Myelodysplastic Syndromes: Translating Genetics to Clinical Practice - Episode 3

Diagnostic Challenges: When Is It MDS

Transcript:Mikkael A. Sekeres, MD, MS: You’re getting at a little bit to what Rami referred to earlier as “the clonal hematopoiesis of indeterminate potential,” people who have these clonal abnormalities and may not have overt blood abnormalities or something that needs to be acted upon immediately. I think we’re learning more and more about this. And what I’ve heard from some pathologists is that they are geared up to make a diagnosis of MDS, even in the absence of dysplasia, if there are enough of these molecular abnormalities that have been detected that are consistent with the diagnosis of MDS. I don’t know how others feel about that.

Jamile Shammo, MD: I would have a problem with that, and I will be perfectly honest with you because I do think that it’s very important to include mutational analyses in certain diagnoses. But when it comes to MDS, the only one that I will probably think it to be specific to is the SF3B1. I mean, this is in 80% of patients who have refractory anemia with ring sideroblasts. But even then you should be able to see a ring sideroblast—which is the best feature of this specific entity. But, aside from that—and I know we’re going to be talking a lot about this later—what about people who have perfectly normal blood counts and they had DNA methyltransferase mutation? Does that make them people who have MDS?

I think nowadays the MDS diagnosis is strictly linked to the identification of dysplasia. The problem is that the identification of dysplasia varies from one observation to the next. To your point, often you may have to do another bone marrow so that the pathologist will appreciate the dysplasia at that time. And that’s in the absence of cytogenetic or blasts. So, it leads you to believe that the diagnosis of low-risk MDS is actually where the problem lies. Because we have blast cytogenetic abnormalities, then I think it’s relatively much easier to make a diagnosis. That’s one challenge.

The other challenge is in hematologists seeing those patients that have cytopenias and for the primary care physician to actually send you that patient for evaluation. Often, there is some degree of either age bias or perhaps multiple comorbidity bias. Well, if they have those many illnesses, it’s natural or okay or acceptable for them to have cytopenias? I’m not sure I would agree with that. So, that’s my take on diagnosis.

Ellen K. Ritchie, MD: I see that a lot, too, that patients are often referred relatively late. If they have heart failure or they have a little renal insufficiency, their doctor is not actually sending them for analysis. But a lot of times, the anemia is contributing to their own comorbidities. So, it’s really, I think, a real challenge for us to communicate to the primary care physician when it’s appropriate to refer a patient for a myelodysplastic syndrome evaluation.

Rami S. Komrokji, MD: Let’s look at it from other side. So, obviously, as you mentioned, there are patients that are not cytopenic at all that will have a clonal hematopoiesis maybe because of aging, but those tend to happen in very low clonal burden frequency. The variant allele frequency for those patients, like if they have a TET2, is going to be 5% or 10%. Those are probably a little bit different. But, if you have a patient who is cytopenic and you have some of the characteristics of mutations, and you are just lacking the dysplasia, currently we put those patients as “clonal cytopenia of unknown significance.” I think we have to learn a little bit more about that group and how they behave, because they could end.

The cutoff now between those two groups is just presence of some dysplasia—which we all discussed. That’s sometimes very difficult to find, and it depends on the hematopathologist’s experience. So, I think there will be a role for the mutations down the road. If you have somebody, like in CMML (chronic myelomonocytic leukemia), we know that there are certain patterns of mutations—like a TET2, SRSF2 mutation—in the setting of monocytosis with a variant allele frequency not justified for 10%. We just need some time to learn about those patients. When do they declare themselves? But I think down the road, we are going to be using a form of diagnostic criteria.

Mikkael A. Sekeres, MD, MS: If I could take the liberty of trying to sum up what we’ve just said: anemia or other cytopenias are not just a normal consequence of aging, and these people should be referred to a hematologist for a bone marrow biopsy to get a determination of whether or not they have MDS. At the same time, detection of an isolated molecular abnormality in the setting of no cytopenias or a mild cytopenia doesn’t make an MDS diagnosis. However, in the right clinical setting—and, as you mentioned—spliceosome mutations or other molecular abnormalities with persistent monocytosis and cytopenias may get you close enough that you’re more comfortable making that MDS diagnosis. Are there any blood tests you send on patients referred to you, what I call the “I don’t want to be an idiot test,” to rule mimics of MDS?

Rami S. Komrokji, MD: I think that’s a great question. So, for a while, we were doing even, as a reflex. We checked LGL (large granular lymphocytic), PNH (paroxysmal nocturnal hemoglobinuria) profile on almost every patient—at least the lower-risk patients where the blasts are not increased—and then obviously the simple things. We sometimes forget, but Vitamin B12 deficiency…I’ve seen cases where sometimes even the megaloblastoid changes were called AML (acute myeloid leukemia), and it was B12. So, you always have to make sure that we really ruled all the nutritional deficiencies. There are cases of LGL that could present like MDS, less likely PNH. I’m not sure about testing in every single patient unless there is hemolysis. I think sometimes a couple deficiencies could be a great mimic for MDS, and in the right setting, one could think of that as well.

Ellen K. Ritchie, MD: I think looking at medications is also very important. Many, especially older patients, who tend to be diagnosed with myelodysplastic syndrome have multiple specialists or multiple doctors they’re going to who are giving them multiple medications. Someone might not be sitting down with them and looking to see exactly what they’re taking. And many medications actually can cause cytopenias. So, I think it’s very important, particularly in elderly patients, that you spend some time looking at their medications, because many times we find that maybe a new medication that was introduced to their regimen is influencing their blood counts.

Jamile Shammo, MD: I’ll add to that list: viral infections, hepatitis, and HIV infections can certainly cause dysplasia. Remember the epidemic of MDS in HIV patients? It didn’t pan out to be. But those are all dysplastic changes induced by infection or autoimmunity.

Mikkael A. Sekeres, MD, MS: Also, I will add that the age group in whom STDs are increasing with the greatest frequency are those in their 70s and 80s, those nursing home residents. So, we can’t eliminate HIV just because of age.

Transcript Edited for Clarity