-stage oncology solid tumors before leaving and going to Zai Lab, a China-based company, for a couple of years, leading their global development group as CMO, and then most recently, Mirati, November of 2022, I think, until the acquisition by BMS in January, and a few months with BMS and then emancipation in April of this year, and where I worked, did some consulting on a couple of boards, including ALX, and then some changes in the C-suite and saw an opportunity to do one more, at least one more operational role, and very excited.
I know the team very well, the management team very well. Enjoyed the board. The board has some great people on it as well. I think there's also some interesting data that you're aware of and further data that will be coming down the first half of next year. I think there's an opportunity to contribute, given my background with IO, clinical drug development in solid tumors, and the opportunity that we have here with CD47.
Yeah, that was fantastic. But just to clarify, you were on the board beforehand? And so that's a lot of where the familiarity came from. And then you sort of stepped out from the board and just also joined day-to-day as Chief Medical Officer.
Stepped off the board and now.
Stepped off the board. Okay. Fantastic. Jason, same thing. You were on the board as well. Obviously very intimate with the company and was ready to step in as CEO and has been executing from that point. Okay. Fantastic. So given that you're fresh as well, but also the most recent announcement earlier this year was obviously the randomized phase two gastric data. That was sort of the last big announcement. The company has been a bit quiet since then. But maybe you could just talk about since that result, maybe just describe 30 seconds or a minute of the results for those who might not be familiar. Wall Street had a very negative reaction to the results. Do you continue to see stuff there and promise? What have you learned and gleaned from those results?
And what are the next steps they're appreciating? We'll get to head and neck in a second.
Yeah, sure. So in July, I think, as everyone saw, we announced data from our ASPEN-06 study. This is the first randomized study in CD47 in solid tumors. As many of you know, we're testing our combination agent across three different categories. One, anticancer antibodies. And that's really what the ASPEN-06 study meant to deliver in terms of proof of concept. Two, with checkpoint inhibitors. And we'll talk about our head and neck study, our first-line head and neck study with Pembro. And then three, with ADCs. And our mechanism, we think, has relevance across all of those combinations. And as a reminder, we're developing a combination agent. Our agent, by design, is meant to separate the don't-eat-me signal from the eat-me signal. And what we've learned and what we've seen is that we need both.
So for us, the ASPEN-06 data, again, this was 127 patients randomized in second- and third-line gastric. What we saw was over a 40% ORR versus 26%. And again, this is combining Evo with Traz or Herceptin. And what we were, I'd say, most excited about with that data readout was what we were seeing in terms of DOR. We had a durability of over 15 months, which was double that of control and double other benchmarks like the DESTINY study with Enhertu . So I think that gave us a lot of hope, a lot of conviction. And at the time, what we've said about PFS, it was just immature. And again, that's also indicative of an IO agent.
Are you going to have PFS to be reported in 2025?
Yes. That will be the next update.
Okay. So just to summarize, you reported a statistically significant. I can't remember the p-value on the report.
ORR?
Yeah.
Yeah.
Statistically significant, 40% versus 26%. But the street reacted extremely negative because at the interim analysis, the delta was wider. And then the delta came together, although it's still 40% versus 26%. And people trying to do reverse math could try to back into what the second half of the study showed versus the first half, in which case there was an implied wide difference. Is there a difference in the populations in the first part of the study versus the second? Or what would you explain to that? Because when the differences are stark, people say it was all driven by the first half.
Yeah. I think we're looking at the totality of the data. And again, as you pointed out, Mike, what we saw was a significant delta. I think what we've learned is that we need both. We need CD47 blockade. And in this case, we need expression of our combination agent. We need the eat-me signal working in order to drive ADCP and drive phagocytosis. If you just give CD47 blockade alone without an active agent, in our case with a dead Fc, we will have no activity. And I think what's really important about what we saw interim, post-interim, and full is when we had that equation right, meaning we had strong HER2 expression, the drug was very active.
We do have a HER2 in that combination.
Yep, we do. And what we know in gastric, and it's really important, is that it's a very heterogeneous tumor. So this is not our kind of back-solving into what we saw. But we know that gastric is very heterogeneous. And we know that we had a mix of patients who had an archival biopsy and a fresh biopsy.
Can you explain that? So this refreshes me a little bit. So despite the strong results and the controversy around the difference in the first half versus the second half, you went back and you determined that there was a significant difference based on fresh biopsy versus old biopsy, which could be indicative of HER2 expression being old.
One key point.
Talk to that and the confidence around that as confirming your hypothesis.
So one key point, because what you described it as, well, you went back to look, it was actually prospectively placed in the study. So this was something because of that hypothesis that Jason had mentioned, that the idea was that having HER2 there so that you have two ways to lock the cell together, both the HER2 antibody as well as Evo, that that would enhance activity. Recall that the.
It makes sense given the mechanisms of the drug. You have to have HER2 expression to bind in order to drag the antibody.
Correct. So as a result of that, looking to see whether the expression of HER2 was actually there was an important component. So that was prospectively.
Prospectively final protocol.
In the protocol and then came up and showed that for that, that response rates held true. And the important point again, that I know Jason was going to do, but I figure I'm standing up here, I have to say something.
Sure.
Is that.
I want to hear from the Chief Medical Officer.
The response rate actually held true first half and second half, so what you didn't see when you didn't look at HER2 expression, it seemed to go away. However, when you looked at that, both pre and post-interim analysis, it held true.
So what percent of the study was fresh biopsy versus old biopsy?
That's what's interesting here because the interim population had about half of the patients with fresh. The post-interim population had closer to a third. We know after, and again, this was good for patients, not necessarily good for how our study played out in Wall Street's eyes. The second half of the study, after you put up a 30% delta in second-line gastric, is going to enroll patients very quickly. Those sites were not going to wait around for a fresh biopsy. We saw the rate of fresh decline. I think what's really important is, and the question we've got is, that's great, ALX, how else can you look at HER2 expression? What we've been doing since July is we have ctDNA, circulating DNA now from every single one of these patients. Great. Is that going to hold up?
Meaning patients that have HER2 circulating in the bloodstream, are we going to see the same efficacy advantage? And as a reminder, with the fresh population, again, with HER2, strong HER2 signal, we had a 55% ORR versus 23%, I believe, on the control. So an absolute screaming signal in every way in which our statistician looked at this.
55 versus 23 was the response rate, and I can do math on one half and one third, so less than one half of the study had fresh biopsy, but it was 55 versus 23.
No change in the control arm response rate when you looked at the FISH otherwise.
Makes sense.
But it's nice to see as well.
It's important because people could say, oh, well, look, your combination where you have a HER2 targeting agent is working better. So what? But to Alan's point, it's important. One is regardless of how you look at HER2 expression on the control arm, IHC 3+, IHC 2+ did not matter. It was all in the 20s. So what we know we're doing and what we're bringing is something very different than Herceptin. And all of these patients have seen Herceptin. And that's what we think is so exciting because the world needs other agents to go after these tumors.
So based on that analysis, because I want to make sure we budget for time here, in the months that have followed, have you talked with the agency? Do you plan to run another study? What is the plan? Or it's on pause pending the PD-1 data, which we're going to get to in a second. What is the message to investors on?
Yeah, I think the message to investors, again, has been we're going to wait for mature data, PFS. Same question would be what you'd hear from the agency, which is ORR is nice, but let's see PFS. And we want to go to the agency with the strongest.
So is the PFS going to be A, would you look at the total group? B, obviously, you're going to look at a proportion of the group, which was the fresh biopsies, which is less than half. Now, remind me how many patients, but general rule of thumb for the CMO is smaller the numbers, we want to be a little bit more cautious on median PFSs. But to what degree should we look at the total PFS and then the PFS based on only fresh biopsies?
Yeah. I mean, what we're going to share is PFS on an ITT basis. So all 127. PFS on the fresh population, which is 48. And then PFS on ctDNA, which is going to be a higher number. We know that. And that's what we're planning to share. And again, what we know is when there was one other study run with only fresh biopsies, which was the DESTINY, one of the DESTINY-Gastric02, I believe, and HER2. Our data, we think, is superior to that and across all metrics so far. And so I think that is going to lead to some really interesting conversations with.
What would you want to see in PFS? 20%-30% improvement in median PFS on a phase 2?
Yeah.
What does that mean? Take it to partners, take it to Wall Street, seek a financing?
I think take it to the agency is the first stop. And if you look at the Rainbow study or if you look at the Destiny study, I mentioned PFS of, call it, one and a half to two months has been enough to drive an approval. So that for us is the bar. That's what we want to see. Of course, the hazard ratio is very important too.
I would say that, putting my thinking hat on, I would say that if a group had 55% response versus 20-something, and then overall it was 40% versus 26%, based on the scans, which are probably what, every one or two months?
Two months.
Two months. Just based on some math, you could probably assume that if you're having a response and it's a confirmed response, by definition, you made it an extra two months for those people. So the median could get to two months delta. So I'm going to predict the median could hit a two-month delta, and the FDA generally doesn't say, no way, dude, you can't run another study. They'd say, find the money and go run another study. So if it's a two-month median, you will take that. The agency will say, go for it. It's a safe drug.
Yeah. I mean, there is, again, if you look at what drove Enhertu's approval, it was a single-arm study, largely OUS in this patient group. So I think to have randomized data with 127 patients in second-line and third-line gastric is a different story than first-line gastric. There is not much coming for these patients. So I think we have a really strong argument for the agency around what this could mean for patients.
Did you just say that the other drug was approved on a single-arm study?
Correct.
Are you implying that there is a probability that, heck, this is registrational?
I mean, I would never play FDA.
Peter Marks is here on Thursday, so we can ask him.
You should ask him.
He's on here on Thursday. We will ask him the question. But no, all jokes aside. On one hand, you know if they view it as compelling, could be registrational. On the other hand, they could say, run another study.
They could, and again, it's a phase 2/ 3 combined design, so we've already built in what this phase 3 looks like, so again, I think in that case, it's running the base case plan that we've been pursuing all along.
So when is that data coming? ASCO? Is that a reasonable?
It's coming first half.
Coming first half. All right.
And just one other point about the study also is, so it's randomized. So there's almost, in a way, two shots on goal that we haven't discussed with the FDA. But if you think about it, we can look at the active arm, compare it versus historical control in that particular setting. And in that case, your traditional accelerated approval is looking at the points with confidence intervals for overall response rate and DOR. In a non-randomized setting, the FDA doesn't pay attention to PFS and OS. They look at DOR. But in a randomized setting that we have, we have that opportunity to show not only a difference in a median, but a hazard ratio we'll obviously be able to have.
Right. Your point is, in one sense, traditionally, just the fact that you have, what was it, 50, 60 number of patients in the drug arm?
We have 65.
60-something. Okay. That you could compare that against natural history. Number two, you have a control arm. You have statistics around that. If there was a PFS that was statistically significant and your overall response rate is statistically significant, that's already been reported, that's better than open-label data. So we'll have a discussion.
Ram Pac.
Right. Okay. So that's first half. And then so we'll look for that. I think we've discussed that enough. So what about other additional data, more randomized control data, the largest head and neck randomized program, I believe, going on in biotech right now?
Yep.
Tell us about those two studies and when is that data coming?
Yeah. So we're, again, testing the CD47 mechanism across two different signaling paths. One via the macrophage with anticancer antibodies we just talked about, two via dendritic cell signaling. So the story with Pembro is can we release two brakes? In essence, blocking CD47 should better activate T cells. And when combined with Pembro, it should be essentially like a dual checkpoint blockade. And what got us excited about this was our phase 1b study in head and neck, where we had both first- and second-line patients. We showed around a 40% ORR across both of those groups. But again, we're developing an IO agent. So the part that was most exciting for us was the survival benefit. And there we had an over 80% 12-month OS. And that sort of tail is what drove the Pembro approval. If you look at KEYNOTE-048, it was all about.
Didn't really work on responses.
Did not.
But ultimately, the tail drove it a survival benefit. So they approved it, but it did not have a benefit on response rates for Pembro.
Right. And so what we said is, let's go right into large randomized studies. So that's ASPEN-03 and ASPEN-04. One is Evo plus Pembro versus Pembro mono. The other is Evo Pembro chemo versus Pembro plus chemo.
That would be a low PD-1.
Correct.
Right. So both high PD-1 and low PD-1, there's two randomized studies going on.
Well, and it also depends on disease burden and who gets chemo. It's not necessarily. But anyway.
But there's two studies.
Yes. The first patient was enrolled in January of 2022. There's about over 300 patients across those two studies, randomized two to one. So to your point, this will be the largest randomized readout in head and neck since the KEYNOTE-048 study, we believe. And we know there's a lot of enthusiasm in that space, given Merus, given Bicara, certainly very promising agents that they both have. And that's what I think has been really good here is that the world has been educated about the need in head and neck.
I think Merus's data is open-label single-arm, isn't it, too?
That's correct.
Open-label single-arm data, and they sport a multi-billion market cap.
Yeah. Well, that's good.
Yeah. Yeah. That does look good. To your point, you have small data sets, but also promising data, and you're going to report out randomized control data. When is that data coming?
That's coming first half as well.
First half as well.
Right.
Okay. So given that, let's look forward to that. Also, at ASCO, I believe this year, there was Padcev combination data. This is on top of ADCs. You're also running a number of other collaborative studies in myeloma and breast cancer Enhertu . What is the status of what you would do with the Padcev combination, which showed high response rates in Padcev alone, or other ADC data that could be coming this year? I'm particularly keen on Enhertu , but tell me what other.
Yeah.
What's going on here with ADCs?
The ADC combination is, I'd say, our third big shot on goal. Again, there was a lot of preclinical data. There was even more at AACR last year supporting the role of CD47 blockade in really holding back the power of an ADC. And it makes sense mechanistically. And so what we've been doing is running two studies with ADCs. One is with Enhertu . So we're looking at combining with Enhertu in breast. Two is the study you just mentioned in second-line bladder with Padcev. We reported that data at ASCO, again, a very early data set. We were in the midst of still enrolling the study. We had a mix of confirmed and unconfirmed responses. Again, early days. But we were very excited about that signal. And so what we're going to do next is wait for that data to mature to get to PFS and.
But you have enrolled more patients since ASCO because that was an early data set. So how many patients was that and how many more patients? And are you committed to reporting a big update on the Padcev?
Yeah. We reported on about 23 patients. The study is set to enroll close to 30. We should have all patients enrolled with confirmed scans, and there I think we want to wait until we get to a good mature data set there as well.
Mature as PFS or mature as DOR? Because that would feel like an ASCO thing too.
Yeah. I think we feel comfortable with first half.
Okay. A lot coming up in the first half. So an update on the gastric results and perhaps talking with the FDA on that. Two randomized controlled head and neck studies reading out and updated more complete Padcev ADC combination data.
Yep. Yep.
In terms of first half for the head and neck data, this is pretty big readouts coming. Can you help us with that? That's not event-driven. That is based on one-year time points, right? These are landmark analyses, not event-driven. So when did it complete enrollment? So that.
Still event-driven. I mean, looking at maturity, it's still going to be based on event-driven.
I thought it was a 12.
A 12-month endpoint of.
That's a 12-month landmark, but it still takes the events to sort of get there, right? I mean, it's a subtle point between it, but.
Okay. So when did the studies complete enrollment? Have you got?
So we haven't commented on completion of enrollment, but needless to say that we're going to be able to report on all of those patients, full confirmed scans, plus I think have a good look at survival.
My team's going to go do an analysis on when that completed enrollment. So I'm going to add 12 months to that and get a little bit closer. But tell us then in the last minute, your balance sheet. Obviously, that's important, and you have a lot of readouts coming, but balance sheet's important.
Yeah. So we have cash. We have about $180 million as of the last quarter. We have done no deals to hamper the asset. We have only $10 million in debt on the books. And that will get us all the way through Q1 of 2026.
Okay, so you can read out all of these results.
Yeah. And we're running the budget right now to try to extend that further.
Got it.
But the goal is to be able to set us up to go through all of these data points.
Certainly the first six months of 2025 to get all this information. And then.