We have one of the longest histories of “immunotherapy use” in treating patients with MDS and AML, in the form of allogeneic stem cell transplantation (a cellular immunotherapy), which has long represented a potentially curative option for eligible patients. The graft-vs-leukemia response of allogeneic transplant is thought to be, in part, responsible for maintaining disease control, allowing for long-term survival in a subset of patients. How to capture these benefits in other contexts is a great area of investigation.

The first-generation immune checkpoint inhibitors (ICIs) target CTLA-4, PD-1, or PD-L1, in general. These have recently been studied in MDS and AML. The CTLA-4 inhibitor ipilimumab has shown some activity as monotherapy, but it remains a challenge to know exactly which patients will respond. Interestingly, some of those who have responded have had fairly limited disease (eg, leukemia cutis), and we are still exploring what it is about the immune system that permits this type of isolated relapse and then response recovery with CTLA-4 inhibition.

Studies have been undertaken to take ICIs targeting PD-1, PD-L1, and CTLA-4 and combine them with other agents to improve response quality and duration. The use of hypomethylating agents such as azacitidine or decitabine may induce expression of motifs on tumor cells that the immune system can recognize, so there is a rationale for administering azacitidine, for instance, prior to administering an ICI (ie, to enhance immune-mediated disease control). However, treatment-limiting toxicities and limited improvements in efficacy to date using the “canonical” ICIs have led to investigational interest in other immune therapies, such as magrolimab and sabatolimab.

Magrolimab binds to CD47 and enhances the phagocytosis of cancer cells by macrophages. An interesting observation with magrolimab is the relatively high response rates that are reported among patients with TP53-mutated disease. These studies involve a small number of such patients, but many have achieved remissions, or at least blast reductions. There are some unique toxicities with this agent that need to be monitored. Notably, CD47 is expressed on red blood cells, and patients may have a hemolytic process when the first doses are administered, requiring transfusions to keep up with the removal of older red blood cells.

Sabatolimab targets TIM-3, a marker of T-cell exhaustion, which is one of the mechanisms of immune escape in cancer. TIM-3 is expressed on T cells, natural killer cells, and leukemic progenitors, and blocking TIM-3 may enhance antitumor immunity. In contrast to PD-1 or CTLA-4, sabatolimab seems to have a different immune phenotype, with fewer instances of pneumonitis or colitis linked to its use. The toxicity profile seems to allow this combination to be administered to older patients or to patients with more comorbidities in whom more intensive approaches may be limited.

Both magrolimab and sabatolimab are in phase 3 trials that are currently enrolling, and, hopefully, we will be able to report out in the coming years.