Conditioning Impacts on Regulatory T-Cells (Tregs) in the Microenvironment of AML: Observations from the Phase 1 Cynk-001-AML-001 Trial

Introduction: Evidence shows that NK cells can exhibit potent anti-tumor activity against acute myeloid leukemia (AML). CYNK-001 is a CD56+CD3- enriched, off-the-shelf, allogeneic natural killer (NK) cell therapy expanded from placental CD34 cells. The use of recombinant IL-2 (rhIL-2) therapy has induced regression of tumors by stimulating the activation and proliferation of NK cells. Low dose RhIL-2 has been shown to selectively expand and enhance the expression of maturation markers on the surface of NK cells, making them more potent killers in vitro and in vivo. CYNK-001-AML-001 is a Phase I multi-dose study evaluating the safety, tolerability, and persistence of CYNK-001 with or without rhIL-2 in adults with primary or secondary acute myeloid leukemia (AML) in morphologic complete remission (CR) with minimal residual disease (MRD) or relapsed/refractory (R/R) AML

Aim: To determine the safety and tolerability of rhIL-2 when added to CYNK-001 and assess the impact of rhIL-2 on CYNK-001 activity and persistence.

Methods This Phase I, single-arm, dose escalation clinical trial (NCT04310592) enrolled patients 18-80 years with R/R AML or MRD-positive CR. Escalating doses of CYNK-001 cells were administered following a standard 3+3 design, with 2 separate arms (MRD+ and R/R AML). Cohort 4a enrolled MRD+ AML subjects, while cohorts 4b and 5b enrolled R/R AML subjects. Cohort 4a and 4b utilized a dose of 1.8 billion CYNK-001 x 3 doses of CYNK-001 without rhIL-2. Cohort 5b utilized same dose of CYNK-001 along with rhIL-2. The decision to administer rhIL-2 in cohorts 6a, 6b, 7a and 7b would be based on review of safety, efficacy, and translational data in Cohorts 5b. In cohort 5b rhIL-2 was administered at a dose of 6M IU on days 0,2,4,7,9,11 and 14. Cohort 4b was used as a comparator.

Results Five patients in cohort 5b received rhIL-2. Treatment was well tolerated with no dose limiting toxicities (DLT's) being observed. Ten adverse events ≥ grade 3 were experienced. The commonest grade 3 AE's were raised AST (2/10) and ALT (2/10). Two grade 4 AE's were reported including febrile neutropenia and a pre-existing GVHD. In cohort 4b, 3 out of 6 patients achieved an objective response (3x MLFS) at day 28. No patient in cohort 5b achieved a response. While there was an apparent higher baseline blast count in cohort 5b which may have contributed to the differences, translational data demonstrated a surge in Tregs without enhancement of T helper or cytotoxic cells or NK cell reconstitution. In cohort 4a, one subject who received rhIL-2 (6MIU/dose) in error showed an increase in Tregs similar to cohort 5b.

Conclusion Treatment with CYNK-001 in combination with rhIL-2 was well tolerated with no dose-limiting toxicities observed. However, this approach does not enhance CD4 helper T cells, CD8 cytotoxic cells or NK cell reconstitution, while stimulated proliferation of Tregs. CYNK-001 proliferation and persistence was not enhanced with rhIL-2 infusion. Based on these data, rIL-2 was excluded from subsequent cohorts in this study.