iwMyeloma 2024 | Understanding mechanisms of resistance to targeted therapies in myeloma

Good morning, we are here in sunny Miami with my colleagues and we will discuss today a summary of this outstanding session about mechanisms of resistance in myeloma to the different targeted therapies. My name is Francesco Maura from University of Miami, and now maybe my colleagues can introduce themselves.

I’m Paola Neri and associate professor at the University of Calgary.

I’m Leif Bergsagel, a professor of medicine at the Mayo Clinic in Arizona.

I’m Charlotte Pawlyn from the Institute of Cancer Research and Royal Marsden Hospital in London.

So, Paula, I think your talk was, quite, interesting in particular, because immunotherapy is definitely the hot topic in most of the myeloma conferences and trials right now.

So can you, you know, summarize a bit why your translational work can be relevant for not just research but for clinicians.

Absolutely, thank you Francesco. So I think what we’re struggling as a clinician is the fact that even if this therapy have been very successful, we still have patients, they unfortunately progress on it. So we have tried to understand, especially on the anti-BCMA T-cell engager space, which was the topic of my presentation, is to understand why these patients progressed, and to your point, can we prevent that for the future?

So we have looked at the tumor component and the factor are associated more to T-cell therapy that’s associated with resistance. And I can briefly say as I show in the talk that there is antigen escape, so the loss of BCMA due to either biallelic loss or monoallelic loss, acquisition of mutation in the extracellular domain of BCMA seems to be the factors that really lead to resistance, especially post T-cell engager.

Of course, talking about T-cell therapy, we have also looked at the T-cell phenotype of these patients prior to starting therapy and during the therapy and we have noticed that unfortunately, there is that T-cell exhaustion or dysfunctionality in the bone marrow of these patients in their T-cells. Unfortunately, they don’t do well on this therapy.

So my point is then going back to your question, we definitely need to interrogate the tumor and immune cells before selecting the therapy for our patients, because we can actually even preventively anticipate if these patients are going to be a good responder or not. And probably we have to monitor these patients as well, because there are certain mutations that seems to be acquired during therapy because of the continual pressure we are providing.

So thank you.

Charlotte, I think lenalidomide is probably one of the most used, if not the most used drug right now in myeloma, and most of the patients that relapse were right now, at least in the United States, under lenalidomide maintenance. So your work on lenalidomide mechanisms of resistance is definitely important beyond the lab.

So can you tell it more about that.

Yes. Thank you – so we have done or I presented here today two different projects that we’ve done in the lab looking at different mechanisms of resistance to IMiDs, more broadly, lenalidomide, clearly, most used clinically. The first was looking at the effect of missense mutations, so we know that there are some genetic changes in cereblon as patients become resistant that would lead to loss of cereblon. So we know that there are some copy number changes and, splicing alterations that would lead to, to a kind of reduction in cereblon expression. But then we also see a subset of patients having a single base or missense alterations where it’s less clear why that should lead to resistance.

And so we took a collection of these mutations and modeled them in the lab to try and understand A) do they all lead to resistance, and B) do they lead to resistance equally across the IMiDs? And then also the novel CELMoDs, the novel IMiD drugs that have been developed more recently. And interestingly, in that analysis, when we model that in cell lines, we see three different patterns. We see some mutations that, although they have been identified in patients, don’t seem to lead to a lack of, degradation of near substrates in cell line models. Some that clearly do and completely obliterate the cereblon function, and they were most commonly found in the kind of tri-tryptophan pocket, the really important portion of cereblon for binding. And then we had this interesting intermediate group where, these mutations led to resistance to IMiDs, but seemed to still enable response to CELMoDs. And for me, these are important things to think about as we move into a clinical world where we’re starting to kind of panel or whole genome or whatever sequencing method we’re going to use in our patients, maybe even multiple different time points of disease. We need to know how we’re going to use that information clinically. And for me, it’s important that we don’t jump to say agent is not going to work because we see, a mutation if we don’t actually know that that is an important event that will prevent that agent being active.

It’s definitely important, and then comes to the last talk, Leif, about these new potential synergistic mechanisms to also overcome eventually some of these.

And also it’s interesting that it’s not over for small molecules, right? We all talk a lot in immunotherapy, but I think there is still space for more targeted therapy.

Yes, well, you know, small molecules really are a backbone of our treatment with Revlimid and Velcade. And so I presented work that was a collaboration between Emory University, the University of Calgary and the Mayo Clinic that really tells a nice story about another mechanism of resistance to lenalidomide, and that it appears that the backbone of our therapy, which is RVd, Revlimid, Velcade, dexamethasone, all three of those drugs target this critical structure in the cell that orchestrates the gene expression: it’s called the super-enhancer. And the cell, as it evolves, can find ways to shift or change, lineage plasticity was a term that that was used, and that allows it to escape from the effects of Revlimid and allows the gene expression program that allows cell survival to continue, even though Revlimid seems to be doing its job, which is to bind cereblon and degrading Ikaros and Aiolos, so this is a novel mechanism of resistance. And so what we showed further is that there may be ways to partially overcome this by combining Revlimid or pomalidomide with drugs that target other aspects of the super-enhancer machinery, one of which is a protein called P300 and a drug called inobrodib. And the beauty of that is that you can use low doses of that drug, which might otherwise be toxic, but when combined with pomalidomide really see very effective cell killing and very long survival in murine models. So I think the work changes our view about how the drugs that we have work in myeloma, and also gives us thoughts about ways that we can make them work better in the future.

So the question that I have, I mean, most for all of you is in the settings where there is a lot of push, a lot of studies presented in different conferences about immune environment and how important is the immune environment. But then at the end of the day, most of the mechanisms of resistance that we keep finding are cell intrinsic.

So what’s your point about, you know, the opportunity to investigate, you know, both compartments and the idea that immune environment maybe is not really the main target of our therapy, it’s just like a consequence of the tumor.

Can I start maybe from an IMiD point of view. So I think, immunomodulatory drugs, we knew that they targeted the immune microenvironment, but perhaps we focus a little less on the mechanism of that and what might be changing as patients become resistant to IMiDs. And we know that IMiDs act in T-cells and NK cells, they lead to Ikaros and Aiolos degradation in T-cells, for example, that leads to upregulation of IL-2 and increased T cell activity. And that, at least in part, contributes to the activity of IMiDs in patients. We presume that both mechanisms are kind of active, and it’s important in terms of going forward in their immunotherapy space that actually that lends itself to be an excellent combination partner for some of the immunotherapies as well, because it does have that direct activity outside of the myeloma cell intrinsic realm.

But the relative contribution to resistance of both of those compartments, I think we understand a little bit less, we’re trying to look at some of that through modeling it in various different ways, but the interplay between those two things, I think we understand less.

Well, probably to add to the two, Francesco, right now we have data on the relapsed/refractory myeloma setting, right. So we have to see what we’re going to see in the newly diagnosed. Is it a bit different because patients are naive to therapy for example? So maybe the resistance mechanisms will be a bit different and there are also all these new data coming from combinations. So if we combine, for example an anti-BCMA T-cell engager with an anti-CD38 or with an IMiD, are we going to see development of hopefully not too many mechanisms of resistance? Maybe mechanism of resistance will be different, so I think right now what we have shared is just a reflection of where we are at the moment. But maybe in a few months, a year we may see something different because of the different way we use this therapy.

You know, I’d like to ask you a question, Francesco, there are a number of presentations from your laboratory. What was the important message?

Across all the presentations?

No, you can choose the one that you think is most important.

I think I think the general pattern that we are trying to push for the community is that, you know, like FISH is basically a horse, target sequencing is steam engine and whole genome sequencing is the modern car. And so because myeloma is a journey, you want to start with the right equipment and I think you want to start with the car and not with a horse if you want to travel from Miami to New York.

So I think our point is whole genome sequencing is not just informative right now because we can identify, as you know, Sarah and others show like alterations that are important for prognostication. But I think it’s also important to understand the evolution of the disease and things that we still don’t know because we never looked into. And so I think that it’s still a gold mine that needs to be investigated.

There’s so much data we can still pool like from the genetics, somatics, integration of the two. And so, that’s more or less why we work so hard on that because we feel there is a lot of space to learn out of whole genome sequencing data.

You mentioned that an individual patient can put or you could examine the potential outcome of an individual patient on different treatment regimens and come up with a prediction. Is that a tool that you could put in the hands of a patient?

So we release the tool as a website but it’s for research only. And that’s clearly highlighted. The interpretation of the model is complicated, so we have a guide.

And that’s why we think it’s only for researchers, and even for researchers sometimes is difficult. We get emails from people that ask why I get these results. So I think in the future every aspect of our society will be driven by data. It can be like how we regulate traffic and how we treat our patients. So the doctor will rely less and less on their experience and more and more on these tools based on thousands of patient information. And so the idea to develop this tool in an academic setting, I think, is fascinating, and an exciting challenge. And if we don’t do, companies will do.

And I think it’s inevitable that 20 years from now, doctors will decide treatment based on data and not based on their experience or guidelines established by experts. The field is going to change. And so I think we need to contribute as academic scientists to that and possibly lead that in order to make the tools accessible to everybody, and not just for high top cancer center in the United States. So coming back to the tumor plasticity, which was like the first talk, how do you think the tumor plasticity change affects not just the, you know, MYC or IRF4, the sensitivity to lenalidomide, but also the sensitivity to the next line, like are these plasticity, also involved in reduction of BCMA or CD38? Like, do you see that patients with certain alterations might be less sensitive after like some state of plasticity and dynamics?

So I can give you a hypothesis, I can’t give you data on that. But good question.

You know, in prostate cancer, when they have this lineage plasticity, the cells look different, they talk about neuroendocrine differentiation. And I don’t think we really see that in multiple myeloma, although we do see some ectopic gene expression. The cells I think more or less look the same. But the mechanism, that I talked about today is I think the mechanism of dexamethasone resistance may have contributed to the mechanism of IMiD resistance. A lot of our therapies target the plasma cell phenotype, which is governed by the super enhancer. And so I predict that you might see less of or slightly lower expression, if not dramatically lower expression of some of the targets that are plasma cell specific.

I think that there are some data about lenalidomide refractory patients, even most of the patients in trials with immunotherapy, CAR-T and bispecific, are exposed to lenalidomide. Not everybody is refractory and the refractory seems to be associated with worse outcome usually multivariate model. Is that correct? So that would make the case?

Absolutely these are again, a reflection of where we are right now based on the fact that these patients are coming off all from using IMiDs, and then the question is which dose of IMiDs are they being exposed to? Is that a full resistant? Or maybe if we use a CELMoD to Charlotte’s point, we can rescue still these patients because they have we know more much more affinity to cereblon.

Well I think that was a great discussion. Thank you all.