Franziska Schaedeli Stark (1), Clarisse Chavanne (1), Sian Lennon-Chrimes (2), Cheikh Diack (1), Michael Derks (2)
(1) Roche Pharmaceutical Research and Early Development, Pharmaceutical Sciences, Roche Innovation Center Basel, Switzerland; (2) Roche Innovation Center Welwyn, UK
Objectives: Balovaptan is a vasopressin receptor 1a antagonist in development to help improve socialization and communication in autism spectrum disorder (ASD). The first study conducted in paediatric participants was a Phase 2, randomized, double-blind, 24-week, placebo-controlled study to investigate the efficacy, safety and pharmacokinetics (PK) of balovaptan in children and adolescents with ASD aged 5–17 years (aV1ation [1]). A PK cohort for dose confirmation preceded recruitment to the main cohorts in adolescents (13–17 yrs) and children (5–12 yrs). We present the model-informed approach used to support the initial selection and the confirmation of age-adjusted doses in aV1ation to achieve a homogenous, adult-equivalent exposure distribution in paediatric participants aged ≥5 years as a prerequisite for the comparison of efficacy with the previously reported adult study (VANILLA [2,3]).
Methods: Selection of initial doses in aV1ation, equivalent to 4 mg and 10 mg in adults, was guided by a PBPK model-based approach [4]. Four paediatric age bands (15–17, 12–14, 8–11 and 5–7 yrs) with adjusted doses were proposed according to the PBPK-predicted ratio of paediatric CL/F relative to CL/F in adults (CLratio). The study design included an initial cohort of ~24 adolescents (randomized 1:1:1) recruited for dose confirmation based on week 2 PK sampling. Once doses were confirmed, full recruitment was opened for adolescents to the main study part at the confirmed doses, the PK cohort for younger children (5–12 yrs) was opened for dose confirmation prior to opening full recruitment in children. A preliminary population PK (PopPK) model in adults was used to obtain empirical Bayes estimates (EBEs) of paediatric individual PK parameters and exposure. If the individual fit to the data was deemed satisfactory, EBEs of CLratio and AUC at steady-state were compared with the PBPK model predictions, and dose was confirmed if the median per age band was within the 90% CI of PBPK model-predicted median [4]. Along with comparing the medians per age band, the overall trend of CLratio vs. age in the cumulative data was considered for the dose decision at each review step.
Results: The PBPK model predicted CLratios of approximately 95% (15–17 yrs), 70% (12–14 yrs), 50% (8–11 yrs) and 30% (5–7 yrs), leading to initial doses of 3, 5, 7 and 10 mg for the 10 mg eq arm and 1.5, 2, 3 and 4 mg for the 4 mg eq arm. Recruitment was slower than anticipated, multiple PK reviews were performed before sufficient information was available for dose confirmation in adolescents. The CLratio in adolsecents was as predicted for participants aged 15–17 yrs, and in the upper part of the predicted 90% CI for participants aged 13–14 yrs. This was considered acceptable; the main study was opened to adolescents with initial per-protocol doses and dose confirmation was deferred until PK data from children (5–12 yrs) were available to better understand the relationship between age and CLratio. Once PK data from the younger participants were available, the overall trend of age-dependent decrease in CLratio appeared less than predicted. Median CLratio was 97% (15–17 yrs), 95% (13–14 yrs), 80% (8–12 yrs), and 59% (5–7 yrs), the final doses to continue aV1ation were confirmed to be: adult doses (4 mg and 10 mg, respectively) in participants aged ≥8 yrs, and 70% of adult doses (3 and 7 mg, respectively) in participants aged 5–7 yrs. Preliminary PK data from this study were used to estimate the age effect on CL/F in an updated PopPK model, and results are supportive of the proposed age-based dose regimen; however, revisions of age bands for paediatric dose adjustments in future clinical studies may be envisaged, based on simulations using the updated PopPK model.
Conclusions: A model-informed “learning-confirming” approach successfully supported the selection and confirmation of paediatric age-adjusted doses for balovaptan. The mismatch between PBPK model-predicted and observed age effect on CL/F, although still within 2-fold, was unexpected, and underlying mechanistic reasons have not yet been identified. The complex study design turned out to be sensitive to recruitment rate, and some of the adolescents received only 70% of the targeted dose due to delayed final dose confirmation. PK assessment in a separate study prior to efficacy assessment might have been more efficient to achieve the objectives of the aV1ation study.
References:
[1] ClinicalTrials.gov NCT02901431
[2] ClinicalTrials.gov NCT01793441
[3] Bolognani F et al. Sci Trans Med, 11:eaat7838, 2019.
[4] Schaedeli Stark F et al. PAGE 24 (2015) Abstr 3553 [www.page-meeting.org/?abstract=3553]
Reference: PAGE () Abstr 9435 [www.page-meeting.org/?abstract=9435]
Poster: Drug/Disease Modelling - Paediatrics