iwMDS 2022 | Updates in classification and risk stratification in MDS

Rafael Bejar:

Hi, welcome. My name is Rafael Bejar. I’m a physician-scientist at UC San Diego, and I’m here at the first International Working Group for MDS in Miami, Florida. And with me, our panelists that shared the stage with me when we just had a discussion about novel changes to MDS classification, MDS risk stratification, and germline lesions that can lead to MDS and other myeloid malignancies. So I’ll have my co-panelists introduce themselves and we’ll have a discussion about what we talked about. So start with you, Dr Nazha.

Aziz Nazha:

Thank you, Rafael. I’m Aziz Nazha. I’m a staff physician at Thomas Jefferson University.

Jane Churpek:

I’m Jane Churpek. I’m an assistant professor at the University of Wisconsin, Madison.

Robert Hasserjian:

And Robert Hasserjian. I’m a hematopathologist. I work at Mass General Hospital in Boston.

Rafael Bejar:

So, Robert, maybe I’ll start with you. You had a really important discussion about the novel classification system for MDS coming from a couple of different sources. What are some of the key highlights that you’ve presented to us today?

Robert Hasserjian:

So I’ve talked about two updates to the classification system that have come out. There are two independent classifications that have been released, so that produces challenges because we have two classifications we have to now understand. The good news is that both groups that developed these, they’re basically very similar and they overlap a lot. So it’s gratifying that two groups that have looked at the evidence accumulated since the last classification have essentially come to many of the same conclusions in how they handle many of the features, but there are some different features.  So I highlighted some of the different features in how some of the entities are handled, some minor differences in how some of the genetically defined entities are created, and also some differences in blast thresholds that define a new group. And one of the classifications, the International Consensus which I was involved with, which is called AML, MDS/AML and Overlap Group, which is maybe one of the more controversial entities that was created. So I went over the differences and also emphasized many of the similarities in the two classifications.

Rafael Bejar:

And as I understand it, there were a couple new entities that were described basically on their molecular features, for example, the TP53 mutant group. Can you tell me a little bit about that?

Robert Hasserjian:

Yeah, that was an entity that was based on some recent data and, again, essentially identical in the two classifications defining a very high risk-group of MDS that has a biallelic inactivation of the TP53 gene, usually through mutation, and then a deletion. And it represents a very aggressive MDS subtype. And unfortunately these patients have a very poor prognosis. And the hope is by identifying this group maybe novel therapies can be developed that target this group. And I think it’s good to separate these out from other MDS. And I think it’s an important group to recognize.

Rafael Bejar:

So we’ll be seeing this new classification put in place in the near future then.

Robert Hasserjian:

In the near future. As I said, both papers have been released recently, the WHO produces as a blue book later this year and so I think certainly later this year, we’ll start transitioning to the new classifications in terms of our terminology.

Rafael Bejar:

So Jane, one of the new updates in the WHO classification included an update to the germline predisposition genes or diseases. Can you tell us a little bit more about that?

Jane Churpek:

Sure. So it actually is kind of a nice feature because before there was a table but it was unclear exactly how to combine that with the underlying myeloid neoplasm pathology. So now it’s kind of a nice template where you can put the MDS/AML diagnosis at the beginning, attach the germline variant at the end. It also, I think, leaves open the concept of CHIP or CCUS that are occurring in those germline mutations or potentially even some unique pathologies to those germline disorders for the future, as we start to recognize those. So the list of genes is expanded and then this kind of paradigm for how to attach these together were updated.

Rafael Bejar:

And one of the things you mentioned is that we were probably underestimating the rate of germline mutations in our patients with MDS. How is it that you decide whether or not you’re going to do germline testing in your patients?

Jane Churpek:

So I have experience doing this over time and the MDS NCCN guidelines have definitely started to incorporate how should we actually do this. So there are clinical features that you can recognize, including individuals who have another close relative who also has a similar blood disorder, a strong family history of other cancers, or the patient has multiple tumors themselves. There can also be some other syndromic features like pulmonary fibrosis occurring with MDS is suggestive of a telomere syndrome, for example. So those clinical features are one way that we can identify patients.

Jane Churpek:

But now that we’re doing much more molecular classification and testing, we are testing a lot of genes that we want to find the acquired mutations, like P53, but a subset of those are actually germline mutations. And so we can then scan the mutations that we’re identifying there. Those that seem to be heterozygous, so a variant allele frequency 40 to 60% or higher, we can suggest those patients should be getting germline testing. The challenge is always implementing that because there’s a bit of subjectivity in it.

Rafael Bejar:

Sure, it’s understandable. One of the things I was struck by was that even patients who develop MDS in their 70s could still have a germline predisposition and may not necessarily have had a prodrome or something that told you that there was something wrong with their blood system beforehand. So you mentioned the gene DDX41. How prevalent was it?

Jane Churpek:

Yeah. So about 2.6% of MDS patients in the recent IPSS-M publication had a DDX41 mutation. There were several publications before this that pretty much in the majority of MDS and AML patients where you find a DDX41 mutation in their MDS or AML cells, almost all of those are germline. So that one is much more clear. A lot of the other mutations, they’re usually going to be acquired, but there’s always a small subset and picking out which ones those are is really important.

Rafael Bejar:

And besides the obvious implication for other family members, I guess there’s a very big implication for people who might be potential bone marrow donors, right?

Jane Churpek:

Yeah, this is also an area that data is accumulating. What should we do? Is it an absolute contraindication? Because in the past we didn’t really do this testing. So there are examples of using someone who has a germline mutation and their recipient did well for a number of years and they might not have done well with MDS or AML if you just let them on their natural course without a transplant.

Jane Churpek:

So for each one of these, we still have a lot of work to do to figure out should you do it or shouldn’t you. The general rule of thumb is, if you have alternative options, probably having someone who doesn’t have that MDS or AML predisposition is the better donor, but it has to still be individualized, especially when there are a lack of donors.

Rafael Bejar:

Makes sense. And, Aziz, you mentioned testing for somatic mutations and how they can influence our assessment of patient risk. How is it that we can best do that going forward?

Aziz Nazha:

So I think next generation sequencing becomes a standard on all the patients. What we see now with MDS and through our work and others, we have demonstrated that those somatic mutations have significant impact on overall survival and leukemia transformation. And in the question how do you integrate them with the clinical variables, certainly they have added benefit on the top of a clinical variables, but the question how do we integrate them and how many of those genes have significant impact and how many should you include in the scoring system or not.

Rafael Bejar:

So would you say that doing genetic sequencing is really something that should be part of the standard of care?

Aziz Nazha:

I think so. Now we have more evidence in terms of the diagnosis of the disease and the differentiation between MDS and CCUS, but at the same time, it could provide some prognostic impact. Now is it beneficial on all the patients? Probably not, but doing the test might yield more information in terms of the diagnosis and the prognosis of the patients.

Rafael Bejar:

And is it fair to say that we can use this information to reassess the patient prognosis even after they’ve been diagnosed, so perhaps after therapy or even further along in the disease course?

Aziz Nazha:

Potentially. Although, I mean, this is kind of a work in progress and how do you assess whether these evolving mutations, we talk about some of those, whether germline they started already, and then what’s really, I think, unclear how often do you monitor. Like once you diagnose the patient and you start the therapy, do you monitor those mutations during therapy? Certainly in AML we have more evidence of monitoring minimal residual disease has an impact on overall survival and other outcomes. We still need to probably define this better in MDS, whether that minimal residual disease with some of our therapies that we have have an impact later on on the disease and survival.

Rafael Bejar:

That makes sense. And your comment actually reminded me of something else there that’s new in the classification system. So, Robert, sometimes we sequence our patients because we think they might have MDS. We do that bone marrow biopsy and the pathologist tells us that they don’t meet criteria for MDS, yet they do have some of these mutations. So that now forms a new category that we call clinical cytopenias of undetermined significance or CCUS. And now they’re formally recognized, right?

Robert Hasserjian:

That’s correct. Yes. CCUS have been recognized in both the WHO and the ICC classifications as has CHIP. And I think the thing that separates it is morphologic dysplasia. However, morphologic dysplasia, as I said in my talk, is an imperfect feature, right? It’s somewhat subjective. Let me put it this way. If I know the patient has four mutations, pathogenic high VAF, and I look at a marrow, I might see more dysplasia than if I knew the patient had no mutations. That’s the real truth.

So I think we have to combine them both together, ultimately, to arrive at the best classification with the patient. But having the CCUS group that can be further studied, I think, makes us feel these patients at least won’t be lost to follow-up. Once we know they have mutations, they’ll be followed carefully, until they transition to MDS by developing dysplasia.

Rafael Bejar:

And perhaps when we learn more about that, we may actually broaden the umbrella of MDS, to include-

Robert  Hasserjian:

Certainly, it might come to that, yeah.

Rafael Bejar:

So I had the privilege of presenting the revision to the revised IPSS, the International Prognostic Scoring System. It was the International Prognostic Scoring System-Molecular that does what Aziz described, where it takes somatic mutations that we identify in our patients and incorporates them with clinical variables, such as blast count in the bone marrow, cytopenias, and the number of blasts in the bone marrow, to try to come up with the best predictor of a patient’s prognosis. And this is important because we use this information to help us identify the best therapy for patients with MDS.

Rafael Bejar:

So the IPSS-M, which was just published a couple of weeks ago, I think will be a replacement or a substitute for the IPSS-R that now incorporates additional information and, as you mentioned, that if somatic mutation testing is really becoming part of the standard of care, this is one of the ways that we’ll be able to use that information in all of our patients. And I think it’s a great advance. So be on the lookout for the implementation, I should say, of the IPSS-M coming into the near future and perhaps being incorporated into clinical trials, into NCCN guidelines, and other things of that nature. I think it is going to become part of what we do for all of our patients.