TRANSCRIPTION:

DR MORGANSTEIN: So this is a 78-year-old gentleman who’s a retired major in the Army. I initially met him in the Intensive Care Unit when his hemoglobin was 4. And he’d been, over the course of time — he’s a tough guy, didn’t want to go to the doctor and got progressively weaker, more short of breath. Went to the ER. Had a hemoglobin of 4. Initially got transfused up. Had a GI evaluation, was negative. I did a bone marrow pretty quickly on him and he was found to have myelodysplasia. He had refractory anemia. Cytogenetics were normal.

DR LOVE: Could I just ask, because I know certainly a common route for MDS is anemia. This is kind of interesting. I don’t know how often you see people with hemoglobin of 4. How long had it had been since he was functioning well? And what was his baseline functioning?

DR MORGANSTEIN: He’s an incredibly functional guy. He was riding his Harley up until he came into the hospital. And then he had a significant other who noticed that he was getting more and more fatigued. He was shockingly relatively asymptomatic for a guy, 78-year-old with a hemoglobin of 4.

DR LOVE: Any comorbidities?

DR MORGANSTEIN: No. Nothing. He’s healthy as could be.

DR LOVE: Had he, like, sort of avoided doctors in the past?

DR MORGANSTEIN: Not purposely.

DR LOVE: How long do you think it was that he was kind of the trajectory down?

DR MORGANSTEIN: If you talk to him, at least a year, if not more. He just got progressively unable. He was working on his house at the shore. He had just gotten more and more unable to do things.

DR LOVE: Before we go on with the case, Harry, I’m curious about the idea of the presentation of MDS and the diagnosis of it. How often is it delayed and missed?

DR ERBA: The time course in this patient is very illustrative. If this patient had an acute drop in his hemoglobin down to 4, at the age of 78 he would have been symptomatic. I don’t care how good a shape he was in. So I think the fact that he was actually able to present feeling fairly healthy but with a hemoglobin of 4, just some fatigue, tells you that this has been going on for a while. So it leads you away from things like acute GI bleed, which, I agree, needed to be ruled out.

In terms of how the diagnosis might be delayed, we don't do routine CBCs to find when a cytopenia might have started. So a lot of times, our patients come and they don’t have prior hemoglobins or blood counts to help us in that regard. So the clinical history becomes really important.

DR LOVE: Guillermo, what do we know about the diagnosis of MDS? Any thoughts about how often it might be missed or delayed?

DR GARCIA-MANERO: So first, I agree with Harry. I mean, this probably has gone of for probably longer than a year.

The diagnosis is complicated. And actually, a couple of years ago we reported in Blood our experience comparing what we call the “outside” bone marrow with the inside bone marrow. And this is not to say that MD Anderson bone marrow is better than the outside, because actually our referral pattern is not just from community to — there’s maybe people actually that come from even higher centers than us. And the reason I mention of that paper, it was around a 20% discrepancy rate. Now, the reviewers made us tone that down. And I think if you look at that paper, it says like 12% or something like this. But overall, it’s like around 20%.

The reason I did that, the studies, because it’s painful. Sometimes the patient had had a bone marrow a couple of weeks earlier by any of you. And then they say, “Why do I have to do this?” Plus, actually, it’s not cheap, either. So I wanted to see what this second bone marrow meant. And our surprise, actually, that if you have the slides, not just by the reports, you find around a 15%, maybe even a 20% discrepancy rate, usually going from a lower risk to a higher risk type of situation.

DR LOVE: So can you describe what his marrow looked like and what his workup showed?

DR MORGANSTEIN: Yes. So his marrow, he had refractory anemia. His cytogenetics were normal. Once he got out of the hospital, he was somebody who we saw relatively frequently and required, fairly often, transfusions. So we tried him on erythropoietic therapy. He had absolutely no response. So we abandoned that fairly quickly. And then, because he was needing transfusions about once a month if not more, we started him on lenalidomide. And initially started him on 10 mg. Pretty quickly went down to 5 mg because of cytopenias. And then at that point, he maintained on lenalidomide for well over a year. Did great. Became transfusion-independent. Quality of life was great. Hemoglobin actually went up remarkably high.

At about a year and a half, he started to develop some other mild cytopenias, neutropenia, mild thrombocytopenia. So we stopped. And this is about a year and a half ago. He’s been transfusion independent. Good quality of life and doing well.

DR LOVE: So Harry, can you talk a little bit about your take on this case? Would you have managed the patient the same way?

DR ERBA: Let me come back first to the diagnosis of the myelodysplastic syndrome. You say that it showed myelodysplastic syndrome, but was it unilineage.

DR MORGANSTEIN: So it was just refractory anemia, not multilineage cytopenias.

DR ERBA: Okay. And prognostically, that is very important. Then the other issue is, myelodysplastic syndrome remains a diagnosis made by morphology. There’s no exception to that. It doesn’t mean that a patient can’t have a myelodysplastic syndrome, a myeloid stem cell disorder that is not detected by the pathologists. And I think that’s one of the problems we have in the discrepancy between these readings is, how much dysplasia does a pathologist need to see? By definition, it’s 10% of the cells in the lineage have to have dysplastic features. Very few reports give you the number of cells that are actually dysplastic. And you’re relying on the experience of the pathologist reading it.

In terms of this patient, you’ve ruled out other causes of the anemia. He’s been transfused up. He clearly was symptomatic. And so I would have tried an erythropoietic stimulating agent first. I would often check baseline endogenous EPO level. Quite often in older patients it is low, and it predicts for a better chance of response. Then the question is, how long do you give a patient on the erythropoietic stimulating agent before you stop it? And I think that’s something we really need to focus on. I see a lot of patients referred who are continuously getting ESAs but also getting transfusions of 2 units of red cells every couple of weeks. And I just think we’re spending a lot of money without any real benefit for the patient.

So in this case, the situation where they were ESA refractory, that diagnosis was made. And then the question is, what do you choose next? And we could talk about lenalidomide in this situation. That’s 1 option. The other option would have been a hypomethylating agent such as azacitidine or decitabine.

DR LOVE: So I want to get Guillermo’s take on this. But just briefly, what do you think you would have done with this patient?

DR ERBA: Good question. I’d tend not to use lenalidomide to be quit honest with you in this patient. Which means that this patient would not have benefitted from an academic referral.

DR LOVE: So what agent would you have used, if any?

DR ERBA: I would have actually used azacitidine, and I would have used a 5-day schedule. And the reason for that is the following: If you look at what is the expected response rate with lenalidomide, it’s about 25% in terms of transfusion independence. Not bad.

DR MORGANSTEIN: That’s pretty good.

DR ERBA: That’s right. And I’m not disagreeing with the choice because it’s clearly the simpler choice. Taking a pill is much easier than coming into the doctor’s office 5 days in a row. And that will go on indefinitely as long as the patient has response.

If you look at the rate of predicted transfusion independence with azacitidine, it’s somewhere around 40% to 50%. For example, the study that was done by the Texas Oncology Group, where they compared 5-day dosing to the 7-day and a 10-day dosing. With the 5-day dosing, in mostly patients like this with low-grade MDS, there was a 40% to 50% transfusion-independence rate.

So statistically you’re more likely to get a benefit from azacitidine. But I cannot argue with attempting a trial of lenalidomide, as you did in this patient, and seeing a rapid response. I think what we need are the biomarkers that allow us to decide which of these 25% of patients are truly going to be the responders. And we can talk more about that.

DR LOVE: Guillermo?

DR GARCIA-MANERO: So before we go through this, my assumption is that he probably had a very nice platelet count.

DR MORGANSTEIN: His baseline?

DR GARCIA-MANERO: Yes.

DR MORGANSTEIN: Yes.

DR BRENNER: Not a matter of high?

DR MORGANSTEIN: Not high.

DR GARCIA-MANERO: And then the second thing, I’m pretty sure you allowed hemolysis in this patient?

DR MORGANSTEIN: Oh, every patient. Yes, from the get-go.

DR GARCIA-MANERO: Yes. So first of all, this is an extremely unusual situation, almost reportable, actually. So I totally agree with Harry.

So first of all, the compound lenalidomide is not really approved for this indication. So to be honest, I sometimes will consider this type of approach. Sometimes some insurances will approve this, and they will deliver drugs. Sometimes it doesn’t work out. And I never understand what goes in the mind of these people.

But, so Harry is right. Like most people, you go by the book. But I think Neil is right.

So, for instance, I’m going give you an extreme phenotype of this. So these are rare diseases. They actually consider mixed myelodysplastic/myeloproliferative neoplasm that is called refractory anemia with ringed sideroblasts and thrombocytosis RARS-T. This disease, rare, half of the patients have JAK2 mutations.

They, I can tell you from experience, very nicely respond. Similar to what you’re showing here to lenalidomide. And I started using lenalidomide when I could in this context, based on the regional data from the lenalidomide trial. So when you look at the regional studies that led to the approval of this compound, there were 2 key characteristics that were associated with response: (1) 5q minus. Everybody knows this. But there is something that, to me, is even more important, actually, that is patients had a platelet count to 100,000, they did significantly better.

So that data, people don’t talk about this. But that’s probably an important predictor of response to lenalidomide than having a 5q. Indeed, actually, on this study called MDS-005, that is a Phase III randomized study asking this question. One of the questions that I ask, or was part of that study is, can we look at the effect of platelet numbers on the response to lenalidomide?

So I actually agree with both of you. If I have a patient with good platelet count, with anemia as the sole manifestation, if I could, actually, I will try to give them a trial of lenalidomide the same way you did. It may or may not happen. And again, this is an intervention that is really not indicated right now by the FDA.

DR LOVE: What dose, incidentally?

DR GARCIA-MANERO: We use 10 mg of the compound. Like what Neil did.

And then again, if I look back at my experience with this RARS-T that probably are in 10, 12 cases, so it’s not humongous, but they do very well with the compound. And these people don’t have a 5q alteration at all.

DR MORGANSTEIN: Could I ask a question? Do you check JAK2 on everybody now with the myelodysplasias? Because I don’t think he was checked, I’ll be honest with you.

DR GARCIA-MANERO: Right. So every institution probably has different ways. The panel that we use right now is a 28-gene next-gen panel that covers JAK2.

DR MORGANSTEIN: Okay.

DR LOVE: Can is ask why?

DR GARCIA-MANERO: Why what?

DR LOVE: Why you want to know JAK2.

DR GARCIA-MANERO: So first of all, there’s a subset of patients with MDS/MPN features who have this kind of kinase mutations. Actually, the frequency of JAK2 mutation in MDS that may be MDS/MPN, some of this overlaps and it’s around 10%.

Now, this is actually particularly important in chronic myelomonocytic leukemia, CMML. Now, this is another disease that was in MDS that now is branching a little bit outside. But you will agree that we tend to see these together.

That group of patients actually tends to have, also, JAK2 mutations. And there are some trials with JAK2 inhibitors for this group of patients.

DR LOVE: Erik?

DR RUPARD: Yes. This has been a mystery to me for a while, and apparently to Neil Morganstein as well. When I look at the data, I see that azacitidine, in a couple of studies, has a 10% to 15% response rate. And lenalidomide, we now have 2 studies — MDS002/MDS005 — with an identical 26% response rate. It works more quickly. It’s easier to take, obviously. It doesn’t require a 7-day regimen or even a 5-day regimen. And, yet, both of you guys who obviously know what you’re talking about, and NCCN have azacitidine above lenalidomide. And I don’t understand why exactly. So what am I missing?

DR GARCIA-MANERO: I’ll take that if you allow me.

So first of all, the study that led to the approval of lenalidomide is for people with this 5q alteration. So again, if you were in Europe, okay, where they’re really strict about this, you would not able to do this.

You did the right thing. What I was trying to answer, and maybe we went tangential because of this JAK2 question, is that I think when you look at response to lenalidomide there are 2 main characteristics. One everybody knows because it’s a Board question, is 5q. That’s good. But what I was trying to emphasize is if you look at the studies that you are quoting, lack of thrombocytopenia. So people, for instance, who have a platelet count less than 50, even if they are transfusion-independent for the platelets, they actually do not respond very highly to the lenalidomide. Okay?

DR LOVE: And is the reason that they can’t tolerate it a different biology?

DR GARCIA-MANERO: It’s a different biology of the disease. And if you look at those papers, if you have a platelet count over 100,000, that’s when you’re getting these responses that you’re quoting on the 02 study, the 03 study and so forth.

If you have thrombocytosis, that’s the extreme phenotype of this MDS/MPD that I was talking about, this RARS-T. Actually, the response rate, in my experience, is almost universal. But so you’re correct. The issue is that let’s forget about the indication.

This compound, actually you’re quoting 2 types of response. So while you’re talking about response with azacitidine, you’re talking about complete remission. So with azacitidine, a complete remission is, depending on the study, from 10% to 20%. Something like this. Similar, probably, to the decitabine. But these are complete remissions.

So that means the blast dropped to less than 5. They really had very significant thresholds of hemoglobin and platelet count. Actually, technically a complete remission in lower-risk MDS is actually pretty difficult to achieve because, by definition, most patients have a low percentage of blast. And to achieve these levels of hemoglobin, over 12, 13, whatever it is, this is not trivial. It’s very difficult.

So when you’re talking about response with lenalidomide, this 25% for this group, because actually the response rate is significantly higher in the true del(5q)-altered subset. You’re talking about what we call hematological improvement E, erythroid. So what you’re basically looking at is the rate of erythroid response. So as the patient becomes transfusion-independent, if they were dependent, do you go below 10? Eleven? Twelve grams of hemoglobin? That’s not a complete response.

DR LOVE: So what are the numbers for that?

DR GARCIA-MANERO: It’s what Harry mentioned. So if you look at transfusion erythroid responses with aza nucleosides, either with azacitidine or with decitabine, actually, they are in the realm around 40%. So they actually compete positively with this data in the non-del(5q) subset with lenalidomide.

And then the other thing that you have to keep in mind, because obviously it worked, but that lenalidomide does not do anything for your platelets. And for sure it doesn’t do anything for your white count. Sometimes rarely, but I may see in my practice a patient who has a 5 alteration where the main manifestation of the disease is platelet transfusion dependency. Okay? And then doc gives him or her lenalidomide because they have a 5q, when, in fact, the drug doesn’t do anything for your platelets. If anything, actually you may have this issue of neutropenia.

So what you have to understand is that is a really nice drug. In the context of 5q, this is probably the best therapy we have but is monolytic. It’s affecting only your red cell series, and it works better in people who have this 5q and if they have good platelets.

DR LOVE: So before I ask Warren for his thoughts, I just want to clarify. The integration of platelet count into the algorithm for this type of patient, Harry, do you agree with that?

DR ERBA: I do. I agree with Guillermo’s comments about the responses.

DR RUPARD: But is there a clinical difference for the average person, the 75-year-old, having a complete response versus becoming transfusion independent? It’s all palliative.

DR ERBA: It is all palliative. And that’s why I’m saying that azacitidine/decitabine are viable options with 40% to 50% transfusion independence. Because you’re exactly right — it is palliative therapy. Now you could argue, is it really palliative to be making your patient come into your office 5 days out of the month? On the other hand, lenalidomide also has toxicities, including the risk of thromboembolic disease, rashes, GI problems and cytopenias.

And so I’ll second the point made by Guillermo — thrombocytopenia is an incredibly important feature to look at when you’re thinking about using lenalidomide. Now our patient had preserved platelet count and unilineage dysplasia, so not an issue for this patient. But if you look at the factors that predict for response, even in deletion(5q), patients with thrombocytopenia are much less likely to respond.

There was a very small study done by the French of about 50 patients with higher-grade MDS where they gave single-agent lenalidomide, 10 mg a day, for intermediate-2 and high risk, and they saw transfusion independence in a small subset of the patients. It turned out that all of those patients had intermediate- and high-risk disease only on the basis of blast count — if they were thrombocytopenic they didn’t respond and if they had any other cytogenetic changes, except 1 patient with trisomy 8, they didn’t respond.

So I think the real issue for the use of lenalidomide in MDS is understanding which patients are going to benefit. And as I said, we don’t have a biomarker that allows us to do that. There have been a lot of attempts at trying to identify these patients.

I think part of the issue is lenalidomide probably has pleomorphic effects on the marrow. So in deletion(5q), it truly is cytotoxic. You can actually get, in deletion(5q) syndrome, cytogenetic remissions in those patients.

However, in all of the other patients where we’re seeing response, we don’t really know what’s the pathogenesis of the disease that’s leading to the response. It may be immune mediated. It may be something else. There were gene expression profiles, unfortunately, that disagreed. One gene expression profile said, if the marrow had low expression of erythroid genes that that patient was more likely to respond. And then there was another gene expression profiling study that said, no, it’s really the inflammatory immune markers that predicted which patients were to respond. And I think both could end up being true depending on the patient population you’re seeing. I think our challenge is that when we just look at the morphology of the marrow, we don’t really, with the exception maybe JAK2, we don’t really have a marker for how to identify these patients.

So I’m not disagreeing with the choice of lenalidomide. It’s hard to disagree with success.

DR LOVE: Warren?

DR BRENNER: We know in myeloma that cereblon is a marker of efficacy for the IMiDs. Is there any data in MDS?

DR GARCIA-MANERO: Yes. So the group from Ben Ebert in Boston has now published a number of really high-impact papers, not only in myeloma but also in MDS. And had a major plenary presentation at ASH 2 years ago where they’re invoking this pathway in response to myelodysplastic syndrome.

There’s also a gene called CK51 that is in this pathway, that apparently it’s also maybe associated with response to this compound. But this is not yet, that I’m aware of, maybe Harry knows better, a biomarker for response. Is it?

DR ERBA: Yes. No. It’s still not being used as a biomarker for response. But there was a French study in low-grade/low-risk myelodysplastic syndrome patients who were EPO refractory and transfusion dependent. And patients were randomized between lenalidomide alone or lenalidomide and ESA. Exact same design as we have in the SWOG/ECOG study that’s about to close.

They showed, in that group of patients, that the patients who got lenalidomide with ESA seemed to have a trend for a higher rate of erythroid response. Did not meet statistical significance. But when they looked for markers that could predict cereblon, I think it was a promoter mutation or polymorphism, one of those was linked to the response. So still not ready for prime time.

DR GARCIA-MANERO: Okay. So I actually don’t think this is palliative care.

If you look at the data on the prototypical, that this 5q minus disease, if you achieve a complete cytogenetic response, that is actually not uncommon. This patient has a significantly longer expected survival. And I can tell you, in my practice I have patients like this now 10 years later with a hypomethylating agent or lenalidomide. And for this 78-year-old gentleman to be alive 3 or 4 years later, I don't see this as palliative care. I think this actually transformed his natural history. He would not be here alive with 4 grams of hemoglobin.

So I think we need to think a little bit in terms of this type of concept. I think, honestly, in this patient there was something else cooking. And I think Harry alluded to this. I was asking you at the beginning, what were his platelets? Was there any level of hemolysis? I think we had some type of inflammatory process, something was going on at this point. Because I think what is unusual about the case is not that he had such a nice response. What is a little bit unusual is that he maintained this response without therapy. That’s what is unusual here, like you stopped treatment and now he’s a year and a half without intervention maintained. That to me suggests that there was something else that has actually been recognized more and more where these patients with MDS have some type of inflammatory condition, something like this that is happening. And I can tell you that many times you treat the primary disease. That is possible. This is totally hypothetical that the len could be effecting at that time and this goes away.

So and then to answer your question, and I’m going to throw a bomb here, okay, I would not give an ESA to this patient. So if this guy comes to my practice, okay, I’ve never seen any data that — everybody does it. I’m not saying it’s bad. It’s like religion, I guess is good. I don’t know. But in my practice, I will not use it. Why? Because I have never seen any data that these growth factors do anything for the survival of your patients. As Harry said, the response rate is low and I have no idea why we use these compounds. But that’s maybe for another discussion.

DR MORGANSTEIN: And now he’s becoming thrombocytopenic. His most recent update.

DR LOVE: Wow!

DR MORGANSTEIN: His platelet count is running about 75, 80.

DR LOVE: Now he’s getting thrombocytopenic, you still like your theory of some kind of inflammatory process?

DR GARCIA-MANERO: No, this is 3 years.

DR MORGANSTEIN: This is about yes, almost 3 years.

DR GARCIA-MANERO: So this is the natural history of this disease.

DR LOVE: So you weren’t saying before that he did not have MDS? Or were you?

DR GARCIA-MANERO: No, no. He had MDS. But he has something else on top of that.

DR LOVE: Ah! I see.

DR ERBA: And, if I may, there is emerging data on the role of the innate immune system in the pathogenesis of myelodysplastic syndrome. And so there is an interaction between the myeloid stem cell and the immune system that has yet to be fully elucidated.

DR GARCIA-MANERO: Exactly.

DR ERBA: And I think it may explain not only some of the responses that we see to immunomodulatory agents but maybe even the long-lasting response. So similar observations were made when we used thalidomide, where you would give a pulse of 3 months and that’s all the patient could tolerate, and then 6 months later they’re in a remission. So we’ve seen these. But as of yet, we haven’t been able to explain fully why it happens in the individual patient.

Just because I can’t leave the bomb exploded on the field, okay, there’s only 1 study that showed a survival benefit in myelodysplastic syndrome and that was the azacitidine versus conventional care regimens in higher-risk/higher-grade patients. You’re right. There’s no data showing that ESAs, or lenalidomide, even in deletion 5q in a randomized study show a survival benefit.

So I applaud your enthusiasm. I love it. But I agree that when I think of treatments for patients with myelodysplastic syndrome, I’m thinking more in a palliative line and I think it’s important to keep that distinction from, example, acute myeloid leukemia, where you’re willing to give very intensive therapies for the goal of extending survival and cure. So it’s a little bit different.

Important clinical issues in the management of MDS

Case discussions

Audio version