Abstract 846: Utilizing quantitative systems pharmacology (QSP) model to understand cytokine release syndrome (CRS) following administration of epcoritamab in DLBCL patients

Abstract

Abstract Background: Epcoritamab is a novel bispecific antibody that targets CD3 and CD20 and induces T-cell-mediated cytotoxic activity against CD20+ malignant B cells (1). Epcoritamab’s pharmacokinetics (PK), pharmacodynamics, and safety profiles were investigated in GCT3013-01 escalation study across a wide range of doses. CRS is an adverse event of special interest for T cell engagers such as Epcoritamab. As such, step-up dosing (SUD) has been investigated in study GCT3013-01 escalation to minimize release of cytokines and mitigate severe CRS symptoms. Seventeen permutations of SUD were tested, with doses ranging from 0.004 to 60 mg. Based on observed CRS data, a SUD regimen of 0.16/0.8/48 mg was selected. The aim of the analysis was to develop a QSP model capable of predicting cytokine release which can be used to identify potential optimal SUD regimen. Method: Leveraging Epcoritamab preclinical and clinical data, we have developed a QSP platform that captures the observed tumor responses and allows for systemic predictions of biomarkers with the selected SUD regimen. A minimal physiologically based PK model was linked to submodels of T cell activation, proliferation, and cytotoxicity of target-bearing cells, to create an integrated QSP model of the DLBCL system. Epcoritamab-mediated cross-linking of target(s) and CD3 receptor to form trimer complexes drives cell trafficking and cytokine production as well as inducing death of target-bearing cells. The model described the PK, cellular effects, and observed cytokine profiles within the plasma, and simulated similar effects in lymph node(s), tumor, and other peripheral body tissue. Result: After SC administration of Epcoritamab, transient and modest elevation of selected cytokines was observed. Patients’ cytokine profiles were categorized into three different trends based on Cycle 1 data. By accounting for the interaction of receptors and ligands at the nano-scale level (time and physical size) we characterized the dose response and dynamics of T cells, B cells, IL6, and IL10 in patients enrolled in the study by using the QSP model. Conclusions: This analysis integrated underlying mechanisms to predict various systemic biomarkers dynamics including T cells, B cells, IL6 and IL10 following administration of Epcoritamab. The well-studied pathways and mechanisms allow implementation of the model in disease indications other than DLBCL.