Radiotherapy of extramedullary acute myeloid leukemias overcomes resistance to checkpoint blockade that is enhanced by natural killer cell depletion

Abstract Introduction: Immune checkpoint inhibitors have been relatively ineffective in acute myeloid leukemias (AML), likely due to immune tolerance induced by the systemic disease as well as the chronic immune suppression after cytotoxic therapies. Radiotherapy of solid tumors has been shown to induce anti-tumor immunity and facilitate tumor regression. Since extramedullary leukemias may facilitate antigen presentation in the draining lymph nodes, similar to solid tumors, we hypothesized that local irradiation of cutaneous, extramedullary leukemias will stimulate anti-leukemia T cell responses to increase survival with checkpoint inhibitors. Methods: All animal experiments were conducted in accordance with the Declaration of Helsinki and complied with institutional IACUC guidelines. C57BL/6 mice were injected both subcutaneously and intravenously with the syngeneic murine AML cell line C1498 to model extramedullary lesions and systemic leukemias, respectively. Subcutaneous tumors were irradiated with a single dose of 8 Gy using the Precision X-ray XRAD SmART system. Mice were treated with 200 ug of anti-PDL1 (clone 10F.9G2) or anti-CTLA4 (clone UC10-4F10-11) at the start of radiation and every four days for 12 days. To deplete CD8+ T cells, CD4+ T cells, or NK cells, mice were treated with anti-CD8 (clone YTS-169), anti-CD4 (clone GK1.5), or anti-NK1.1 (clone PK136) neutralizing antibodies i.p. at the start of irradiation and every 3 days. To inhibit lymphocyte egress from tumor-draining lymph nodes, mice were treated with 25 μg of FTY720 i.p. every 2 days for 14 days. Two-way ANOVA and Student's t-test were used for paired-wise comparisons. The Kaplan-Meier plot estimated survival, and groups were compared using the log-rank test. Results: In mice bearing systemic C1498 leukemia with irradiated subcutaneous tumors, treatment with anti-PDL1 but not anti-CTLA4 significantly increased survival of mice after irradiation of extramedullary leukemias (median survival: not reached vs. 25.5 days; p=0.0025). By contrast, irradiation of subcutaneous tumors with 8 Gy alone did not increase survival compared to non-irradiated mice (median survival: 24 vs. 23 days; p=0.30). Mice treated with immune checkpoint inhibition and tumor irradiation had significantly fewer detectable circulating leukemia cells (P <0.01). Depletion of CD8+ T cells but not CD4+ T cells reduced the survival of mice treated with radiotherapy and anti-PDL1. By contrast, depletion of NK cells increased the survival of mice treated with anti-PDL1 and irradiation. Improved survival required the combination of irradiation of subcutaneous leukemia, checkpoint blockade, and NK cell depletion because treatment of mice with anti-PDL1 alone or anti-PDL1 and anti-NK1.1 without irradiation did not increase survival compared to control mice receiving isotype antibodies. Improved survival depended on inhibition of lymphocyte egress, as mice treated with FTY720 had worse survival compared to control-treated mice. Conclusions: Irradiated extramedullary AML improved survival and cleared systemic leukemia in mice treated with checkpoint blockade. Depletion of NK cells further enhanced survival, indicating that NK cells regulated anti-leukemia immune responses induced by radiation and anti-PDL1. These results suggest that focal radiotherapy and immune checkpoint inhibition may enhance anti-leukemia immune responses in AML patients.