Sally Babiker 1, Oscar Della Pasqua 1
1 Clinical Pharmacology & Therapeutics Group, University College London (London, United Kingdom)
Introduction: Tuberculous meningitis (TBM) manifests as the most devastating form of paediatric tuberculosis, causing high mortality rates and neurological sequelae in 50% of patients despite treatment[1]. Current dosing strategies for first-line drugs, including pyrazinamide (PZA), are largely extrapolated from pulmonary tuberculosis (PTB) in adult subjects, and increasing evidence indicates that World Health Organization (WHO)’s weight-banded dosing recommendations regularly result in suboptimal exposure in children, particularly in younger populations[2].In 2022, WHO conditionally proposed a 6-month regimen comprising increased doses of isoniazid (H) and rifampicin (R), with pyrazinamide (Z) and ethionamide (Eto) (6HRZEto), as an alternative option to the standard 12-month regimen (2HRZ-Ethambutol/10HR). Nevertheless, the pharmacokinetic (PK) comparability of these regimens remains uncertain, particularly at the central nervous system (CNS), where the main infection foci occur[3,4]. Hence, this analysis aimed to use model-informed approaches to (i) extrapolate adult PZA plasma and cerebrospinal fluid (CSF) PK to paediatric patients with TBM, (ii) evaluate exposure-matching between adults and children receiving WHO weight-banded regimens, and (iii) compare systemic and CNS exposures between the standard 12-month and intensive 6-month regimens to inform dose optimisation.
Methods: Despite differences in the endpoints used to assess treatment response in PTB vs TBM, extrapolation was implemented assuming that the PKPD relationship for PZA was sufficiently similar between adults and children. Moreover, processes determining the plasma to brain ratio were retained from adults, assuming similar CNS penetration between populations[5]. Therefore, we first conducted a literature review to identify a pharmacokinetic model for PZA with suitable parameterisation for the extrapolation of disposition properties from adults to children. The selected model included fat-free mass as the main covariate, with its effect on disposition parameters based on allometric principles. The model was further adapted to incorporate an enzyme maturation function, describing age-related changes in metabolic clearance in children aged <2 years[6]. Following a preliminary evaluation of the model performance in a previously described adult and paediatric population, a representative virtual paediatric cohort of 5000 subjects (median (min–max) age: 9 (0.17–17.9)years; weight: 23 (4–87)kg) was generated for subsequent assessment of systemic and CNS exposure in children following different dosing scenarios. Given the importance of realistic simulation results, baseline characteristics in this cohort accounted for malnourishment and stunted growth[7]. Two WHO weight-banded dosing regimens were evaluated: (a) standard 12-month regimen and (b) 6-month intensive regimen applied to the weight-eligible subset (<35kg) of the same virtual cohort. Adult steady state exposure, assessed as AUC₀–₂₄h (median [90% confidence interval]) was used as a reference therapeutic target range and subsequent exposure-matching assessment. Additionally, the proportion of children with exposures below or above the adult reference median and 90% CI was computed and compared between regimens. Finally, CSF:plasma AUC ratios were calculated across a relevant paediatric weight-range and compared to the adult reference median (95% CI). Results: Based on regimen (a), predicted AUC₀–₂₄h estimates in plasma and CSF were 336.6 [221–543]mg·h/L and 367.6 [243.8–588.7]mg·h/L, respectively. Similar estimates were predicted for regimen (b), where AUC₀–₂₄h in plasma and CSF were 341.5 [255.5-460.7]mg·h/L and 374.2 [281.9–502]mg·h/L, respectively. Relative to the adult reference exposure range, the proportion of children below the adult 5th percentile was 31.1% vs. 20% for plasma and 24.8% vs. 13.3% for CSF. Exposure ranges also differed significantly with increasing body weight despite the use of weight-banded regimens. Interestingly, for both regimens, simulated CSF:plasma AUC ratios across the different paediatric weight-bands were consistently above the adult reference median of 1.05 (95% CI 0.99–1.09) but demonstrated a modest decline with increasing weight. Conclusions: Irrespective of the numerous initiatives supporting novel therapy development for PTB, limited attention has been given to this less common but more severe clinical presentation of disease in children. Our analysis represents the first attempt to evaluate the dose rationale for TBM using extrapolation principles, considering available plasma-CSF pharmacokinetic data in adults. These results show that current WHO weight-banded dosing frequently leads to reduced exposures in children, particularly in the youngest groups, which are mostly affected by this condition. Whilst the use of an intensive 6-month regimen reduces the proportion of subjects with very low exposure, it does not ensure adequate exposure to PZA. Further efforts are needed to ensure all paediatric patients are exposed to optimised dosing regimens, enabling treatment shortening in this vulnerable population. References: 1. Marx GE, Chan ED. Tuberculous meningitis: Diagnosis and treatment Overview. Tuberculosis Research and Treatment. 2011;2011:1-9. doi:10.1155/2011/798764 2. Kwara A, Yang H, Martyn-Dickens C, et al. Adequacy of WHO weight-band dosing and fixed-dose combinations for the treatment of TB in children. The International Journal of Tuberculosis and Lung Disease. 2023;27(5):401-407. doi:10.5588/ijtld.22.0591 3. Wasmann RE, Masini T, Viney K, et al. A model-based approach for a practical dosing strategy for the short, intensive treatment regimen for paediatric tuberculous meningitis. Frontiers in Pharmacology. 2023;14:1055329. doi:10.3389/fphar.2023.1055329 4. Sulis G, Tavaziva G, Gore G, et al. Comparative Effectiveness of Regimens for Drug-Susceptible Tuberculous Meningitis in Children and Adolescents: A Systematic Review and Aggregate-Level Data Meta-Analysis. Open Forum Infectious Diseases. 2022;9(6):ofac108. doi:10.1093/ofid/ofac108 5. ICH guideline E11A on pediatric extrapolation - Scientific guideline | European Medicines Agency (EMA). European Medicines Agency (EMA). Published September 3, 2024. https://www.ema.europa.eu/en/ich-guideline-e11a-pediatric-extrapolation-scientific-guideline 6. Calderin JM, Wasserman S, Resendiz-Galvan JE, et al. Population pharmacokinetics of pyrazinamide and isoniazid in plasma and cerebrospinal fluid from South African adults with tuberculous meningitis. Antimicrobial Agents and Chemotherapy. 2025;69(8):e0009925. doi:10.1128/aac.00099-25 7. Wasmann RE, Svensson EM, Walker AS, Clements MN, Denti P. Constructing a representative in‐silico population for paediatric simulations: Application to HIV‐positive African children. British Journal of Clinical Pharmacology. 2020;87(7):2847-2854. doi:10.1111/bcp.14694
Reference: PAGE 34 (2026) Abstr 12219 [www.page-meeting.org/?abstract=12219]
Poster: Drug/Disease Modelling - Paediatrics