Eva Degraeuwe (1,2), Elke Gasthuys (1), Paulien Debruyne (2), Ann Raes (2), Johan Vande Walle (2), An Vermeulen (1), Louis Sandra (1)
(1) Laboratory of Medical Biochemistry and Clinical Analysis, Faculty of Pharmaceutical Sciences, Ghent, Belgium (2) Department of Internal Medicine and Paediatrics, Ghent University, Ghent, Belgium
Objectives: Lisinopril, an angiotensin-converting enzyme (ACE) inhibitor, is the most frequently prescribed antihypertensive therapeutic in the pediatric population, indicated primarily in children with mild to moderate renal impairment [1-3]. The population pharmacokinetic (PK) models available in scientific literature are not based on data reflecting the day-to-day clinical pediatric population, limiting their applicability [4,5]. We would like to identify covariate relationships significantly influencing lisinopril exposure in hypertensive children with mild to moderate renal impairment, ultimately aiming to challenge or confirm the currently implemented dose and regimen strategy.
Methods: During the SAFEPEDRUG project, a dose-escalation pilot study investigating lisinopril PK was performed [6]. Lisinopril was administered once daily (doses ranging from 0.05 mg kg-1 to 0.2 mg kg-1) at steady-state. Peak and trough samples were drawn during visits within an interval of 4 hours after dosing. Samples were analysed via an internally developed LC-MS/MS bioanalytical method. Patients were aged between 1.9-17.9 years (median: 13.5 years), had a total bodyweight (BW) between 9.62 and 97.2 kg (median: 53.2 kg) and an estimated glomerular filtration rate (eGFR) between 55.5 and 180 mL min-1 1.73m-2 (median: 99.9 mL min-1 1.73m-2 with 3 patients having an eGFR lower than 60 mL min-1 1.73m-2). A one-compartment model with 1st order absorption and 1st order elimination and a proportional error model best described the data. Absence of intravenous PK data rendered it impossible to estimate the oral bioavailability. Inter individual variability (IIV) was quantified for elimination clearance (CL) only. Model building was performed based on improvement in objective function values (OFVs), Standard error estimates, goodness-of-fit evaluations, prediction corrected visual predictive checks (pcVPCs), convergence assessments and parsimony. All data were analyzed using Monolix version 2020R1 (Lixoft®, France) and R version 3.6.2.
Results: A total of 46 plasma samples from 13 children with primary or secondary hypertension were collected. Parameter estimates of ka (0.077 h-1 [9%], typical value [relative standard error]), V (32.9 L 70kg-1 [37%]) and CL (22 L h-1 [8%]) show good predictive ability. Significant covariate effects (|ΔOFV| ≻ 3.84, Χ2-test for 1 degree of freedom) of BW on volume (V) and the CL and eGFR on CL were identified. The covariate effects of BW on V and CL were implemented using fixed allometric theory: allometric exponents of 0.75 for CL and 1 for V, centered around a BW of 70 kg [7]. The effect of eGFR on the CL was best described by a power law relationship and was centered around 105 mL min-1 1.73m-2. 8% of the IIV of CL was explained by the addition of this covariate effect, leaving 49% unexplained. Effects of BW and eGFR on CL were visualized using deterministic simulations. These results are in accordance with Thomson et al., highlighting the superiority of accounting for both BW and eGFR on CL compared to either solely accounting for BW or eGFR [8]. The eGFR covariate effect on CL is physiologically plausible, given that lisinopril undergoes no metabolism and is 100% renally excreted [9].
Conclusion: Lisinopril dose and regimen adjustments for pediatric patients should consider eGFR on top of BW adjustments. Extending the model to characterize pharmacodynamic effects (capturing blood pressure changes) is required to further identify target attainment and subsequently inform dosing.
References:
[1] Raes A, Malfait F, Van Aken S, France A, Donckerwolcke R, Vande Walle J. Lisinopril in paediatric medicine: a retrospective chart review of long-term treatment in children. J Renin Angiotensin Aldosterone Syst. 2007 Mar;8(1):3-12. doi: 10.3317/jraas.2007.004.
[2] Snauwaert E, Vande Walle J, De Bruyne P. Therapeutic efficacy and safety of ACE inhibitors in the hypertensive paediatric population: a review. Arch Dis Child. 2017 Jan;102(1):63-71. doi: 10.1136/archdischild-2016-310582.
[3] Barker CIS, Standing JF, Kelly LE, Hanly Faught L, Needham AC, Rieder MJ, de Wildt SN, Offringa M. Pharmacokinetic studies in children: recommendations for practice and research. Arch Dis Child. 2018 Jul;103(7):695-702. doi: 10.1136/archdischild-2017-314506
[4] Xie F, Van Bocxlaer J, Vermeulen A. Physiologically based pharmacokinetic modelling of lisinopril in children: A case story of angiotensin converting enzyme inhibitors. Br J Clin Pharmacol. 2021 Mar;87(3):1203-1214. doi: 10.1111/bcp.14492.
[5] Rashid M, Sarfraz M, Arafat M, Hussain A, Abbas N, Sadiq MW, Rasool MF, Bukhari NI. Prediction of lisinopril pediatric dose from the reference adult dose by employing a physiologically based pharmacokinetic model. BMC Pharmacol Toxicol. 2020 Jul 29;21(1):56. doi: 10.1186/s40360-020-00429-y.
[6] Bruyne P De, Meesters K, Walle J Vande. SAFE-PEDRUG: A NEW STRATEGY FOR PAEDIATRIC DRUG RESEARCH. Arch Dis Child. 2016;101(1):e1.29-e1. doi:10.1136/archdischild-2015-310148.35
[7] Anderson BJ, Holford NH. Tips and traps analyzing pediatric PK data. Paediatr Anaesth. 2011 Mar;21(3):222-37. doi: 10.1111/j.1460-9592.2011.03536.x.
[8] Thomson AH, Kelly JG, Whiting B. Lisinopril population pharmacokinetics in elderly and renal disease patients with hypertension. Br J Clin Pharmacol. 1989 Jan;27(1):57-65. doi: 10.1111/j.1365-2125.1989.tb05335.x.
[9] Beermann B. Pharmacokinetics of lisinopril. Am J Med. 1988 Sep 23;85(3B):25-30. doi: 10.1016/0002-9343(88)90346-4.
Reference: PAGE 29 (2021) Abstr 9766 [www.page-meeting.org/?abstract=9766]
Poster: Drug/Disease Modelling - Paediatrics