Andreas Lindauer (1), Erno van Schaick (1), Pilar Pizcueta (2), Laura Rodríguez-Pascau (2), Marc Martinell (2)
(1) Calvagone SAS, France, (2) Minoryx Therapeutics S.L., Spain
Introduction:
Leriglitazone is a full, selective and brain penetrant peroxisome proliferator-activated receptor gamma (PPARγ) agonist. The compound is a promising candidate for addressing rare neurodegenerative diseases including X-linked adrenoleukodystrophy (ALD) [1]. The cerebral form of X-ALD (cALD) manifests in boys and men and is characterized by demyelinating brain lesions that become rapidly progressive leading to death in 2-4 years; while adrenomyeloneuropathy (AMN) occurs in adult men (and 80% of heterozygous women) and affects spinal cord and peripheral nerves [2,3,4]. In addition, leriglitazone is also investigated for the treatment of Friedreich’s Ataxia (FRDA).
This work outlines a population pharmacokinetic (popPK) analysis derived from studies in patients and healthy volunteers, focusing on simulating pediatric dosing strategies.
Objectives:
- To develop a popPK model using data from 3 studies in adult and pediatric patients (AMN, FRDA and cALD) and 2 studies in healthy subjects.
- To simulate different dosing strategies in children (weight-based, fixed, concentration-guided) to achieve a target AUC of 136-204 µg.h/mL.
Methods:
Model development was initially conducted on a dataset including 28 healthy subjects and 103 adult patients (2136 PK samples). Later the model was updated with additional data from 15 boys with cALD (134 samples) and 250 samples from a mass-balance study in 11 healthy subjects. In total of 2524 concentration measurements from 156 patients receiving leriglitazone as an oral solution were available for analysis.
Following an exploratory analysis, one- and two-compartment models with linear elimination and first-order absorption with/without lag-time were explored using NONMEM v7.5 with FOCE+INTER. Automated stepwise covariate modeling (SCM) using a forward addition/backward deletion procedure as implemented in PsN (v5.0.0) was applied.
Simulations were conducted using Campsis (v1.3.1) [5], with the demographic covariates for boys (2-17 years) sampled from the NHANES database [6]. The following dosing scenarios were explored: 1) flat dose regimens per age group, 2) body-weight-based dosing, 3) dose-adjustment based on simulated trough concentrations, 4) dosing according to the parameter-covariate equation of the final model, and 5) dosing according to derived look-up tables with age, weight and height ranges.
Results:
A two-compartment model with first-order absorption and linear elimination provided a good fit to the available data in this heterogeneous population including boys as young as 3 years up to 66-year-old adults, weighing up to 138 kg.
The typical rate of absorption was fast when MIN-102 was taken without food (5.33 1/h), however was reduced by 83.5% when taken with food. Furthermore, food reduced the bioavailability by 4.1%. Body weight was included in the model as a covariate on clearance, inter-compartment flow and volume terms by means of allometric scaling with fixed exponents. Further covariates were identified in the SCM: Age was significantly associated with clearance (the younger the lower), as was sex (18.8% lower CL in women). Body mass index (BMI) was significantly correlated with (relative) bioavailability (larger values with higher BMI).
Simulations showed that concentration-based dose individualizations lead to a low variability in AUC of about 23% CV, with 62.2% of pediatric patients within the target AUC range. A priori dose individualization using dosing tables accounting for age, weight and height achieved a similar percentage of patient within the AUC target range as a covariate-based formula (57.5% and 58.9%). Fixed dosing with age-groups (2-5, 6-11, >12 years) and body-weight-based dosing achieved only about 50% target attainment with a tendency for under-exposure.
Conclusions:
The population PK model described well the concentration data of a diverse group from young children to older adults. Food intake significantly slowed drug absorption and slightly reduced bioavailability. Body weight impacted drug clearance and distribution, with age, sex, and BMI also affecting drug metabolism.
For targeted dosing, concentration-based adjustments showed the least variability in drug exposure, with target AUC attainment in over 60% of pediatric patients. Dosing tables based on age, weight, and height performed similarly well, while fixed and body-weight-based dosing performed poorly.
References:
[1] Pizcueta et al., Int J Mol Sci. 2023 Feb 6;24(4):3201
[2] De Beer at al .; Neurology. 2014; 83(24):2227-31
[3] Raymond et al. ; Biol Blood Marrow Transplant. 2019; 25(3):538-48.
[4] Van Geel et al. Ann Neurol. 2001; 49:186-94.
[5] Luyckx et al., PAGE 30 (2022) Abstr 9991 [www.page-meeting.org/?abstract=9991]
[6] National Health and Nutrition Examination Survey database (NHANES, 2017-2018), https://wwwn.cdc.gov/nchs/nhanes/
Reference: PAGE 32 (2024) Abstr 11087 [www.page-meeting.org/?abstract=11087]
Poster: Clinical Applications