Alejandra Schiavo (1,2),Cecilia Maldonado (1), Marta Vázquez (1), Pietro Fagiolino(1), Iñaki F. Trocóniz (3,4), Manuel Ibarra (1)
(1) Department of Pharmaceutical Sciences. Faculty of Chemistry. Universidad de la República. Montevideo, Uruguay. (2) Postgraduate program, Faculty of Chemistry. Universidad de la República. Montevideo, Uruguay.(3) Pharmacometrics and Systems Pharmacology Research Unit, Department of Pharmaceutical Technology and Chemistry, School of Pharmacy and Nutrition, University of Navarra. Pamplona, Spain. (4)IdiSNA; Navarra Institute for Health Research, Pamplona, Spain.
Introduction: Valproic acid (VPA) is a narrow therapeutic index drug widely prescribed for the treatment of epilepsy, psychiatric disorders and migraine. As a branched short-fatty acid, VPA β-oxidation involves its transport into the mitochondrial matrix using the “carnitine shuttle” via formation of valproylcarnitine as intermediate. This route accounts for ~40% of total VPA clearance. Additionally, valproylcarnitine down-regulates cellular carnitine uptake. In turn, long term high-dose VPA therapy or acute VPA overdose can cause carnitine depletion, interfering with the urea cycle and leading to an increase in ammonia levels. A parallel consequence is the relative increase on VPA ω-oxidation, a minor elimination route that produces a metabolite, 4-ene-VPA, that blocks the first step of the urea cycle. Hyperammonemia is then a frequently reported VPA dose-dependent induced toxicity, with an overall incidence of ~40% [1,2]. Despite being a reversible adverse event following the stop of VPA treatment, it might lead to severe complications. Supplementation of L-carnitine is empirically used to reduce ammonia levels, although the implemented dosage regime and its impact on VPA pharmacokinetics are not yet quantitatively characterized from a population approach.
Objectives:
- Develop a QSP model characterizing VPA-induced carnitine depletion and subsequent hyperammonemia.
- Evaluate the efficacy of carnitine supplementation (CS) in acute and chronic exposure to high VPA levels.
- Propose an optimal dosage regime for CS according to VPA treatment.
Methods: A population pharmacokinetic model for VPA developed in NONMEM 7.4 (Icon plc.) with data from an average bioequivalence study was expanded integrating literature data to describe its saturable plasma protein binding and the kinetics of different metabolites (4-en-VPA, VPA glucuronide, VPA-carnitine and 2-en-VPA), carnitine, ammonia and fatty acids. The QSP model was developed using mlxR 4.1.0 package (Lixoft) in R 3.5.3 (R-Project). Model validation was performed using reported and clinically available data for different VPA dosage regimens, evaluating model accuracy for total and free VPA, its metabolites, ammonia and carnitine at steady state. Relative changes were modeled in case of elements for which experimental observations were not available. A local sensitivity analysis was conducted for all the parameters of the model evaluating the impact of a +/- 20% change on total and free VPA, ammonia and carnitine mean metrics (area under the concentration-time curve, maximum and minimum concentrations at steady state). Effect of VPA dosage over ammonia levels and CS antidote efficacy was assessed by simulations from the final model.
Results: A QSP model characterizing the impact of VPA free plasma concentrations in ammonia blood levels was developed and validated. The model comprises 17 ordinary differential equations and relates twelve elements through 39 reactions. Pharmacokinetic nonlinearities in VPA total and free concentrations due to saturable binding in plasma and capacity limited metabolism were well captured by the model. Assuming a normal range of 15-55 uM for ammonia in plasma, the hyperammonemia predicted incidence for three different VPA dosage regimens was in accordance with previous reports: 33% (500q12H), 61% (500q8H) and 86% (1000q12H). In relation with the impact of CS, after including this in this model we found a beneficial effect in reverting the VPA-induced hyperammonemia under long-term treatments with 500 to 2000 mg/day. Results show that under a given daily VPA dose, a 2-fold higher l-carnitine daily dose administered following the same interdose interval would be effective without affecting VPA efficacy. In addition, the use of CS as an antidote after an acute intoxication with VPA would also be effective. Different dosage regime for the use of CS as antidote were assessed and compared. A L-carnitine loading dose of 100 mg/kg followed by 15 mg/kg every 4 hours for one day would return ammonia levels to baseline within 30 hours of treatment.
Conclusions: A QSP model to quantitatively study the mechanisms behind VPA-induced hyperammonemia was developed. This model reflects the benefit of CS in both chronic and toxic VPA administration. Moreover, it was possible to propose an efficient carnitine dose according to VPA dose in chronic treatments.
References:
[1] M. Vázquez et al., “Hyperammonemia associated with valproic acid concentrations”, BioMed Research International, vol. 2014, Article ID 217269, 7 pages, 2014.
[2] C. Maldonado et al., “Carnitine and/or acylcarnitine deficiency as a cause of higher levels of ammonia”, BioMed Research International, vol. 2016, Article ID 2920108, 8 pages, 2016.
[3] W Nasreddine et al., “Determinants of free serum valproate concentration: A prospective study in patients on divalproex sodium monotherapy”, Sizure 59, pg. 24-27, 2018.
Reference: PAGE 29 (2021) Abstr 9810 [www.page-meeting.org/?abstract=9810]
Poster: Drug/Disease Modelling - Safety