Back Back
CD47 Therapy in MDS and AML
Experts Address Questions on Therapeutic Targeting of CD47 in MDS and AML

Released: July 19, 2023

Activity Information

Activity

Progress
1
Course Completed
Continue
Key Takeaways
  • Anemia is the primary adverse event associated with magrolimab treatment because of CD47 expression on red blood cells.
  • Consistent dosing of magrolimab every 2-2.5 weeks is key to improving response rates.
  • Magrolimab may interfere with blood typing, but additional studies are needed in this area.

In this commentary, Michael Heuser, MD, facilitates a discussion with Naval G. Daver, MD, and Amer Zeidan, MBBS, who address questions on therapeutic targeting of CD47 in myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML) submitted by a healthcare professional (HCP) audience during a live symposium at the European Hematology Association 2023 Annual Congress. The questions focus on data on magrolimab, an anti-CD47 monoclonal antibody that currently is being examined in combination with azacitidine and other agents for the treatment of high-risk MDS or AML.

Is CD47 expression different on TP53-mutated AML blasts compared with TP53 wild-type blasts?

Naval G. Daver, MD:
Thus far, based on analysis from our center and others, we have not found a difference in CD47 expression or in expression of other prophagocytic markers, including CAL-R, SLAM-7, and phosphatidylserine.

In the presentation, you noted that anemia is the predominant adverse event associated with magrolimab and occurs within the first 10 days of treatment. What have you observed in terms of tolerability or adverse events after the first 10 days of treatment?

Naval G. Daver, MD:
Anemia appears to be the predominant adverse event, as magrolimab binds to CD47 expressed on red blood cells (RBCs). Daily labs are needed during those first 10 days of therapy. Trials are ongoing to further examine the safety of magrolimab. Once the initial 7-10 days of anemia is managed, the cycles run quite smoothly, and rarely do we see delays in treatment. We have not seen prolonged neutropenia or thrombocytopenia, or long-term gastrointestinal or liver function issues.

Because macrophage activation can induce hemophagocytosis, do you find that patients receiving CD47-targeted therapy who have a rapid, severe decrease in hemoglobin also have a drop in platelets or an increase in ferritin levels? Would you manage ferritin daily and potentially give the patient dexamethasone or a similar agent if you think it is a type of hemophagocytic syndrome?

Naval G. Daver, MD:
I do not consider this reaction to be a hemolytic anemia because hemolysis markers, such as haptoglobin, reticulocytes, and lactate dehydrogenase are not increased like a typical autoimmune hemolytic anemia. What we think is happening is that these are tagged RBCs coated with the CD47 antibody that are being removed by the macrophage sink in the reticulo-endothelial system, such as the spleen. We do not quite understand why the macrophage activation by magrolimab does not evolve or spill over into a hemophagocytic lymphohistiocytosis macrophage activation syndrome. This seems to be different from what we have seen with immune checkpoints that activate T-cells and, when they are overactivated, may result in the various “-itis” adverse events, such as pneumonitis and other immune-related adverse events. We have seen a few cases treated with magrolimab in which we suspected pneumonitis, but after extensive work-up these were eventually thought to be most likely occult/viral pneumonia. We have not seen true macrophage activation syndrome or hemophagocytic lymphohistiocytosis with magrolimab in my experience. Initially, we measured ferritin levels, but they did not seem to increase with treatment. The anemia occurring in patients receiving magrolimab seems to be separate from the degree of macrophage overactivation.

Does magrolimab interfere with blood bank typing in a way similar to daratumumab?

Naval G. Daver, MD:
We do not know whether magrolimab interferes with blood bank typing, but this is something we plan to examine. Preclinical and monotherapy trial data found coating of the RBCs with CD47 antibody. Reports from blood bank experts indicated that they were not able to determine a full extended type and crossmatch with magrolimab on board, but major ABO types still could be identified. This led to RBC genotyping at screening prior to the start of magrolimab, similar to what we do for daratumumab treatment. During the COVID-19 pandemic, occasionally we could not find fully genotype-matched blood for some patients, so we transfused them with blood matched for the most common minor ABO types, and these patients have done fairly well.

Michael Heuser, MD:
But is it obvious to a transfusion medicine specialist that patients have received a CD47 antibody?

Naval G. Daver, MD:
Yes. Currently, trial protocols state that RBC genotyping should be performed, and we should try to obtain genotype information as frequently as possible, as recommended in the protocols. I think there may be scenarios in the future as we learn more where this extended genotype crossmatch may not be absolutely needed but we do not have enough data or knowledge on this yet.

Is it problematic for a patient to receive a transfusion before receiving magrolimab, considering that would add more potential targets via RBCs expressing CD47 and thus a potential loss of efficacy?

Amer Zeidan, MBBS:
I would not be very concerned about this. The concept of RBC pruning where RBCs expressing CD47 are removed is quite interesting. From a safety perspective, transfusions are important to help manage short-term anemia during the initial 10 days of magrolimab therapy. If HCPs must interrupt magrolimab treatment, then they have to repeat priming doses, and each time therapy restarts, anemia must be monitored and managed because although it is generally mild, there are certainly possible outliers, particularly among older patients.

Naval G. Daver, MD:
As Dr Zeidan noted, breaks in treatment longer than 4 or 5 weeks require repeating priming, which is difficult and ideally should be avoided. Another thing to note regarding magrolimab is that it is important to perform frequent dosing. Although a 3- or 4-week treatment regimen might be convenient, we observed response rates drop using a 4-week regimen during the phase I dose optimization cohorts, and a high receptor saturation dropped off by the third week. I think it is important in the initial phase to keep close to a magrolimab dosing regimen of 2 or 2.5 weeks. It is important to decouple and continue magrolimab treatment, even if there are delays in initiating therapy with azacitidine or azacitidine/venetoclax.

Are transfusions from recently collected blood superior to transfusions using older blood?

Naval G. Daver, MD:
We know that older RBCs express more CD47 than younger ones. We have not been able to formally study this question because of low availability of blood, particularly during the COVID-19 pandemic, but in our experience transfusions from older blood work without any major concerns.

Is there any benefit to combining immune checkpoint inhibitors with magrolimab?

Amer Zeidan, MBBS:
More clarity on the use of magrolimab in MDS is needed, but I think it would make sense, as they use different pathways. That said, immune checkpoint inhibitor development in treatment of myeloid malignancies is very challenging for multiple reasons. With solid tumors, if you suspect a patient has pneumonitis, you could obtain a biopsy of the lung and confirm pneumonitis. In AML and MDS, there may not be available platelets, and you cannot obtain a biopsy, so it is not as straightforward. This is one of the biggest challenges that we have had. For patients who are heavily immunosuppressed, HCPs should be cautious before using combination treatment with immune checkpoint inhibitors. Hopefully, once we have more than 1 effective agent, we may have additional options in combination therapy.

As triplet and doublet regimens with magrolimab are being examined, do you think criteria will be identified to suggest that a patient might benefit from just doublet therapy vs triplet?

Naval G. Daver, MD:
For patients with TP53 mutations, my sense is that azacitidine/magrolimab combination therapy may be sufficient and that the addition of venetoclax may add limited efficacy, but we need to see the data from the ENHANCE-3 randomized study to make more definitive conclusions. In patients without a TP53 mutation, the combination of azacitidine/venetoclax has demonstrated good response and tolerability, so that has been the backbone of therapy that we are trying to build and hopefully improve on. Of the many triplet combinations we have studied, azacitidine/venetoclax/magrolimab appears to have the least cumulative myelosuppression, neutropenia, and thrombocytopenia.

Your Thoughts?
Do you consider clinical trial enrollment on trials of CD47-targeted therapies for your patients with MDS or AML? Answer the polling question and join the conversation by adding a comment in the discussion section.

Continue

Poll

1.

How often do you discuss participation in clinical trials examining novel treatments such as CD47-targeted therapy with your patients with MDS or AML?

Submit