Anna Largajolli1, Stephen Duffull1, Francesco Bellanti1, Federica Alessi2, Sara Cazzaniga2, Paolo Bettica2
1Certara, 2 Italfarmaco SpA.
Introduction: Duchenne muscular dystrophy (DMD) is a severe progressive muscular disease [1]. DMD is caused by a reduction in dystrophin protein responsible for muscle cell integrity in skeletal and smooth muscles and affects standard motor activities with degenerative loss of function. The onset of symptoms occur early in childhood and reduces life expectancy to 20-30 y. Givinostat is an oral histone deacetylase inhibitor recently approved in the US for the treatment of DMD in patients aged =6 years, based on findings from the phase 3 EPIDYS study. A DMD disease progression (DP) modelling framework and Clinical Trial Simulation tool, has been developed for several clinical endpoints by C-Path’s Duchenne Regulatory Science Consortium (D-RSC) [2]. Objectives: (1) To validate and update the D-RSC DMD modelling framework with 4-stair climb (4SC) data from EPIDYS and the ongoing long-term safety and tolerability study (LTSE). (2) To assess differences in the time course of DMD progression between patients receiving standard of care (SoC; i.e. corticosteroids) or givinostat (in combination with SoC). Methods: The analysis used nonlinear mixed-effects modelling (NONMEM v7.5). To align with the D-RSC model, 4SC was transformed into velocity climb (1/s). 4SC disease progression has been modelled using a Chapman-Richard growth linked to a sigmoid Emax function [2], accounting for effects of baseline age and baseline 4SC on the DP. The published DMD disease progression model of 4SC was first validated with clinical data from patients in the EPIDYS study receiving SoC. The model performance was assessed using visual predictive check (VPC) diagnostics. Simulations were then conducted in 1000 virtual patients using baseline demographic data from the available studies, comparing the updated (givinostat) and published (SoC) models; 500 replicates were simulated accounting for interindividual variability. Virtual patients were followed for 5 years, and the effects of givinostat and SoC were compared using the median curve with 95% CI. Results: Data from a total of 357 patients were included in the analysis, with 296 receiving givinostat and 61 receiving SoC. Patients with missing baseline age or baseline 4SC were excluded from the analysis. In addition, the M3 method was used to account for the impact of patients becoming nonambulant during the study and therefore unable to perform the 4SC assessment. VPC diagnostics showed good agreement between model predictions and observed data for both SoC-treated patients from EPIDYS and givinostat-treated patients from EPIDYS and the LTSE. When re-estimating the DP model with data from givinostat-treated patients only, the most impacted parameters were the DP50 (age at which 4SC reaches half of its maximum decrease) increasing from 10.8 to 12.2 years, and gamma (hill coefficient of the Imax relationship) reduced from 15.5 to 6.9, in both cases suggesting a slower disease progression in givinostat-treated patients. Results of the 5-year simulations showed that givinostat was associated with a delay of approximately 2 years in the DMD progression curve for 4SC starting at 9 years of age. At 14 years of age, the model showed that patients treated with givinostat were able to climb 4 stairs faster (approximately 0.1 1/s) than those receiving only SoC. Conclusions: The D-RSC DP model was validated successfully with data from the EPIDYS study and extended to quantify the impact of givinostat therapy on 4SC. The results of the simulations demonstrated a delay in DMD progression, as measured by decline in 4SC, in patients treated with givinostat compared with SoC.
[1] Angulski et al. Frontiers Physiol 2023;14:1183101 [2] Lingineni et al. CPT Pharmacometrics Syst Pharmacol. 2022;11:318–332
Reference: PAGE 33 (2025) Abstr 11524 [www.page-meeting.org/?abstract=11524]
Poster: Drug/Disease Modelling - Paediatrics