A phase 1b/2 study of azacitidine with PD-L1 antibody avelumab in relapsed/refractory acute myeloid leukemia

Background: Patients with relapsed/refractory (R/R) acute myeloid leukemia (AML) have limited treatment options. In preclinical models of AML, inhibition of the PD-1/PD-L1 axis demonstrated antileukemic activity. Avelumab is an anti-PD-L1 immune checkpoint inhibitor (ICI) approved in multiple solid tumors. The authors conducted a phase 1b/2 clinical trial to assess the safety and efficacy of azacitidine with avelumab in patients with R/R AML.

Methods: Patients aged ≥18 years who had R/R AML received azacitidine 75 mg/m² on days 1 through 7 and avelumab on days 1 and 14 of 28-day cycles.

Results: Nineteen patients were treated. The median age was 66 years (range, 22-83 years), 100% had European LeukemiaNet 2017 adverse-risk disease, and 63% had prior exposure to a hypomethylating agent. Avelumab was dosed at 3 mg/kg for the first 7 patients and at 10 mg/kg for the subsequent 12 patients. The most common grade ≥3 treatment-related adverse events were neutropenia and anemia in 2 patients each. Two patients experienced immune-related adverse events of grade 2 and grade 3 pneumonitis, respectively. The overall complete remission rate was 10.5%, and both were complete remission with residual thrombocytopenia. The median overall survival was 4.8 months. Bone marrow blasts were analyzed for immune-related markers by mass cytometry and demonstrated significantly higher expression of PD-L2 compared with PD-L1 both pretherapy and at all time points during therapy, with increasing PD-L2 expression on therapy.

Conclusions: Although the combination of azacitidine and avelumab was well tolerated, clinical activity was limited. High expression of PD-L2 on bone marrow blasts may be an important mechanism of resistance to anti-PD-L1 therapy in AML.

Lay summary: This report describes the results of a phase 1b/2 study of azacitidine with the anti-PD-L1 immune checkpoint inhibitor avelumab for patients with relapsed/refractory acute myeloid leukemia (AML). The clinical activity of the combination therapy was modest, with an overall response rate of 10.5%. However, mass cytometry analysis revealed significantly higher expression of PD-L2 compared with PD-L1 on AML blasts from all patients who were analyzed at all time points. These data suggest a novel potential role for PD-L2 as a means of AML immune escape.