Promising Safety and Efficacy Results from an Ongoing Phase 1b Study of Pembrolizumab Combined with Decitabine in Patients with Relapsed/Refractory (R/R) Acute Myeloid Leukemia (AML)

BACKGROUND: For AML patients (pts) deemed ineligible for intensive cytotoxic chemotherapy, hypomethylating agents (HMA) have been established as standard treatment. Decitabine 20 mg/m2 on days 1-10 per 28 day cycle has demonstrated single agent efficacy with 15% complete remission (CR)/CR with incomplete hematologic recovery (CRi) in R/R AML (Khan et al. Leuk Lymphoma 2017). In a phase 2 trial, 10-day decitabine with venetoclax demonstrated overall response rate (ORR) of 62% in R/R AML (Dinardo C et al. Lancet Haematol 2020). However, aberrant upregulation of PDL1 and PD-L2 has been observed in 25% and 33% of CD34+ samples from AML pts respectively (Plass C et al. Semin Oncol 2008). Pts resistant to epigenetic therapy had relative higher increments in gene expression of PD-L1, PD-L2 and CTLA4 compared with pts who achieved response, suggesting that PD-1 signaling may be involved in resistance mechanisms to HMA. In a prior phase 2 trial evaluating pembrolizumab and azacitidine in R/R AML, responses among 29 evaluable pts demonstrated ORR 17.2% (Gojo I et al. Blood 2019). In this phase 1b trial, we set out to test the combination of pembrolizumab and decitabine (NCT03969446).

METHODS: Pts [≥18 years old (yo); ECOG ≤1] with R/R AML not eligible for allogeneic hematopoietic cell transplantation (alloHCT), received decitabine 20 mg/m2 on days 1-10. If in CR/CRi, decitabine was reduced to days 1-5 of cycles 3+. Pembrolizumab was given on days 1 and 22 in all cycles. Cycle length was 42 days. This phase 1b study has two arms, AML and myelodysplastic syndromes (MDS) with primary objectives of safety, tolerability, recommended phase 2 dose (RP2D) and to obtain preliminary estimates of CR/CRi rate using the “3+3” design. Here, we provide interim safety and efficacy data for the first 6 pts on the AML arm.

RESULTS: As of July 1, 2022, 6 pts were enrolled to the AML arm cohort 1. The six treated and evaluable pts received a median of 2 (range: 1-7) cycles of treatment. At baseline, the median age was 78 yo (range: 72-81) with median 2 (range: 1-4) prior lines of regimens, and median 35% (7-65) bone marrow blasts. All pts had previously progressed on decitabine and venetoclax. All-non-hematologic grade 3 toxicities were reversible. No dose limiting toxicity, ≥ gr4 non-hematologic toxicity, or treatment related deaths were seen. Most common Grade (Gr) 3-4 treatment-related hematologic toxicities were: anemia (100%), febrile neutropenia (83%), lymphopenia (100%), and thrombocytopenia (100%). Most common infectious toxicities included Gr 3 lung infections (50%). Three pts (50%) achieved CR/CRi, 1 pt (16.7%) achieved partial remission, and 2 pts (33%) achieved stable disease. ORR was 66.7% in this cohort. The median follow-up time is 19.8 months (4.1-22.6), with 4 pts still alive at the data cut -off. One pt was able to bridge to alloHCT.

CONCLUSIONS: The combination of pembrolizumab and decitabine in the AML arm cohort 1 of this study in pts with R/R AML was deemed safe with manageable side effect profile and encouraging anti-leukemic activity and offers potential for promising therapy for R/R AML that progress even after prior HMA therapy. Furthermore, this study demonstrates the effectiveness of the 10-day decitabine schedule. The study is now open for the MDS arm, with treatment of the same schedule of pembrolizumab in combination with decitabine 20 mg/m2 on days 1-5 of each cycle. The study has also been amended to include AML cohort 2 to add venetoclax to the decitabine/pembrolizumab regimen and pts are now being enrolled on this arm.