 Most clinical trials that I’ve seen have used CR, or complete response, as the primary endpoint. Why is that? Why is there this focus on CR in MDS [myelodysplastic syndrome]? What’s the distinction between a CR and a bone marrow CR?

 The response criteria in MDS are complicated, to say the least. CR has traditionally been the perfect response where the blood counts are near normal and the bone marrow blasts are less than 5%. But in MDS, 5% is still a part of the CR. In AML [acute myeloid leukemia], it’s not. Nonetheless, it’s lower blasts, so less than or equal to 5%, with near normalization of blood counts and absence of blasts. The dysplasia is not part of the criteria. The reason for using CR as opposed to survival outcomes is quite obvious. They want a quick endpoint to get drugs approved, and maybe some of these endpoints correlate with survival. They just haven’t been validated as much until recently. Marrow CR, on the other hand, is having 5% or less blast in the bone marrow. The caveat is you have to have at least a 50% reduction in the blast percentage. If you’re starting with 6% blast and get to 5%, it’s not considered marrow CR, in the absence of count normalization.

Another endpoint that can be added to marrow CR is whether you have hematologic improvement. Marrow CR without hematologic improvement, where the patient is still pancytopenic, requiring blood and platelet transfusion, is clinically a very different response from someone who has become transfusion independent and also has a marrow CR.

The MDS Clinical Research Consortium recently published on the validation of the IWG [International Working Group] response criteria in higher-risk MDS. What they found is that any time you get the blasts low but also need that hematologic improvement, the survival outcomes are going to be better and longer. Whereas marrow CR without any hematologic improvements, stable disease, or progressive disease are the responses where patients are going to do quite poorly. The definition of 

 is also tricky. How many weeks to use is also complex. I usually have to look at those tables every time I’m assessing response in MDS.

Do you believe we should expect and strive for more than just a CR, or looking at event-free survival or overall survival as primary trial endpoints or things that are meaningful?

 Yes. Yazan, fortunately or unfortunately, I’m old enough that I’ve seen a lot of clinical trials focused on CR. They do it because, reasonably, there had been retrospective studies that have shown that CR is an interim marker of a clinically meaningful benefit like overall survival. The assumption is if a patient achieves a CR, that means they must live longer. But in fact, there are a number of studies that have shown even a doubling of CR in patients with higher-risk MDS or older adults with acute myeloid leukemia, which is biologically almost identical without any difference in overall survival. It’s been disappointing. I don’t think CR alone is enough to say that there’s a drug that benefits a person.

The FDA uses the loose criteria of “lives longer or lives better.” If you can show that a drug allows someone to live longer and live better with the balance of safety, then that’s a drug that’s useful to have on the market. I’m focused on overall survival. I still think that should be the standard for higher-risk patients with MDS. That being said, we hold ourselves to a very high bar, and I think higher than a lot of solid tumors, where they’ve standardized progression-free survival or response duration as an interim marker of a clinically meaningful benefit. Even the FDA has hedged about saying progression-free survival itself may be clinically meaningful, although I don’t completely believe it in the absence of a patient also having an improved quality of life.

To sum it up, overall survival should be the standard in randomized studies exploring clinically meaningful benefits in patients with higher-risk MDS. I’ve been part of a study that’s focused on event-free survival, the event being either death or transformation to acute myeloid leukemia. Once a patient has higher-risk MDS and transforms to acute myeloid leukemia, that person deteriorates rapidly. The combined endpoint that constitutes event-free survival is clinically meaningful to patients. Unfortunately, we’ve been shown over and over again that CR alone doesn’t cut it.

I like thinking about quality-of-life endpoints as well. The problem is that the instruments we’ve been using to assess quality of life really stink when it comes to MDS, the most common of which is the EORTC QLQ-C30 [European Organization for Research and Treatment of Cancer Quality of Life Questionnaire version 3.0]. It doesn’t get at what a patient with MDS feels. The QUALMS [Quality of Life in Myelodysplasia Scale] instrument, which was developed by Gregory Abel at Dana-Farber Cancer Institute, is much more accurate at assessing quality of life. But what I’d love to see in future studies is the use of wearables to assess quality of life. Let’s use that. We can quantify what their quality of life is with real numbers in real time.

Videos

Experts in the treatment of MDS explore clinically meaningful treatment endpoints, including complete responses, bone marrow complete responses, and overall survival outcomes in MDS.

MDS: Clinical Primary Endpoints

 Yazan, given the general tolerability of these drugs, are there patients in whom you wouldn’t use them with higher-risk disease? If so, how do you treat those patients?

 Yes. That’s more of a rare occurrence. In general, I really try to incorporate HMA [hypomethylating agent] therapy. For patients who are reluctant, I usually try to offer them 1 cycle. There’s no harm in switching or stopping therapy if they don’t tolerate it, for quality-of-life reasons, or other reasons that the patient chooses not to go on therapy. There are quite limited options as standard of care to offer for those patients, so I give supportive management with blood checks and transfusion support. If they have some form of proliferative disease, I might incorporate low-dose Hydrea to just maintain the blood cell counts, not drop them too much, and then support their bone marrow with platelets and red blood cell transfusions.

You can always extrapolate from AML [acute myeloid leukemia] therapies if they’re not interested in clinical trials. I usually try to offer clinical trials up front for any patient with newly diagnosed MDS [myelodysplastic syndromes] when available. In the relapsed/refractory setting or post-HMA era, I check for targetable mutations. There are early clinical trial data for using enasidenib or ivosidenib in patients with MDS, and they seem to work OK. We don’t have many patients who were treated with them, but they might be an option to try to use off-label for those with no clinical trial options. But I try to limit my use of that, because we do have some trials available for those patients. It’s definitely an area of unmet need, whether up front or in the relapsed/refractory setting, so I tend to talk about clinical trials at any point in time when we’re switching therapy.

Drs Madanat and Sekeres discuss specific patient characteristics affecting selection of treatment for MDS.

MDS Treatment: Patient Considerations

 Yazan, given that these patients tend to be older, what toxicities do you look for in patients who get azacitidine or decitabine?

 That’s a great point. Just to comment on the oral decitabine, I’ve had to use it a few times, but only for convenience purposes for patients who can’t come to our infusion center on a daily basis. Despite the maybe less robust data with efficacy or survival, or the lack of a true randomized, placebo-controlled trial, convenience tends to be a big part for some of the patients who live quite a distance away from our center.

In terms of toxicity, those agents are generally quite well tolerated. That’s why they’re considered low-intensity chemotherapy when we talk to patients with AML [acute myeloid leukemia]. In general, fatigue tends to be quite a big component of it, but that’s always multifactorial, whether it’s the anemia, the disease, the therapy, or all of it contributing to fatigue. That’s something I usually talk to patients about.

For those who get subcutaneous azacitidine, they can get the local inflammatory reaction, which oftentimes can be mitigated if we switch to IV [intravenous] if it’s not very tolerable, and that can also result in some skin rashes here and there. GI [gastrointestinal] toxicity tends to be quite manageable. Whether that’s nausea, vomiting, diarrhea, constipation, or a mix of the above, they all tend to be quite mild and manageable for most patients. I’ve rarely had to switch therapy or stop or change because of those adverse effects. They’re usually managed in the outpatient setting. About 10% to 15% of patients get arthralgias. Sometimes you question whether the disease is contributing to those, or if it’s truly a drug toxicity or adverse effect. In terms of organ function, like kidneys and liver, mild liver function abnormalities can happen, but the kidneys are usually good with those agents. I’m happy with them.

I’m interested to know how you would tailor each of those therapies for a given patient in a particular setting, and some of the factors that you may weigh to select the final treatment choice.

 I already revealed my cards on this a little. I go with azacitidine first because the data are just so strong supporting its use up front in higher-risk MDS [myelodysplastic syndromes]. I haven’t been won over yet by oral decitabine, just because of the approval path they took to get it based on PK [pharmacokinetics] and PD [pharmacodynamics] data. Now that I’m practicing in Miami, I have a different patient population, and I have some patients who are here from South America to initiate their therapy in Miami and then want to go back to South America for a few months, where they may not have the same availability of azacitidine. Just because they have to go back for green card reasons, I will debate using oral decitabine in those patients, but only because I have no other options.

All things being even, I go with azacitidine first. I don’t criticize people for using decitabine, because it was a study design problem that led to the decitabine study that was conducted in Europe to not have the same results as azacitidine. For lower-risk disease, I will consider truncating the courses of azacitidine and decitabine down to 5 days, or even 3 days, based on trial work that we’ve done through the MDS Clinical Research Consortium.

Yazan Madanat, MD, explores toxicity and adverse events impacting treatment selection for patients with MDS.

MDS Treatment-Related Toxicity and AEs

 Mikkael, can you tell us about treatment options for higher-risk MDS [myelodysplastic syndrome], going into detail about clinical trial efficacy and how you would select 1 over the other in your practice?

 Sure. It’s easy to remember. I’m proof of principle that you don’t have to be a genius to know all the trial data in MDS because there aren’t a lot of great trial data. There’s only 1 prospective randomized trial that has ever shown a survival advantage in higher-risk MDS, and that’s the AZA-001 study that you mentioned, Yazan. In that trial, patients with higher-risk disease as defined by IPSS [International Prognostic Scoring System] were randomized to receive azacitidine or conventional care regimens. Conventional care regimens consisted of 60% of patients receiving best supportive care. The remaining 40% received either low-dose cytarabine or classic 7+3 intensive AML [acute myeloid leukemia]–type therapy.

On that study, patients who were randomized to receive azacitidine survived a median of 24 months compared with 15½ months for those who received conventional care regimens. That difference was statistically significantly different. That was dramatic. This is now the standard of care. There’s no arguing about it.

Azacitidine and decitabine are very similar structurally and in how they behave biologically. In fact, 1 is a prodrug for the other. A similar study that was conducted in Europe looked at decitabine in higher-risk patients with MDS. Patients were randomized to getting decitabine or best supportive care. It wasn’t exactly the same as the AZA-001 study, because it wasn’t conventional care regimens, but it was similar to the 60% of patients on AZA-001 who got best supportive care. In the decitabine study, however, the median survival for those who received decitabine was 10.1 months vs 8.5 months for those who got best supportive care. That was not a significant difference between those 2 arms.

This has had us wringing our hands for years. Why is it that we had such a dramatic survival benefit in 1 study but not the other? The answer is in the control groups. If you look at the control group on the AZA-001 study, they lived for 15½ months. For those on the decitabine randomized trial, they lived for only 8½ months. That has to get to patient selection on the studies. These were simply very different patient populations who were enrolled in 1 study vs the other.

Another explanation for the big difference between the 2 studies is that I always say there must have been some magic dust that they sprinkled on those patients in the AZA-001 study, because no randomized trial has ever been able to recapitulate that median survival of 24 months for patients who received azacitidine. Much more typical is the median survival that we saw in the United States in North American Cooperative Group studies that compared azacitidine to azacitidine plus entinostat on the ECOG study, or azacitidine plus lenalidomide or vorinostat in the North American Intergroup Study. On both of those studies, the median survival for a patient receiving azacitidine with higher-risk MDS was about a year and a half. That is what our comparison should be.

Those are the trial data justifying our use. Yazan, you and I are pretty supportive of using azacitidine as first-line therapy in patients with higher-risk MDS. We have this new oral decitabine-cedazuridine combination. It was approved by the FDA in a study that was a little complicated in how it was structured. Patients with higher-risk disease or intermediate 1 risk disease—lower risk was included in that—were randomized to receive oral decitabine or IV [intravenous] decitabine. There wasn’t a placebo control group. There wasn’t another arm getting another different type of therapy. They both got decitabine. They got that for 1 cycle, and then they could switch over and get IV decitabine.

When you look at the response rates on this trial, it was about what we’ve seen for decitabine or azacitidine. When you look at the PK [pharmacokinetics] and PD [pharmacodynamics] data, they’re very similar for oral decitabine and IV decitabine. When you look at the median overall survival, it’s about a year and a half which, in my mind was a little low for a study that included patients with lower-risk MDS. I would have thought it would have been in the 20- to 24-month range. That being said, we’re seeing similar outcome data for oral decitabine vs IV decitabine, but we haven’t seen a true comparative study of oral decitabine vs conventional care regimens or placebo. I’m still voting for higher-risk MDS. My level 1 data is with treating them with azacitidine up front.

Mikkael A. Sekeres, MD, discusses his approach to treatment in patients with high-risk MDS incorporating efficacious clinical trial data.

High-Risk MDS: Clinical Efficacy Data

 We’ve been talking a lot about the biology of the disease, risk stratification, and genetics. We tend to talk a lot about that with MDS [myelodysplastic syndrome] because, frankly, we don’t have a lot of treatment options, so we spend a lot of time talking about the biology. Once someone has a diagnosis of higher-risk MDS, what are the clinical implications? How do you start to think about treatment?

 That’s a great point. And to comment on the risk score that you mentioned that’s coming out, this would be fantastic for better personalizing therapy and recommendations for those patients. We get those NGS [next-generation sequencing] panels and molecular testing for all those patients. It’s become a standard of care, but we don’t fully know how to incorporate them, other than we know that certain mutations have higher risk. But which patients should go on to transplant and which patients may be OK just monitoring and watching those counts at the time they would maybe have an AML [acute myeloid leukemia] progression?

This is where those clinical implications come into play. From a treatment perspective, when I see patients with higher-risk disease, the first question I ask myself is, are they going to go on to get an allogeneic transplant, and are we going for a curative therapy? At the 2020 ASH [American Society of Hematology Annual Meeting], they presented nice data that were also presented at the TCT [Transplantation & Cellular Therapy] Meetings for patients up 75 years of age who proceeded with an allogeneic transplant. That showed a pretty clear survival advantage for those patients. As you know, getting to transplant may take a few months. That process is undergoing, but getting the disease burden under better control would be the most common approach to therapy rather than watching the patients, even if their blasts were acceptable for transplant, to prevent that AML progression and maybe reduce the disease burden, though some argue against that.

My general approach would be to start with hypomethylating agent therapy. Nowadays, we have 2 older ones: the azacitidine and IV [intravenous] decitabine and then oral decitabine, or decitabine-cedazuridine, which has been approved in the past year as an oral option for patients with higher-risk MDS. Hypomethylating therapy is the standard of care in this day and age. Getting an allogeneic transplant or having that patient meet with a transplant physician getting the HLA [human leukocyte antigen] typing in process up front would be my way to go, and then reassessing the disease at 4 to 6 months, depending on the transplant plans.

If they don’t have any plans to go to transplant, I’d delay it up to 6 months, because most patients would have some form of response by then if they’re going to respond. If the transplant things are ready, we may push the marrow a month or 2 earlier. That’s to decrease AML transformation and to prolong survival. As you’re aware, the trial that really showed the survival advantage was with azacitidine—the AZA-001 trial—and that would be the preferred option for me as a starting therapy.

Experts in the treatment of MDS discuss the clinical implications and treatment selection for patients with high-risk MDS.

Clinical Implications of High-Risk MDS

 This brings us to talking about prognostic and predictive factors that you think might be relevant for our patients with higher-risk disease.

 It almost seems silly to say, but this is one of my favorite questions I ask residents and fellows: What’s the greatest risk factor for survival with MDS [myelodysplastic syndromes]? Because it’s so obvious, people usually skip over it, but No. 1 is age. If you have somebody who’s presenting with MDS at the age of 98, regardless of whether that person has MDS, they probably don’t have a long predicted overall survival compared with somebody who’s presenting with MDS at age 55, which is relatively uncommon.

Secondly, we think about how the MDS arose. Therapy-related MDS is often called out as a poor-risk disease, but it’s a complicated issue. First, it’s complicated because how do we define therapy-related MDS? It’s a clinical diagnosis that somebody received something years earlier that could cause DNA damage, and now that person has MDS. Whether that agent they received, be it chemotherapy or radiation therapy, actually led to the MDS is not known biologically. What’s known is they had something in the past—let’s say it’s lymphoma or breast cancer—received treatment for that with radiation therapy or chemotherapy, and then developed MDS 7 years later.

The reason this has become so complicated is because we don’t know if that person actually had a CHIP [clonal hematopoiesis of indeterminate potential] abnormality prior to the diagnosis of that solid tumor or lymphoma, and then received chemotherapy and radiation therapy, and if that CHIP abnormality would have evolved naturally, or if it was nudged along by the radiation therapy and chemotherapy that person subsequently received. Then we get into the issue of adjusting for the poor-risk features that are sometimes associated with therapy-related disease. If somebody has a 

 abnormality or chromosome 7 abnormality, regardless of whether that’s therapy-related MDS or de novo MDS, that person has a poor prognosis.

When we do these adjustments, it turns out that based on a study published out of Massachusetts General Hospital about a decade ago, folks who were exposed to radiation therapy for previous cancers, when adjusted for risk factors for MDS, didn’t have much worse of an outcome than patients with de novo MDS who had some of those same karyotypic abnormalities and other risk factors. Still, how I try to summarize therapy-related disease, because it’s complicated, is that if someone has what we think is true therapy-related disease, where previous chemotherapy or radiation therapy either started the clock, moving a genetic landslide that leads to MDS, or advanced a clock that had already started toward MDS, they tend to have worse genetic markers and therefore they tend to have worse outcome.

What this leads to is the third major risk factor that you alluded to, Yazan, the molecular makeup of folks who have MDS. We’re getting much more sophisticated about this. We know that there are certain abnormalities like 

 mutations that portend a worse prognosis. How we incorporate these into the IPSS [International Prognostic Scoring System] and revised IPSS is also very complicated. Aziz Nazha, MD, whom you mentioned before, published a paper incorporating molecular abnormalities into the IPSS and showed that it provides additional accuracy over the clinical parameters alone. He also has a paper coming out looking at machine learning and artificial intelligence, where it takes it up a notch in looking at genetic and clinical factors. The machine learning approach to looking at prognosis is both more accurate than the IPSS or revised IPSS, even with molecular abnormalities incorporated, and it’s dynamic, meaning it can be used at day 0 of a person’s diagnosis, or 1, 2, or 3 years later, and maintain its accuracy. Also, it seems to be agnostic to therapy. That will help us risk stratify a lot better, particularly incorporating molecular abnormalities.

Mikkael A. Sekeres, MD, explores prognostic and predictive factors inherent to high-risk MDS, including age and molecular abnormalities.

High-Risk MDS: Prognostic and Predictive Factors

 Yazan, can you go into a little more detail about how risk is assessed? You mentioned there are a lot of different risk stratification systems we could use. Do you think one is clearly better than the others? What do you use practically in the clinic?

 That’s a great question. It’s a challenge to figure out what the best tool is because the IPSS [International Prognostic Scoring System] was the first and was devised in 1997. That relies on the chromosomal abnormalities, number of cytopenias, and blast percentage. But it’s quite outdated in terms of how it classifies disease at that point. It included patients with 21% to 30% blasts, where nowadays, we would classify them as having AML [acute myeloid leukemia]. It only included patients who were treated with supportive care at that time. It also has only been validated at diagnosis. It definitely has its limitations. The easy part is that it groups patients into 4 disease risks. The low and intermediate-1 can be grouped together as lower-risk, and then intermediate-2 and high can be grouped together as higher-risk disease.

When it was revised in 2012 with the IPSS-R [revised] model, it focused more on the cytogenetic risks and had more thorough cytogenetic detail with 5 cytogenetic risk categories, which gives it more points to evolve into higher-risk disease, even if the patient doesn’t have as many cytopenias or a high blast percentage. That made it a little more complex and added an intermediate category. Sometimes in clinical trials, it’s counted as lower-risk disease, while in others it’s counted as higher-risk disease, which creates some confusion as to what to do with the intermediate-risk category. One study did an analysis where they looked at a score of 3.5 and said that 3.5 or less is lower risk, but higher than that, we should group them together as higher risk.

I usually use both the IPSS and the IPSS-R and see if they correlate, because then I know I’m dealing with 1 disease risk. If they’re not correlating, you always question which is going to be more accurate. Some of the work Aziz Nazha, MD, published is looking into which score really validates a diagnosis. He presented 3 nice cases where each of the scoring systems had their deficiencies. It’s always validated up to 70% accuracy, or 80% at best, but there are always some inaccuracies in those scoring systems. I use them as a guide, but not a definitive prediction that a patient is going to progress at this time point.

Then there are the WHO [World Health Organization] Prognostic Scoring System and the MD Anderson Scoring System. The advantage of the WHO system is that it’s a dynamic scoring system, so it can be applied at diagnosis and at later time points. It also includes de novo and therapy-related MDS [myelodysplastic syndromes]. However, it’s not widely used. The issue is that it doesn’t account for cytopenias other than anemia, so it’s not a great scoring system for all patients with MDS. The MD Anderson Scoring System can be a little complex to use as a quick guide in the clinic, but it also includes therapy-related MDS. The most commonly used scoring systems are the IPSS and its revised version. We’re looking forward to hopefully a molecular IPSS score at some point.

Yazan Madanat, MD, discusses available MDS risk assessment tools and provides insight into his preferred clinical assessment methods.

MDS: Risk Assessment Tools

Drs Madanat and Sekeres comment on the various risk stratification factors and prognostic criteria used in the classification of MDS disease.

Clinical Considerations in the Diagnosis of MDS

Yazan Madanat, M.D and Mikkael A. Sekeres, M.D. discuss considerations in the classification of myelodysplastic syndrome taking into account disease heterogeneity.

Yazan Madanat, MD

Articles