068: Pharmacokinetic Analyses & Population Pharmacokinetic Modeling of Maplirpacept in Patients With Newly Diagnosed Acute Myeloid Leukemia

Statement of Purpose, Innovation or Hypothesis: Maplirpacept is a fusion protein directed against CD47 and is in development for hematologic malignancies including acute myeloid leukemia (AML). Here, we conducted non-compartmental analysis (NCA) to estimate PK parameters, and population PK (PPK) analyses to describe the PK of maplirpacept in untreated AML patients in combination with either azacitidine (AZA) or AZA and venetoclax (VEN), and to identify potential covariates that impact maplirpacept PK.

Description of Methods & Materials: The dataset for the analyses was based on the phase 1a/1b dose-escalation and expansion trial of maplirpacept in patients with advanced hematological malignancies (NCT03530683). Patients with newly-diagnosed TP-53 mutated AML received either 8, 16, or 24 mg/kg QW maplirpacept in combination with approved doses of AZA, while elderly/unfit patients with newly diagnosed TP-53 wild type AML received either 8, 16 or 24 mg/kg QW maplirpacept in combination with approved doses of AZA and VEN. Serial serum samples for PK assessments were collected at pre-specified time points and concentrations measured using a validated assay. PK parameters were estimated using oNCA version 2.7.8, Pfizer’s proprietary tool for NCA. NONMEM version 7.4.3 was used for PPK modeling and stepwise covariate model selection was performed using Perl-speaks-NONMEM version 4.9.0. Post-processing of outputs was completed with R version 4.1.3. Model selection was based on the assessment of objective function values, visual inspection of diagnostic plots, precision of the parameter estimates, prediction-corrected visual predictive checks, and decrease in residual variability. Based on clinical relevance and/or visual inspection of data, potential covariates were selected and tested for statistical significance.

Data & Results: As of 20 March 2023, data from 53 patients with at least one measurable PK sample were available in the AML cohorts. A total of 508 measurable serum maplirpacept concentrations from these patients were included in the analyses. Of the 53 patients 23 received maplirpacept in combination with AZA at three dose levels, while 30 received maplirpacept in combination with AZA and VEN at 2 dose levels. Data from maplirpacept at 24 mg/kg in combination with AZA and VEN was not available at the time of the analyses. The single and multiple dose PK parameters of maplirpacept are summarized in Table 1. Generally, maplirpacept exhibited a greater than dose-proportional increase in systemic exposures likely due to target-mediated drug disposition. A 3-compartment PK model with simultaneous linear and non-linear clearance components described the PK of maplirpacept. The final parameter estimates for Vmax (Maximum elimination rate), Km (Concentration at which elimination is half maximum), CLL (Linear Clearance), V1, V2, V3 [volume terms for central (1) and peripheral compartments (2,3)], inter-compartment clearance Q1, Q2 are 2.01 mg/hr, 14.7 mg/L, 0.023 L/h, 5.54 L, 6.12 L, 9.35 L, 0.094 L/hr, 0.0068 L/h respectively. Baseline body weight and sex were determined to be significant covariates on V1. Women had a 20% lower V1 compared to men. Baseline body weight was not a significant covariate of clearance.

Interpretation, Conclusion or Significance: Maplirpacept exhibited non-linear PK likely due to target-mediated drug disposition. The lack of body weight effect on maplirpacept clearance supports a fixed dosing strategy in future development of maplirpacept.

Disclosures: Yibo Wang was a Pfizer employee at the time of abstract development.

Citations/References:

Additional Information/Authors: NA