Population PK models aim to describe the change in drug concentration over time for a specific population. The populations in population PK modelling often refer to subjects in a clinical trial of a potential drug candidate. Population PK models are frequently described by non-linear mixed effect (NLME) models, that including both random and fixed effect components. The fixed effect components 𝜽 (THETA) portray typical parameter values in the population while the random effects components 𝜼 (ETA) allow for the incorporation of inter-individual variability (IIV) on the typical population value. The IIVs are therefore an important element of NLME models, but the estimation of the IIVs can be time consuming and become a limiting factor for more complex models. Linear approximation of the IIV’s has been suggested as a way to reduce the estimation time whilst maintaining robustness. The aim of this project was to evaluate and compare the estimation time and robustness of the IIVs for the linear approximation of parameter estimation errors in NLME models compared to those estimated in non-linear models. Population PK NLME models were developed for two datasets of phenobarbital and moxonidine. The datasets contained different levels of complexity such as number of subjects, datapoints and route of administration. The models were developed within R-studio using the assembler and Pharmpy packages and evaluated in NONMEM 7.5. Based on the objective function values (OFVs), obtained in the model building processes, selected models were linearised using Pearl speaks NONMEM (PsN). The estimated 𝜀′𝑠 and run-time of the linearised models were compared to their non-linearized counterparts. For all the models a reduction in run-time could be observed but with a slight variation in the estimations between the linearised and non-linearised models. The biggest run time reduction was seen in the oral transit compartment models for moxonidine with a 3100-fold reduction in estimation time. The estimation time reduction displayed could more quickly provide valuable information regarding the chosen error models of more complex models and while parameters estimated may not be identical to the non-linearised models, they should be sufficient during the model building phase.

<div><b>Background:</b> Bupropion, an atypical antidepressant and smoking cessation aid, is associated with wide inter-subject variability in its efficacy and safety. Variability in response to bupropion therapy is thought to be driven by variability in metabolism. Bupropion undergoes complex phase 1 and 2 stereoselective metabolism. Though bupropion`s pharmacology is not fully understood, much of it is thought to be due to its metabolites, specially, S, S-hydroxybupropion. In vitro studies (functional assays measuring IC50 at dopamine transporter-DAT, norepinephrine transporter-NET, various subtypes of nicotinic receptors-nAChR) and mouse models (forced swim test to assess antidepressant effect, antinociceptive models to assess antagonism of nicotine effects) indicate S, S-hydroxybupropion to contribute more towards efficacy as an antidepressant and smoking cessation aid than racemic bupropion and R, R-hydroxybupropion, respectively. Both pharmacokinetics (PK) and pharmacodynamics (PD) of bupropion and its metabolites are complex and reported to be stereoselective. As bupropion is known to act on multiple central nervous system (CNS) targets (DAT, NET nAChR), understanding CNS disposition (target site) is critical to explain variability in bupropion`s therapeutic and toxic effects. </div><div><b>Objective: </b>The objective of our study was to characterize the exposure of bupropion enantiomers and corresponding phase 1 metabolite diastereomers in plasma and brain in a surrogate non-clinical species, and to subsequently develop animal-to-human-translational population-PK and Physiologically Based PK (PBPK) models to predict human brain concentrations of bupropion and its active metabolite S, S-hydroxybupropion. Application of these PK modeling approaches to map the time course of unbound brain concentration can then be compared to in vitro potency measures at DAT, NET and nAChRs to predict target engagement over time (PD). Establishing relationships between plasma PK, target site PK along with PD would elucidate possible cause(s) of inter-patient variability to bupropion therapy. </div><div><b>Methods: </b>The first step towards development of a CNS model was to identify a nonclinical species with phase 1 metabolism closest to humans. To accomplish this, hepatic microsomal incubations of four species-rat, mouse, non-human primates (NHPs) and humans were conducted separately for the R- and S-bupropion enantiomers, and the formation of enantiomer-specific metabolites was determined using LC-MS/MS. Intrinsic formation clearance (CLint) of metabolites across the four species (rats, mice, NHPs, humans) was determined from the formation rate versus substrate concentration relationship. </div><div>Racemic bupropion (10 mg/kg) and preformed S, S-hydroxybupropion (2 mg/kg) were administered subcutaneously to adult male Sprague Dawley rats (n = 24/compound). Brain and plasma were collected from rats (n = 3) at eight time points for 6 hours and analyzed using a chiral LC-MS/MS method. Rat plasma protein and brain homogenate binding studies were conducted for all analytes to correct for unbound fraction using equilibrium dialysis method.</div><div>A plasma-brain compartmental pharmacokinetic approach was used to describe the blood–brain-barrier transport of both bupropion and S, S-hydroxybupropion. Also, a 2-compartment permeability-limited brain model consisting of brain blood, brain mass compartments was developed and incorporated into a whole body physiologically-based pharmacokinetic (PBPK) parent-metabolite model for bupropion and S, S-hydroxybupropion. Both population PK and PBPK modeling approaches were subsequently translated to humans to predict human plasma and brain site exposure and its relationship to DAT and NET IC50 potencies.</div><div><b>Results: </b>The total clearance of S-bupropion was higher than that of R-bupropion in monkey and human liver microsomes. The contribution of hydroxybupropion to the total racemic bupropion clearance was 38%, 62%, 17%, and 96% in human, monkey, rat, and mouse, respectively. In the same species order, threohydrobupropion contributed 53%, 23%, 17%, and 3%, and erythrohydrobupropion contributed 9%, 14%, 66%, and 1.3%, respectively, to racemic bupropion clearance. Hepatic microsomal incubation studies indicated non-human primates to be the appropriate species to model CNS disposition. However, the cost and limited pharmacokinetic and pharmacodynamic data in NHPs were insurmountable barriers to conducting in vivo studies in NHPs. After considering multiple factors, such as the formation of reductive metabolites (higher in rats than mice), which are also thought to contribute to bupropion`s therapeutic efficacy, availability of microdialysis data measuring bupropion and dopamine, norepinephrine levels in brain extracellular fluid (ECF) and other in vitro potency evaluations in rats, rat was chosen as the surrogate species to model bupropion`s disposition.</div><div>In rats, unbound plasma and brain exposures and plasma clearances of both R and S-bupropion were similar. The exposure to parent was higher (50 to 100-fold) than to metabolites. The exposure of oxidative metabolites (R, R- and S, S-hydroxybupropion) was 2 to 3-fold higher in brain and plasma than reductive metabolites (R, R- and S, S-threohydrobupropion, S, R- and R, S-erythrohydrobupropion). Hepatic clearances of R- and S-bupropion scaled from in vitro rat hepatic microsomal incubation studies were 3-fold and 25-fold lower than their respective in vivo unbound apparent clearances. This could possibly be due to substantial contribution of metabolic pathways not characterized in this in vivo study and/or possible extrahepatic disposition in the rat. The unbound brain to unbound plasma AUC0-6h ratio (Kp,uu) of R- and S-bupropion were 0.43 and 0.38 respectively. Kp,uu of oxidative metabolites (R, R- and S, S-hydroxybupropion) and reductive metabolites (R, R- and S, S-threohydrobupropion) were close to 1. Kp,uu of S, R-erythrohydrobupropion was 0.43 and that of pre-formed S, S-hydroxybupropion was 5.</div><div>With respect to population PK modeling of both bupropion and S, S-hydroxybupropion, a plasma-brain compartmental model structure with time dependent change in brain influx clearance was required to adequately characterize the BBB transport of parent and this active metabolite. Using a physiologically-based pharmacokinetic model (PBPK) approach too, incorporation of active efflux and carrier mediated uptake terms in addition to passive permeability was necessary to adequately characterize brain disposition of bupropion and S, S-hydroxybupropion. Both modeling approaches (population-PK and PBPK) when translated to humans indicated that the predicted human brain exposures fall below the reported DAT and NET IC50 measures of bupropion and S, S-hydroxybupropion. </div><div><b>Conclusion: </b>Specific to our work in the rat, the discrepancy between in vitro scaled hepatic clearance and in vivo plasma clearance of R and S-bupropion suggests alternative non-CYP mediated clearance pathways and/or extra hepatic disposition of bupropion. Both translational PK models indicate active process such as efflux transporter or carrier mediated uptake could be involved in bupropion`s disposition in the brain. Variability in expression of these speculated active/carrier mediated transporters could possibly cause variability in response. Also, other CNS targets could contribute to bupropion`s therapeutic efficacy, elucidation of which would require further investigation.</div><div><br></div>