For patients with myelodysplastic syndromes (MDS) or low-blast secondary acute myeloid leukemia (AML), allogeneic hematopoietic cell transplantation (AHCT) is the only potentially curative treatment option. However, debate persists around the balance of benefits and risks associated with disease reduction using hypomethylating agents or induction chemotherapy prior to transplantation. Furthermore, if disease reduction proves valuable in this setting, what is the best approach?
Despite the availability of promising new treatment options for MDS, their role in pretransplant cytoreduction has not been established.
In this edition of Drawing First Blood, ASH Clinical News asked Corey Cutler, MD, MPH, and Guido Kobbe, MD, to weigh in on the question, “Should patients with advanced MDS or low-blast AML undergo disease reduction (or debulking) with chemotherapy before undergoing AHCT?” Dr. Cutler was asked to argue in support of disease reduction, and Dr. Kobbe was asked to argue against.
Corey Cutler, MD, MPH: We know that transplant outcomes correlate with pretransplant blast counts, International Prognostic Scoring System (IPSS), and revised IPSS (IPSS-R) scores. The one thing we can do prior to transplant to try to influence outcomes is lower blast counts, thereby reducing IPSS and IPSS-R scores. Debulking might also reduce allelic burden and adverse molecular features in patients with advanced MDS. Given these data, I think patients with MDS and low-blast AML should undergo debulking chemotherapy prior to transplant if possible.
Also, if transplant is delayed for any reason beyond our control, then having an ongoing strategy for disease reduction will prevent the patient’s disease from advancing too far and, perhaps, increasing the risk of the patient never making it to the transplant unit.
Guido Kobbe, MD: I would argue against debulking prior to AHCT in this setting because the early studies that demonstrated the correlation between transplant outcomes and pretransplant blast counts specifically compared lower-risk patients with higher-risk patients.
If you compare a patient who already had an attempt of debulking with a patient who had upfront transplant, you will not find a link between higher blast counts and outcomes. The outcomes are more likely a function of genetic risk, which is determined by karyotype and, of course, additional mutations.
Another argument against debulking, at least for now, is that we do not have efficient tools for real debulking in higher-risk patients. Conventional chemotherapy, as well as hypomethylating agents, do not perform well in high-risk patients. In addition, administering chemotherapy to these higher-risk patients puts them at risk for severe complications, which could also lead to patients not arriving at transplant.
Finally, a third argument against pretransplant disease reduction is the management of relapse. Looking at patients who relapse after transplant – and relapse is a major problem for MDS – we found a correlation between response and being untreated before transplant.1 The idea behind this is that, with pretreatment, you select for chemotherapy-resistant clones that adapt to the cytotoxic stress they are exposed to. When those patients’ disease relapses, there is hardly a chance for long-term survival.
“[An] argument against debulking, at least for now, is that we do not have efficient tools for real debulking in higher-risk patients.”
–Guido Kobbe, MD
In contrast, patients who do not undergo pretransplant disease reduction and whose disease later relapses after transplant respond better to therapy with hypomethylating agents and donor lymphocyte infusion, resulting in reasonable long-term remission rates.
Dr. Cutler: I have to question that last point, because the patients who undergo debulking and do not do well after transplant are those who we could predict would not do well after transplant regardless. It is possible to avoid transplant if we know that a patient is destined to not do well based on their response to chemotherapy. Those whose disease progresses are the least likely to be helped by transplant. Those may be the patients who should be directed to receive innovative new therapeutics or clinical trials prior to undergoing a transplant.
Dr. Kobbe: We also need to acknowledge that all the data we have so far – and we have looked at a lot of patients, especially in Germany – are from the pre-venetoclax era. The probability of achieving a remission with these agents is far below 20%; in high-risk patients with complex karyotype, the remission rate is similar to that with conventional cytotoxic therapy. The BCL2 inhibitor venetoclax may transform pretransplant therapy in this setting. We still do not know that.
Dr. Cutler: I agree with you completely that venetoclax might be a game changer. It might make pretransplant debulking therapy very effective, significantly increasing the number of patients whose disease will respond to transplant.
Dr. Kobbe: In the VidazaAllo trial, which compared survival and response rates with azacitidine and azacitidine induction followed by AHCT in older patients with newly diagnosed high-risk MDS, 31% of patients discontinued the study prematurely due to progressive disease during the debulking strategy with hypomethylating agents.2 All participants were transplant-eligible at inclusion, but nearly one-third did not reach transplant even though they were scheduled only to receive four courses of azacitidine.
Dr. Cutler: I would not advocate for a fixed number of cycles prior to transplant. My argument is that, when you meet the patient, you cannot predict when transplant is going to occur. This is particularly true now in the context of the COVID-19 pandemic. If we do not perform debulking, and it takes three to four months to plan transplant and obtain a donor, some patients might have progressive disease or may develop comorbidities or infections, making them ineligible for transplant.
I agree that one-third of participants is a substantial portion, but I don’t think we need to give four rounds of therapy automatically. Instead, start the patient and move to transplant when the time is right.
Dr. Kobbe: If you use only azacitidine without venetoclax as a disease reduction strategy, the median number of cycles needed to achieve optimal response and reach the debulking goal is about four. In patients who have neutrophil counts hovering at the border or just above 500 cells/µL or who have low platelet counts, there is a high probability that administering hypomethylating agents will push patients into the “danger zone” for a period of two to three months before they achieve a meaningful response.
Venetoclax might change the conversation because the maximum number of cycles needed to achieve response with venetoclax is short, about 1.3. We currently do not have real data on venetoclax as a pretransplant debulking strategy.
There also has never been a randomized clinical trial directly comparing debulking against no debulking. The so-called EBMT 2×2 study started in the 1990s but was closed 15 years ago because of low recruitment. Several retrospective studies have compared chemotherapy with an upfront transplant strategy; none found significant differences in overall survival, recurrence-free survival, or non-relapse mortality. However, all these studies looked at patients at the time of transplant. One must keep in mind that the dropout rate among patients who had received chemotherapy was at least 10 to 15% on the way to transplant. That was not calculated in the results.
We recently published our single-center retrospective data, showing that there was no difference in recurrence-free survival with either strategy.3 Taking into account the improved responses after relapse, there was at least a trend toward improved overall survival in untreated patients.
Dr. Cutler: Another study from 2017 suggested that pretransplant marrow responders do the best,4 but none of these trials are adequately powered to address the question of disease reduction. I agree with Dr. Kobbe, though, that there have been no prospective, randomized trials to help answer this question.
“If we do not perform debulking, and it takes three to four months to plan transplant … some patients [may become] ineligible for transplant.”
–Corey Cutler, MD, MPH
The main issue here is that the randomization needs to occur at the time the physician decides whether a patient should undergo transplant. As Dr. Kobbe said, the common flaw in the retrospective analyses is that they only looked at patients who received transplant. You need to start the clock at the time when patients are deemed to be candidates; if not, you miss those patients who you wanted to transplant, but who developed comorbidities or died before transplant occurs.
In the most recent Blood and Marrow Transplant Clinical Trials Network Study 1102, we looked at the influence of hypomethylation on transplant outcomes.5 In a Cox regression model, one of the important factors that influenced whether patients responded well to transplant was their past response to hypomethylating therapy. In other words, if an individual did not have a good response to hypomethylating therapy, his or her outcome after transplant was worse. That supports the argument that those are people who should not be moving right to transplant. They should be offered alternative therapies first, before transplant.
Dr. Kobbe: With all the retrospective studies, the main argument is that we may be talking about different patients. Patients who do not receive hypomethylation and have stable disease will likely do well even without debulking. In those who have progressive disease, physicians will intuitively start therapy.
This is the standing argument when I speak against pretransplant disease reduction. With this in mind, we looked at the genetic signature of our patients and looked for unfavorable mutations, trying to make a link between biologic signature of the disease and influence of pretreatment.6
In our recently published results, we saw that, particularly in patients who have unfavorable karyotypes or mutations, outcomes were better without pretransplant disease reduction.
To definitively answer the debulking question, we have to look in great detail at response to salvage therapy after relapse. So far, there are only a few papers demonstrating that emergence of resistant clones during azacitidine treatment is the basis of relapse after post-transplant relapse. That is an important thing: The longer the debulking strategy lasts before transplant, and the more evolutionary pressure is put on the leukemic stem cells, the more resistant cells will be selected. This must be investigated in a larger group study.
Dr. Cutler: The next question is defining the optimal approach to pretransplant disease reduction. What are the best agents? What is the value of venetoclax? What are the goals of debulking?
The key message of this discussion is that we need proper randomized studies. At the end of the day, we both would agree that a proper randomized trial using an active agent is what is required to definitively answer this question. We have reached the time when we are able to do that.
- Rautenberg C, Bergmann A, Germing U, et al. Prediction of response and survival following treatment with azacytidine for relapse of acute myeloid leukemia and myelodysplastic syndromes after allogeneic hematopoietic stem cell transplantation. Cancers (Basel). 2020;12(8):2255.
- Kroeger N, Sockel K, Wolschke C, et al. Prospective multicenter phase 3 study comparing 5-azacytidine (5-aza) induction followed by allogeneic stem cell transplantation versus continuous 5-aza according to donor availability in elderly MDS patients (55-70 years) (VidazaAllo Study). Blood. 2018;132(supplement 1):208.
- Schroeder T, Wegener N, Lauseker M, et al. Comparison between upfront transplantation and different pretransplant cytoreductive treatment approaches in patients with high-risk myelodysplastic syndrome and secondary acute myelogenous leukemia. Biol Blood Marrow Transplant. 2019;25(8):1550-1559.
- Yahng S-A, Kim M, Kim T-M, et al. Better transplant outcome with pre-transplant marrow response after hypomethylating treatment in higher-risk MDS with excess blasts. Oncotarget. 2017;8(7):12342-12354.
- Nakamura R, Saber W, Martens MJ, et al. A multi-center biologic assignment trial comparing reduced intensity allogeneic hematopoietic cell transplantation to hypomethylating therapy or best supportive care in patients aged 50-75 with advanced myelodysplastic syndrome: Blood and Marrow Transplant Clinical Trials Network Study 1102. Abstract #75. Presented at the 2020 American Society of Hematology Annual Meeting, December 5, 2020.
- Rautenberg C, Germing U, Stepanow S, et al. Influence of somatic mutations and pretransplant strategies in patients allografted for myelodysplastic syndrome or secondary acute myeloid leukemia Am J Hematol. 2021;96(1):E15-E17.