Check(point) or checkmate for acute myeloid leukemia?

In this edition of Haematologica, Stroopinsky and colleagues report on the application of a personalized vaccine derived from fusing leukemia cells to autologous dendritic cells (fusion vaccine) as a potential tool to overcome checkpoint blockade in acute myeloid leukemia (AML).¹ AML is an aggressive hematologic malignancy with poor long-term outcomes despite recent treatment advances. The only potential curative therapeutic option for intermediate- to high-risk AML is allogeneic stem cell transplantation through the induction of a graft-versus-tumor effect, demonstrating the importance of cell-based immunotherapy.²

Antibodies that block the programmed cell death protein 1 (PD-1) or programmed death ligand-1 (PD-L1) inhibitory pathway have led to improvements in progression-free survival and overall survival in several solid tumors and Hodgkin lymphoma, encouraging multiple clinical trials in other hematologic malignancies including AML.³ Daver and colleagues reported the overexpression of clinically targetable checkpoint inhibitor receptors, PD-1 and OX40, in the bone marrow of patients with AML, making checkpoint inhibition an interesting therapeutic option to study further. However to date, checkpoint inhibitor therapy for AML has yielded disappointing results.^4,5

In an immunocompetent murine AML model using the TIB-49 murine AML cell line genetically altered to express luciferase and mCherry, Stroopinsky and colleagues confirmed that treatment with anti-PD1, anti-TIM3 or antirepulsive guidance molecule b (RGMb) antibodies as single agents had little therapeutic efficacy compared to isotope controls. To overcome this resistance to checkpoint inhibition, these investigators tested a combinatorial approach whereby a fusion vaccine was given in combination with checkpoint inhibitors. Personalized vaccines derived from patient-derived AML cells fused with autologous dendritic cells have been previously tested in 17 patients who achieved complete remission after chemotherapy.⁶ Rosenblatt and colleagues observed that vaccination induced an increase in circulating T cells recognizing leukemia-specific antigens that persisted for more than 6 months with 12 of 17 patients remaining alive without recurrence at a median follow-up of 57 months. This study demonstrated that personalized vaccination of AML patients could induce expansion of leukemia-specific T cells which may have the potential to protect against leukemia relapse.

In the study reported in this issue of the Journal, Stroopinsky and colleagues hypothesized that the combination of a personalized fusion vaccine and checkpoint inhibitor therapy could elicit a unique synergistic response whereby vaccination would induce leukemia-specific T-cell populations while checkpoint inhibition would enhance the function and persistence of these antileukemic T cells.¹ Using an immunocompetent murine AML model, cohorts of mice were vaccinated 24 h after being inoculated with murine leukemia cells followed by treatment with immune checkpoint inhibition every 3 days for a total of six doses. Rapid AML progression occurred by day 29 in all the control mice, which required euthanasia. Mice treated with checkpoint inhibition alone showed a modest improvement in survival compared to the control cohort, but all required euthanasia by day 44. Two of five mice treated with the personalized vaccine alone remained leukemia free at day 90 of leukemia inoculation. Remarkably all the mice treated with vaccination and checkpoint blockade remained alive and leukemia free at day 90 after leukemia inoculation. Stroopinsky and colleagues showed that mice treated with the personalized vaccine alone had variable expansion of tumor reactive T cells, but mice treated with the combination of personalized fusion vaccine and checkpoint blockade demonstrated robust expansion of circulating tumor-specific CD8⁺ T cells. The enhanced expansion of tumor-specific T cells following vaccination and checkpoint blockade was confirmed in the splenocytes of mice euthanized 17 days after leukemia inoculation, which showed that the combination of vaccine with checkpoint inhibition resulted in induction of tumor-specific immunity with prevention of leukemia engraftment. The investigators further found that combination treatment with the fusion vaccine and checkpoint inhibition induced a T-cell memory response and increased clonal diversity along with a statistically significant decrease in CD4⁺/CD25⁺ FOXP3⁺ T regulatory cells compared to treatment with either the fusion vaccine or checkpoint inhibition alone. Stroopinsky and colleagues also demonstrated that the combination approach provided long-term protection from leukemia relapse even after re-challenge via retro-orbital inoculation of a lethal dose of leukemia at day 90 after treatment.

Important questions remain, such as whether these impressive pre-clinical results can be replicated in a broader range of murine leukemias. It is also unclear whether this combination approach, which utilizes a fusion vaccine created with a “snapshot” of the leukemia at diagnosis, can elicit effective and long-term immune responses against the genetically complex and clonally heterogeneous leukemic populations that characterize AML in humans. Relapses of AML are often due to the emergence of treatment-resistant clones that may be undetectable at diagnosis and may not be sufficiently immunogenic in this model. Nonetheless the study reported in this issue of the Journal provides a strong scientific foundation for a clinical trial of combination therapy using personalized fusion vaccines and checkpoint inhibition. Results are eagerly awaited to determine whether this novel approach can finally check(point) or even checkmate AML.

Footnotes

Correspondence

AARON P. RAPOPORT - arapoport@umm.edu

Disclosures

No conflicts of interest to disclose.

Contributions

SN and APR co-wrote the manuscript.