This perspective article draws a distinction between some of the well-known drug classification systems and a "Chemistry Classification System" (CCS). Rather than have drug classification based on some simple properties like solubility and permeability or route of systemic elimination, a CCS results in more than four or five classes and each class has distinct properties that impact formulation development. This perspective provides and outline of 13 classes, but a CCS is a flexible system that introduces a thought process for classification. The number of classes is not rigid, and chemists are encouraged to adapt these methods to their own situations. A CCS utilizes machine-learning models and artificial intelligence (AI) to estimate physicochemical properties that result in unique, frequently observed dissolution, Absorption, Distribution, Metabolism, and Excretion (ADME) properties to guide formulation development.

The purpose of this study was to develop a physiologically based pharmacokinetic (PBPK) model predicting the pharmacokinetics (PK) of different compounds in pregnant subjects. This model considers the differences in tissue sizes, blood flow rates, enzyme expression levels, glomerular filtration rates, plasma protein binding, and other factors affected during pregnancy in both the maternal and fetal models. The PBPKPlus™ module in GastroPlus® was used to model the PK of cefuroxime and cefazolin. For both compounds, the model was first validated against PK data in healthy non-pregnant volunteers and then applied to predict pregnant groups PK. The model accurately described the PK in both non-pregnant and pregnant groups and explained well differences in the plasma concentration due to pregnancy. The fetal plasma and amniotic fluid concentrations were also predicted reasonably well at different stages of pregnancy. This work describes the use of a PBPK approach for drug development and demonstrates the ability to predict differences in PK in pregnant subjects and fetal exposure for compounds excreted renally. The prediction for pregnant groups is also improved when the model is calibrated with postpartum or non-pregnant female group if such data are available.

Uridine 5′-diphospho-glucuronosyltransferases (UGTs) are expressed in the small intestines, but prediction of first-pass extraction from the related metabolism is not well studied. This work assesses physiologically based pharmacokinetic (PBPK) modeling as a tool for predicting intestinal metabolism due to UGTs in the human gastrointestinal tract. Available data for intestinal UGT expression levels and in vitro approaches that can be used to predict intestinal metabolism of UGT substrates are reviewed. Human PBPK models for UGT substrates with varying extents of UGT-mediated intestinal metabolism (lorazepam, oxazepam, naloxone, zidovudine, cabotegravir, raltegravir, and dolutegravir) have demonstrated utility for predicting the extent of intestinal metabolism. Drug–drug interactions (DDIs) of UGT1A1 substrates dolutegravir and raltegravir with UGT1A1 inhibitor atazanavir have been simulated, and the role of intestinal metabolism in these clinical DDIs examined. Utility of an in silico tool for predicting substrate specificity for UGTs is discussed. Improved in vitro tools to study metabolism for UGT compounds, such as coculture models for low clearance compounds and better understanding of optimal conditions for in vitro studies, may provide an opportunity for improved in vitro–in vivo extrapolation (IVIVE) and prospective predictions. PBPK modeling shows promise as a useful tool for predicting intestinal metabolism for UGT substrates.

Highly variable disposition after oral ingestion of acyclovir has been reported, although little is known regarding the underlying mechanisms. Different studies using the same reference product (Zovirax ®) showed that Cmax and AUC were respectively 44 and 35% lower in Saudi Arabians than Europeans, consistent with higher frequencies of reduced-activity polymorphs of the organic cation transporter (OCT1) in Europeans. In this study, the contribution of physiology (i.e., OCT1 activity) to the oral disposition of acyclovir immediate release (IR) tablets was hypothesized to be greater than dissolution. The potential role of OCT1 was studied in a validated physiologically-based biopharmaceutics model (PBBM), while dissolution of two Chilean generics (with demonstrated bioequivalence) and the reference product was assessed in vitro. The PBBM suggested that OCT1 activity could partially explain population-related pharmacokinetic differences. Further, dissolution of generics was slower than the regulatory criterion for BCS III IR products. Remarkably, virtual bioequivalence (incorporating in vitro dissolution into the PBBM) correctly and robustly predicted the bioequivalence of these products, showcasing its value in support of failed BCS biowaivers. These findings suggest that very-rapid dissolution for acyclovir IR products may not be critical for BCS biowaiver. They also endorse the relevance of cross-over designs in bioequivalence trials.

The aim of this study was to use a combined in vitro-in silico approach to develop a physiologically based pharmacokinetic model (PBPK) that predicts the bioavailability of albendazole (ABZ), a BCS class II/IV lipophilic weak base, and simulates its main metabolite albendazole sulphoxide (ABZSO) after oral administration of the current marketed dose of 400 mg in the fasted state. In vitro data was collected from solubility and dissolution tests performed with biorelevant media and transfer tests were carried out to evaluate the supersaturation and precipitation characteristics of ABZ upon gastric emptying. These in vitro results were used as biopharmaceutical inputs together with ABZ physicochemical properties including also permeability and in vitro metabolism data and information gathered from different clinical trials reported in the literature, were used to enable PBPK models to be developed using GastroPlus™ (version 9.7). As expected for this weak base with pKa = 3.6, ABZ exhibited a pronounced pH dependent solubility, with the solubility and extent of dissolution being greater at gastric pH and dropping significantly in the intestinal environment suggesting supersaturation and precipitation upon gastric emptying, which was confirmed by the transfer model experiments. PBPK models were set up for heathy volunteers using a full PBPK modeling approach and by implementing dynamic fluid volumes in the ACAT gut physiology in GastroPlus™. When coupling in vitro data (solubility values, dissolution rate and precipitation rate constant, etc.) for ABZ and with fitted values for the Vdss and liver systemic clearance of the sulfoxide metabolite to the PBPK model, the simulated profiles successfully predicated plasma concentrations of ABZ at 400 mg dose and simulated ABZSO at different ABZ dose levels and with different study populations, indicating the usefulness of combing in vitro biorelevant tools with PBPK modeling for the accurate prediction of ABZ bioavailability. The results obtained in this study also helped confirm that ABZ behaves as a BCS class IV compound.

Drug design and development requires collaboration among scientists with diverse expertise. In early-stage drug design, establishment of promising drug candidates requires integration of structure-based molecular design with biopharmaceutics and ADMET (absorption, distribution, metabolism, excretion, and toxicity) principles. Much of this work is increasingly computational. Thus, students in a drug design course should learn the basis for and effects of candidate small molecules binding to therapeutic targets and the importance of simultaneous understanding of ADMET characteristics that will largely determine if the molecule can achieve a therapeutically relevant concentration at the target site and become an orally deliverable drug. This second aspect of education in drug design has been relatively overlooked compared to structure-based design. Here, we describe a course using sophisticated simulation and modeling software, GastroPlus and ADMET Predictor, which are commonly used in the pharmaceutical industry. Our teaching approach has evolved over a decade in early graduate and advanced undergraduate settings. We use a combination of short lectures, software demonstration, and hands on use, which allows students to "design a drug" with incorporation of structural and ADMET considerations. Student feedback indicated that the use of the software and the course design were helpful in enhancing their understanding of the importance of biopharmaceutics properties and ADMET in drug design. GastroPlus and ADMET Predictor have recently become more accessible to educators, and our experience using this software in an educational setting may be helpful for instructors who wish to develop a similar course.

During non-clinical and clinical development of a new molecular entity (NME), modeling and simulation (M&S) are routinely used to predict the exposure and pharmacokinetics (PK) of the drug compound in humans. The basic methodology and output are generally understood across all functional disciplines. However, this understanding is mostly restricted to traditional methods such as those in simplified kinetic models and void of adequate mechanistic foundation to address questions beyond the observed clinical data. In the past two decades, alternative and more mechanistic methods, particularly for describing absorption, distribution, excretion and metabolism (ADME) of drugs have been developed and applied under the general umbrella of physiologically-based pharmacokinetic (PBPK) methods. Their mechanistic nature gives the ability to ask many other questions which were not traditionally asked and provide some logically and evidenced-based potential answers. Whilst traditional PK methods are mainstream and understood by most scientists, mechanistic absorption models alongside other PBPK approaches are still deemed eclectic, despite making significant strides in the fundamental science as well as regulatory acceptance. On November 3rd, a short course was held at the annual American Association of Pharmaceutical Scientists (AAPS) meeting in San Antonio, Texas. The different talks were tailored to provide a basis or rationale for the subject, introduction to fundamental principles with historical perspective, a critique of the state-of-the-art, examples of successful application of the methods across different phases of the drug development process and the specific standards these mechanistic models should meet to be fully reliable from a regulatory perspective.

In vitro dissolution experiments are used to qualitatively assess the impact of formulation composition and process changes on the drug dosage form performance. However, the use of dissolution data to quantitatively predict changes in the absorption profile remains limited. Physiologically-based Pharmacokinetic(s) (PBPK) models facilitate incorporation of in vitro dissolution experiments into mechanistic oral absorption models to predict in vivo oral formulation performance, and verify if the drug product dissolution method is biopredictive or clinically relevant. Nevertheless, a standardized approach for using dissolution data within PBPK models does not yet exist and the introduction of dissolution data in PBPK relies on a case by case approach which accommodates from differences in release mechanism and limitations to drug absorption. As part of the Innovative Medicines Initiative (IMI) Oral Biopharmaceutics Tools (OrBiTo) project a cross-work package was set up to gather a realistic understanding of various approaches used and their areas of applications. This paper presents the approaches shared by academic and industrial scientists through the OrBiTo project to integrate dissolution data within PBPK software to improve the prediction accuracy of oral formulations in vivo. Some general recommendations regarding current use and future improvements are also provided.

Oral drug absorption is a complex process depending on many factors, including the physicochemical properties of the drug, formulation characteristics and their interplay with gastrointestinal physiology and biology. Physiological-based pharmacokinetic (PBPK) models integrate all available information on gastro-intestinal system with drug and formulation data to predict oral drug absorption. The latter together with in vitro-in vivo extrapolation and other preclinical data on drug disposition can be used to predict plasma concentration-time profiles in silico. Despite recent successes of PBPK in many areas of drug development, an improvement in their utility for evaluating oral absorption is much needed. Current status of predictive performance, within the confinement of commonly available in vitro data on drugs and formulations alongside systems information, were tested using 3 PBPK software packages (GI -Sim (ver.4.1), Simcyp (R) Simulator (ver.15.0.86.0), and GastroPlus (TM) (ver.9.0.00xx)). This was part of the Innovative Medicines Initiative (IMI) Oral Biopharmaceutics Tools (OrBiTo) project. Fifty eight active pharmaceutical ingredients (APIs) were qualified from the OrBiTo database to be part of the investigation based on a priori set criteria on availability of minimum necessary information to allow modelling exercise. The set entailed over 200 human clinical studies with over 700 study arms. These were simulated using input parameters which had been harmonised by a panel of experts across different software packages prior to conduct of any simulation. Overall prediction performance and software packages comparison were evaluated based on performance indicators (Fold error (FE), Average fold error (AFE) and absolute average fold error (AAFE)) of pharmacokinetic (PK) parameters. On average, PK parameters (Area Under the Concentration-time curve (AUC(0-tlast)), Maximal concentration (C-max), half-life (t(1/2))) were predicted with AFE values between 1.11 and 1.97. Variability in FEs of these PK parameters was relatively high with AAFE values ranging from 2.08 to 2.74. Around half of the simulations were within the 2-fold error for AUC(0-tlast) and around 90% of the simulations were within 10-fold error for AUC(0-tlast). Oral bioavailability (F-oral) predictions, which were limited to 19 APIs having intravenous (i.v.) human data, showed AFE and AAFE of values 1.37 and 1.75 respectively. Across different APIs, AFE of AUC(0-tlast) predictions were between 0.22 and 22.76 with 70% of the APIs showing an AFE > 1. When compared across different formulations and routes of administration, AUC(0-tlast) for oral controlled release and i.v. administration were better predicted than that for oral immediate release formulations. Average predictive performance did not clearly differ between software packages but some APIs showed a high level of variability in predictive performance across different software packages. This variability could be related to several factors such as compound specific properties, the quality and availability of information, and errors in scaling from in vitro and preclinical in vivo data to human in vivo behaviour which will be explored further. Results were compared with previous similar exercise when the input data selection was carried by the modeller rather than a panel of experts on each in vitro test. Overall, average predictive performance was increased as reflected in smaller AAFE value of 2.8 as compared to AAFE value of 3.8 in case of previous exercise.

PURPOSE:The purpose of this study is to show how the Ocular Compartmental Absorption & Transit (OCAT™) model in GastroPlus® can be used to characterize ocular drug pharmacokinetic performance in rabbits for ointment formulations.METHODS:A newly OCAT™ model developed for fluorometholone, as well as a previously verified model for dexamethasone, were used to characterize the aqueous humor (AH) concentration following the administration of multiple ointment formulations to rabbit. The model uses the following parameters: application surface area (SA), a fitted application time, and the fitted Higuchi release constant to characterize the rate of passage of the active pharmaceutical ingredient from the ointment formulations into the tears in vivo.RESULTS:Parameter sensitivity analysis was performed to understand the impact of ointment formulation changes on ocular exposure. While application time was found to have a significant impact on the time of maximal concentration in AH, both the application SA and the Higuchi release constant significantly influenced both the maximum concentration and the ocular exposure.CONCLUSIONS:This initial model for ointment ophthalmic formulations is a first step to better understand the interplay between physiological factors and ophthalmic formulation physicochemical properties and their impact on in vivo ocular drug pharmacokinetic performance in rabbits.

Developing mathematical models to predict changes in ocular bioavailability and pharmacokinetics due to differences in the physicochemical properties of complex topical ophthalmic suspension formulations is important in drug product development and regulatory assessment. Herein, we used published FDA clinical pharmacology review data, in-house, and literature rabbit pharmacokinetic data generated for dexamethasone ophthalmic suspensions to demonstrate how the mechanistic Ocular Compartmental Absorption and Transit model by GastroPlus™ can be used to characterize ocular drug pharmacokinetic performance in rabbits for suspension formulations. This model was used to describe the dose-dependent (0.01 to 0.1%) non-linear pharmacokinetic in ocular tissues and characterize the impact of viscosity (1.67 to 72.9 cP) and particle size (5.5 to 22 μm) on in vivo ocular drug absorption and disposition. Parameter sensitivity analysis (hypothetical suspension particle size: 1 to 10 μm, viscosity: 1 to 100 cP) demonstrated that the interplay between formulation properties and physiological clearance through drainage and tear turnover rates in the pre-corneal compartment drives the ocular drug bioavailability. The quick removal of drug suspended particles from the pre-corneal compartment renders the impact of particle size inconsequential relative to viscosity modification. The in vivo ocular absorption is (1) viscosity non-sensitive when the viscosity is high and the impact of viscosity on the pre-corneal residence time reaches the maximum physiological system capacity or (2) viscosity sensitive when the viscosity is below a certain limit. This study reinforces our understanding of the interplay between physiological factors and ophthalmic formulation physicochemical properties and their impact on in vivo ocular drug PK performance in rabbits.

Cytosolic sulfotransferases (SULTs), including SULT1A, SULT1B, SULT1E, and SULT2A isoforms, play noteworthy roles in xenobiotic and endobiotic metabolism. We quantified the protein abundances of SULT1A1, SULT1A3, SULT1B1, and SULT2A1 in human liver cytosol samples (n = 194) by liquid chromatography-tandem mass spectrometry proteomics. The data were analyzed for their associations by age, sex, genotype, and ethnicity of the donors. SULT1A1, SULT1B1, and SULT2A1 showed significant age-dependent protein abundance, whereas SULT1A3 was invariable across 0-70 years. The respective mean abundances of SULT1A1, SULT1B1, and SULT2A1 in neonatal samples was 24%, 19%, and 38% of the adult levels. Interestingly, unlike UDP-glucuronosyltransferases and cytochrome P450 enzymes, SULT1A1 and SULT2A1 showed the highest abundance during early childhood (1 to <6 years), which gradually decreased by approx. 40% in adolescents and adults. SULT1A3 and SULT1B1 abundances were significantly lower in African Americans compared with Caucasians. Multiple linear regression analysis further confirmed the association of SULT abundances by age, ethnicity, and genotype. To demonstrate clinical application of the characteristic SULT ontogeny profiles, we developed and validated a proteomics-informed physiologically based pharmacokinetic model of acetaminophen. The latter confirmed the higher fractional contribution of sulfation over glucuronidation in the metabolism of acetaminophen in children. The study thus highlights that the ontogeny-based age-dependent fractional contribution (f(m)) of individual drug-metabolizing enzymes has better potential in prediction of drug-drug interactions and the effect of genetic polymorphisms in the pediatric population.

The application of preclinical in vitro and in silico models can help formulation scientists to predict the in vivo performance of a drug in an early stage of oral drug product development. An important aspect is that these models should include equations that represent mechanisms that are biorelevant and are sensitive to changes in parameter values. Human gastrointestinal physiology involves many processes that change as a function of time. In this work, a dynamic fluid and pH model was applied in GastroPlus™ to simulate intraluminal and systemic concentrations of the weak base posaconazole in a biorelevant manner. Simulated results were compared with observed data, extracted from a previously reported human in vivo gastrointestinal aspiration study. Three different formulations were explored (i.e., 1 solution [20 mg dose strength] and 2 suspensions [both 40 mg dose strength]). Simulated results were compared and in line with the observed results for different intraluminal (e.g., precipitated fraction) and systemic parameters (e.g., plasma Cmax). The optimization of the advanced compartmental and absorption transit model related to fluid dynamics and dynamic pH in this work creates perspectives to validate this model with other reference data derived from aspiration/magnetic resonance imaging studies.

When medicinal chemists need to improve bioavailability (%F) within a chemical series during lead optimization, they synthesize new series members with systematically modified properties mainly by following experience and general rules of thumb. More quantitative models that predict %F of proposed compounds from chemical structure alone have proven elusive. Global empirical %F quantitative structure-property (QSPR) models perform poorly, and projects have too little data to train local %F QSPR models. Mechanistic oral absorption and physiologically based pharmacokinetic (PBPK) models simulate the dissolution, absorption, systemic distribution, and clearance of a drug in preclinical species and humans. Attempts to build global PBPK models based purely on calculated inputs have not achieved the <2-fold average error needed to guide lead optimization. In this work, local GastroPlus PBPK models are instead customized for individual medchem series. The key innovation was building a local QSPR for a numerically fitted effective intrinsic clearance (CLloc). All inputs are subsequently computed from structure alone, so the models can be applied in advance of synthesis. Training CLloc on the first 15-18 rat %F measurements gave adequate predictions, with clear improvements up to about 30 measurements, and incremental improvements beyond that.

When medicinal chemists need to improve oral bioavailability (%F) during lead optimization, they systematically modify compound properties mainly based on their own experience and general rules of thumb. However, at least a dozen properties can influence %F, and the difficulty of multiparameter optimization for such complex nonlinear processes grows combinatorially with the number of variables. Furthermore, strategies can be in conflict. For example, adding a polar or charged group will generally increase solubility but decrease permeability. Identifying the 2 or 3 properties that most influence %F for a given compound series would make %F optimization much more efficient. We previously reported an adaptation of physiologically based pharmacokinetic (PBPK) simulations to predict %F for lead series from purely computational inputs within a 2-fold average error. Here, we run thousands of such simulations to generate a comprehensive "bioavailability landscape" for each series. A key innovation was recognition that the large and variable number of p Ka's in drug molecules could be replaced by just the two straddling the isoelectric point. Another was use of the ZINC database to cull out chemically inaccessible regions of property space. A quadratic partial least squares regression (PLS) accurately fits a continuous surface to these thousands of bioavailability predictions. The PLS coefficients indicate the globally sensitive compound properties. The PLS surface also displays the %F landscape in these sensitive properties locally around compounds of particular interest. Finally, being quick to calculate, the PLS equation can be combined with models for activity and other properties for multiobjective lead optimization.

New approaches are needed to assess the effects of inhaled substances on human health. These approaches will be based on mechanisms of toxicity, an understanding of dosimetry, and the use of in silico modeling and in vitro test methods. In order to accelerate wider implementation of such approaches, development of adverse outcome pathways (AOPs) can help identify and address gaps in our understanding of relevant parameters for model input and mechanisms, and optimize non-animal approaches that can be used to investigate key events of toxicity. This paper describes the AOPs and the toolbox of in vitro and in silico models that can be used to assess the key events leading to toxicity following inhalation exposure. Because the optimal testing strategy will vary depending on the substance of interest, here we present a decision tree approach to identify an appropriate non-animal integrated testing strategy that incorporates consideration of a substance's physicochemical properties, relevant mechanisms of toxicity, and available in silico models and in vitro test methods. This decision tree can facilitate standardization of the testing approaches. Case study examples are presented to provide a basis for proof-of-concept testing to illustrate the utility of non-animal approaches to inform hazard identification and risk assessment of humans exposed to inhaled substances.