Myeloid Malignancies: The Next Frontier for Immunotherapy

Building on the success of chimeric antigen receptor (CAR) T-cell therapies in lymphocytic leukemia and certain lymphomas – and the advances seen with checkpoint inhibitors in solid tumors – researchers are aggressively pursuing new immunotherapy agents for myeloid malignancies.

Several knotty challenges accompany the development of immunotherapies for myeloid malignancies, however, including difficulties with tumor-specific antigen targeting and a hostile marrow microenvironment. Without a suitable target antigen, for example, it is more likely that immune approaches will result in off-target, off-tumor toxicity and reduced anti-leukemic effectiveness.

“We are still searching for the ideal acute myeloid leukemia (AML) antigen, and there may not be one,” Naval Daver, MD, from the University of Texas MD Anderson Cancer Center in Houston, told ASH Clinical News.

“In acute lymphocytic leukemia (ALL), the antigens CD19 and CD22 are more or less restricted to the ALL blast population,” he explained. “But in AML, antigens such as CD33 and CD123 are expressed on the majority of myeloid blasts and leukemic stem cells (LSC) and also are found on normal hematopoietic stem cells (HSCs), progenitor cells, and monocytes and macrophages.”

When clinicians target these antigens on AML cells, they also run the risk of killing normal stem and myeloid progenitor cells, “which can result in prolonged aplasia, profound cytopenia, and other problems,” Dr. Daver continued. “In addition, the treatment may stimulate macrophages and monocytes, resulting in early cytokine release syndrome (CRS), even before it is able to kill the bulk of the AML cells.”

How far down the pipeline are the various immunotherapeutic approaches for myeloid malignancies? And what challenges will investigators face as they try to bring these therapies into routine clinical use? ASH Clinical News spoke with hematologists and immunotherapy trialists for answers.

Tremendous Need for New Approaches

Myeloid malignancies are a heterogeneous group of clonal disorders, characterized by excessive proliferation, abnormal self-renewal, and/or differentiation defects of hematopoietic cells and myeloid progenitor cells. They include myeloproliferative neoplasms (MPNs), myelodysplastic syndromes (MDS), and AML.

AML is the most common acute leukemia among U.S. adults and has the highest mortality rate of the major forms of leukemia, accounting for about 1.3 percent of all new cancer diagnoses and 1.8 percent of cancer deaths each year.1 Standard treatments for AML have changed little since the 1970s: induction chemotherapy to reduce the number of blasts (most often using the standard 7+3 regimen of cytarabine and anthracycline), followed by consolidation chemotherapy to destroy residual leukemic cells. The cure rate for primary AML is about 35 percent, and that rate decreases with age.

Another high-grade malignancy that shares some features with AML, blastic plasmacytoid dendritic cell neoplasm (BPDCN), is both rare and unusually aggressive. It mostly affects men (75-90% of patients) and often presents initially as skin lesions. BPDCN is likely under-recognized and can be misdiagnosed as cutaneous lymphoma. It accounts for less than 1 percent of all hematologic malignancies.

“There is an urgent unmet need for relapsed/refractory AML and BPDCN,” explained Elizabeth Budde, MD, PhD, from the City of Hope National Medical Center in Duarte, California. “For patients with AML who have a recurrence after allogeneic hematopoietic cell transplantation (alloHCT), the prognosis is very poor. For those with BPDCN, there are no approved therapies and really no accepted standard of care.” Median overall survival (OS) for these patients is only 12 to 14 months without HCT, she noted. Dr. Budde’s group is testing CAR T-cell therapy candidates in both AML and BPDCN.

CAR Ts Go Target Hunting

CAR T-cell therapies use harvested T cells expanded and re-engineered to express a recombinant receptor that targets a tumor-specific protein. The retrained T-cells are infused back into the patient’s body to proliferate and eradicate cancer cells.

In August 2017, tisagenlecleucel became the first CAR T-cell therapy approved by the U.S. Food and Drug Administration (FDA) and is indicated for the treatment of children and young adults with B-cell precursor ALL that is refractory or in second or later relapse. The second CAR T-cell product to market was axicabtagene ciloleucel. In October 2017, axicabtagene ciloleucel was approved for the treatment of adult patients with relapsed or refractory large B-cell lymphoma. Both revolutionary therapies target CD19-expressing cells.

“We are still in the early days of CAR T-cell therapies for myeloid leukemia, in part because we just don’t have a target that is as pristine as CD19,” Dr. Budde said. “Any of the antigens we target might also be found on normal stem cells, making myeloablation or prolonged neutropenia a major concern. We also need the correct antigen to make the CAR T cells persist in the hostile microenvironment inside the bone marrow.”

Dr. Budde’s group is testing an experimental CAR T-cell product targeting CD123, which is often overexpressed on AML blasts and LSC–enriched cell subpopulations compared with expression in normal HSCs and myeloid progenitors.2 The trial has a built-in rescue strategy requiring all patients to have an identified donor or stem cell source for alloHCT. The CAR T-cell treatment is expected to serve as a bridge to potentially curative alloHCT.

“So far, we have treated six patients who had refractory AML following alloHCT,” Dr. Budde reported. “They have all tolerated the treatment well, and we have not seen any treatment-related myeloablation.” The researchers also noted that this CAR T-cell product produced “promising anti-leukemic activity.”

In the BPDCN cohort, they’ve treated two patients. The first is in complete remission more than 60 days post-infusion. “There are only about 60 patients diagnosed each year with BPDCN, so we are hoping to have more of those patients referred to City of Hope for CAR T-cell therapy,” said Dr. Budde.

“We have yet to see any severe graft-versus-host disease, neurologic toxicity, or dose-limiting toxicity [associated with the CD123-targeting approach], but we started with a very conservative dose so as we escalate, we have to be careful,” she said.

Researchers at City of Hope have also observed = CRS in their trial patients – 4 with grade 1 CRS, and 1 with grade 2 CRS.  But Dr. Budde feels confident that her dedicated CAR T-cell team can identify and manage CRS successfully.

Among the dozens of ongoing CAR T-cell trials worldwide, only a handful are for AML, MDS, or other myeloid malignancies. CD33 is considered the most promising myeloid-associated target, but there are CAR T-cell products in development targeting other antigens or antigen combinations that have shown promising preclinical results, including CD123. The possibility of combining CAR T cells with immune checkpoint inhibitors to enhance anti-leukemic effects also is under investigation.3

Assembly Line CARs

Just as the Ford Model T is generally regarded as the first efficiently produced, affordable, and reliable automobile that opened travel to the masses, an off-the-shelf, allogeneic CAR T-cell product could open immunotherapy to much larger populations.

“The biggest roadblock today is related to the production of the ‘one-off’ CAR T-cell products,” Dr. Daver explained. “If we can get an off-the-shelf product that works and has a similar safety profile to the custom-designed cells, we could stockpile 15 or 20 doses – or even more in large institutions. When a suitable patient comes in, we could just admit him or her and administer the therapy, like we do with antibody drug conjugates or immune checkpoints.”

It remains to be seen whether the “universal” products being tested now are as safe and effective as autologous CAR products, he noted. Cellectis, a French biotechnology company, is developing allogeneic CAR T-cell products, including UCART123 for AML and BPDCN and UCART19 for ALL, and was the first sponsor to open a clinical trial of allogeneic T-cells in late 2017. However, while the UCART19 ALL product is showing clinical activity, the UCART123 product ran into some early trouble.

In September 2017, the FDA placed a clinical hold on phase I studies of UCART123 in patients with BPDCN and AML following a patient death related to CRS and lung infection. In November, protocol adjustments allowed the trials to be restarted with a lower dose. Since then, additional patients have been safely dosed and updates are awaited.3

How to Harness T Cells

Beyond CAR products, there are a few other immunotherapies that direct the body’s endogenous T cells to fight against tumors. According to Dr. Daver, two promising approaches being heavily pursued are bispecific antibodies and immune checkpoint inhibitors.

Bispecific antibodies are monoclonal antibodies engineered to have dual receptors: one binds to a T-cell’s CD3 receptor and the other targets the malignant cells. A number of AML-associated antigens are under investigation, Dr. Daver noted, including CD33, CD123, and CLL1.

Immune checkpoint receptor blockade is also being studied in AML and other myeloid malignancies, though results have been disappointing thus far. Immune checkpoint inhibition – or removing the inhibitory signals on the T cells that the cancer has produced – is widely used in solid tumor oncology with multiple agents having received FDA approval, and is more recently finding its way into hematology. Investigators have reported encouraging results with anti–programmed death 1 (PD-1) antibodies in classical Hodgkin lymphoma and anti-PD-1 in combination with other agents in non-Hodgkin lymphoma and multiple myeloma.4-6

Checkpoint inhibition as a monotherapy for AML and MDS has unfortunately shown limited clinical efficacy. To boost effectiveness, investigators are looking at combining checkpoint inhibitors with hypomethylating agents (HMAs) such as azacitidine and decitabine.

“We know azacitidine and decitabine have their own anti-tumor activity, but we also know that they modulate the immune system by multiple mechanisms, including upregulation of inhibitory immune checkpoint molecule expression, specifically PD1/PD-L1, which are targets for the immune checkpoint inhibitors,” Dr. Daver explained.7 Several trials at his center and elsewhere are evaluating the efficacy of HMAs combined with either CTLA-4, PD-1, or PD-L1 blocking antibodies, including ipilimumab, nivolumab, pembrolizumab, and durvalumab.

Reports presented at recent medical meetings suggest that toxicities associated with these agents are manageable for most patients. For example, the combination of azacitidine and nivolumab produced what trial investigators felt was an “encouraging” response rate and survival in patients with relapsed AML and poor-risk features, compared with historic survival rates in this population.8

“Looking at this combination in patients with newly diagnosed or relapsed MDS, we’re seeing very similar data,” Dr. Daver said, noting that approximately 75 percent of patients in the frontline setting responded to this treatment regimen.

“With the immune checkpoint inhibitors, we see less toxicity than with the CAR T-cell products,” he added. “And the good thing is that patients usually respond quickly to steroids and do not require the use of more potent immunosuppressive agents.”

More experience with these agents will continue to improve their clinical use, Dr. Daver said. “Over time, we’ve become adept at identifying the toxicities and treating them early with steroids, and most patients can continue to receive the PD-1 or CTLA-4 inhibitor. Awareness, early recognition, and rapid treatment are very important with immunotherapies in general.”

Naked No Longer: Directed ADCs

T-cell harnessing works best in patients who have an adequate population of “manipulable” T cells, such as frontline and early salvage patients, Dr. Daver noted.

Other options under investigation are the directed antibody-drug conjugates (ADCs), which consist of a monoclonal antibody chemically linked to a drug. This class of agents combines the cytotoxic activity of chemotherapy drugs with the selectivity of targeted monoclonal antibodies. In AML, the most commonly used targets for ADCs are the myeloid surface antigens CD33 and CD123.

“ADCs are not dependent on a patient’s own T-cell population to fight against a tumor, but they act more like cytotoxic therapies and target a particular antigen,” explained Dr. Daver. “These monoclonal antibodies are linked to a toxic payload, either bacterial or chemical, and once the antibodies attach to the AML antigen, they are basically internalized into the AML cells and the toxin is released, resulting in cell death.”

Many of the first-generation monoclonal antibodies evaluated in AML were “naked,” in that they relied on antibody-dependent, cell-mediated cytotoxicity, as rituximab and other anti-CD20 antibodies do successfully in lymphomas.  However, in AML, naked antibodies demonstrated limited antileukemic activity. Now, researchers are focusing on developing conjugated antibodies engineered to deliver a bacterial, viral, or chemical payload to leukemic blasts and/or LSCs.

Last year saw the re-approval of the CD33-targeted ADC gemtuzumab ozogamicin (GO) for frontline and relapsed AML. GO was initially approved by the FDA in 2000, but was voluntarily withdrawn from market in June 2010 when a phase III trial of GO plus chemotherapy failed to show benefit over chemotherapy alone. However, after several European phase III trials showed improved event-free survival, and in some studies, improved OS when GO was added to chemotherapy, the drug was approved in September 2017 for newly diagnosed or relapsed/refractory CD33-positive AML and in pediatric patients at least 2 years of age.9

“GO is already being widely used in combination with induction chemotherapy for AML, or as a single agent in relapsed AML, but there are newer ADCs in development that could offer additional options,” Dr. Daver said. “In the next few years, we may have 20 to 40 percent of AML patients being treated with an ADC added to induction therapy or in the relapsed setting.”

Overall, the future looks bright for immunotherapy for myeloid malignancies. Besides GO, Dr. Daver also thinks the immune checkpoint inhibitors will play a vital role in AML treatment, as will CAR T-cell therapies – particularly if a great allogeneic product is developed. —By Debra Beck


  1. Surveillance, Epidemiology, and End Results. “Cancer Stat Facts: Leukemia – Acute Myeloid Leukemia (AML).” Accessed March 28, 2018, from
  2. Budde L, Song JY, Kim Y, et al. Remissions of acute myeloid leukemia and blastic plasmacytoid dendritic cell neoplasm following treatment with CD123-specific CAR T cells: a first-in-human clinical trial. Abstract #811. Presented at the 2017 ASH Annual Meeting, December 11, 2017; Atlanta, GA.
  3. “FDA Lifts Clinical Hold on Cellectis Phase 1 Clinical Trials with UCART123 in AML and BPDCN.” Accessed April 3, 2018, from
  4. Daver N, Basu S, Garcia-Manero G, et al. Phase Ib/II study of nivolumab in combination with azacitidine (AZA) in patients (pts) with relapsed acute myeloid leukemia (AML). Abstract #763. Presented at the 2016 ASH Annual Meeting, December 6, 2016; San Diego, CA.
  5. Merryman RW, Armand P, Wright KT, Rodig SJ. Checkpoint blockade in Hodgkin and non-Hodgkin lymphoma. Blood Adv. 2017 November 10.
  6. Rosenblatt J, Avigan D. Targeting the PD-1/PD-L1 axis in multiple myeloma: a dream or a reality. Blood. 2017;129:275-9.
  7. Daver N, Boddu P, Garcia-Manero G, et al. Hypomethylating agents in combination with immune checkpoint inhibitors in acute myeloid leukemia and myelodysplastic syndromes. Leukemia. 2018 February 22. [Epub ahead of print]
  8. Daver N, Garcia-Manero G, Basu S, et al. Nivolumab (nivo) with azacytidine (AZA) in patients (pts) with relapsed acute myeloid leukemia (AML) or frontline elderly AML. Abstract #1345. Presented at the 2017 ASH Annual Meeting, December 9, 2017; Atlanta, GA.
  9. S. Food and Drug Administration. “FDA approves Mylotarg for treatment of acute myeloid leukemia.” Accessed April 4, 2018, from