A number of early phase studies have been conducted to assess a possible role of the established immune checkpoints (ie, PD-1, PD-L1, and CTLA-4) in MDS/AML. This has been important work, and we have learned that the benefits of traditional immune checkpoint inhibitors (ICIs) in MDS/AML, if any, are modest.

With immuno-myeloid therapy, I think that we are looking beyond those traditional ICIs that have been approved by the US Food and Drug Administration to treat so many different types of solid tumors. The need to look beyond the traditional ICIs may relate to the type of T-cell exhaustion and immune dysfunction that we commonly seen in patients with myeloid malignancies. And so, a goal with immuno-myeloid therapy is to find ways to circumvent T-cell exhaustion, and this is an important rationale behind strategies such as TIM-3 inhibition.

Natural killer cell–mediated immunotherapy is another interesting research direction. There are allogeneic “off-the-shelf” approaches to replace natural killer cells, and there have been some early trial data in AML that appear encouraging. This is yet another way of potentially providing some of the favorable outcomes of a stem cell transplantation, without the conditioning regimen and toxicity, resulting in an immune system with new functionality.

The need for new strategies is great, particularly for patients with TP53-mutated MDS/AML. We have decades' worth of data showing that patients with TP53-mutated AML do not do well with standard intensive chemotherapy. We were really hoping that azacitidine or decitabine plus venetoclax would be more effective for this group of patients, but, unfortunately, higher response rates with the combination have not translated to longer overall survival in TP53-mutated AML.

The general experience with traditional ICIs in myeloid malignancies has been one of limited responses, as Dr DiNardo outlined. Ipilimumab, which targets CTLA-4, has some activity, and there are some limited reports of leukemia cutis in the post-transplant setting that responded to traditional ICIs. However, a recent phase 2 study showed no benefit and some added toxicity when the PD-1 inhibitor durvalumab was added to azacitidine in the first-line treatment of patients with high-risk MDS. 

I think that we would all like to see advances in immunologic therapy, whether we are talking about bispecific antibodies, monoclonal antibodies, antibody-drug conjugates, or cellular therapies such as chimeric antigen receptor T-cell therapies. One of the challenges in development has been related to finding an effective intervention that does not cause excess toxicity to the healthy myeloid compartment. 

Hematopoietic stem cell transplantation is the ultimate immunotherapy, and we would like to be able to extend some of the benefits of this treatment to patients who cannot proceed to transplant. Additionally, in those who do proceed to transplant, we would like to be able to augment the donor cells in an attempt to improve outcomes and reduce relapses. These are the areas in which I believe that emerging immuno-myeloid therapies may have a role. 

TIM-3 (sabatolimab) and CD47 (magrolimab) are novel immune targets that may be able to improve upon the results with azacitidine in high-risk MDS and AML. In a patient who proceeds to allogeneic stem cell transplant, one of the potential concerns is that adding an immune-based agent could result in excess graft-versus-host disease. In our research, there was a cohort of patients who were treated with sabatolimab and proceeded to transplant, and we did not see a significant signal of excess graft-versus-host disease. Still, more experience with immunologic therapies in the transplant setting is needed. We are learning new ways to target the immune system and the dysregulated immunity of myeloid malignancies. Hopefully, this will translate to newer treatment options for patients who proceed to transplant and for those who do not.

The point that Dr Brunner raised about the immune system being dysregulated in MDS is an important one. In fact, failures of past studies in which we tried to engage the immune system in MDS have been attributed to this dysregulation. There is a clear distinction between lower- and higher-risk MDS in terms of immune function. There is also an association with autoimmunity. We analyzed a large cohort of patients with MDS to further study this association and found that autoimmune diseases, including hypothyroidism, were prevalent in MDS. Additionally, those patients with autoimmune diseases had better overall survival and less AML transformation.  

One can think of immunotherapy in so many different ways. Regarding the classical ICIs that are used in the treatment of many different solid tumors, I agree that they are not making inroads in the myeloid world. But then we also have some agents in development that we have been discussing here that work differently. For example, we have magrolimab, which is a macrophage ICI. It targets CD47, which functions as an antiphagocytic or a “do-not-eat-me” signal, enabling CD47-expressing cells to evade phagocytic elimination by macrophages and other phagocytes. Then we have sabatolimab. It targets the TIM-3 inhibitory receptor, which is involved in regulating both innate and adaptive immune responses. TIM-3 is expressed on immune cells and is also aberrantly expressed on leukemic stem cells, but not on normal stem cells. Both magrolimab and sabatolimab are being investigated in combination with azacitidine.

And, finally, there is the potential for antibodies alone, antibody-drug conjugates, bispecific antibodies, and chimeric antigen receptor T-cell therapy. All of these agents are in early phase trials, and I think that we must learn more about which groups of patients stand to benefit from these therapies.