All right. Welcome everyone to 2026 Jefferies Global Healthcare Conference in London. My name is Roger Song, Senior Analyst, covers biotech in the U.S. It is my pleasure to have the fireside chat with Allogene Therapeutics, and we have CEO, Zach Roberts. Welcome, Zach.
Thank you, Roger. Great to be here. Thanks for the opportunity.
Awesome. All right, Zach, I want you to open up with some high-level overview of Allogene, and we know you have a lot of exciting things already happened, and also in the future. Thank you.
Yeah. Allogene is an allogeneic cell therapy company. We're developing CAR T cells for patients with frontline, post-frontline large B-cell lymphoma, using a very novel strategy in which we're treating patients who appear to be in a clinical remission but remain positive for minimal residual disease. That's our lead program. It's a pivotal study. It's called ALPHA3. And we also have a very exciting and differentiated program in autoimmune disease in which we are using a novel product designed specifically for this indication set that is the world's first dual-targeting CAR T, CD19, CD70, also, of course, healthy donor-derived allogeneic product.
Okay, great. Those are the two major topics we can touch today. For cema-cel, this ALPHA3, I think you have an interim analysis building, and then potentially going to be a meeting next year. Give us a little bit more teaser, what exactly you're going to see, and then what kind of statistics, the numbers you are targeting?
Yeah. The study's been open for a couple of years now. I would say that we have had a very nice year so far in 2026 for enrollment. We released some data back in April of this year. It was a pre-planned interim analysis in which we looked at efficacy as measured by MRD clearance. As a reminder, everybody who comes into the ALPHA3 study is MRD positive, and so the patients were examined in the two arms. Observation is the control arm versus treatment with cema-cel is the experimental arm. And we looked at both safety and efficacy, and the efficacy there was assessed by the rate of MRD clearance.
And we saw a very nice differential between the two arms. 58% of the patients in the cema-cel arm went from MRD positive to MRD negative, and only 16% of the patients in the control arm went from MRD positive to MRD negative. We had an extremely clean safety profile, no cases of CRS, no cases of ICANS. Everybody was managed as an outpatient, including in community practices where CAR T is not generally given. No toci, no steroids. This has really, I think, validated the hypothesis of ALPHA3. The next analysis that is planned for the study is the one that you mentioned that was currently assumed to occur middle of 2027.
That is looking not at MRD clearance, but at event-free survival, which is the primary endpoint of the study. Then we have the primary analysis that is scheduled for 2028, or we assume it will happen in 2028. Both are event-driven. For this upcoming potential analysis next year, we haven't gone into much detail about the number of events or the alpha allocation between these two analyses, aside to say that a relatively small amount of alpha is allocated for that, because we wanted to preserve most of the alpha for the primary analysis.
As we are conducting the study and executing, really our focus is on running a very clean study, one that is not affected by any kind of bias or that would impact the integrity or the data interpretability, because ultimately we believe that this will be a positive study when it is all said and done.
We believe it will also lead to the approval of cema-cel. The most important thing from our perspective is just making sure that that data set, when it is finally generated, is a BLA quality, and we will have a favorable outcome with the FDA.
Yeah, absolutely. Investors are interested in those details, but not necessarily, sometimes you not need to disclose because you may change the trial integrity. Maybe just on the potential outcome, right? If the DSMB look at the data, then what would be the recommendations, and how are you going to disclose the outcome to the street?
Yeah. You brought up the Data Safety Monitoring Board. This is a DSMB-governed trial. It has a randomized pivotal study. They will be the ones to conduct the analysis at the time of the event trigger, and they will be unblinded to the results. We will only know how many events have accrued. We will not know in which arm.
Okay.
When we hit the number of events that we're targeting and we submit the data to the DSMB, they'll evaluate the data, and there are three main possibilities.
One, in our view, the most likely possibility is that they will tell us that the study is on track and we should continue on to the primary analysis as planned.
There is a chance that they could come back and say it's statistically significantly positive.
Then they may, depending on what they see, they could make a recommendation that we stop the study or, and open it for efficacy and release the results. Then, of course, there's always a remote chance that they tell us that it's futile. We don't expect, based on the early data, that that is a likely outcome, but in the spirit of completeness, that is something else that they could say.
Got it. For the second outcome, which probably is the most interesting and optimistic outcome, if they think that the result hits statistics, you say it also depends on the magnitude of the result or the effect size you will decide to unblind. Then also that unblinding means you can talk with the FDA, maybe potential file based on that interim result, because this is a pivotal study. If you unblind, then what that mean to the overall filing strategy?
Yeah. So it's a great question, and it's one that we get asked about a lot. If the recommendation is to unblind the study, effectively, that will be the sum total of the data that we have to communicate with the FDA. Because once the results of that study are known, it's going to be very difficult to get patients to randomize into a study where they have a 50% chance of ending up in the control arm, which they may have just learned from a press release, has worse outcomes, right?
That's right.
Additionally, patients who were already enrolled but had not yet experienced disease progression, when they find out that they are in the control arm and they would have done better if they were in the treatment arm, they may call their oncologist and say, "I would like to receive some treatment," right? That does put us in a position where if that is statistically significant, it needs to be a registrational package on its own, because we will not be able to expect much data after that.
Got it. Okay. So that result needs to be strong enough, you think—
Correct.
—it can be registrational.
Correct.
Okay. Got it. All right. Then, in terms of Mitch, just remind us, what is the study design for the hazard ratio? Then, can we disclose the total EFS events need to be, not the interim?
We haven't said anything about the number of events at either the interim or the primary. We have disclosed that the study is designed, powered to demonstrate a difference of a hazard ratio of 0.5. That's essentially a doubling of EFS in the treatment arm relative to control.
Got it. What I tried to get at is that in order for interim to hit statistics, it needs to be, hazard ratio needs to be very high, because the alpha spend is low.
Correct.
If your primary is 0.5, it doesn't need to be like 0.2.
Certainly to maintain a power that people would expect. That's exactly right, Roger. It needs to be overwhelmingly positive.
Okay. Got it. And then, understanding you are not get access to any of the data, but have you seen any blind data in the relatively regular base, and particularly for the safety, dropouts, anything you can say at this point?
Well, I mean, safety is always monitored in real time in all of these trials. If there are significant events in any of our programs, we know about that right away, usually within 24 hours. I will say that that has not happened in ALPHA3, and the safety data that we shared back in April at the interim analysis is consistent with what we've seen in the months that have followed. We are only looking at data at the time of pre-specified analyses, so there's no additional efficacy data or anything that I have access to.
Yeah, sure. Okay. Recently, EPCORE, the bispecifics plus R-CHOP, reported a pretty interesting hazard ratio in the first line LBCL. So how do you think about the MRD positive will involve with those new regimen? And then the second question, second part of the question is, have you enrolled any patient with such regimen?
Yeah. So, the study is designed so that any standard regimen that's used in frontline currently, if you are MRD positive at the end of that regimen, then you're eligible for ALPHA3. So really, in practical fact right now, that means R-CHOP and polatuzumab R-CHP. Those are really the only two approved regimens.
Yeah.
Of course, there's variants of R-CHOP out there, and those are eligible too. There's no bispecific patients that have come into the study. However, one could imagine that if such a regimen is approved and the study is still enrolling, and they were MRD positive at the end of that regimen, they would be eligible. That is viewed as how this will work in the end, because one of the key features of ALPHA3 is that it doesn't really matter what happens up front. If you are MRD positive at the end, it means that whatever you got didn't really work. It didn't work all the way. That's where the cema-cel could add value. We don't much care which regimen these patients receive in the frontline setting. Now, the other question that you asked is that going to change the MRD positive dynamics?
I would say, it would, based on the hazard ratio. Of course, we haven't seen any of the other data.
Yeah.
We don't know what the PFS actually was. We don't know what the safety profile looked like. There's a lot we still don't know about that. But we did expect that this was going to be a positive outcome. Those are powerful, potent regimens. However, they are toxic and complex, and significantly more toxic than R-CHOP or polatuzumab R-CHP. Based on conversations we've had, as well as some quantitative market research, we actually feel like the uptake of these regimens is going to be fairly restricted.
Okay.
And really for select patients at select centers. One of the realities that we're grappling with ALPHA3, is that almost all of these patients who are newly diagnosed with lymphoma are diagnosed in the community and treated in the community.
These regimens that are toxic and complex really are not community-friendly regimens. They're probably going to be for patients who are referred in for some reason, they have very high-risk disease, or they're super sick when they're first diagnosed, they'll get sent to a referral center. Maybe those patients will get an EPCORE-based regimen. But for the rank and file, they're going to continue to get pola-R-CHP and R-CHOP. We expect, even once these regimens are approved, assuming they do get approved, that it's going to have a pretty minimal impact on the overall MRD positivity, probably about 5%-10% on the overall rate across the board of MRD positivity.
Got it. Okay, great. For the MRD testing, do you see it will evolving into a greater adoption? Then how ALPHA3 data will change the landscape as well?
Another great question, and I think really drives at the dynamic nature of this space. We've actually done two rounds of market research. One was a qualitative round, which was about 30 physicians, and then the more recent one was about 150 physicians. Just one snippet of information from that research is that when we first asked the 30 physicians how frequently were they using MRD, that number was about 20%. Then when we asked the 150 physicians, it was closer to 40%. So even in the few months between those two rounds of research, it's not the entire story about the dynamics of the field right now, but we're seeing already substantive uptake in MRD, and that's also corroborated in the quarterly calls by the MRD companies saying that they're just seeing really rapid uptake of MRD, specifically in lymphoma.
Your question around will the availability of cema-cel change those dynamics, 100%. In fact, we actually tested that in our research, and it is the current lack of a treatment decision that you can make based on a positive result, which is keeping some physicians from ordering the test in the first place. Everyone orders the test now who orders it because they hope it's going to come back negative, because that's great news for the patients, very good prognostic information. The problem is if it comes back positive, then that's a very difficult conversation that you have with the patient because you don't have anything to offer them.
That is right.
cema-cel will solve that if it is approved.
Yeah. Awesome. All right, great. In terms of the ALPHA3 enrollment geography or diversity of the patient, how many of them are U.S., global, Asian? Another thing, another angle is the academic versus the community. What is the balance of the enrollment so far?
Yeah. At last update, I think we said we have over 85 sites globally. That includes the U.S., Canada, Australia, and South Korea. I can say now that the Asia-Pac region is fully onboarded. All the sites that we had selected are now open to enrollment, and they have done a really stellar job at coming online very rapidly. They are screening patients extremely efficiently and enrolling patients very well. We have been quite pleased to see that. The next goal that we have set for ourselves is to have approximately 100 sites online by the end of this calendar year, and Europe is going to be opening in the first half of next year. We will have a lot more news as time goes on around the final configuration and footprint of the study. As for the breakdown of community versus academic-
Traditionally, we've said it's about 50/50. I do think it's probably a little bit more than 50% academic now as some of the sites abroad have come on. There's not really the same bifurcation of care between community and academia in Korea and Australia as there is in the U.S. So all of those sites really are considered to be more academic sites. So we're starting to see a little bit of a shift towards a more academic balance, but that's long been expected, and we're still seeing terrific activity from the community sites that are on board.
Yeah. Got it. But in terms of U.S. and then maybe not Canada, those kind of Western country, they're still relatively 50/50 definitely?
Yeah. It's certainly in the U.S. We have a lot of community practices in the U.S., and they were early. They saw the value of this. This is in many cases the first opportunity they have to offer CAR T to their patient population. So, the value proposition of the ALPHA3 trial and of cema-cel in consolidation has only grown in appeal to those practices. So we're seeing super engagement from those practices here in the States.
Yeah. Because as you said earlier, a lot of patients are managed in the community, so you want to make sure they can use this cema-cel.
Correct. The overarching goal of this trial is to actually provide access to CAR T to patients who historically do not have access. It is a stat that I often bring up, which is only about 15%-20% of patients in the U.S. who could benefit from a CAR T are ever offered a CAR T, ever receive a CAR T. That is a really terrible state of affairs, and one where I think a disruptive technology like cema-cel and disruptive design like ALPHA3 can really solve that problem, because as we are validating in the study, we can actually deliver these products where patients receive their care rather than making the patients go to where the care is available.
Yeah. Got it. Last question related to the cema-cel. You are definitely the leader and the first one, maybe the only one still doing this setting, first-line consolidation. Any other potential allogeneic CAR T on your radar can be also applicable to this setting or any plan you know? We do not know, but maybe you know.
I think there are some investigator-sponsored studies that are beginning to crop up that are now public, and so we are seeing this. I think that is highly validating for our idea here of frontline consolidation. So we welcome the company. We are, as you point out, well ahead of everybody else, and that is because we saw this field developing back in 2023 based on the strength of that early data from MRD and realized that there was a big opportunity here. So we moved quickly, and I think that is a tremendous benefit for us. Even if we do see some early phase studies coming online, it is certainly not something that we think is going to threaten us in a commercial context.
Excellent. Okay. We should talk about this autoimmune as well, and as you said, you are the only CD19 and CD70 in the autoimmune disease at the CAR T. Recently we saw the data from ACR, which is pretty interesting, and then you do see a lot of the good clinical activity. My question is, I think I may ask you that before, but I just want to put that on the record. The correlation between the B cell depletion and the clinical activity, how you explain that? I know you, maybe it is also related to CD70, but also maybe related to the reset. We just don't know the threshold to kind of reset the pathogenic kind of B cell. So tell us a little bit more about the correlation.
Yeah, I think what our data really highlights is that there is still a lot that we as a field don't understand about what an immune reset is. I think that in the early days, it was quite easy to look at peripheral B cells and seeing them go to zero for 28 days. That was all the explanation we needed to see these profound and durable responses in patients who receive CAR T for autoimmune disease. It is not only our data, by the way, that is starting to provide a little bit of hints that maybe there is more to this story. We see peer-reviewed publications coming out now that is zeroing in on specific subpopulations of B cells that may be driving the pathogenesis in these indications. So I think there is still a lot more for us to learn.
I will point out, it is not like we saw no change in the B cells in the no lymphodepletion arm. We did see about a 30%-40% reduction. It tended to be transient, but clearly there is some activity there. You also brought up the CD70 component. So one of the key aspects and one of the reasons that we included the CD70 is because we know that this is a key player in the formation of immune synapses, in the activation of T and B lymphocytes, and it is also expressed on antigen-presenting cells. So it is increasingly likely, as we learn more about this molecule and its central role in the activation of an immune response, that modulation of B and T cells that express CD70 is likely to have a disproportionate effect on the outcome of the disorder. Now, this is still early days.
There is a lot for us to learn. We have lots of preliminary correlative work that we are working on internally right now that wasn't quite ready for prime time in the poster, so we didn't include it, but we look forward to developing that storyline in the coming months. At the end of the day, taking a step back, you look at our data, and these patients are incontrovertibly getting better, having significant improvements in their symptoms in a very difficult to treat disease of diffuse cutaneous scleroderma. We are able to do it without imparting significant immunosuppression in these patients that would come with the Flu/Cy regimen.
Yeah. Got it. Okay. I know you have a lot of internal data not yet released. Would that be including the CAR T expansion? Because at least that's another PK or slash kind of PD or kind of
Yeah, that's another kind of very interesting area, Roger. Of course, the peripheral blood compartment is a very easy thing to assess and an easy thing to measure. I think the field has maybe not pushed as hard as we should is we check one spot, we get what we think we're supposed to get, and then we move on. We have already learned in the field of autoimmune CAR T that we don't see the same degree of expansion and persistence that we see in oncology, that most of the action probably is in the tissues, not in the peripheral blood. So we started at very low doses, as you recall. So the first doses that we started at was 20 million cells flat dose.
Then we went to 40 million cells and then 80 million cells , and then we also disclosed last week that the next dose level is 160 million cells. We are still well below our peer companies in terms of the overall level of dose, particularly as it applies to allogeneic. So that story is obviously something that we're watching quite keenly and again, looking for the areas within the body that could be generating the right PK or cellular kinetics, cellular activity that would lead to the observations that we're making at the bedside.
Yeah. Got it. All right. You mentioned that you keep those escalating, and next is the 160 million cells , I think, 320 million cells is also planned. So what will be the next steps, right? You're going to continue to escalate, we're going to see data, and then when you're going to start to do those expansions, do you need more full data from the escalation before you can move into the expansion?
The study was written with great flexibility, so we can actually expand any cohort at any point once that cohort and that regimen has been deemed safe to proceed based on an independent safety review team who looks at all the data. We have ultimate flexibility about which cohorts we want to expand versus continuing to dose escalate. We can actually do both at the same time. Right now, internally, we are evaluating the data that we shared at ACR, as well as additional data that has accumulated since the data cut that went into the ACR presentation. We are starting to make some strategic decisions about what to do with the study. I will call out specifically whether we continue to dose with the Flu/Cy regimen.
I think some of the events in recent weeks around potential fludarabine toxicity in CAR T and autoimmune has certainly put that in perspective, whether that is the right risk-benefit for those patients. Then, of course, we also need to decide, are we going to continue to enroll all three indications that are currently eligible for the basket study? That is lupus, myositis, and scleroderma. We have some near-term decisions about how to refine and tailor the development strategy here, including whether and at what dose we expand. Look for some news updates early in the new year as we advance that program.
Got it. Dose, conditioning regimen, indication.
Indication.
They are all TBD in the next kind of step, and then you will potentially expand this to a—
Correct. Yeah. Ultimately here, of course, we're looking to build an evidence base that would support a phase II.
That's exactly my last question. How and when you're going to move into the pivotal, because the CAR T-cell therapy, it's great in clinical, but eventually you translate that into the business. When we can see that?
Yeah. I would say, look to 2027 for those updates. The study, one thing that has been great about this trial since the day we opened it is that we have had no problems enrolling it. We enrolled those 18 patients that we featured at the ACR in about 10 months from the first patient enrolled until the data cut. We've had now several additional patients already enroll in the trial, so we've got a waiting list of patients ready to come in. We have kind of the resources to continue to develop this asset and make very informed decisions. I don't want to get too far ahead of myself about when we might make those decisions, but rest assured, it will come in the next few months.
Excellent. Thank you so much, Zach. Thank you, everyone.