Francois Pierre Combes
Novartis Pharmaceuticals Corporation, One Health Plaza, East Hanover, NJ, USA
Objectives
Epilepsy is a common feature in patients with tuberous sclerosis complex (TSC), occurring in up to 85% of patients; approximately 60% of these patients remain treatment-refractory. Everolimus, an oral mammalian target of rapamycin (mTOR) inhibitor, is approved in Europe as an adjunctive therapy for patients aged ≥2 years with TSC-associated treatment-refractory partial-onset seizures, based largely on results from the EXIST studies, notably EXIST-3 study (NCT01713946). Analyses were performed to assess everolimus exposure and efficacy: For exposure, a physiologically based pharmacokinetic (PBPK) model was built and qualified based on pharmacokinetic (PK) data from pooled studies, in which the youngest observed patient would have started everolimus at approximately age about 1 year. For efficacy, the EXIST-3 study, where the youngest patient started everolimus at the age of 2 years, was the only study performed in this indication. Consequently, all built models are qualified for patients aged ≥1 year for PK and ≥2 years for pharmacodynamic (PD). As TSC-associated seizures can also affect children between 6 months and 2 years, a modeling and simulation (M&S) approach was taken to predict the expected exposure and corresponding efficacy (short-term via population PD [3], and long-term via linear mixed effect models) of everolimus as adjunctive treatment in patients aged between 6 months and 2 years with TSC-associated treatment-refractory seizures. The aim of this approach was to understand whether the recommended dose of 6 mg/m2 would bring patients aged between 6 months and 2 years old to the everolimus target range of 5-15 ng/mL, resulting in a reduction from baseline in seizure frequency (RSF).
Methods
The M&S study consisted of a framework of 3 models that were used together to enable the prediction of reliable patient exposure metrics (trough concentrations, Cmin) in children between 6 months and <1 year, and efficacy metric (RSF) in children between 6 months and 2 years. A PBPK model using SimCYP® simulation software was developed to predict Cmin over a 2 year period in pediatric patients initiating everolimus at an age between 6 months and 2 years. A total of 200 patients (replication of 10 trials with 20 patients each) were simulated. A population PD (PopPD) model and a linear mixed effect model were used to predict respectively short-term (after 6 months of treatment) and long-term (after 24 months of treatment) efficacy in terms of RSF, using the PBPK-predicted daily Cmin values.
Results
The PBPK model predicted that the simulated Cmin at the end of 24 weeks QD treatment for simulated patients of ages between 6 months and 2 years was in the range of 7.7-10.5 ng/mL, well within the targeted range of 5-15 ng/mL used in older patients. Predicted concentrations in younger patients who started everolimus before the age of 2 years were slightly higher than those observed in older children and adults included in EXIST-3. Using this predicted Cmin, the PopPD model predicted that the RSF would be of at least 66.1% (5th-95th percentiles: 50.3-75.8%) for a patient starting everolimus at the age 2 years and 77.8% (5th-95th percentiles: 60.6-87.6%) for a patient who started everolimus at an age of 6 months. Similar results were found when using the linear mixed effect model which used time-normalized predicted Cmin computed over 12-week intervals as the exposure covariate. Based on the long-term model-based analysis, the RSF observed in EXIST-3 was 21.39% (95% confidence interval [CI]: 13.30-28.74) for a 2-fold increase in time-normalized Cmin. Every additional 12 weeks of exposure to everolimus was predicted to have a modest, but significant effect on SF, resulting in a further decrease of SF by 5.64% (95%CI: 3.54-7.70) per period. Additionally, a 0.5-fold lower baseline SF was predicted to reduce RSF by 49.41% (95%CI: 45.68-52.89). The parameter estimates from this model and the time-normalized PBPK predicted Cmin were used to predict long-term efficacy. On an average, about a 50% RSF was predicted at the first 12 week time point and a reduction of around 70% at the final time point (2 years).
Conclusion
Based on the results of the M&S study, everolimus at the dose of 6 mg/m2 is anticipated lead to an efficacious treatment in children aged 6 months to 2 years, with concentrations within the recommended target range.
References:
[1] French JA, et al. Adjunctive everolimus therapy for treatment-resistant focal-onset seizures associated with tuberous sclerosis (EXIST-3): a phase 3, randomised, double-blind, placebo-controlled study. Lancet. 2016 Oct 29;388(10056):2153-2163. doi: 10.1016/S0140-6736(16)31419-2. Epub 2016 Sep 6.
[2] Leong R, et al (2012) Regulatory experience with physiologically based pharmacokinetic modeling for pediatric drug trials. Clin Pharm Ther; 91: 926-931.
[3]Plan EL.modeling and simulation of count data, CPT Pharmacometrics Syst Pharmacol. 2014 Aug 13;3:e129. doi: 10.1038/psp.2014.27.
Reference: PAGE 27 (2018) Abstr 8516 [www.page-meeting.org/?abstract=8516]
Poster: Drug/Disease Modelling - Paediatrics