iwAL 2025 | Approaching TP53-mutated AML: current challenges, ongoing research & novel therapeutic strategies

Eunice Wang:

Hello. Thank you for joining us. I am here at the seventh annual International Workshop on Acute Leukemia, where we have just completed a session talking about one of the most deadly subsets of acute myeloid leukemia, which is P53 mutant disease. So I’d like to introduce my panel members here. I’m Eunice Wang from the Roswell Park Comprehensive Cancer Center in Buffalo, New York.

David Sallman:

I’m David Sallman from Moffitt Cancer Center in Tampa, Florida.

Gabriel Mannis:

I’m Gabe Mannis from Stanford Cancer Institute in Palo Alto, California.

Andrew Wei:

Andrew Wei from the Peter MacCallum Cancer Center and Royal Melbourne Hospital in Australia.

Eunice Wang:

So I’d like to start out with Andrew. Dr. Wei was talking to us about some of his research that his laboratory and clinical service is doing, and particularly wanted to just lead off with the discussion about the role of p53 in clonal hematopoiesis and how that leads to acute leukemia development. So could you give us some insights into that, as well as some of your trials or interventions that you’ve been investigating?

Andrew Wei:

Yeah, so just in view of the great difficulties in making clinical inroads into AML, we’ve been looking at the possibilities of seeing what causes AML, the early drivers, and whether potentially some pre-emptive approach might be more effective in the future. And so in order to do this, we’re just trying to learn more about the disease, one from mouse models and also from the study of clinical samples. The key principles at the moment are that p53 clonal hematopoiesis is present in the community, but doesn’t always lead to AML. And so the first principle is that we should be cautious about doing very extreme things, including allografting, based on clonal hematopoiesis alone. The second principle is, can we identify cooperating hits that might drive AML? And this is still very much an early area of research, and perhaps looking at longitudinal samples will be really important to study what are the events which lead towards AML. AML, in our experience is very frequently biallelic, especially when you look at the single-cell level. Defining biallelic AML is difficult in clinical practice because we don’t have single-cell technologies on hand and we’re unfortunately limited to more imperfect guidance such as high VAF, chromosome plus mutation or two mutations, which doesn’t always tell us what’s going on at the single-cell level. And the third area of research is what’s the role of the immune system in p53 mutant AML, and are there ways for us to positively manipulate that to our clinical advantage? So I think these are all still very early elements of research, and we have a long way to go.

David Sallman:

I think, Andrew, I think it’s, you know, the challenge with the single-cell, like how many, multi-hit cells matter. Is it 1%, 5%, X percent? I think a couple groups are looking at this question, then what is the threshold that we need to be quite aggressive? I mean, as you alluded to, it’s rare in excess blast MDS AML to have single hit. We had around eight patients that were quote-unquote single hit, at least by bulk, and their outcomes to allo, and this was in the setting of high-risk MDS-AML were actually really great. We actually had no, we had one relapse out of that. So if we intervene there with allo, it may matter a lot. But like, as you said, we don’t want to be giving excessive toxicity for clones that may not matter. So I guess, how do we do this? Is this an international collaboration to build much larger data sets? It’s a difficult question to tease out. And obviously, once you miss the window, you’ve missed the window.

Andrew Wei:

So I think there’s obviously discussion at the moment with respect to the future classification and whether the current algorithms are, how useful they are. I think the first thing would be with the threshold of p53 needing to be above 10% at the moment, if the patient already has an excess of blasts, I do feel that perhaps that might be the first thing to go, it’s a bit like having a threshold VAF of FLT3 that we used to have. I think having p53 VAF alone defining AML, I don’t think that’s the right way to go either for the reasons I’ve mentioned before, and that some blast threshold plus a detectable p53 abnormality might be the lowest threshold in order just to classify this group of patients in the easiest way, manner, without trying to make things too artificial.

Eunice Wang:

So, Dr. Mannis, I know you talked a little bit about the poor outcomes of p53 mutant patients overall, especially following ven-aza. So, do you guys, we talked a little bit about transplant. Do you guys take these patients directly to transplant, or how do you manage p53 mutant patients? You put them directly on a clinical trial and then transplant, or what, how do you manage this?

Gabriel Mannis:

Yeah, I mean, I think this is by far the most challenging discussion session of this meeting, just because there really isn’t a right answer. And I think, as evidenced by the discussion, everybody does it a little bit differently. The majority of our transplanters do require that patients with p53 mutated disease be in some form of remission. And depending on the transplanter or you ask, they may require MRD negative by flow, cytogenetic remission, less than 5% blast, but maybe not great count recovery. And so, you know, I think it is not standardized even within our center. You know, I think we don’t know whether patients benefit from treatment before transplant. I think in general, our practice is to give some treatment. We know that there doesn’t seem to be a survival benefit to adding venetoclax to upfront hypomethylating agent therapy, although I would say, especially with patients with high blast percentage, it’s been our practice to use ven to try and get them into some sort of hematologic remission, if not in advance of transplant, at least to decrease their transfusion burden as sort of a palliative approach. But, you know, I’m not even sure that that’s the right thing to do. I think it just highlights, you know, all of the controversy about what to do with these patients, highlights the fact that it’s just a very challenging patient population, and we need to come together as a group, understand the biology, and then use those biological insights to develop safer, more effective treatments for these patients.

Eunice Wang:

So I think, you know, we know chemotherapy really maybe even feeds the disease, right? not really effective BCL2 inhibitors, et cetera. But I also was really intrigued by looking at all of the data that these p53 mutant patients are really immunodeficient. Like, they don’t respond to CAR-T therapy. They were struggling with the failure of magrolimab to think about immune checkpoints for these individuals. But I think there’s some data now about maybe in the post-transplant setting kind of exploiting the anti-leukemic activity of the transplant from the immune system, from the donor immune system to kind of go in there and eradicate the disease. Because we know inherently, intrinsically, these p53 immune patients, there’s not so much we can do to help them maybe, but maybe we could help like the graft-versus-leukemic effect. And I think you had some interesting data, I think, from your studies or from looking at maybe even identifying those patients that are about to relapse.

David Sallman:

Yeah, so I think this is, it’s kind of two main, you know, maybe paths. I think one is kind of Andrew’s path. Can we intervene before? I think what’s the frontline option? That’s maybe a separate discussion we could have in a moment. But I think trying to get these patients to transplant, but critically, currently at best, we have 20% long-term survival. And it’s probably in multi-hit even somewhat, you know, somewhat worse than that. So I think what we know, if we look by multiple different MRD technologies, I mean, every patient is positive pre-transplant, which is a challenge. I do think the deeper molecular response, although given some controversial, has some prediction of outcome. But even post-allo, at our early assessments, many of these patients, around 80% in one of our post-transplant maintenance studies were still positive. And this is often preserved more in the LSC fraction based on data from Steve Chung and I think, shown by a couple of others as well. So anything that we can do to try and buy time and augment GVL is critical. There’s a couple of maybe just easy international collaborations. Can we think about pathways to where we taper patients off immunosuppression very rapidly? We think about early HMA maintenance, which I’ll talk in a moment, and even things like prophylactic DLI. There’s a number of cellular therapies that we’ve heard about this congress, gamma-delta T-cells, NK cells, CAR T-cells, other strategies. And again, we know that effector to target ratios matter. And although as p53 may have its own challenge, can we hyper-focus? There’s a couple of post-transplant cellular therapies that are focused on low-level MRD disease, but p53 may be a really bright subset. The other thing, and again, this is something I think relatively easy to implement is considering post-HMA, post-transplant HMA maintenance. Now, as an all-comer, you know, trial, there’s essentially been one randomized trial that was negative, but really we had no data in molecular subsets. And although we’re trying to grow our cohort, currently our survival seems to be dramatically longer in patients that have early, you know, HMA maintenance. As far as the mechanism and how it’s beneficial, I think there’s still ongoing work. But I do think really one of the bottom lines is that all steps throughout a patient’s journey with p53 is we need to be out of the box. If you’re just kind of doing standard therapy, unfortunately, outcomes are not as we would like them to be.

Eunice Wang:

Have you done the HMA? We’ve been doing HMA in alternating with DLI. You know, have you done that? And what are your thoughts? And we’ve seen some good results with that.

David Sallman:

I think, so our group has had a lot of interest in basically pairing HMA and DLI. I’d say we’ve done this more in the setting of MRD-positive disease or mixed chimeras. So if we’re going to go with DLI, we’ll often do HMA therapy and 7 to 10 days later do the DLI. I think it’s a really interesting question. Could we think about… We have not prophylactically done that in an MRD-negative state, but I think that could be a consideration to look at.

Eunice Wang:

And then you were talking, Dr. Mannis, about tagraxofusp and maybe that drug could have some p53 independent mechanisms of action, I mean, in early studies in combination with chemo?

Gabriel Mannis:

Yeah. So we have a study looking at the first salvage for patients who don’t respond or lose response to venetoclax and a hypomethylating agent, which independent of TP53 status is a very difficult to treat patient population. And of course, that population is enriched for patients with TP53 mutations. And so in thinking about how to design a trial for these patients, considerations are that it needs to be something that is tolerable in an elderly unfit population. It doesn’t have overlapping mechanisms of resistance with ven-HMA. And then we’ve been incorporating tagraxofusp, which is a CD123-targeting drug that’s conjugated to a diphtheria toxin. And so we have just launched this trial. It’s only been open for a few months. But it’s in part predicated on some data that Andy Lane’s group at Dana-Farber has shown looking at the triplet of aza-ven and TAG in the frontline setting. And, you know, although there were a small number of TP53-mutated patients, I think maybe a dozen patients, they saw a 50%, 60% response rate. but the survival looks potentially longer than what we might see with standard ven-HMA. And there are some hypotheses about why, in particular, the diphtheria toxin may be active in TP53 rather than it being an effect of the CD123 targeting. I think that’s still hypothetical, but we are hopeful that maybe this, you know, provides some benefit to these patients.

Eunice Wang:

And I know, like, Dr. Wei, you talked a little bit about STING agonists and the role of maybe interferons. How do you propose that those are going to affect p53-mutant disease?

Andrew Wei:

I think they are a very interesting category of drugs based upon two papers, one in Cancer Cell 24, where STING agonists have been demonstrated to have a direct apoptotic effect on AML cells, which is p53 independent and a second paper in Blood by Marion Subklewe that shows that STING agonists can augment the activity of immune targeting drugs and so I think both of those I think are sufficient to suggest that it would definitely be worth looking at the possibility of STING agonists in AML. The huge blockade at the moment is that many drug companies which did have STING agonists have rightly or wrongly put them on the shelf because of their lack of activity in solid cancers. But I would, I think we would collectively strongly make a plea to pharma that they should really explore the possibility of STING agonists in AML because the biology is extremely compelling and it would be a travesty for industry and also for patients if these STING agonists are incredibly effective, and it took us 10 years to find this out because they’re all locked up in pharmaceutical drug cupboards.

Eunice Wang:

Well, you have some newer drugs as well, Dr. Sallman. I know you had mentioned some that are also coming maybe off the shelf. Arsenic, for example?

David Sallman:

Yeah. So I think, you know, we’ve done a lot of work, you know, and there’s been international collaborations looking at P53 reactivation. You know, the eprenetapopt Phase III trial, you know, narrowly missed CR. Actually, we still need to see these Phase III data. But actually mechanistically, you know, there may be other agents that can be repurposed. Arsenic for certain structural variants can reactivate P53. And actually, I believe the trial is just launched within the GFM led by Thomas Cluzeau. And we’re actually looking on trying to see if we can identify, you know, what is the full breadth of variants that could be reactivated? And I think there’s still some questions. So that’s possible because we have arsenic. There’s also oral arsenic formulations going for approval in APL that could be utilized. There are the, you know, Y220C, which is the most sensitive of all variants to be potentially reactivated. And there’s a drug called rezatapopt to potentially other agents that has pretty exciting data in solid tumor presented a couple times, including in aggressive diseases like small cell carcinoma. So just launched is an early Phase I trial with aza in both relapsed and frontline settings across AML and MDS, kind of initial patients just going on, opened at MD Anderson, opening at our site as we speak, and potentially Sloan Kettering. So I think there could be some unique Y220C. Now, this is 1%, so definitely if we start to see activity, again, really need to be sort of international, you know, collaboration from that perspective.

Eunice Wang:

All right, great. So we had a great session, and I appreciate everyone taking the time to listen to our discussion and looking forward to further advances for this particular subset of AML.