Expert Opinions on Common Clinical Challenges in AML

Published on October 14, 2020 in Treatment

Naval Daver, MD
Associate Professor
Department of Leukemia
The University of Texas MD Anderson Cancer Center
Houston, Texas
Jessica K. Altman, MD
Professor of Medicine
Robert H. Lurie Comprehensive Cancer Center
Northwestern University
Chicago, Illinois
Alexander Perl, MD
Associate Professor of Medicine
Abramson Cancer Center
University of Pennsylvania
Philadelphia, Pennsylvania


Despite advances in treatment for acute myeloid leukemia (AML), many clinical challenges remain, such as choosing the best treatment course for patients with TP53-mutated disease or maintenance regimens for FLT3-mutated AML or minimal residual disease (MRD)–eradicating strategies or optimizing approved agents, among others. This article summarizes some of the audience questions from the MediCom accredited activity, New Advances in AML Live Interactive Webcast Series. Read more below to see how program chair Naval Daver, MD, and faculty panelists Jessica K. Altman, MD, and Alexander E. Perl, MD, responded to audience members’ most compelling clinical questions.

What maintenance do you give for FLT3-mutated AML?

Jessica K. Altman, MD: If you recall, the RATIFY trial,1 which was the trial looking at the combination of 7+ 3 with midostaurin or placebo for newly diagnosed patients and then utilizing it and midostaurin in consolidation, that clinical trial allowed midostaurin maintenance as well. But the FDA did not approve the utilization of midostaurin in a maintenance setting, and midostaurin does not have an approval in that setting. Any of my patients with a FLT3 mutation who are fit for stem cell transplant and have an appropriate donor, I take to stem cell transplant. And in those patients, there has been a recent study that closed looking at gilteritinib vs placebo for maintenance post-stem cell transplant in that setting, and we're waiting for results from that trial to determine the benefit of gilteritinib vs placebo post-transplant.2 Those results are eagerly awaited.

In the absence of a clinical trial, based on European data3 if patients can tolerate it, I frequently use sorafenib, but as mentioned, we're eagerly waiting results from other studies. And I think for patients who are not transplant patients, it really kind of depends how I treated them in the frontline setting. If they're receiving an azacitidine/venetoclax-based regimen, we are generally continuing that until progression, or if they happen to receive intensive induction/consolidation therapy and weren't able to go to stem cell transplant. Again, it's not clear what an optimal maintenance approach is, and if I can't utilize gilteritinib or another agent in maintenance, those are patients that I follow extraordinarily closely and try to see if they develop any evidence of molecular recurrence and then treat those patients at that time. But I admit that's an area where we don't have perfect information about what to do.

When giving venetoclax to an older patient, knowing that venetoclax is highly associated with low blood counts, how do you assess the patient to rule out relapse vs toxicity? How often should I do a bone marrow aspiration?

Naval Daver, MD: This is an extremely important question, and I think something that still has not yet potentially percolated into the community setting as much as it has been in the large and very large academic cancer centers that have been doing venetoclax now for almost 4 to 5 years in AML. The combination of azacitidine or decitabine with venetoclax is extremely active. I don't think there's any questions about the efficacy of that combination with the complete response/complete response with incomplete neutrophil recovery (CR/CRi) rates of 65% to 70%, as well as low early mortality rates usually of 6% to 7%,4 and in our experience at MD Anderson is even lower, 1% to 2%, because of routine use of prophylactic antifungals/antibacterials, as well as close monitoring in patients during the initial stay.

However, it must be noted that the combination is certainly more myelosuppressive than azacitidine alone,4 and I think just understanding that one big difference, that how you're used to giving azacitidine every 4 to 5 weeks with limited monitoring during cycle in your patients for the last 2 decades, this approach absolutely will not be safe in azacitidine/venetoclax. These patients cannot be seen every 4 to 5 weeks. We, for example, admit them usually for the first 7 to 10 days of therapy at least. We do follow the ramp-up as recommended. We do hydration. We do tumor lysis prophylaxis. We then do bloodwork, usually at least twice a week outpatient because we do see significant thrombocytopenia/anemia often requiring transfusion during the first 1-2 cycles. Nothing that cannot be managed, but something that does need to be closely monitored. You cannot just see the patient after 3 or 4 weeks like with azacitidine alone. And we always do a bone marrow at the end of cycle one. The exact timing may be debatable, I would say somewhere close to 28 days at the end of cycle 1. But you really need to do that end of cycle bone marrow because that is the only way you're going to know whether the low counts that the patient has at that time are because of persistent AML or because of myelosuppression from the azacitidine/venetoclax. What we see is about 70% to 80% of patients on the day-28 bone marrow will be hypocellular/acellular, often with little or no residual leukemia. So, they have achieved a marrow remission, but they're myelosuppressed because of continuous venetoclax. So, in those cases, we interrupted the venetoclax treatment. We give a 14- to 18-day break from venetoclax allowing for neutrophil platelet recovery. If neutrophils do not recover in 14 to 18 days on their own after waiting, we then give filgrastim for 3 or 4 continuous doses. Usually with that approach, we almost always get their counts up, and for the subsequent cycles, we are dropping the venetoclax to 21, and more and more frequently, even down to 14 days per cycle. And with that approach, we're seeing more and more people having count recovery, achieve full CR, and have less infections. So, it's very critical to check end-of-cycle bone marrow and use that to modulate your next course of action. Of course, if at the end of cycle 1, bone marrow shows persistent AML with high blast, then you proceed directly to cycle 2 because there's active disease causing low counts. But that is the minority of cases. Only 15% to 20% of patients in our experience have been truly primary refractory. So, I think that is very important when using and approaching the venetoclax-based treatments.

Is CPX-351 effective in patients who don't undergo hematopoietic stem cell transplant?

Naval Daver, MD: The answer to that question is yes, in fact, the paper that was published in the Journal of Clinical Oncology (JCO) in 2018 by Jeff Lancet and colleagues5 did show that even in patients who did not go the stem cell transplant, there was an improvement in survival. The difference is not as clearly significant as compared to patients who got CPX-351 and made it to allogeneic stem cell transplant, but there still appears to be a clinical benefit. I think the optimal sequence of events, and we are learning more and more that in AML, it is not one intervention or event but a sequence of events (total therapy approach), that results in improved overall survival. This has been shown with FLT3, for example, giving 3+7 midostaurin, doing stem cell transplant in remission, and doing post-transplant FLT3 maintenance is really when you start seeing dramatic improvements in survival. It's not any one of those interventions alone per se, but those all add up when done together and improve your cure rate. And the same analogy is now seen with therapy-AML and secondary MRC-related AML with CPX-351. CPX-351 seems to give higher remission rates, 48% vs 35%, and less toxicity as shown by lower early mortality, 13% vs 21%. This probably resulted in patients having a better overall condition, less comorbidities when they go to stem cell transplant. Also, probably was associated with a deeper depth of remission, and so once you take those patients to transplant, they tend to do better post-transplant because the two factors that determine transplant outcome are how fit the patient is when they go in for transplant, how well they're doing in general, as well as how deep the remission is before transplant. And so that's why we see them when we take people to transplant after CPX-351 as compared to after 3+7, we see much better post-transplant 2-year survival and beyond. In fact, at ASCO 2020, there was a long-term follow up showing the patients who got CPX-351 and went to transplant had a 4- to 5-year survival of almost 52% compared to 20% to 25% after 3+7; this was with 5-year follow up.6 So, this is really remarkable where a very, very high-risk group of patients aged 60 years and older with secondary or therapy-related AML could have at 5 years up to 50% survival, confirming long-term benefit and cure potential when you give CPX-351 followed by allogeneic stem cell transplant optimally.

How do you treat AML after myelodysplasia that's secondary to presumably chemotherapy?

Alexander E. Perl, MD: The treatment of AML that arises from MDS that's secondary to chemotherapy is quite challenging. The main way you want to treat this is with transplant. And the big question is how do you get your patients to transplant fit with controlled disease and then give them the best chances of doing well after transplant? It sort of depends on what chemotherapy was already delivered, whether the patient is cured of other cancer, such as adjuvant therapy for a solid tumor or curative therapy for a lymphoma then years later they develop MDS and then AML. In that setting, we're often quite aggressive and do use CPX to treat these patients. Because thus far, that study has the best data that's a potentially curative approach.6 And I think we have to think in terms of the long run for patients, not just look at what's the easiest thing to do in the short run. In our current situation, coronavirus has certainly weighed into our decision making, but an important take-home from coronavirus is not to undertreat patients who are curable. I do think those with secondary leukemias with MDS following prior chemotherapy are still are curable if they get aggressive therapy. So, we are still using CPX on those patients who are fit. There are patients for whom I'm not using CPX, if in particular they've had anthracycline. Patients who have had breast cancer adjuvant therapy or lymphoma therapy might have difficulty getting full-dose intensity. Even though CPX is a liposomal formulation of daunorubicin and cytarabine, it still has daunorubicin in it and there is still a risk of cardiotoxicities. So, in those situations, as well as in patients who are potentially less likely to go straight to transplant or we have more questions about whether they be fit enough for transplant, we're using venetoclax-based strategies, typically venetoclax and azacitidine. New data that were published since our presentation was filmed are that a comparison between venetoclax and azacitidine as compared to the hypomethylating agent alone was superior in terms of response and survival.4 That was just presented at the EHA 2020 meeting published this summer. Our typical approach is to combine venetoclax and azacitidine for patients who are not going to get CPX.

What do we do with patients who are suboptimally responding to low-intensity therapy such as a patient who receives venetoclax and a hypomethylating agent and still has persistent blast at day 28?

Alexander E. Perl, MD: I think this is true whether they're receiving venetoclax and hypomethylating agent or they're receiving a more intensive therapy, which is if you think what you're doing has efficacy, you should keep giving it until you think you've received the adequate amount of therapy to say it's working or it's not working. And for induction chemotherapy, that's two cycles of intensive therapy. We don't declare patients to have failed induction chemotherapy—really, it’s the therapy that has failed them, it's not the patient's fault. But we would not say that therapy was ineffective until they receive two induction cycles, assuming they were fit enough to receive a second cycle. Now, the timing on that cycle is often debated. Should patients get it right at day 14? Should they wait until count recovery? That's something that I think is variable and I think we do have a lot of data on what the best approach is here, and there's more than one way to skin a cat. I know that in Europe and most of the world, the second induction cycle is done after some degree of count recovery, generally after day 21 and typically on day 28. In the US, more typically, we've given a second induction cycle right after the bone marrow comes back after day 14, but I don't think we have a good comparison of one versus the other to say which is the right way to treat patients. This is also true for venetoclax-based regimens. If you look at the number of cycles needed to get patients into remission, while many patients enter complete remission or really cleared they marrow blasts by day 28, it was by no means all patients, and while 80% of patients who are destined to go into remission ultimately went into remission after the second cycle, 20% of patients needed more than two cycles, and only 40% to 50% of patients had achieved that degree of response after the first cycle. So, it's not unusual that you have to give a second cycle of venetoclax and azacitidine to get a patient to respond. I would not stop if you saw bone marrow reductions in blasts, but not complete elimination. This is not, however, azacitidine alone where we treat patients for six months before we change gears. I really do think you need that degree of treatment with a single-agent hypomethylating agent to say, “I have given this patient adequate therapy, if it were going to work, it would work by now,” and if it hasn't worked, we should change to something different so long as the patient is tolerating therapy and not really falling by the wayside while receiving it in a way that could be benefited by alternative therapy. I think if you've chosen a therapy, you need to give it enough time to work, and generally, I will give 2 cycles of venetoclax and azacitidine before saying it's working or it's not working. It's a rare patient that I want to give a third cycle to. If they really didn't have good treatment alternatives, I might, but only if we're really starting to see that they're basically responding but they have a few residual blasts. And in that setting, it may be that they just don't recover good counts after that therapy. They probably have some degree of underlying MDS that we just can't make go away that fast. But I don't like to stop therapy until I really think we've given it enough time.


  1. Stone RM, Mandrekar SJ, Sanford BL, et al. Midostaurin plus chemotherapy for acute myeloid leukemia with FLT3 mutation. N Engl J Med. 2017;377(5):454-464.
  2. National Institutes of Health, US National Library of Medicine, Web site. A study of ASP2215 (gilteritinib), administered as maintenance therapy following induction/consolidation therapy for subjects with FMS-like tyrosine kinase 3 (FLT3/ITD) acute myeloid leukemia (AML) in first complete remission. Available at: Accessed September 10, 2020.
  3. Burchert A, Bug G, Fritz LV, et al. Sorafenib maintenance after allogeneic hematopoietic stem cell transplantation for acute myeloid leukemia with FLT3-internal tandem duplication mutation (SORMAIN). J Clin Oncol. 2020;38(26):2993-3002.
  4. DiNardo CD, Jonas BA, Pullakat V, et al. Azacitidine and venetoclax in previously untreated acute myeloid leukemia. N Engl J Med. 2020;383(7):617-629.
  5. Lancet JE, Uy GL, Cotes JE, et al. CPX-351 (cytarabine and daunorubicin) liposome for injection versus conventional cytarabine plus daunorubicin in older patients with newly diagnosed secondary acute myeloid leukemia. J Clin Oncol. 2018;36(26):2684-2692.
  6. Lancet JE, Uy GL, Newell LF, et al. Five-year final results of a phase III study of CPX-351 versus 7+3 in older adults with newly diagnosed high-risk/secondary AML. J Clin Oncol. 2020;38(suppl):Abstract 7510.

Last modified: October 12, 2020