Yumi Cleary1,2, Heidemarie Kletzl1, Paul Grimsey3, Nicolas Frey1, Hanna Silber1, Cordula Stillhart4, Agnes Poirier1, Katja Heinig1, Stephen Fowler1, Kayode Ogungbenro2, Leon Aarons2, Aleksandra Galetin2 and Michael Gertz1
1Roche Pharma Research and Early Development, Pharmaceutical Sciences, Roche Innovation Center Basel, Basel, Switzerland; 2Centre for Applied Pharmacokinetic Research, School of Health Sciences, University of Manchester, Manchester, UK; 3Roche Pharma Research and Early Development, Pharmaceutical Sciences, Roche Innovation Center Welwyn, UK; 4Formulation and Process Sciences, F Hoffmann-La Roche Ltd. Basel, Switzerland.
Objectives: Spinal muscular atrophy (SMA) is a severe, autosomal recessive neuromuscular disease caused by homozygous deletions and/or mutations in the survival of motor neuron-1 (SMN1) gene, resulting in insufficient expression of SMN protein [1]. A second gene, SMN2, produces only low levels of functional SMN protein. Risdiplam is an orally administered, centrally and peripherally distributed SMN2 pre-mRNA splicing modifier that increases the levels of functional SMN protein levels [2]. Patients with Type 1–3 SMA are generally diagnosed in early childhood. The efficacy and safety of risdiplam is currently being investigated in infants, children and adults with SMA. Population and physiologically-based pharmacokinetic (PPK and PBPK) modeling was performed to characterize risdiplam PK in patients with SMA.
- PBPK modeling, established on healthy volunteers’ PK data, and used to predict the PK in patients with SMA (adults and children) for selection of the appropriate starting doses. Additional applications of the model including extrapolations to other populations and drug-drug interaction (DDI) risk investigations was envisaged.
- PPK analysis was performed on sparse PK data from patients with SMA to characterize absorption, distribution, metabolism and elimination (ADME) and to assess variability and covariate effects. The post-hoc PK parameters assisted in reparameterization of the PBPK model for patients with SMA.
Methods: PBPK modeling was performed by integrating in vitro, preclinical and clinical PK data from healthy adults using SimCYP version 18. The PBPK model parameters were scaled to children by incorporating age-dependent physiological information after verification of the model using the adult PK data. PPK modelling was performed using NONMEM version 7.4. Linear one- to three compartmental disposition models as well as various absorption models were investigated. Demographics, disease types and severity scales were investigated in the covariate assessment. The absolute bioavailability was predicted by the PBPK model to approximate the systemic PK parameters from post-hoc apparent PK parameters of the PPK model. Subsequently, the clearance (CL) and volume of distribution (V) were compared between the PPK and PBPK models.
Results: Plasma samples collected from 327 subjects (age range: 2 months to 52 years) were included in the analysis. A PPK model consisting of an absorption model with three transit compartments and two– compartment disposition model describes the risdiplam PK data well. Between-subject variability was estimated for the apparent clearance (CL/F), central volume (Vc/F) and transit absorption rate constant (ktr). Time-varying effects of age and body weight were incorporated as covariates to describe the variability in the heterogeneous population and the age-dependent changes in risdiplam CL/F and Vc/F.
- Absorption: a high bioavailability was predicted by the PBPK model, indicating that systemic CL and V could be approximated from apparent PK parameters estimated by the PPK analysis.
- Distribution: a maturation function was needed for Vc/F in the PPK model to account for the age effect in children < 1 years old. The PBPK model predicted a lower V in infants due to differences in tissue composition, in line with the PPK modelling.
- Metabolism / Elimination: the body weight-normalized post-hoc CL/F estimate was higher in children than adults. This finding could be described by a hyperbolic ontogeny functions [3] for the metabolizing enzymes relevance for risdiplam in the PBPK modeling.
Conclusions: The PPK model successfully characterized the risdiplam PK from infant to adult patients with SMA, and identified age-dependent CL/F guided the selection/implementation of ontogeny functions for the PBPK modeling. This is particularly critical when the PBPK model is intended to be used for investigating PK in unstudied conditions such as different populations and DDIs to ensure safety and efficacy of risdiplam treatment.
References:
[1] Lefebvre S, Burglen L, Reboullet S, Clermont O, Burlet P, Viollet L, et al. Identification and characterization of a spinal muscular atrophy-determining gene. Cell. 1995;80(1):155-165.
[2] Ratni H, Ebeling M, Baird J, Bendels S, Bylund J, Chen KS, et al. Discovery of risdiplam, a selective survival of motor neuron-2 ( SMN2) gene splicing modifier for the treatment of spinal muscular atrophy (SMA). J Med Chem. 2018;61(15):6501-6517.
[3] Upreti VV, Wahlstrom JL. Meta-analysis of hepatic cytochrome P450 ontogeny to underwrite the prediction of pediatric pharmacokinetics using physiologically based pharmacokinetic modeling. J Clin Pharmacol. 2016;56(3):266-283.
Reference: PAGE () Abstr 9434 [www.page-meeting.org/?abstract=9434]
Poster: Drug/Disease Modelling - Absorption & PBPK