Management of Hypomethylating Agent (HMA) Failure in Myelodysplastic Syndromes (MDS)

Dr. Jamile Shammo
Jamile M. Shammo, MD, FASCP, FACP
Professor of Medicine and Pathology
Section of Hematology and Stem Cell Transplantation
Division of Hematology/Oncology
Rush University Medical Center
Chicago, Illinois

Join Dr. Jamile Shammo as she explores challenges and advances in the management of hypomethylating agent (HMA) failure in myelodysplastic syndromes (MDS). Topics will include potential mechanisms of resistance, distinguishing between HMA failure and HMA-related cytopenia, current standard care following HMA failure, and notable clinical trials for treatment of MDS patients with HMA failure.

Can you describe the current challenges around managing high-risk MDS patients who fail on treatment with hypomethylating agents (HMAs)?

Hypomethylating agents including azacitidine and decitabine represent the standard of care for the treatment of patients who have intermediate-2 and high-risk disease. Results of the AZA-001 trial published back in 2009 showed for the first time that there was a prolongation in survival with HMA treatment in this particular patient population.1 This was a paradigm shift, and it changed how we think about the importance of active therapy for those patients. However, we realized that there was no cure with such treatment, and that only half of patients would reap any benefits from those approaches, despite the perceived benefit from treatment with HMAs and their impact on the natural course of disease. Furthermore, the outcome of patients who have failed such treatment is dismal as they have a very short survival, estimated between 4-8 months, depending on the study you look at.

Part of the problem is that we don't truly understand the mechanism of resistance to HMAs. There's been a lot of work that focused on that particular issue, mostly lab work, to understand exactly what happens on a cellular level when patients stop responding to HMAs, or even in those who, to begin with, don't respond to HMAs. There is really no single answer to this question, nor is there actually a practical mechanism to explain how to overcome this particular predicament. This could explain in part why we don't have a second-line treatment for myelodysplastic syndromes (MDS) despite the multitude of clinical trials that have taken place in this particular space.2

A few years back, we collaborated with researchers from the University of Chicago and examined methylation patterns in various cells that we had obtained from MDS and AML patients, including those that have had a good response to azacitidine and those who had failed such treatment.3 The study demonstrated that there's a particular pattern to methyltransferases that interacts with various cellular proteins that form distinct complexes and that these active chromatin structures in azacitidine-sensitive leukemia or MDS cells and in resistant cells could potentially be exploited pharmacologically. Interestingly, next-generation sequencing (NGS) could not identify a particular mutations or genetic alterations in association with 5-AZA-resistant chromatin structure in MDS/leukemia cells. Therefore, additional studies are needed to identify the genetic alterations and the chemical modifications underlying the drug responsive/resistant chromatin structures and to determine the potential of using RNA:m5C/RCMTs and their associated chromatin structures as therapeutic targets in MDS/AML.

I think perhaps that in an ideal world, and once patterns of resistance are delineated, we may be better equipped to identify a therapeutic target that a particular patient may benefit from after failing frontline therapy. You can imagine how complicated this may be on a patient-by-patient basis. That also means, in more general terms, that in order for us to answer that question, and have better outcomes in this patient population, we truly should encourage participation in rationally designed clinical trials aiming to incorporate molecular targets in their treatment schema. This is the current standard of care, although there are other potential options, such as switching to an alternative HMA or administering intensive chemotherapy with allogeneic hematopoietic stem cell transplant.

Meanwhile, until we are able to better understand MDS biology and patterns of resistance, the MDS scientific community has focused on improving the quality of response by investigating combination trials as a management strategy in frontline setting of MDS. Multiple frontline combination trials focusing on combining HMAs with other agents, such as pevonedistat, venetoclax, magrolimab and others, are currently ongoing. It is intuitive for oncologists to believe that combination treatment may be better than single-agent therapy, but we need to have randomized phase 3 data that support this notion before it is translated into practice.

What about patients who have failed or progressed through initial line of therapy? How can we manage those patients accordingly, and how should they be channeled into clinical trials so that we can ultimately identify a second-line treatment?

How do you distinguish between HMA failure, or a true loss of response or progression, as opposed to a cytopenia related to the HMA itself?

We, as practitioners, do need to be very careful about designating a patient as an “HMA failure.” Often, patients with MDS experience prolonged cytopenias that could be the result of repeated cycles of chemotherapy. If lengthening the chemotherapy cycle results in improvement in cytopenias, then their cytopenias were related to chemotherapy, not necessarily to disease progression. In that case, the patient should be able to continue therapy with HMAs. By contrast, if cytopenias continue to downtrend despite lengthening treatment cycle, then indeed they are the result of disease progression. Holding chemotherapy for 1-2 weeks may allow for bone marrow recovery. A bone marrow biopsy may be needed if counts do not recover at that point. The International Working Group (IWG) criteria for disease progression can then be applied in order to designate a patient as an HMA failure.4

What are some of the challenges in achieving favorable clinical outcomes for patients with intermediate- and high-risk MDS who are known to have failed HMAs?

As of yet, we do not have second-line treatment options for patients who have already failed HMAs. Therefore, everyone who meets criteria for HMA failure should be considered for rationally designed clinical trials, if they are eligible. Now, there's obviously a challenge in this patient population, which typically tends to be older and with comorbidities. Most of the clinical trials are run at academic medical centers that are often in urban areas, so there are a lot of logistical issues that perhaps preclude patients from participating in clinical studies. In fact, a member of my research team is currently investigating this particular issue to identify the reasons behind patients’ lack of participation in clinical trials. It's very interesting to see why some of the patients don't participate—it goes beyond comorbidities and a lot of it is logistical.

If there is no clinical trial available to the patient, and if the patient is clinically fit with high-risk MDS/normal karyotype, and has already been evaluated for an allogeneic transplant—which is the recommended course of action for anyone who has high-risk MDS at the outset, and we typically recommend that this takes place at the very early stage of evaluating this disease. I don't think that intensive chemotherapy for this patient population is unreasonable, provided allogeneic stem cell transplantation takes place at some point. For me, I rarely consider intensive chemotherapy without proceeding to allogeneic transplant in the treatment of high-risk MDS.

In the absence of a clinical trial, switching to a different or an alternate HMA has been reported, at least anecdotally. There are very few trials and the inclusion criteria for those trials is all over the board, so it's really difficult to draw any meaningful conclusions when you switch from one agent to the next. You could argue that the different HMAs perhaps have different mechanisms of action, and perhaps you could justify switching from one to the other. In my opinion, it's very possible that perhaps it is the continuation of treatment with a hypomethylator that may result in improvement, or perhaps result in a response that maybe you didn't see early in the treatment course.

In addition, now that we have an oral option that's available (decitabine and cedazuridine), HMA treatment may represent a viable option for those who, in the setting of a COVID-19 pandemic, want to minimize their visits to the clinic.5 I think it's also a reasonable option for patients who may not have responded to IV or subcutaneous azacitidine. So, I might entertain that option in the absence of a clinical trial.

It may be helpful to obtain a next-generation sequencing study which would be very helpful to identify a mutation/targeted therapy or identify a clinical trial for this patient. However, I do not recommend adding a targeted treatment outside of the setting of a clinical trial for those patients.

What are some of the more notable recent clinical trials that evaluated treatments for MDS patients with HMA failure?

The second-generation HMA guadecitabine created a lot of excitement, initially, because it was a drug that was rationally designed to allow a prolonged exposure intracellularly to decitabine. In the initial studies, it appeared that there were reasonable responses for people who had failed HMAs. We participated in ASTRAL-3 study that evaluated MDS patients who had failed HMA therapy. The ASTRAL-3 study compared 28-day cycles of guadecitabine to either low-dose cytarabine, standard intensive chemotherapy (7+3), or supportive care only. Unfortunately, the trial did not meet the primary endpoint of overall survival. This validates why it is important to conduct phase 3 clinical studies despite promising phase 2 studies.

How about venetoclax? What is the potential role of venetoclax in MDS?

Venetoclax is a promising drug. In the initial phase 1b study including patients with relapsed/refractory MDS, there were two arms, looking at single-agent venetoclax (400 mg or 800 mg every day for 28 days) or venetoclax escalated 100 mg, 200 mg, 400 mg for 14 days of every 28 days plus azacitidine for 7 days. Patients had to have either azacitidine or decitabine for a minimum of 4 cycles to be eligible. In the initial report in 2019, the investigators first determined the phase 2 recommended dose.6 At the 2020 American Society of Hematology (ASH) annual meeting, they presented extended data on the combination arm.7 For 37 patients evaluable for response, they reported a 40% rate of complete response (CR) plus marrow CR (mCR) with this combination. At 12 months, 65% of those patients were alive. We have to be very careful, because about 37% of those patients discontinued the study. Patients can have prolonged cytopenias. But compared to the very dismal outcomes that we see in patients who have had HMA failure, the 12-month estimate of overall survival of 65%, and of 78% among patients who achieved mCR is remarkable.

This is a small phase 1b study, so again, we need to be very careful, but I am excited about this. We need long-term follow-up data. Also, there is a frontline clinical trial, as I mentioned before, evaluating combination therapy in frontline setting compared with single-agent therapy. The questions are two-fold: first, would you get better responses with the combination in the frontline? And secondly, if venetoclax is added to azacitidine in the relapsed/refractory setting, could those patients be salvaged—could you prolong their responses and essentially reverse the state of resistance with BCL2 inhibition? More to come on that—we’ll just have to wait and see.

What about other approaches that have been investigated recently, such as the use of immune checkpoint inhibitors or bemcentinib?

Dr. Garcia-Manero and others have done a tremendous amount of work in clinical trials looking at immune checkpoint inhibitors, either alone or in combination with azacitidine, both in the frontline and the relapsed setting. It makes sense to put those treatments together, because it appears that HMAs have anti-tumor immunity enhancing effects that may abrogate some of the evasion mechanisms to immune checkpoint blockade therapy.8 However, the study results are such a mixture. In one case, you can see clinical trials with a zero response rate, and in others, you see some improvement. More data are needed.

Bemcentinib is an inhibitor of AXL, which is a surface membrane protein kinase receptor that mediates drug resistance and decreases anti-tumor immune response.9 The phase 2 BERGAMO trial looked at bemcentinib in high-risk MDS patients who were refractory to either azacitidine or decitabine.10 Overall response rate, the primary endpoint in this study, was 33.3%, or 7 out of 21 patients, but if you look at the designation of the responses, 1 patient had a CR, 3 had a CR with incomplete hematologic recovery, and 3 had stable disease. I realize that this is really very early data, and it's a difficult patient population, so maybe we can say that there's a signal in here. We may need to understand a little more about the patients who responded to understand how this agent would be studied in the future.

What can be done to encourage participation in relevant MDS clinical trials?

I think that patients truly need to team up with their MDS expert and with their treating physician. They need to have realistic expectations that once their disease truly fails, then they have a major decision to make. Is allogeneic transplant a consideration? If not, then the question is whether a clinical trial is a consideration? If it is, then what is available, and are they willing to participate? My suggestion is to see what is offered in your area/local academic center. The MDS Foundation is one option to find clinical trials, and is another.

If patients decide not to be treated in a clinical trial, then the question is, what is their goal? Supportive care, iron chelation, antibiotics, growth factors? It needs to be an honest and a straightforward conversation with their treating physician as to goals of care.

What is the most important thing a practicing clinician can do today to address the clinical challenge of treating a high-risk MDS patient who has failed HMA therapy?

Every effort must be made to optimize every patient's care from the beginning. You must absolutely understand the disease characteristics at baseline, and determine the plan of care from the outset, with consideration for stem cell transplantation early on. I think you have to have the plan laid out and understand the disease characteristics from the get-go—that way there are no surprises in the middle of the road.


  1. Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al. Efficacy of azacitidine compared with that of conventional care regimens in the treatment of higher-risk myelodysplastic syndromes: a randomised, open-label, phase III study. Lancet Oncol. 2009;10(3):223-232. doi:10.1016/
  2. Kubasch AS, Platzbecker U. The wolf of hypomethylating agent failure: What comes next? Haematologica. 2019;104(8):1505-1508. doi:10.3324/haematol.2019.222794
  3. Cheng JX, Chen L, Li Y, et al. RNA cytosine methylation and methyltransferases mediate chromatin organization and 5-azacytidine response and resistance in leukaemia [published correction appears in Nat Commun. 2018 Jun 6;9(1):2286]. Nat Commun. 2018;9(1):1163. Published 2018 Mar 21. doi:10.1038/s41467-018-03513-4
  4. Cheson BD, Bennett JM, Kantarjian H, et al. Report of an international working group to standardize response criteria for myelodysplastic syndromes. Blood 2000; 96 (12): 3671-3674. doi:
  5. FDA approves combination of decitabine and cedazuridine for myelodysplastic syndromes. FDA press release. July 7, 2020. Accessed on March 15, 2021.
  6. Zeidan AM, Pollyea DA, Garcia JS, et al. A Phase 1b Study Evaluating the Safety and Efficacy of Venetoclax As Monotherapy or in Combination with Azacitidine for the Treatment of Relapsed/Refractory Myelodysplastic Syndrome. Blood. 2019;134(Supplement_1):565-565. doi:10.1182/blood-2019-124994
  7. Garcia JS, Wei AH, Borate U, et al. A Phase 1b Study Evaluating the Safety and Efficacy of Venetoclax in Combination with Azacitidine for the Treatment of Relapsed/Refractory Myelodysplastic Syndrome. Abstract #3109. Presented at the 2020 ASH Annual Meeting, December 7, 2020.
  8. Daver N, Boddu P, Garcia-Manero G, et al. Hypomethylating agents in combination with immune checkpoint inhibitors in acute myeloid leukemia and myelodysplastic syndromes. Leukemia. 2018;32(5):1094-1105. doi:10.1038/s41375-018-0070-8
  9. Zhu C, Wei Y, Wei X. AXL receptor tyrosine kinase as a promising anti-cancer approach: Functions, molecular mechanisms and clinical applications. Mol Cancer. 2019;18(1):153. Published 2019 Nov 4. doi:10.1186/s12943-019-1090-1093.
  10. Kubasch AS, Peterlin P, Cluzeau T, et al. Efficacy and Safety of Bemcentinib in Patients with Myelodysplastic Syndromes or Acute Myeloid Leukemia Failing Hypomethylating Agents. Abstract #1287. Presented at the 2020 ASH Annual Meeting, December 5, 2020.

This activity is supported by educational grants from Bristol-Myers Squibb and Taiho Oncology, Inc.