Population pharmacokinetic analysis of Satralizumab in NMOSD patients and healthy volunteers.
Leonid Gibiansky (1), Hajime Ito (2), Sian Lennon-Chrimes (3), Sebastien Jolivet (4), Patricia Sanwald Ducray (4), Nicolas Frey (4), Hanna Silber Baumann (4)
(1) QuantPharm LLC, North Potomac, US, (2) Chugai Pharmaceutical Co., Ltd., Tokyo, Japan, (3) Roche Innovation Center Welwyn, UK, (4) Roche Innovation Center Basel, Switzerland
Objectives: Interleukin-6 (IL-6) has been implicated in the immunopathology of neuromyelitis optica spectrum disorder (NMOSD). Satralizumab is a subcutaneously (SC) administered monoclonal antibody (mAb) that binds to and blocks the IL-6 receptor (IL-6R) in a pH-dependent manner. Satralizumab was engineered to be recycled back into the circulation via the neonatal Fc receptor (FcRn), increasing its serum half-life and resulting in prolonged inhibition of IL-6R signalling. The objective was to characterize the pharmacokinetic (PK) properties of satralizumab in healthy subjects and NMOSD patients.
Methods: Data from a Phase 1 study in healthy volunteers (HV) and two pivotal Phase 3 studies in NMOSD patients (SAkuraSky [add-on to immunosuppressive therapy] and SAkuraStar [monotherapy]) were used to characterize the PK through non-linear mixed effects modeling. The Phase 1 study was a single dose study with a range of doses (30-240 mg) given either through intravenous (IV) or SC administration. The Phase 3 studies were multiple dose studies with a 120 mg dose administered SC every 4 weeks (q4w) with initial loading doses administered at 0, 2, and 4 weeks. The base model development and initial covariate analysis was based on the HV and SAkuraSky studies. The model was re-estimated once SAkuraStar was available. The starting model was selected based on prior knowledge about mAb PK in similar compounds including the expectation of observing both linear and saturable elimination. A covariate analysis was performed to investigate factors explaining variability in PK parameters, including demographic variables, formulation (pre-filled syringe (PFS) vs vial) and the presence of anti-drug antibodies (ADA). A full covariate model approach was adopted, meaning that all covariate relationships to be investigated were included simultaneously followed by stepwise reduction of the model to arrive at a parsimonious model. NONMEM was used for the analysis. Model performance was assessed through statistical and graphical assessments.
Results: Data from 226 individuals (72 HVs and 154 NMOSD patients) and 4715 concentration samples were included in the analysis. The data covered up to 4 years of treatment in some patients. ADA were present in 41% (SAkuraSky) and 71% (SAkuraStar) of NMOSD patients at the end of the double-blind period. A two-compartment PK model with parallel linear and target-mediated (Michaelis-Menten) elimination and first-order SC absorption accurately described the concentration-time course of satralizumab in both HVs and patients with NMOSD. For a reference subject (patient with NMOSD, weighting 60 kg without detected ADAs), clearance, central volume, inter-compartment clearance, and peripheral volume were estimated at 0.060 L/day, 3.46 L, 0.336 L/day, 2.07 L, respectively. Absolute bioavailability and the absorption rate constant were estimated at 0.854 and 0.251 day-1. The maximum Michaelis-Menten elimination rate and Michaelis-Menten constant were estimated at 0.455 µg/mL/day and 0.462 µg/mL, respectively. The effective half-life was approximately 30 days at steady state (ss). The longer half-life compared to most conventional mAbs is the result of the modifications made to the molecule to allow pH-dependent binding to IL-6R and to enhance the recycling properties through the FcRn receptor (Chugai proprietary Recycling Antibody® technology). The following covariate relationships were identified based on the covariate analysis: bodyweight on the clearance and volume parameters according to allometric scaling (power of 0.75 for clearance and 1 for volumes), ADA on bioavailability (-13%) and on linear clearance (+45% in presence of ADA), disease status on linear clearance (+96% in HVs), formulation on linear clearance (+9% with PFS). The predicted receptor occupancy (RO) was high, on average >95%, throughout the dosing interval at ss in both ADA-negative and ADA-positive patients.
Conclusions: The PK of satralizumab following IV and SC administration was well described by a two-compartment model with the first order absorption and parallel linear and saturable elimination. The longer half-life due to applied antibody recycling technology supports a q4w dosing interval. The RO was high throughout the dose interval in the majority of patients regardless of the ADA status. No impact on safety and no clear impact on efficacy was observed due to ADA. In addition, the lack of meaningful impact of the formulation supports the use of the marketed PFS formulation.