I’ll ask you to briefly comment on what’s exciting you of the compounds out there. There are several phase 3 trials, very exciting compounds both in lower and higher risk. I’ll ask you to give us a few minutes of what’s exciting you.

Well, I think we should be clear, we still only have 5 drugs in our MDS [myelodysplastic syndromes] space. We mentioned the oral hypomethylating agent, azacitidine, decitabine, luspatercept, and lenalidomide, which we haven’t spent a lot of time on because it’s for patients with del(5q) [deletion 5q], but I think we are entering a really exciting time in the myeloid space. By my count, we’re going to have 4 phase 3 trials open in a year or 9 months from now. One in lower-risk disease, which is looking at a novel mechanism to augment hemoglobin in patients. And then one for higher-risk disease, looking at 3 novel mechanisms, all seemingly compared to azacitidine backbone alone, which is analogous to the acute myeloid leukemia space. We’re looking to target BCL2 inhibition with venetoclax, which is a compound many people are familiar with. Then there are a couple of other pathways, like magrolimab, or other ways to target different aspects of myeloid maturation pathways that are deranged in patients who have excess blasts. I think we’re really going to start to see changes to the landscape and that these patients have more options. Once they’re approved, something else that excites me is that we really need to understand how to sequence these drugs. I’m not sure your thoughts on those topics.

 Yes, I totally agree. I think it’s very exciting that, as you mentioned, we have several trials, but we also have huge unmet needs for our patients that we still have to work hard on. Obviously, in the lower risk, for example, even with the drugs approved like luspatercept, we are trying to see if there’s benefit outside the approval. The COMMANDS study is ongoing looking at up-front ESA [erythropoiesis-stimulating agent] failure. We have some interesting data with a compound called entinostat. In higher risk, as you mentioned, we have 3 drugs that look interesting: magrolimab, APR-246, venetoclax. In the lower risk, I do think we’ll have to sequence those therapies. You mentioned briefly now about lenalidomide for del(5q). We use it sometimes outside, non-del(5q), for patients who are just anemic. But when we look at the sequence, we see higher responses if they were used before hypomethylating agents, for example. I think the sequence in the lower risk matters. I will be sequencing those treatments.

I think the exciting thing in the higher-risk disease is that now we probably, at least in the phase 2 or early phase 3 data, we are doubling the responses to hypomethylating agents. That will be an opportunity for us to maybe take more patients to transplant. I’m starting to try to introduce this concept of what I call total therapy for patients; we have to factor all those things in: the up-front treatment, the transplant, the maintenance of transplant, and maybe it’s also a time for us to introduce the minimal residual disease MRD status in MDS. It’s something we never used to think of because we could not achieve it. Now with the new treatments, I think we’ll start to think of a continuum for the patients, about how we are going to sequence those treatments, or whether we’re going to try to achieve MRD, take patients to transplant, or do maintenance after transplant. It’s really an exciting time for me. Now we can think of things we could not think of because we have better compounds, and hopefully, by next year, we’ll have a couple of those approved for our patients.

I think that’s tremendous. You did mention a drug that I should have highlighted about clinical trials, and it gets back to tying together everything we’ve talked about. It is a heterogeneous disease, we both do molecular sequencing, and TP53 is a very, unfortunately, unfavorable mutation for these patients. You mentioned APR-246, which was a phase 3 study that’s already completed accrual, but we don’t have the readout yet. As I see the future to improved outcomes for patients with MDS, it’s a path like this: We define them very clearly, we have a sense of the biology of whatever mutation they have, and then we rationally develop drugs and trials that are able to target what we think is driving the disease biology to improve outcomes for patients.

In that schema, do you have any advice for community oncologists for how to handle a patient who is newly diagnosed with MDS in their clinics?

 I think we tried to cover most of this now, and we went through the steps, diagnosis, risk stratification, trying to set the goals of management, thinking of it, as you mentioned, as a long-term treatment for those patients, and thinking of a transplant. All those things factor in. I would say also at time of failure of therapy, if somebody was on an ESA or on an HMA [hypomethylating agent] in higher risk, I think it’s always a good idea to pause, reevaluate the disease, sometimes get some molecular data. For example, we know that HMA failure in higher-risk MDS is associated with worse outcome, 4 to 6 months survival. Traditional intensive chemotherapy doesn’t work, and if we can take those patients to transplant, it’s helpful. Every now and then, we repeat the molecular testing and we see a new molecular mutation or a mutation that we didn’t know about, like IDH1 or IDH2 mutations. There are now drugs available for that if patients are transforming to AML [acute monocytic leukemia].

I would say it’s really going through the sequence of things, like establishing diagnosis, risk stratification and ongoing, and in a fashion of setting the goals and selecting the treatments accordingly. It’s really an exciting time. For the patients with TP53, as you mentioned, I think those patients don’t do well with intensive chemotherapy; they do better with HMAs. Both APR-246, which is the p53 regulator, and magrolimab, the anti-CD47, have shown some promising activity there. When possible, those patients should be referred for a clinical trial, but if not, I always prefer treating those patients with HMAs rather than intensive chemotherapy.

 Good. I think doctors like Dr Komrokji and I are always more than happy to help and see these patients. There are many other large MDS referral centers around if that is of use for even a one-time consultation, just to map out a little bit of a plan and then return to the community to receive their HMA or transfusional support. I really enjoyed talking to you today, Rami. It’s been fun thinking about our patients together, and thanks everybody for your time.

Videos

The Future of MDS Management

 I’d love to transition back to some of your expertise on therapies that can stave off transplant or transfusion burden. You had mentioned luspatercept-aamt a moment ago. Can you tell me a little bit more about that? How does it work? Is it just supportive care? What do you think? 

As you know, luspatercept-aamt is the first drug to be approved in more than 10 years in MDS [myelodysplastic syndromes]; that’s very exciting. We have a new option for our patients. In my experience, whenever we get a new drug approved in a disease, we get some momentum. Hopefully this is the beginning of getting many other drugs. We’ve worked on this compound for several years and a prior similar compound. Luspatercept-aamt is a fusion trap protein; it’s a monoclonal antibody that will neutralize or bind the ligands for a pathway called TGF [transforming growth factor]-beta in the serum before they bind the receptor. It turns out that the TGF-beta pathway is important in patients with MDS, evident by overexpression of some of the downstream markers of that pathway that contributes to the myelosuppression. Some of the ligands of that pathway affect the terminal erythroid differentiation. Erythroid-stimulating agents help with the early stage of erythroid differentiation, but there is a regulation of the terminal erythroid differentiation that some of the ligands of the TGF-beta pathway act as negative regulators, like GDF11, for example.

Luspatercept-aamt will bind those ligands and release that terminal block on erythroid maturation, and now we’re starting to use the term erythroid maturation agent. Basically, it’s a fusion trap protein that traps those ligands and helps with the erythroid maturation. It was tested in phase 1/2 studies, with the notion or the signal that maybe there are higher-risk patients with ring sideroblasts or SF3B1 mutation. That led to the MEDALIST study [NCT02631070] that randomized patients with lower-risk MDS, ring sideroblasts, and transfusion dependence after ESA [erythropoiesis-stimulating agent] failure to luspatercept-aamt versus placebo, and met the primary end point of transfusion independence. There was also benefit of hematological improvement with transfusion reductions in patients who were not heavily transfusion dependent and significant increase in the hemoglobin. It’s typically an injection given every 3 weeks. There is no doubt that for that subset of patients, I think it’s going to be a very helpful treatment. Maybe I’ll ask you because you also have been involved in this, how do you use those drugs? Where are you positioning them? What’s the dosing with those medications, and how has your experience been since the approval?

 I think you had a great point that we have enthusiasm with luspatercept-aamt. I had always done diagnostic iron stains, but I feel like I’ve been ordering them more and rigorously encouraging my hematopathology colleagues to count the ring sideroblasts quite precisely, not in just a range. I have been fortunate to have treated patients both on trial and then since its approval in April. I had a lot of patients with known ring sideroblasts, patients with SF3B1, sitting in the queue, and it’s been very gratifying. Certainly there are patients who had been densely red cell transfusion-dependent for many months and sometimes years, that we’ve initiated this therapy in. It’s a subcutaneous injection. One milligram per kilogram is the starting dose given as a shot every 3 weeks. After the first couple of doses, you can augment to 1.33 mg/kg and then 1.75 mg/kg to get a response. I will say in my hands, in the real-world population, I have had to dose increase the majority of patients. That could be my patient population, but I actually think it’s relatively safe to do, and I try to stay quite attentive to the schedule to do so, to make sure that we can increase their hemoglobin to safe levels and maintain some transfusion independence.

You sent me a patient who lived in Florida who now lives up here, who we’ve just toggled along, and it’s gone very well. Certainly, no drug is without adverse effects, but as you alluded to, the mechanism as an erythroid maturation agent, it is well tolerated. A few patients have commented to me on loose stools. None of my patients have had life-threatening or life-altering diarrhea. I don’t know if you’ve had any, but I’ve had a couple of patients who thought they could have had lower extremity swelling and some shortness of breath related to it, and I think that may have correlated. Again, none of these for my patient population have been life altering in any way, but the demonstrable increase in hemoglobin has been good. Have you had any other safety signals or things that are pearls as you manage it that you might mention?

No, I think you covered most of it. I totally agree with you. I think one of the things that’s important to stress is the escalation of the dosing. I’m seeing a lot of patients, we send them back to the community oncologist, and they come back after 3 or 4 months on the same dose. After each couple of doses, if there is not a good response, one should escalate up to the maximum dose, which is 1.75 mg/kg. I think the escalation is key.

As you mentioned, my experience has been very similar to yours, and most of the patients have to escalate. If you look at the data from the study and later on, it’s been presented by Ryan Platzbecker, MD, that if patients are heavily transfusion dependent, there are really not many responses on the 1 mg/kg. We should probably be going for higher doses, especially if patients are heavily transfusion dependent. So it’s key to escalate the dose quickly, after 2 injections, go up to the next dose. I don’t see a lot of dose-dependent toxicity, and I don’t see the adverse effects increasing as we increase the dose. 

My experience with the adverse effects is very similar to yours. Fatigue, that’s typically in the first couple of rounds. In a couple of patients, it was profound, but most patients tolerated it well. Some aches; I’ve had a couple of patients mentioning diarrhea, but nothing that’s like what we experience with using FLT3 inhibitors, for example. Leg edema, some shortness of breath that temporarily could correlate with the treatment. But in general, in 80% of the patients it’s very well tolerated. I think the key is rapid escalation of the dose, and once you get the higher dose, and after 2 or 3 injections if there is no response, there is no point to use it.

The other clinical pearl is introducing the drug early a little bit. When you look at the data, those patients who were heavily transfusion dependent, more than 6 units of blood, we do see some benefit in transfusion reductions. The rate of transfusion dependency was definitely lower. In my practice, I think we’re now in transition because many of the patients, as you mentioned, have been waiting. Many of the patients have had prior hypomethylating agents, lenalidomide, which is a population that was not even studied on the trial, but we use it and we have some success. In patients, classically what our position is, is somebody who had ESA, and is starting to have ESA failure, before becoming heavily transfusion dependent. Once they’re starting to be transfusion dependent by the label, obviously, is to introduce the drug before they start needing 6 to 8, so using it early on is probably going to yield better responses in the patients.

Overview of Luspatercept-aamt as MDS Treatment

 In terms of other patients who have lower-risk disease, tell me a little about how you manage those patients specifically and if they have ring sideroblasts. Maybe this is a very small subset compared with the broad lower risk I just mentioned.

There has been quite an interest recently in the MDS [myelodysplastic syndrome] with ring sideroblasts, or SF3B1 mutation, as we learn more about that category of the disease. Those are unique, and we started talking about the heterogeneity of the disease. The MDS ring sideroblasts, or SF3B1, is a unique subtype of MDS with a distinct mechanism of ineffective erythropoiesis and treatment. I looked recently in our database, looking at the patients with MDS ring sideroblasts; around 16% to 20% of the patients with MDS will have ring sideroblasts. Around 18% of the patients in our database at least had an SF3B1 mutation, so it’s not that uncommon. The diagnosis by the WHO [World Health Organization] was changed. Historically, if we had more than 15% ring sideroblasts, we called that MDS with ring sideroblasts. Now if we have more than 5% ring sideroblasts, or we have an SF3B1 mutation, then those can be called MDS with ring sideroblasts, and the WHO goes into unilineage dysplasia or multilineage dysplasia that, in reality, doesn’t matter. Once they have SF3B1 mutation, the behavior is probably the same.

There’s a new proposal by International Working Group, a paper by Luca Malcovati that was just published in Blood—a very nice paper actually—saying that SF3B1-mutant disease should be its own separate category with certain criteria. Those patients, in general, have favorable outcomes, good overall survival, low rate of acute myeloid leukemia transformation but ongoing ineffective erythropoiesis that most of those patients will end to be transfusion dependent over time. In our experience, almost half those patients, when we see them referred to us, are already transfusion dependent. Over time most of them become transfusion dependent, and the outcome in our experience for patients who are transfusion dependent is worse.  

Historically, we treated them again with all the available options, like ESAs [erythropoiesis stimulating agents]. Patients actually do pretty well with ESAs and have good responses. There are responses to lenalidomide, even in the 5q, probably sometimes in combination. In our experience, there is a little less response to hypomethylating agents than other subtypes. Also in our experience, albeit a small subset of patients, there are no responses to immunosuppressive therapy like ATG [anti-thymocyte globulin] or ATG-cyclosporine combination. I tend to shy away from immunosuppressive therapy in patients with ring sideroblasts, or SF3B1 mutation, and use the hypomethylating agents as a last-ditch effort. If I run out of everything, start with ESAs, and I think about which we’ll be probably discussing, we’ll spend a little more time talking about luspatercept. Luspatercept is probably becoming positioned as the first line for patients with ring sideroblasts after ESA failure.

Before we discuss more into the luspatercept and the excitement about a new drug approved for MDS, maybe we can talk a little more about the higher-risk disease and role of transplant because, no doubt, the transplant remains the only curative option for those patients. What’s your take on transplant, its role, and when to consider it?

 I wear 2 hats in our myeloid group, and I am an active transplanter. I’ll state that disclosure at the beginning. Your point is well taken. Azacitidine, ESAs, luspatercept, and all these other drugs that we’ve mentioned that have really advanced care for these patients are not a potential path to cure. I am always very careful to couch transplant like that to an MDS patient and their families. In allogeneic bone marrow transplant, which can be a bone marrow stem cell source or a peripheral blood stem cell source, we tend not to use an umbilical cord source in MDS. It is a potential path to cure, but it is really not for every patient. I do personally believe that any MDS patient, certainly with higher-risk disease, should have at least a onetime consultation with an active transplant physician to evaluate whether they truly are a candidate. I like to explain to a patient that there are disease-related factors, there are transplant-related factors, and then there are patient-related factors that bring us to the ultimate decision as to whether we proceed with transplant.

The higher-risk patients really must achieve a hematologic remission. I firmly believe that all the data support that you have to have a measure of disease control for a patient with higher-risk disease, usually evidenced by decrease of the blasts to less than 5%, hopefully elimination of the cytogenetic abnormality, and decrease the allele frequency of their molecular mutations. To feel as if you’re offering them a therapy that truly is high reward because it comes with high risk. That brings us to the next issue. The majority of transplant patients are of an age range where we consider only reduced-intensity conditioning regimens, but they’re still quite challenging. You have to ask a patient to take out a year of their life, even if they’re going to have a successful transplant outcome, because the potential toxicity is quite high.

Then there are patient-related factors. They may not want to do that. They may be more comfortable with transfusional support or hypomethylating support with response, or feature a clinical trial. They’re very nuanced discussions, but that’s the reason early referral to ask the question and think about a donor for an individual patient could be really important. Those are complex decisions, and that’s probably as much as we can say about it for that today.

The Role of Hypomethylating Agents and Transplantation

 When do you start treatment in those patients? We could talk about both lower and higher risk.

 I definitely initiate treatment. Let’s start with the higher-risk patients. Sometimes these are the most obvious. Certainly, in a patient with excess blasts, densely transfusion dependent, recurrent infections in the setting of neutropenia need to begin therapy. I always describe to the patients that the blasts are a space-occupying lesion in the bone marrow not allowing room for good hematopoiesis, and we must approach that proactively. Nobody wants to start chemotherapy, but if they’re having those clear clinical consequences of their disease, it’s time to start therapy.

It’s a separate issue in patients who have very preserved counts. I actually think you and I have talked about this on a number of occasions; we practice similarly. Excess blasts alone are a point of discussion, depending on whether you’re planning moving ahead to transplant or not. It’s the depth of cytopenias that require recurrent health care interactions that usually get me to start a hypomethylating agent. We’ll talk about clinical trials later, but from a standard-of-care perspective, for higher-risk disease, we have azacitidine and we have decitabine. I won’t say I use them interchangeably. I tend toward azacitidine a bit more in my personal practice. We have a great deal of experience with that here at Sidney Kimmel Comprehensive Cancer Center for many years. I start it fairly soon. For lower-risk patients, again it’s driven similarly by what are the clinical consequences of their disease. Let’s take a lower-risk anemic patient. Are you someone who uses a lot of erythropoiesis stimulating agents [ESAs] if the patient is anemic and has lower-risk disease?

 In general, in the lower-risk disease, my trigger to pull a treatment is the cytopenias as you mentioned. It’s really the degree of the cytopenia that dictates. I don’t think there is a magical number to say that I will pull the trigger on treatment. A majority of the patients, when we are treating lower risk, we are treating anemia. Most patients are anemic. Sometimes they may have concomitant thrombocytopenia and neutropenia that may, down the road, dictate your choice of therapy. But in a majority of the patients, we are treating anemia. If patients are neutropenic or at least just pure neutropenia, I rarely have treated those patients. If they have thrombocytopenia, the options are limited. In some younger patients, I’ll consider immunosuppressive therapy and hypomethylating agents. Sometimes if it’s a truly lower-risk disease, nowadays I move to a thrombocytopenia bag before hypomethylating agents. But for a majority of the patients, anemia is the major indication to treat, and I start to do that with erythroid-stimulating agents. 

For those patients, when you look at the data of the predictive models of response to erythroid-stimulating agents, it’s usually in an endogenous level. If they are more than 500—and some use a 200 cutoff with their transfusion burden if patients are getting more than 2 units per month already—the chances of response to erythroid-stimulating agents drop to probably below 10%. Once I see the patient’s hemoglobin is drifting down and they are starting to be symptomatic, I would start treatment before they get transfusion dependent, usually with erythroid-simulating agents. Sometimes you could say maybe hemoglobin of 9 g/dL is my cutoff, but really there is no magical threshold there. I use it because I don’t want them to get to 8 or 7 g/dL, when they will need blood transfusions.

I give it a trial of 8 to 12 weeks. Sometimes I go to the highest dose. Some people like to add granulocyte colony-stimulating factor because it increases the erythroid response. If it’s working, then I continue with that. If they had an adequate trial of 12 weeks or so and it’s not working, then I will move next. One thing is always this: We move to the next option, then we could talk about this later. Maybe sometimes there are situations where you keep the erythroid-stimulating agents, but in the majority of cases they should be stopped. I see patients that are getting blood every 2 weeks, and they are still on Aranesp or Procrit after a year, with the notion that maybe it’s making the blood transfusions less. I don’t think that’s the right thing we should do, so we should stop them in majority of the cases.

In the lower-risk disease, we have a few options. We have the immunosuppressive therapy, we just talked about the ESA. We talked about lenalidomide and the hypomethylating agents. Particularly in the United States, there was a longtime track record of using hypomethylating agents in the lower-risk disease. Can you discuss the role of hypomethylating agents, in both lower- and higher-risk disease—the 2 agents, and some excitement about some oral drugs coming out?

I alluded to this earlier in higher-risk disease, but I actually do use both azacitidine and decitabine in lower-risk disease. Certainly, you know we both participated in a clinical trial of randomizing lower-risk patients to different schedules of azacitidine or decitabine, as few as 3 days, and I really do tailor this to the individual patient when I’m doing it in an off-trial fashion. There absolutely is a role for these hypomethylating agents in patients who are transfusion dependent. Receiving azacitidine or decitabine can diminish or at least decrease that transfusion burden, and I find it very useful for those patients. There’s always that period of time when you initiate a hypomethylating agent, that the myelosuppression from the drug can escalate the transfusion burden for the first 1 to 2 cycles during that day-10-to-22 period of a 28-day cycle. If I can push myself and the patient through, and we are patients, then really the response rate after 4 to 6 cycles can be transfusion reduction or even complete hematologic improvement even in the lower-risk patients.

The same principles apply to the higher-risk patients. Sometimes we have a little higher mountain to climb in that we must decrease the blast burden with our hypomethylating agents as well. One of the holy grails of myelodysplastic syndrome therapy, in general, was the ability to give an oral hypomethylating agent, and there was 1 approved just at the end of August for higher-risk disease. Its brand name is Inqovi, and it’s the combination of an oral cytidine deaminase inhibitor and oral decitabine. It’s a pill that has the right ratio of those 2 drugs to enable an individual patient to get the same levels as they would if they were getting their decitabine through intravenous form.

I actually think the community and even myself who participated in the clinical trials are still going to be learning how to do it. You give it for 5 days of every 28, it reaches steady state, and it has very similar kinetics and has very good response rates after a few cycles of hematologic improvement and then partial remission, even complete remissions. The ability, especially during the coronavirus era to get a treatment at home so you don’t have to come in daily to the clinic is really a very patient-friendly therapy. It’s something I’m incorporating into my practice more now that it’s commercially available.

Initiating First-Line Treatment  

 Maybe we can start talking about the supportive care management that you use for your patients and move on to the treatment options.

 Sure. Just briefly, I sometimes smile when patients ask if I’m going to treat them and transfuse them, as if those are 2 separate buckets. To me, they go hand in hand. Anemia is the predominant cytopenia in many of these patients, so red cell transfusional support is key. I try to be judicious, though I also think anemia and health care–related interactions for these transfusions have to be balanced with the patient’s level of fatigue. Most of my patients are transfused somewhere between a hemoglobin level of 7 and 8 g/dL. For patients whose ferritin is creeping up from transfusional iron overload, if they can tolerate it, I am someone who believes in chelation. The chelators can be hard to tolerate at least in my practice from a GI [gastrointestinal] perspective, so I am not militant about it, but I do have attention to follow their ferritins and keep them on a regular transfusional course. Is that similar to how you monitor iron and keep these things going for quality of life?

Yeah, absolutely. I have the same experience. Patients sometimes think those things are not hand-in-hand: treatment and supportive care, and it’s important. Obviously, transfusions are a mainstay of what we do, and they do have impact, as you mentioned, on the long-run iron overload, the quality of life for patients coming every week or 2 for blood transfusions. I feel an economic factor that we don’t factor in sometimes is that transfusions are costly, and there is a lot of need for blood. That’s an important factor. I follow the same as you with the blood transfusions and platelets when needed if they are very low, below 10 per mm3. I tend not to use many growth factors for neutropenias. In the lower risk, I don’t think there’s any convincing evidence that they alter the outcome. Every now and then I’ll do prophylactic antibiotics if patients had recurrent infections, but not as a standard from the get-go for every patient. For iron overload, I think you and I are almost in the same camp. I do consider it, I discuss it with patients, and I’m not very rigid about it. The data are out there. Several retrospective studies suggest survival advantage, and they altered the size because they were retrospective. Plus, we know that high iron levels and high ferritin correlate with worse outcome. In the last couple of years, we heard from the TELESTO trial, which looked at the composite end point survival and hospitalization, and there was some benefit. My read on all the evidence is that there is probably benefit, but it should be individualized because of the adverse effects you mentioned.

Some guidelines say you should consider adding chelation above 1000 ng/mL ferritin, and some say above 2500 ng/mL. I do somewhere in between based on the trajectory of the course of the disease. Sometimes, if I have a very good response to treatment, I take that as an opportunity. If I know that the patient had gone through all the other treatments, I do iron chelation. I try not to introduce 2 variables together, so if I’m starting a treatment, I let it go for a few months, because if we get adverse effects, I don’t know if it’s from adding chelation or the treatment. I do consider it and discuss it with many of the patients. The issue with the iron chelators, especially the oral ones, is that there is some renal toxicity, and some GI adverse effects, so one has to keep that in consideration.

The Role of Supportive Care 

 Comorbidities, as well as the individual human’s goals, aren’t incorporated into the IPSS-R [Revised International Prognostic Scoring System] either, so that can certainly be a limitation.

The other thing is that for the IPSS-R, perhaps a woman who had breast cancer and got very aggressive cytotoxic chemotherapy for that malignancy or a gentleman who had prostate seeds, radiation, or even docetaxel for their prostate cancer can have treatment-related MDS [myelodysplastic syndromes]. I rely a little less on the IPSS-R in those patients. Is that how you think about it?

I totally agree. We’ve moved mostly to the revised IPSS-R. There are so many different models out there. For somebody in the community who is busy seeing patients, you need to adapt 1 system and move with it. The IPSS-R does a pretty good job, much better than the IPSS. We are finally moving to incorporate that even to clinical trial because for some time there was a challenge. All the trials are based on the IPSS and in real life are using the revised IPSS.

As you mentioned, the only possible dilemma in the IPSS-R is that intermediate-risk group. I like what you do. I rely on the molecular profile, transfusion burden, presence of circulating blasts, and patient age to decide how I’m going to treat the intermediate group. The other thing is that I don’t think they capture the kinetics of the dynamics of the disease. Sometimes I’ll see some patients, and they are intermediate risk. I’ll say, “Let’s see how the disease will behave over the coming 3 to 6 months.” You see those patients who will have progressive profound cytopenia, so you know the disease is acting up, and some patients will actually stay steady. There is some element of disease dynamics or kinetics that those models don’t capture.

As you mentioned in the therapy-related MDS, those models did not include patients with therapy-related MDS. We tried to look at the large cohort here of almost 500 patients, and it does predict outcome. Stage to stage, a very low-risk therapy-related MDS will do worse than the de novo MDS, but you can’t segregate patients in the therapy-related MDS on that. The challenge is that almost 70% or 80% of patients with therapy-related MDS are going to be higher, or very high risk, because it’s mostly dictated by the cytogenetics, high-risk mutations like TP53. Generally, if a patient is lower—a therapy-related MDS but doesn’t have blasts, doesn’t have bad cytogenetics—it’s reasonable to label them in the lower-risk MDS bucket and go stepwise. I deal with that event with more caution, but they do a little worse, unfortunately, than de novo MDS, even stage to stage if they are the same stage. Part of that is probably the comorbidities from down the line in the disease.

I’m obviously aware of the comorbidities; they are very important in the higher risk. You’re trying to decide transplant or not, so the comorbidities play a very important role. There’s a question I ask: How do I use them dynamically or restage those patients? It’s challenging because those models were clinically developed at diagnosis, and I don’t think they’ve been validated in terms of restaging. I don’t really do a formal restaging. I look at the blast percentages increasing—that’s going to higher risk if we acquire new cytogenetic abnormality or mutation or develop profound cytopenia. It’s important to notice that, for example, after HMA [hypomethylating agent] failure, although we label sometimes patients as lower risk, their outcome is not really good unfortunately. With amyotrophic lateral sclerosis, it’s around 18 months for overall survival. We look together, and it has been projected looking at the utility of the risk models after HMA failure, and they really don’t predict the outcome very well. The patterns are different. Maybe some patients sustain that lower-risk MDS with more profound cytopenia and bone marrow failure, and some patients progress to MDS. I don’t use a formal restaging. I think this has fallen in clinical trials after standard-therapy failure because they require restaging, and I don’t think it’s an accurate way to do it. It’s hard to develop newer models for restaging. After there’s failure, we have enough data that even after ESA [erythropoiesis stimulating agent] failure or lenalidomide failure, patients don’t do as well, and we should not be very rigid using only those risk modifications or risk certification at that time.

What I tell my fellows usually is that in any disease, you have to go through those 2 steps, because then you’re going to move to the next step, which is the management. We probably should start moving to talk about the management because we have a lot of interesting drugs to talk about. I will try to make sure the diagnosis is right, I risk stratify the patients, and factor in the patient’s related factors, comorbidities, and social things. That dictates how we manage those patients, and then I move to manage those patients. There is no doubt that in MDS, supportive care is still an important part.

IPSS/IPSS-R Reliability in Various Disease Settings 

 I was wondering if you think about this similarly in terms of risk stratification, from low-risk and high-risk patients. Because the disease course is so different, it gets into this expectation management that I was alluding to.

 Absolutely. I totally agree. Obviously, we’ll talk more about what we use to risk-stratify patients, but in the end, we are putting patients in almost 2 categories: a lower-risk disease and a higher-risk disease. I think it’s important to remember that if we take all MDS [myelodysplastic syndromes] patients, all comers, roughly around 30% will transform to AML [acute myeloid leukemia], much higher in the higher risk obviously and probably lower in patients with MDS ring sideroblasts. It’s probably less than 10%. However, when we look at the data, 50% to 60% of the patients will still unfortunately die from complications of MDS. The fact that some patients are lower risk doesn’t mean that this disease cannot be life-threatening or serious. In lower-risk disease, there’s a lot of interaction and…the cytopenias may exacerbate the comorbidities or the other way.

Together in this consortium, we published some data years ago about looking at what we labeled as lower risk. Even when we labeled patients as lower risk, around 25% of those patients unfortunately died from the disease and died within a couple of years. It could be comorbidities, but like a patient with coronary artery disease, if they become anemic with hemoglobin less than 8.0 g/dL, the cytopenia could exacerbate comorbidities. In general, obviously, the higher-risk disease seems to dominate the course, so those patients are at higher risk of developing AML. Their cytopenias are more profound, we tend to see more thrombocytopenia, and the rapidity of the course of the disease is higher. In the lower-risk disease, the severity and the depth of the cytopenia dictates a lot of the outcome of those patients. There’s no doubt in my mind that the disease does affect the patients both in terms of quality of life and survival, both in the lower- and higher-risk disease, maybe in a different pattern.

The point you bring to the comorbidity is important, and we didn’t pay much attention to it for many years. Now we know fascinating data that some of the somatic mutations that we see in MDS can contribute to inflammation and lots of certain comorbidities. Also, the comorbidities feed into the MDS as well, where again, a patient with coronary artery disease may have a heart attack, but it could be prompted by the severity of the cytopenia and so forth.

We should talk about how we label those patients as lower risk or higher risk. What do you use? How reliable are the risk stratifications? What’s state of the art that you’re practicing now?

 I try to be state of the art, but it’s challenging because I always come back to the same refrain to every patient. I want to give them the best quality of life for the longest quantity. I do personally use the International Prognostic Scoring System–Revised [IPSS-R] most often to couch that particular discussion. When I was training and in my early years on faculty, we did use its earlier predecessor, which was the IPSS [International Prognostic Scoring System]. I have found in my practice that the IPSS has fallen a little behind, and I try to stick concretely to the IPSS-R. I do this for a couple of reasons. The publication was a little more recent, though it’s still further back in our minds. Also, the IPSS-R incorporates morphology and cytogenetics, similar to the IPSS but with increased granularity of detail.

There are more categories of cytogenetics. The blasts are separated out a little more, and there are actually 5 categories. I find it is a helpful starting point as I put people in lower- or higher-risk boxes. Then I think about how to explain to them why we need to be more aggressive in the context of how many blasts they have, if their cytogenetic markers are particularly unfavorable and really suggest they’re much more likely to get leukemia sooner than a patient who doesn’t have that particular cytogenetic marker. One of the best examples is a chromosome 7 abnormality, whereas something like the –Y in a man is a much less bad cytogenetic abnormality.

The IPSS-R is a good tool, but there are limitations. For 1 thing, there’s this intermediate category. In the past, you and I have debated whether we should treat these intermediate patients like low risk or high risk. It is a challenge. That’s where I really lean quite heavily on molecular information, even if someone doesn’t have a lot of blasts. Or perhaps their karyotype is normal but they have mutations that I know are associated with an accelerated path to truly high-risk disease, whether elevated blasts or acute leukemia. I bring that into the discussion with the patient even though it’s not part, at this time, of the formal IPSS-R. We’re going to have another R there soon, or maybe the IPSS-R Molecular. That really will incorporate these data based on large series.

Disease-Course Differentiation: Low Risk and High Risk

 The importance of the diagnosis is related to what I asked you to talk about. It’s like the challenge and the management of MDS [myelodysplastic syndromes] patients as we discussed. They are a very heterogeneous group. If you can talk about what that accurate diagnosis and risk stratification will be doing to the management of the patient. There is not 1 treatment for every patient.

 That’s probably the understatement of the year. I completely agree. We’ve talked a lot about the thorough diagnostics that are required because I actually think 1 of the largest challenges early in a patient’s disease course is understanding the disease biology to the best of our ability with current techniques. A great deal of the bedside management of myelodysplastic syndrome is expectation management. These are patients who we’ve given a chronic disease diagnosis, who can take—back to that heterogeneity—quite a few varied paths over their lifetimes. We’ll talk in a minute about risk stratification, but most patients who hear that they have MDS come to see somebody at a tertiary referral center, perhaps like you or me, naturally believe they have high-risk disease and the end is coming next week. That’s certainly not always true.

I try to spend some time helping them understand the way in which their 3 cell lines may differ from normal and then how we manage each of them differently—perhaps neutropenia, anemia, thrombocytopenia—and work through how to incorporate their other potential comorbidities. The median age at diagnosis is in the late 60s, early 70s. These are people who have accumulated a lifetime of other medical issues. At least in my practice, you can tell me your experience, but I have found that, unfortunately, all of a sudden the triage list—no matter what—puts MDS way ahead of everything else. Suddenly we’re not managing the blood pressure, not following the blood sugars, this, that, and the other. This is a chronic diagnosis for even higher-risk patients, and we need to make sure the whole patient’s body stays as healthy as possible, and that’s after expectation management.

For me, the largest challenge is to make sure that while we focus on appropriate surveillance and treatment of the myelodysplastic syndrome, all the other medical comorbidities continue to get appropriate attention. So whenever we need to intervene, whether it’s next week or 5 years from now for MDS, the patient is in the best possible health to receive whatever intervention we think they need. Sometimes these other medical comorbidities complicate the therapy for MDS.

Challenges of MDS Management

 From your perspective, Dr Komrokji, what do community physicians need to look out for in terms of recognizing some of these symptoms and making sure we bring a patient to their accurate diagnosis?

 As you mentioned, the symptoms are nonspecific in MDS [myelodysplastic syndrome], and that’s always the challenge. When we see those low counts, the differential diagnosis could be wide, from nonhematologic conditions to hematologic conditions. There are so many things that can resemble MDS. For primary care physician doctors, as you mentioned, when I see some of those patients, I see that they, for example, have some macrocytosis that was present a year or 2 before. From my point of view, I would say that whenever you see bicytopenias, with more than 1 cell line affected, that’s probably a red flag or a trigger to have a little more thorough evaluation and to look at the bone marrow and presence of circulating blasts. Obviously, it’s a very good practice to exclude things that resemble MDS or the basic stuff, like nutrition deficiencies, B12, folate, and iron deficiency. There is no doubt that these cytopenias can be caused by other things. It’s important to keep in mind that MDS is not that uncommon actually, and if you think of older patients, it’s probably 1 of the top 5 causes of anemia. It should be on our mind when we are seeing those patients.

That probably leads me to ask you, when you see those patients, what’s the diagnostic work-up that you do?

 I do always have it at the forefront of my mind because just like you, the majority of patients in my clinic carry this diagnosis. I like to give them the benefit of the doubt and make sure they’ve had a thorough internal medicine work-up to make sure we aren’t missing what I think of as the masqueraders of MDS. You alluded to the cytopenias. Certainly, I always check peripheral blood counts, and that does include a complete blood count with differential. The differential is quite important. I also always check reticulocyte count. Sometimes that can lead you down a different path if it’s a more hyperproliferative procedure.

I do tend to get iron studies with a focus on the ferritin for iron storage. Folate levels are something that, when I trained, we were taught to always check—and red blood cell folate as opposed to serum. But this is really falling out of favor as the majority of people, at least in the United States, are not folate deficient. I check this less now. I do check a vitamin B12; this certainly can be a masquerader in some. I personally always also check a baseline erythropoietin level as well as a comprehensive metabolic panel because, occasionally, you’ll pick up some quite significant renal insufficiency that could be leading to the anemia or hepatitis that had been previously undiagnosed.

One of my favorite diagnostic tests to send is a copper level. Every once in a while, you get a truly copper-deficient patient that does not have MDS. Last but certainly not least, it’s very important to me that we get a very good bone marrow evaluation, which includes an aspirate. We send flow cytometry on that aspirate. We also do an iron stain at diagnosis, a reticulin stain, and conventional cytogenetics, as well as molecular testing. This is how I tend to approach it, but I’d be very interested to hear if you always do molecular and genetic testing or agree with the work-up that I suggested.

 Definitely I agree, and it sounds sometimes that we are talking about simple things. To put this in perspective, just a couple of anecdotes. I had a younger patient who presented—originally on the outside they called it acute myeloid leukemia, and he was transferred to another place and sent to start intensive chemotherapy. They get the bone marrow at the other institution, and the blast percentage was 15%. They started on azacitidine for 6 months, and he was pancytopenic. We see the patient, and it was actually megaloblastoid changes. The patient had a severe vitamin B12 deficiency, that the megaloblastoids were resembling myeloblasts, and with vitamin B12 treatment, he improved. To your favorite copper level, I also had seen a patient last year that was 1 year on azacitidine, labeled as MDS ring sideroblasts, but had neutropenia, thrombocytopenia, and some neurological changes. It was severe copper deficiency. For example, in MDS ring sideroblasts, whenever I don’t see 

 mutation, which is very common, I would not treat a patient without checking a copper level. Although what you went through sounds simple and everybody does it, we’ve seen real cases where it really makes a big difference.

Moving on, when we get the patients, most of the time this has been done. Obviously we get the bone marrow, the aspirate biopsy, and currently we do standardly get a next-generation sequencing on those patients. It’s very helpful. It’s not part of the diagnostic criteria, but it definitely can help you make a new diagnosis. It has some prognostic implications, no doubt, and we are moving now that we are incorporating that in how we manage patients, so we do get molecular testing. In the past, we used to go a little bit more, even sophisticated, checking LDH [lactate dehydrogenase] and flow cytometry on every lower-risk MDS patient. In my experience, the need for that is pretty low, but if I see a high large LGH, low hemoglobin, I may get it because this subset of patients that will have it, most of the time they are minute clones and we don’t have to do anything about them. Then we also used to check for LGL [granular lymphocytic] in almost every patient. Around 20% of these patients will have a small LGL clone, but most of the time it’s not significant. However, there are some cases where LGL would resemble MDS, and the diagnosis will be LGL.

We do get a bone marrow aspirate biopsy. We get a molecular profile. In select cases we may look for LGL, paroxysmal nocturnal hemoglobinuria, and the basic stuff you mentioned. It’s really crucial to get those levels of B12, looking at the retic count, some aspect of hemolysis, and so forth. It’s very important. Making a diagnosis is an important step. There is 1 paper that I know about, from The University of Texas MD Anderson Cancer Center, looking at its experience reviewing outside bone marrows. In 20% of the cases, the diagnosis was changed. Sometimes it’s just a subtle difference between the blast percentage, but in some cases it’s extreme, like the ones we talked about. It was copper deficiency and stuff like that.

Ensuring an Accurate Diagnosis

®, “Treatment Options in Myelodysplastic Syndromes.” I am Amy DeZern from Johns Hopkins University School of Medicine in Baltimore, Maryland. Joining me is Rami Komrokji from H. Lee Moffitt Cancer Center & Research Institute in Tampa, Florida. Today we’re going to discuss several interesting topics in the management of myelodysplastic syndrome [MDS], including new therapies. Let’s get started. Rami, could you begin, please, by describing the heterogeneity of myelodysplastic syndrome for us?

 Absolutely. As you know, MDS is a collective term of bone marrow failure. Clinically, they are characterized by cytopenia, a tendency to progress to acute myeloid leukemia, but we definitely see a variety of disease and heterogeneity from biology, molecular mutations, and somatic mutations of cytogenetic abnormalities to the clinical presentation. Obviously, we can think of higher risk and lower risk MDS being almost completely separate behavior, but even among the lower risk, we know that there is a lot of heterogeneity. We have, for example, the 5q patients; they reemphasize on patients with MDS ring sideroblasts, so the 

mutation. There is a lot of heterogeneity even among those groups of patients and the behavior. We see patients who remain asymptomatic for a long time, and we see patients who are very heavily transfusion dependent. As we move forward, we have to think of the heterogeneity because for a long time, we did the clinical trials, including all those patients. We lumped patients as lower risk; some we treated. The reality is that’s not going to be the best way moving forward. I’m sure that all the clinicians have observed this and seen this heterogeneity and the difference in the behavior. Also, there is some host-related factors that they say is the course of the disease or how symptomatic is the disease is, but that’s something always to keep in mind: how heterogeneous is MDS. One day we’re going to be looking at those as separate diseases.

 Yes. You probably also could comment on your experience on how you see the presentation and the heterogeneity in the presentation. How do you get patients diagnosed with MDS? How do they get to you?

 I do think heterogeneity remains the consistent feature, right? Nothing is as stable as that 1 thing. Patients come in lots of different ways, and I do think it happens to matter whether or not they have lower- or higher-risk disease at the time of presentation. Some symptoms that can be seen across all categories of myelodysplastic syndrome can be fatigue, certainly cytopenias. Hopefully we meet them prior to transfusion dependence, though I will say many people come to presentation when they come to the emergency department with an extremely low hemoglobin or a bleeding event in the context of thrombocytopenia. Shortness of breath as the hemoglobin comes down may be the first thing that somebody complains to their internist about or a decreased exercise tolerance, something that prompts a physician, probably not an oncologist or hematologist, to check somebody’s complete blood count and recognize that 1, 2, or even 3 of the cell lines are down, perhaps a macrocytosis, and this prompts a series of events that brings them to your attention or my attention with a bone marrow evaluation that ultimately makes the diagnosis of MDS.

Some other common signs can be easy bruising. I feel like a lot of people do note this when we ask their review of symptoms, though it often comes up even in patients who are not truly thrombocytopenic, and the patients can have petechiae when they have platelets less than 10,000 or 20,000 per mm3. Then there are other people who spend a lot of time focusing on an internal fatigue that is distinct from the degree of anemia, distinct from any other treatments, and they just can’t explain why they don’t have any more of that oomph that they used to have, the get up and go. It’s really a challenge.

Amy DeZern, MD, MHS, Sidney Kimmel Comprehensive Cancer Center

Articles