Simulating realistic patient profiles from pharmacokinetic models by a machine learning postprocessing correction of residual variability.

We address the problem of model misspecification in population pharmacokinetics (PopPK), by modeling residual unexplained variability (RUV) by machine learning (ML) methods in a postprocessing step after conventional model building. The practical purpose of the method is the generation of realistic virtual patient profiles and the quantification of the extent of model misspecification, by introducing an appropriate metric, to be used as an additional diagnostic of model quality. The proposed methodology consists of the following steps: After developing a PopPK model, the individual residual errors IRES = DV-IPRED, are computed, where DV are the observations and IPRED the individual predictions and are modeled by ML to obtain IRESML. Correction of the IPREDs can then be carried out as DVML = IPRED + IRESML. The methodology was tested in a PK study of ropinirole, for which a PopPK model was developed while a second deliberately misspecified model was also considered. Various supervised ML algorithms were tested, among which Random Forest gave the best results. The ML model was able to correct individual predictions as inspected in diagnostic plots and most importantly it simulated realistic profiles that resembled the real data and canceled out the artifacts introduced by the elevated RUV, even in the case of the heavily misspecified model. Furthermore, a metric to quantify the extent of model misspecification was introduced based on the R2 between IRES and IRESML, following the rationale that the greater the extent of variability explained by the ML model, the higher the degree of model misspecification present in the original model.

Donepezil Brain and Blood Pharmacokinetic Modeling after Nasal Film and Oral Solution Administration in Mice.

Intranasal delivery is a non-invasive mode of administration, gaining popularity due to its potential for targeted delivery to the brain. The anatomic connection of the nasal cavity with the central nervous system (CNS) is based on two nerves: olfactory and trigeminal. Moreover, the high vasculature of the respiratory area enables systemic absorption avoiding possible hepatic metabolism. Due to these physiological peculiarities of the nasal cavity, compartmental modeling for nasal formulation is considered a demanding process. For this purpose, intravenous models have been proposed, based on the fast absorption from the olfactory nerve. However, most of the sophisticated approaches are required to describe the different absorption events occurring in the nasal cavity. Donepezil was recently formulated in the form of nasal film ensuring drug delivery in both bloodstream and the brain. In this work, a three-compartment model was first developed to describe donepezil oral brain and blood pharmacokinetics. Subsequently, using parameters estimated by this model, an intranasal model was developed dividing the administered dose into three fractions, corresponding to absorption directly to the bloodstream and brain, as well as indirectly to the brain expressed through transit compartments. Hence, the models of this study aim to describe the drug flow on both occasions and quantify the direct nose-to-brain and systemic distribution.

Implementation of non-linear mixed effects models defined by fractional differential equations.

Fractional differential equations (FDEs), i.e. differential equations with derivatives of non-integer order, can describe certain experimental datasets more accurately than classic models and have found application in pharmacokinetics (PKs), but wider applicability has been hindered by the lack of appropriate software. In the present work an extension of NONMEM software is introduced, as a FORTRAN subroutine, that allows the definition of nonlinear mixed effects (NLME) models with FDEs. The new subroutine can handle arbitrary user defined linear and nonlinear models with multiple equations, and multiple doses and can be integrated in NONMEM workflows seamlessly, working well with third party packages. The performance of the subroutine in parameter estimation exercises, with simple linear and nonlinear (Michaelis-Menten) fractional PK models has been evaluated by simulations and an application to a real clinical dataset of diazepam is presented. In the simulation study, model parameters were estimated for each of 100 simulated datasets for the two models. The relative mean bias (RMB) and relative root mean square error (RRMSE) were calculated in order to assess the bias and precision of the methodology. In all cases both RMB and RRMSE were below 20% showing high accuracy and precision for the estimates. For the diazepam application the fractional model that best described the drug kinetics was a one-compartment linear model which had similar performance, according to diagnostic plots and Visual Predictive Check, to a three-compartment classic model, but including four less parameters than the latter. To the best of our knowledge, it is the first attempt to use FDE systems in an NLME framework, so the approach could be of interest to other disciplines apart from PKs.

In Silico Dosimetry Study of Tc99m-Tetrofosmin in Children Using a Novel PBPK Model in Humans Built from SPECT Imaging Data.

PURPOSE: The aim of this work is to develop a Physiologically Based Pharmacokinetic model (PBPK) for the radiopharmaceutical Tc99m-Tetrofosmin in humans, from literature SPECT imaging data, to carry out in-silico dosimetry studies in children and extrapolate dosing. METHODS: A whole body PBPK model was developed from literature data from humans of Tc99m-Tetrofosmin tissue distribution. A data driven approach to estimate partition coeffects, permeability parameters and clearances was carried out, while some parameters were determined using a standard in silico PBPK method. Paediatric PK data for all tissues were simulated by changing the physiological parameters from the adult to paediatric values. Absorbed and effective doses for children of all ages were calculated using S-values from literature of Tc99m that have been computed from anthropomorphic phantoms. RESULTS: Using the results from each tissue, satisfactory goodness-of-fit was achieved, assessed by visual inspection and a coefficient of determination of R2 = 0.965 while all estimated parameters had good standard errors. Paediatric simulations of Tetrofosmin distribution showed that paediatric profiles are not very different to the those of adults. The effective doses per unit of administered activity for 15 yo, 10 yo, 5 yo and 1 yo children were calculated to be 1.2, 1.7, 2.6 and 4.8 times higher, respectively than the adult value. Based on these calculations maximum administered activity scale more than proportionately to body weight. CONCLUSIONS: A PBPK model of tetrofosmin in adults has been developed from SPECT imaging data and was extrapolated to conduct in-silico dosimetry studies in children of all ages.