Population Pharmacokinetics and Exposure-Response Analyses for Glofitamab in Relapsed/Refractory B-Cell Non-Hodgkin Lymphoma (R/R NHL): Confirmation of Efficacy and CRS Mitigation in Patients with Step-up Dosing

Introduction: Glofitamab (RG6026; RO7082859; CD20-TCB) is a novel ‘2:1’ format T-cell-engaging bispecific antibody that has two CD20 and one CD3 binding domains, enabling increased tumor antigen avidity, rapid T-cell activation, and enhanced tumor cell killing in B-cell malignancies. Clinical data from NP30179 demonstrated that fixed dosing of glofitamab (0.6–25mg) induced high and durable complete responses with a manageable safety profile in heavily pre-treated R/R NHL patients (pts; Dickinson, et al. EHA 2020). Obinutuzumab pretreatment (Gpt) 7 days prior to first administration of glofitamab was shown to be effective in mitigating the risk of cytokine release syndrome (CRS), allowing for rapid escalation of glofitamab to clinically active doses (Dickinson, et al. EHA 2020). We previously investigated population pharmacokinetics (popPK) and exposure-response (ER) relationships for glofitamab in NP30179; NCT03075696 (Djebli N, et al. Blood 2019), where modelling indicated step-up dosing would further mitigate CRS while maximizing efficacy. The present analysis is an update of previous models, including confirmatory data from the first step-up dosing (SUD) pts.

Methods: Pts with indolent (i) or aggressive (a) R/R NHL received glofitamab fixed dosing (0.005–25mg every 2 or 3 weeks) or SUD (n=31, 2.5/10/16 and 2.5/10/30mg) following single Gpt 1000mg on Cycle (C) 1 Day (D) −7 to mitigate CRS. Serial and sparse glofitamab, and sparse G PK data were used to develop a popPK model in NONMEM^® software (v7.4). The cut-off date of April 17, 2020 enabled inclusion of 16 (2.5/10/16mg) and 15 (2.5/10/30mg) SUD pts. Physiologically relevant covariates were investigated for their potential influence on glofitamab PK variability. Using the established G popPK model (Gibiansky, et al. CPT Pharmacometrics Syst Pharmacol 2014), G concentration-time profiles were constructed to estimate glofitamab receptor occupancy (RO%) in the presence of G competing for CD20 receptors over time. The relationship between glofitamab AvgRO% over the first 24 hours and CRS, with a focus on Grade (Gr) ≥2 CRS (defined by ASTCT criteria [Lee, et al. 2019]) was investigated in iNHL and aNHL pts combined. ER relationships between glofitamab time-averaged RO% (AvgRO%) up to C3D1, which is when the first response assessment was taken, and complete response rate (CRR) were characterized in aNHL pts who reached C3D1.

Results: PopPK were analyzed in 230 iNHL and aNHL pts with ≥1 PK sample (fixed and SUD). ER relationships were analyzed in 95 aNHL pts with PK/efficacy data at C3D1, and in 204 iNHL and aNHL pts with PK/safety data. Glofitamab PK were best described using a two-compartment PK model with linear clearance and were comparable in pts with iNHL and aNHL. The effect of bodyweight on volumes and clearances was retained. Positive ER relationships were observed between AvgRO% over the first 24 hours and Gr ≥2 CRS in both iNHL and aNHL pts (p=0.002; Figure 1A), and between AvgRO% up to C3D1 and efficacy in aNHL pts (p=0.008; Figure 1B). Based on previous ER analyses (Djebli, et al. Blood 2019) of data from pts receiving fixed dosing, a SUD regimen (2.5/10/30mg Q3W) was selected to optimize the benefit/risk profile by beginning treatment at a dose to have CRS at manageable levels whilst allowing escalation to a higher dose associated with better clinical response. Updated ER analysis from fixed (n=199) and SUD (n=31) pts predicts an AvgRO% in the first 24 hours of 0.16% (0.10–0.29%), corresponding to a predicted Gr ≥2 CRS rate of 23.3% (20.8–26.8%) in iNHL and aNHL pts, and an AvgRO% to C3D1 of 0.75% (0.49–1.98%) corresponding to an anticipated CRR at Cycle 3 of 46.1% (42.7–53.8%) in aNHL pts. In comparison, clinical data from aNHL and iNHL pts receiving 2.5/10/16 and 2.5/10/30mg SUD (Hutchings, et al. ASH 2020) demonstrated a Gr ≥2 CRS rate of 21.6 % following the 2.5mg glofitamab dose (n=37), and a complete metabolic response rate of 40.6% (n=32).

Conclusions: Glofitamab PopPK and ER relationships for efficacy/safety were updated, including data from SUD pts. These models and emerging SUD clinical data confirm that in NHL pts, the SUD regimen allowed glofitamab escalation up to 30mg to maximize efficacy while minimizing the risk of increased CRS at the first administration. These models are being developed further to support optimal biological-dose selection of glofitamab, both as monotherapy and in combination with other agents.