Contributions
Abstract: PB1726
Type: Publication Only
Background
Hu5F9-G4 (5F9) is a humanized monoclonal antibody targeting CD47, a protective “don’t eat me” signal on cancer cells. Blocking CD47 stimulates tumor cell phagocytosis and activates an anti-tumor T-cell response. Pre-clinically, 5F9 eliminates leukemic disease and induces durable remissions in patient-derived xenograft mouse models.
Aims
The objectives of this analysis were to characterize the pharmacokinetics (PK) and anti-drug antibody (ADA) incidence of 5F9 after single and multiple doses in patients with acute myeloid leukemia (AML).
Methods
Data from the Phase 1 CAMELLIA study (MRC grant: MR/L008963/1) were used for this analysis. Multiple intravenous (IV) doses of Hu5F9 in the range 0.1 – 30 mg/kg were given at twice weekly frequency. PK samples were drawn in all patients after single and multiple doses. A total of 13 AML patients provided PK and ADA data. PK data were analyzed by a noncompartmental approach using the PKNCA package in R language. Population modelling using NONMEM software was performed by combining PK data in AML patients with data from patients with solid tumours and lymphomas. Model-estimated individual PK parameters (linear and nonlinear clearance, and volume of distribution) were used to compare PK parameters in AML vs solid tumour patients. Simulations using the model were used to identify optimal Phase 2 dose in AML patients. Presence of ADA was analyzed using a 3-tiered approach consisting of screening, confirmatory, and titre determination.
Results
Increases in maximum serum concentration (Cmax) of 5F9 were greater than dose-proportional in the dose range 0.1-10 mg/kg indicating nonlinearity in PK. At doses ≥10 mg/kg, increases in Cmax were linear. The PK parameters were consistent with the presence of a CD47 antigen sink, which was saturated at doses ≥ 10 mg/kg. Population PK modelling indicated that PK parameters in AML patients were similar to those estimated in solid tumour patients. Simulations with the model predicted that a maintenance dosing regimen of 30 mg/kg every week after week 2 would result in serum concentrations ≥200 µg/mL, a level where near-maximal CD47 receptor occupancy on blood and bone marrow cells was observed in this study.
ADA was confirmed in 1 subject (7.7%) in this study but had no observable impact on the PK parameters.
Conclusion
5F9 exhibits nonlinear PK in AML patients, typical of a receptor-targeted antibody. The PK in AML patients is similar to that in patients with solid tumours. The immunogenic potential of 5F9 in AML patients is low. Overall, the PK profile of 5F9 is suitable for once weekly dosing in AML patients. Ongoing studies in AML have implemented this dosing regimen (NCT03248479).
Session topic: 4. Acute myeloid leukemia - Clinical
Keyword(s): Macrophage, Phagocytosis, Pharmacokinetic
Abstract: PB1726
Type: Publication Only
Background
Hu5F9-G4 (5F9) is a humanized monoclonal antibody targeting CD47, a protective “don’t eat me” signal on cancer cells. Blocking CD47 stimulates tumor cell phagocytosis and activates an anti-tumor T-cell response. Pre-clinically, 5F9 eliminates leukemic disease and induces durable remissions in patient-derived xenograft mouse models.
Aims
The objectives of this analysis were to characterize the pharmacokinetics (PK) and anti-drug antibody (ADA) incidence of 5F9 after single and multiple doses in patients with acute myeloid leukemia (AML).
Methods
Data from the Phase 1 CAMELLIA study (MRC grant: MR/L008963/1) were used for this analysis. Multiple intravenous (IV) doses of Hu5F9 in the range 0.1 – 30 mg/kg were given at twice weekly frequency. PK samples were drawn in all patients after single and multiple doses. A total of 13 AML patients provided PK and ADA data. PK data were analyzed by a noncompartmental approach using the PKNCA package in R language. Population modelling using NONMEM software was performed by combining PK data in AML patients with data from patients with solid tumours and lymphomas. Model-estimated individual PK parameters (linear and nonlinear clearance, and volume of distribution) were used to compare PK parameters in AML vs solid tumour patients. Simulations using the model were used to identify optimal Phase 2 dose in AML patients. Presence of ADA was analyzed using a 3-tiered approach consisting of screening, confirmatory, and titre determination.
Results
Increases in maximum serum concentration (Cmax) of 5F9 were greater than dose-proportional in the dose range 0.1-10 mg/kg indicating nonlinearity in PK. At doses ≥10 mg/kg, increases in Cmax were linear. The PK parameters were consistent with the presence of a CD47 antigen sink, which was saturated at doses ≥ 10 mg/kg. Population PK modelling indicated that PK parameters in AML patients were similar to those estimated in solid tumour patients. Simulations with the model predicted that a maintenance dosing regimen of 30 mg/kg every week after week 2 would result in serum concentrations ≥200 µg/mL, a level where near-maximal CD47 receptor occupancy on blood and bone marrow cells was observed in this study.
ADA was confirmed in 1 subject (7.7%) in this study but had no observable impact on the PK parameters.
Conclusion
5F9 exhibits nonlinear PK in AML patients, typical of a receptor-targeted antibody. The PK in AML patients is similar to that in patients with solid tumours. The immunogenic potential of 5F9 in AML patients is low. Overall, the PK profile of 5F9 is suitable for once weekly dosing in AML patients. Ongoing studies in AML have implemented this dosing regimen (NCT03248479).
Session topic: 4. Acute myeloid leukemia - Clinical
Keyword(s): Macrophage, Phagocytosis, Pharmacokinetic