Pharmacokinetic time-dependency and covariates modelling of Isatuximab monoclonal antibody in multiple myeloma patients: analysis from pooled phase I/II & phase III studies
FAU J‐B (1), EL‐CHEIKH R(1), BRILLAC C(1), KOIWAI K(1), SEMIOND D(1), CAMPANA F (1), NGUYEN L (1) and GIBIANSKY L (2)
1Translationalmedicine & early development, Sanofi, France. 2 Quantpharm LLC, USA.
Introduction Multiple Myeloma (MM) is a malignant disease characterized by clonal proliferation of plasma cells in the bone marrow and the production of excessive amounts of abnormal proteins, the so-called Myeloma proteins (M-prot). The M-prot are usually monoclonal immunoglobulins of type G (IgG) and less frequently other Ig types or free-light chains. Isatuximab (ISA) is a humanized monoclonal antibody of IgG1 isotype that selectively targets the CD38 antigen overexpressed in malignant plasma cells [1]. A full clinical development program of Isatuximab as monotherapy or in combination therapy is on-going.
Objective: The aim of this work was to characterize the time-dependent PK of free ISA concentrations and to investigate the sources of PK variability from a pooled dataset of phases I/II & phase III clinical studies.
Methods: A total of 476 MM patients treated as single agent or in combination with Pomalidomide/Dexamethasone were analysed. Isatuximab was administered as intravenous infusion over various dosing regimens (QW and/or Q2W at doses ranging from 1 to 20 mg/kg). The population PK analysis was performed using SAEM algorithm for nonlinear mixed-effects model implemented in MONOLIX software (version 2018R1). Several structural PK models including linear and/or non-linear elimination pathways with different time-varying clearance functions were tested. The influence of many baseline demographic and pathophysiological covariates was investigated following univariate and multivariate analyses based on likelihood ratio test and Wald test. Qualification of the population PK model was performed using goodness-of-fit plots and visual predictive checks.
Results: ISA PK was best described by a two-compartment model with parallel linear and nonlinear (Michaelis Menten) elimination and time-varying linear clearance function. At the recommended therapeutic dose, the linear elimination pathway was the main contributor to the total clearance indicating that the target receptor was saturated.
Linear clearance was found to be related to the type of Myeloma (IgG vs non IgG), beta2-microglobulin and body weight while central volume of distribution was found to be related to body weight, gender, formulation and Race (Asian vs others). Myeloma type has the most meaningful impact: patients producing monoclonal IgG (IgG myeloma) demonstrated higher linear clearance than patients secreting other types of immunoglobulins or free-light chains (non-IgG myeloma). In average, a two-fold lower exposure at steady-state was predicted in IgG myeloma compared to non-IgG myeloma. Other covariates retained in the final model showed limited to moderate effect with a maximal variation less than ±30% at steady state exposure compared to the median value. There was no effect of age, renal or liver function impairment.
Linear clearance change over the treatment course was modeled as a sigmoidal Emax function. For a typical patient, the linear clearance decreases 50% from its initial value over the first 8 weeks of treatment. The decrease in clearance was slower in IgG myeloma compared to non-IgG myeloma patients. On average, the linear clearance reaches quasi-steady state after 4 weeks and 10 weeks of treatment in non-IgG myeloma and IgG myeloma patients, respectively.
Conclusion: The type of myeloma proteins production was the main contributor to explain ISA PK variability: faster clearance in IgG myeloma patients is likely due to the competition between high concentration of disease-produced IgG M-protein secreted by myeloma cells and Isatuximab undergoing lower FcRn recycling. Time-varying clearance of ISA was well characterized on a large set of data. As already shown with other therapeutic monoclonal antibodies in oncology [2-3], the decrease of clearance over time might be partly explained by an inflammatory status reduction associated with lowering of protein turnover due to treatment efficacy. Supporting this hypothesis, a close correlation was established between time-varying clearance and clinical response rate. Further development of the current model is warranted to include the mutual interplay between PK change over time and the clinical response measured by longitudinal PD biomarkers evolution (M-Prot).
References:
[1] Deckert J, Wetzel MC, Bartle LM et al. SAR650984, a novel humanized CD38-targeting antibody, demonstrates potent antitumor activity in models of multiple myeloma and other CD38+ hematologic malignancies. Clin Cancer Res (2014);20(17):4574-83
[2] Bajaj G, Wang X, Agrawal S et al.. Model-Based Population Pharmacokinetic Analysis of Nivolumab in Patients With Solid Tumors. CPT Pharmacometrics Syst Pharmacol. 2017 Jan;6(1):58-66
[3] Hongshan Li, Jingyu Yu, Chao Liu et al. Time dependent pharmacokinetics of pembrolizumab in patients with solid tumor and its correlation with best overall response. J Pharmacokinet Pharmacodyn (2017) 44:403–414