iwMyeloma 2024 | Improving our understanding of myeloma biology: mouse models, epigenetics & spatial transcriptomics

My name is Felipe Prosper. I had the pleasure to chair a session on myeloma that was really very interesting. We discussed a number of issues, mainly focused on the development of mouse models for multiple myeloma that can help us really to understand the biology, but also the treatment of this disease. But we also had some very interesting insights into the use of spatial transcriptomics for characterizing the myeloma and the identification of some new high-risk factors that we can actually target as well in these patients. I will let my colleagues to introduce themselves and we’ll continue the discussion.

So my name is Neils Weinhold I’m from Heidelberg in Germany. And I give an overview of recent results in spatial transcriptomics in myeloma, and tried to convince the audience that these kind of analysis are very important to better understand the pathogenesis of myeloma.

My name is Bob Orlowski, I’m the director of the myeloma section at the MD Anderson Cancer Center in Houston, Texas, and the presentation from our team focused on a new mediator of high-risk disease, which is called CLPP, an endopeptidase in mitochondria that’s involved in generating energy for myeloma cells. And there is now a drug in clinical trials which targets a transcription factor which is involved in upregulating CLPP, and therefore the drug downregulates it and causes myeloma cell death.

And basically, the take home message is that even within the high-risk subgroups based on molecular profiling, high expression of this protease is another additional you could call ultra-high-risk feature, and part of that is because of the role in energy production of this protease. So basically a high-risk myeloma cell with lots of energy is worse than a high-risk myeloma cell with low levels of energy.

Well in my case I my talk was on two different things actually: the first one was the role of epigenetics and how we can target epigenetics in myeloma to develop new treatments, and the second part of the talk was more related to the non-immune microenvironment. We’ve been listening in some of the talks about the immune microenvironment, about how the different immune cells and the impact that the myeloma has on the immune cells, and actually how therapy also may actually modify the immune microenvironment of the myeloma to help or to prevent efficacy of the therapy. So what we’ve been looking in my lab is how the other part of the of the microenvironment, which is the non-immune, the mesenchymal, stromal cells and endothelial cells can also participate in regulating the myeloma. I’m sure you’re familiar with words years ago, how MSCs can foster, can stimulate the growth of myeloma plasma cells, but this type of studies were done some time ago. And now we with a new tools, with new technologies, we can really dip much further into the mechanisms by which those things are happening in myeloma.

I think one of the things that impressed me about the session, and you mentioned some of the mouse models, but it’s the faithfulness with which these models are now able to model the real situation in patients.

And I think it argues that we should be doing even more mouse modeling, because if we can predict something will work based on the mouse data, it seems like it has a much better chance of working in people. And that way, maybe we can eliminated the things that don’t look like they’ll work because we have so many exciting new arenas to investigate, we’d like to just focus on what looks like will be working the best.

Yeah, I totally agree, so I was kind of astonished that the changes seen in the mouse model are very similar to what we have seen in our longitudinal study, and also in our spatial study, like an enrichment of stromal cells, endothelial cells, specific subsets of NK cells and macrophages are very, very similar.

And even the growth pattern, according to immunohistochemistry was very similar to what we see in myeloma. So I’m thinking we are on a good way to have now finally proper models to better study the pathogenesis.

I agree, I also think that what you guys are doing with the spatial is very relevant.

We’ve been always arguing with our patients that myeloma is a very patchy disease, but we have never been able to get into not the cell types, but actually the transcriptional programs and the genomics. And with these new technologies, the technologies that allow to interrogate the cells in the context of the other cells, we’re going to be able to learn many things, whether the T-cells that are close to the myeloma cells are different from the T-cells that are in other areas of the bone marrow. And these, I think, are going to be key findings to really develop better therapies, and we haven’t been able to do that yet. So that’s something we need to work on – and the mouse models again is going to help us enormously in that.

What are actually your experience regarding subclones in mouse models, so do you see subclones?

I haven’t done a lot of that. I know Dr Climent has been working with exome sequencing.

And so and it’s actually you see subclones because as a matter as you know, his model is a genetic model, but actually, when you do exome sequencing, you not only pick up the genetic alterations that you introduce in the mice, but you also acquire new alterations.

So in his model, they developed subclone subclones that can be important for the disease so that is going to be a very useful tool as well.

I was going to say there are also now mouse models where you can take the samples from the patients and actually propagate them in the mice, which there aren’t enough of those. I think we all need to probably continue to work to develop more of them, but you pointed out that there are some differences between the mouse models and the human models in terms of, for example, CD38 expression. So I think the more faithful we can even make the models, the better and probably trying to find a way to propagate the human cells and the human microenvironment in the mice is equally important.

One of the things that I know is being done with some of these mouse models is to humanize some of the main targets. So for instance, the model from Martinez-Climent, he has been able to humanize cereblon, he has been able to humanize CD38.

And that is going to give us another opportunity, which is really to test therapies in these mouse models, because that’s the other thing that we need – not only understand the mechanisms, but even to be able to use it for testing new drugs and new combinations, which I think is very important as well.

And when you treat these mice, do you have any data on where residual cells are located?

Yes. Yeah. You can imagine that working in with Bruno and Jesus, there is no chance that we don’t do that. So they are not only doing MRD, but they are also doing CTCs and looking for the CT… I mean, now we are able to do bone marrow aspirations on the mice without killing them, and we are able to bleed the mice without killing them. So Bruno is looking for CTCs and characterizing the CTCs and the bone marrow cells and doing the same kind of experiments that he does with patients. That’s great.

So we’ll see.

It is tedious work and it’s a complicated work, but I hope we’ll be able to come up with at least ideas and new hypotheses.

So will we ever get to a point where we have co-clinical trials? B ecause this has been looked at in other diseases, where you take the patient tumor cells, you put them into a mouse, you randomize a whole bunch of these mice to various therapies, and whichever one works the best is what you give to the patient. I’m not aware of anyone having done that in myeloma yet, but do we think that’s something that’s going to come?

Well, I’m not sure, it’s at the end of the day, how, not feasible, but how practical it will be. I think I like it better the personalization based on some of the genomics and some of the other studies where you can really tell whether a patient, I mean, for instance, what Francesco presented this morning, which is a risk based strategy for patients, but not only for risk of progression of the disease, but actually for risk of response to specific how is the patient going to evolve depending on the therapy? That could be very useful as well, and maybe more practical than just using mice and doing treatment with the patient and with the mice at the same time. But we don’t know –

I guess my concern is we’ve got so many different therapies now that are having 100% response rates that it becomes difficult for clinicians to pick between them, and maybe this could be at least one piece of information to help guide one way or another.

Yeah, that could be. Still, we don’t cure myeloma, so we’re at 100% responses, but we are not there yet.

But we will get there.

We hope so.

I totally agree with you. So we even have bad experience in getting simple parameters like BCMA expression. So we have heard about these biallelic losses of BCMA, and still, clinicians treated patients twice with BCMA products, even although they are also already BCMA negative according to retrospective data, But so they didn’t check for that. And nowadays it’s still very complicated to talk to pathologists and convince them to do these kind of, stainings before BCMA products are applied. But that’s certainly a must for the near future to check at least for the target gene expression.

Yep, I agree.

Not only for the response, also because sometimes we in clinical trials you include patients that heterogeneous in these kinds of things, and you may miss some good therapies because you are not selecting the right patients. So, I’d like to thank both Bob and Neils for their presence here today, and the rest of the speakers of the session. I think it has been very, very interesting and with a lot of new ideas and I’m sure we’ll continue to work in this area and try to provide new alternatives for our patients, which is, at the end of the day, what we are really looking forward.