Modeling of the Pharmacokinetics of the Selective Orexin-1 Receptor Antagonist ACT-539313 with Auto-Inhibition
Khalid Ahmed1,2, Linda Chaba1,3, Amira Ghoneim1,4,6, Jantine Brussee5, Chih-Hsuan Hsin5, Amy Cheung6, Samer Mouksassi6, Goonaseelan (Colin) Pillai1, Andreas Krause5
1Africa Pharmacometrics Training Program (ATP) Pharmacometrics Africa NPC, K45 Old Main Building, Groote Schuur Hospital, Cape Town, South Africa; 2Najran University, Najran, Saudi Arabia; 3Strathmore University, Nairobi, Kenya; 4Future University in Egypt, New Cairo, Egypt; 5Idorsia Pharmaceuticals Ltd, Allschwil, Switzerland, 6Integrated Drug Development, Certara, Princeton, NJ, United States
Objectives:
Binge eating disorder is a condition characterized by recurrent episodes of excessive food intake, accompanied by a sense of loss of control of eating behavior. It is associated with significant impairment of psychological and social functioning and can lead to obesity, metabolic dysfunction, and cardiovascular disease [1]. Currently, treatment includes cognitive-behavioral therapy, pharmacotherapy, and weight management interventions [2-4]. Selective orexin-1 receptor antagonists have emerged as a promising pharmacological approach in treating binge eating disorders [5]. [AK1] [AG2] Berger et al. (2020) reported that treatment with ACT-539313, a novel selective orexin-1 receptor antagonist, elicited CYP3A enzyme inhibition [6]. This study aimed to develop a population pharmacokinetic (popPK) model for ACT-539313 that is linked to auto-inhibition, after single and multiple oral dose administration. Auto-inhibition (i.e., a molecule inhibits enzymes responsible for its own metabolism) and body-size parameters were important features for the popPK model to support potential dose adjustments in the eventual heavier target patient population for the drug.
Methods:
Data from three phase I studies in healthy subjects (single and multiple ascending doses of 10, 30, 100, 200, and 400 mg, 94 participants, body weight 78.5 ± 9.8 kg) and one phase II clinical study (multiple doses of 100 mg, 63 participants, body weight = 101 ± 25 kg) were combined to develop a popPK model to characterize the pharmacokinetics (PK) of ACT-539313 and to describe the influence of intrinsic (body size) and extrinsic (food) covariates on its PK parameters. The model reported by Svensson et al. (2018) was adopted after modification to describe the enzyme turnover [7]. Model development was carried out using Monolix (Version 2021R2). Simulations were conducted using Berkeley Madonna Version 10 to determine the effect of auto-inhibition on exposure (Area Under the Curve – AUC and maximum plasma concentration - Cmax, after 24 h and over a steady-state dosing interval) metrics of ACT-539313. Models with- and without auto-inhibition were fit to assess the differences in estimation of steady-state exposure (AUC, Cmax) if auto-inhibition was neglected.
Results:
A two‐compartment model with first‐order absorption, lag time, and first‐order elimination with auto-inhibition adequately described the plasma concentration-time profile of ACT-539313 after single and multiple dosing in healthy subjects and obese patients. Parameter estimates were as follows: absorption rate constant (ka) = 0.27/h, lag time (tlag) = 0.24 h, volume of distribution of the central compartment (Vc) = 14.98 L, volume of distribution of the peripheral compartment (Vp) =5.54 L, clearance (CL) =19.35 L/h, intercompartmental transfer (Q) = 8.94 L/h , enzyme degradation rate constant (kenz) = 0.61/h, and concentration corresponding to 50% of Imax (IC50) = 65.77 ng/mL, with relative standard errors of 0.69%, 10.3%, 11.3%, 53.4% and 21.2%, respectively.
In fed condition, tlag was delayed by 9.5 minutes (p = 0.007) and ka was increased by 62.9% (p = 0.014). The data and the metrics of body size tested (body weight, body mass index, lean body weight, and fat mass) were unable to support estimation of any covariate model that might be useful for future dosage regimen design of ACT-539313 in obese patients (e.g., the estimated coefficients were negative). The model provided accurate estimates of the concentration-time profiles, with good agreement on graphical and numerical goodness-of-fit criteria and strong correlation between predicted and observed exposure (R2 = 1.00 and 0.99 for AUC and Cmax, respectively). Fitting the same model without auto-inhibition, i.e., ignoring auto-inhibition, resulted in estimated AUC and Cmax at steady state 12.8 and 10.1% lower, respectively.
Conclusions:
The auto-inhibition model provided a good fit for the PK of ACT-539313 and could be helpful in deciding dosage regimens. The model highlighted the relevance of capturing auto-inhibition adequately: steady-state exposure is underestimated if auto-inhibition is not modeled by appropriate model components. Further research for better understanding of the effect of body size on the PK of ACT-539313 may be helpful to ensure that dosages are appropriate for all patients, e.g., a well-controlled phase I PK study in obese subjects.
References:
[1] Agüera Z, Lozano-Madrid M, Mallorquí-Bagué N. et al. A review of binge eating disorder and obesity. Neuropsychiatr. 2021; 35:57–67. doi: 10.1007/s40211-020-00346-w
[2] Wilfley DE, Welch RR, Stein RI, Spurrell EB, Cohen LR, Saelens BE, Dounchis JZ, Frank MA, Wiseman CV, Matt GE. A randomized comparison of group cognitive-behavioral therapy and group interpersonal psychotherapy for the treatment of overweight individuals with binge-eating disorder. Arch Gen Psychiatry. 2002; 59(8):713-21. [AK1] [AG2] [AK3] doi: 10.1001/archpsyc.59.8.713
[3] Wilson GT, Wilfley DE, Agras WS, Bryson SW. Psychological treatments of binge eating disorder. Arch Gen Psychiatry. 2010; 67(1):94-101. doi: 10.1001/archgenpsychiatry.2009.170
[4] Brownley KA, Berkman ND, Peat CM, Lohr KN, Cullen KE, Bann CM, Bulik CM. Binge-Eating Disorder in Adults: A Systematic Review and Meta-analysis. Ann Intern Med. 2016; 165(6):409-20. doi: 10.7326/M15-2455
[5] Piccoli L, Micioni Di Bonaventura MV, Cifani C, Costantini VJ, Massagrande M, Montanari D, Martinelli P, Antolini M, Ciccocioppo R, Massi M, Merlo-Pich E, Di Fabio R, Corsi M. Role of orexin-1 receptor mechanisms on compulsive food consumption in a model of binge eating in female rats. Neuropsychopharmacology. 2012; 37(9):1999-2011
[6] Berger B, Kaufmann P, Koch A, Dingemanse J. Impact of the Selective Orexin-1 Receptor Antagonist ACT-539313 on the Pharmacokinetics of the CYP3A Probe Drug Midazolam in Healthy Male Subjects. J Clin Pharmacol. 2020; 60(7):931-941. doi: 10.1002/jcph.1588
[7] Svensson RJ, Aarnoutse RE, Diacon AH, Dawson R, Gillespie SH, Boeree MJ, Simonsson USH. A Population Pharmacokinetic Model Incorporating Saturable Pharmacokinetics and Autoinduction for High Rifampicin Doses. Clin Pharmacol Ther. 2018; 103(4):674-683
