Development of QSAR models to predict blood-brain barrier permeability

Novel (Q)SAR models for prediction of reversible and time-dependent inhibition of cytochrome P450 enzymes

The 2020 FDA drug-drug interaction (DDI) guidance includes a consideration for metabolites with structural alerts for potential mechanism-based inhibition (MBI) and describes how this information may be used to determine whether in vitro studies need to be conducted to evaluate the inhibitory potential of a metabolite on CYP enzymes. To facilitate identification of structural alerts, an extensive literature search was performed and alerts for mechanism-based inhibition of cytochrome P450 enzymes (CYP) were collected. Furthermore, five quantitative structure-activity relationship (QSAR) models were developed to predict not only time-dependent inhibition of CYP3A4, an enzyme that metabolizes approximately 50% of all marketed drugs, but also reversible inhibition of 3A4, 2C9, 2C19 and 2D6. The non-proprietary training database for the QSAR models contains data for 10,129 chemicals harvested from FDA drug approval packages and published literature. The cross-validation performance statistics for the new CYP QSAR models range from 78% to 84% sensitivity and 79%-84% normalized negative predictivity. Additionally, the performance of the newly developed QSAR models was assessed using external validation sets. Overall performance statistics showed up to 75% in sensitivity and up to 80% in normalized negative predictivity. The newly developed models will provide a faster and more effective evaluation of potential drug-drug interaction caused by metabolites.

Biomimetic Chromatography/QSAR Investigations in Modeling Properties Influencing the Biological Efficacy of Phenoxyacetic Acid-Derived Congeners

A hybrid method-combining liquid biomimetic chromatography techniques (immobilized artificial membrane chromatography and biopartitioning micellar chromatography) and Quantitative Structure-Activity Relationships-was used to derive helpful models for predicting selected biological properties such as penetration through the plant cuticle, the skin and the blood-brain barrier, and binding to human serum albumin of phenoxyacetic acid-derived congeners regarded as potential herbicides. Reliable, high-concept models were developed indicating the lipophilicity, polarizability, and sum of hydrogen bond donors and acceptors as properties that determine the biological efficacy of the title compounds. These models were validated by leave-one-out cross-validation. Modeling the toxicity of phenoxyacetic acid-derived congeners to red blood cells allowed the identification of the most toxic substances as well as those molecular descriptors that determine their hemolytic properties.

Prediction of the Extent of Blood-Brain Barrier Transport Using Machine Learning and Integration into the LeiCNS-PK3.0 Model

INTRODUCTION

The unbound brain-to-plasma partition coefficient (Kp,uu,BBB) is an essential parameter for predicting central nervous system (CNS) drug disposition using physiologically-based pharmacokinetic (PBPK) modeling. Kp,uu,BBB values for specific compounds are however often unavailable, and are moreover time consuming to obtain experimentally. The aim of this study was to develop a quantitative structure-property relationship (QSPR) model to predict the Kp,uu,BBB and to demonstrate how QSPR-model predictions can be integrated into a physiologically-based pharmacokinetic model for the CNS.

METHODS

Rat Kp,uu,BBB values were obtained for 98 compounds from literature or in house historical data. For all compounds, 2D and 3D physico-chemical and structural properties were derived using the Molecular Operating Environment (MOE) software. Multiple machine learning (ML) regression models were compared for prediction of the Kp,uu,BBB, including random forest, support vector machines, K-nearest neighbors, and (sparse-) partial least squares. Finally, we demonstrate how the developed QSPR model predictions can be integrated into a CNS PBPK modeling workflow.

RESULTS

Among all ML algorithms, a random forest showed the best predictive performance for Kp,uu,BBB on test data with R2 value of 0.61 and 61% of all predictions were within twofold error. The obtained Kp,uu,BBB were successfully integrated into the LeiCNS-PK3.0 CNS PBPK model.

CONCLUSIONS

The developed random forest QSPR model for Kp,uu,BBB prediction was found to have adequate performance, and can support drug discovery and development of novel investigational drugs targeting the CNS in conjunction with CNS PBPK modeling.

Transparent Machine Learning Model to Understand Drug Permeability through the Blood-Brain Barrier

The blood-brain barrier (BBB) selectively regulates the passage of chemical compounds into and out of the central nervous system (CNS). As such, understanding the permeability of drug molecules through the BBB is key to treating neurological diseases and evaluating the response of the CNS to medical treatments. Within the last two decades, a diverse portfolio of machine learning (ML) models have been regularly utilized as a tool to predict, and, to a much lesser extent, understand, several functional properties of medicinal drugs, including their propensity to pass through the BBB. However, the most numerically accurate models to date lack in transparency, as they typically rely on complex blends of different descriptors (or features or fingerprints), many of which are not necessarily interpretable in a straightforward fashion. In fact, the "black-box" nature of these models has prevented us from pinpointing any specific design rule to craft the next generation of pharmaceuticals that need to pass (or not) through the BBB. In this work, we have developed a ML model that leverages an uncomplicated, transparent set of descriptors to predict the permeability of drug molecules through the BBB. In addition to its simplicity, our model achieves comparable results in terms of accuracy compared to state-of-the-art models. Moreover, we use a naive Bayes model as an analytical tool to provide further insights into the structure-function relation that underpins the capacity of a given drug molecule to pass through the BBB. Although our results are computational rather than experimental, we have identified several molecular fragments and functional groups that may significantly impact a drug's likelihood of permeating the BBB. This work provides a unique angle to the BBB problem and lays the foundations for future work aimed at leveraging additional transparent descriptors, potentially obtained via bespoke molecular dynamics simulations.

Log BB Prediction Models Using TLC and HPLC Retention Values as Protein Affinity Data

BACKGROUND

The penetration of drugs through the blood-brain barrier is one of the key pharmacokinetic aspects of centrally acting active substances and other drugs in terms of the occurrence of side effects on the central nervous system. In our research, several regression models were constructed in order to observe the connections between the active pharmaceutical ingredients' properties and their bioavailability in the CNS, presented in the form of the log BB parameter, which refers to the drug concentration on both sides of the blood-brain barrier.

METHODS

Predictive models were created using the physicochemical properties of drugs, and multiple linear regression and a data mining method, i.e., MARSplines, were used to build them. Retention values from protein-affinity chromatography (TLC and HPLC) were introduced into the analyses. In both experiments, the stationary phases were modified with serum albumin, which enriched the obtained chromatographic data, and were then introduced into the models with good results.

RESULTS

The conducted analyses confirm that the variables that influence the log BB include high degree of lipophilicity, ionisation capacity and low capability of forming hydrogen bonds. However, the addition of chromatographic data improved the obtained regression results and increased the robustness of the models against an increased number of cases. The linear regression model with chromatographic parameters explains 85% of the log bb variability, whereas the MARSplines model explains 91%. Conclusions: Based on the obtained results, it can be concluded that the use of chromatographic data can increase the robustness of predictive regression models related to penetration through biological barriers.

Synergizing Machine Learning, Conceptual Density Functional Theory, and Biochemistry: No-Code Explainable Predictive Models for Mutagenicity in Aromatic Amines

This study synergizes machine learning (ML) with conceptual density functional theory (CDFT) to develop OECD-compliant predictive models for the mutagenic activity of aromatic amines (AAs) with a fully No-Code methodology using a comprehensive data set of 251 AAs, Leave-One-Out-Cross-Validation (LOOCV), and three distinct data splits. Our research employs the GFN2-xTB method, known for its robustness and speed, to compute descriptors for procarcinogens and their activated metabolites in vacuum and aqueous phases. We evaluate the effectiveness of different theoretical definitions of electrophilicity within CDFT, namely, PSL, GCV, and CDP schemes, and the newly introduced Log QP descriptor to approximate Log P information. SPAARC, RandomTree, and JCHAID* ML methods were used to build explainable predictive models with highly robust internal validation (Avg. Correct Classifications = 76% and Avg. Kappa = 0.29) and external validation (Avg. Correct Classifications = 79% and Avg. Kappa = 0.33) metrics, and the results were compared to those of a two hidden layer Multilayer Perceptron. The results indicate that the second CDP definition for the electrophilicity in both vacuum and aqueous phases and also the newly presented Log QP descriptors are the most important ones for predicting the mutagenic activity of AA (namely ω+VacCDP2+, ω+AqCDP2+, and LogQP1+Vac, respectively). The results indicate that metabolic activation, aqueous solvent properties, and the CDP electrophilicity schemes and Log QP should be considered when building predictive models for the mutagenic activity of AA. This study offers a replicable, No-Code approach to QSAR research, making high-level ML and CDFT applications accessible to a broader audience. Future work will expand these methods to other compound families, enhancing predictive capabilities in the study of mutagenic activities and other biological phenomena.

BrainPepPass: A Framework Based on Supervised Dimensionality Reduction for Predicting Blood-Brain Barrier-Penetrating Peptides

Peptides that pass through the blood-brain barrier (BBB) not only are implicated in brain-related pathologies but also are promising therapeutic tools for treating brain diseases, e.g., as shuttles carrying active medicines across the BBB. Computational prediction of BBB-penetrating peptides (B3PPs) has emerged as an interesting approach because of its ability to screen large peptide libraries in a cost-effective manner. In this study, we present BrainPepPass, a machine learning (ML) framework that utilizes supervised manifold dimensionality reduction and extreme gradient boosting (XGB) algorithms to predict natural and chemically modified B3PPs. The results indicate that the proposed tool outperforms other classifiers, with average accuracies exceeding 94% and 98% in 10-fold cross-validation and leave-one-out cross-validation (LOOCV), respectively. In addition, accuracy values ranging from 45% to 97.05% were achieved in the independent tests. The BrainPepPass tool is available in a public repository for academic use (https://github.com/ewerton-cristhian/BrainPepPass).

Designing drugs optimized for both blood-brain barrier permeation and intra-cerebral partition

INTRODUCTION

With the increasing incidence and prevalence of neurological disorders globally, there is a paramount need for new pharmacotherapies. BBB effectively protects the brain but raises a profound challenge to drug permeation, with less than 2% of most drugs reaching the CNS.

AREAS COVERED

This article reviews aspects of the most recent design strategies, providing insights into ideas and concepts in CNS drug discovery. An overview of the products available on the market is given and why clinical trials are continuously failing is discussed.

EXPERT OPINION

Among the available CNS drugs, small molecules account for most successful CNS therapeutics due to their ability to penetrate the BBB through passive or carrier-mediated mechanisms. The development of new CNS drugs is very difficult. To date, there is a lack of effective drugs for alleviating or even reversing the progression of brain diseases. Particularly, the use of artificial intelligence strategies, together with more appropriate animal models, may enable the design of molecules with appropriate permeation, to elicit a biological response from the neurotherapeutic target.

Modeling the Blood-Brain Barrier Permeability of Potential Heterocyclic Drugs via Biomimetic IAM Chromatography Technique Combined with QSAR Methodology

Penetration through the blood-brain barrier (BBB) is desirable in the case of potential pharmaceuticals acting on the central nervous system (CNS), but is undesirable in the case of drug candidates acting on the peripheral nervous system because it may cause CNS side effects. Therefore, modeling of the permeability across the blood-brain barrier (i.e., the logarithm of the brain to blood concentration ratio, log BB) of potential pharmaceuticals should be performed as early as possible in the preclinical phase of drug development. Biomimetic chromatography with immobilized artificial membrane (IAM) and the quantitative structure-activity relationship (QSAR) methodology were successful in modeling the blood-brain barrier permeability of 126 drug candidates, whose experimentally-derived lipophilicity indices and computationally-derived molecular descriptors (such as molecular weight (MW), number of rotatable bonds (NRB), number of hydrogen bond donors (HBD), number of hydrogen bond acceptors (HBA), topological polar surface area (TPSA), and polarizability (α)) varied by class. The QSARs model established by multiple linear regression showed a positive effect of the lipophilicity (log kw, IAM) and molecular weight of the compound, and a negative effect of the number of hydrogen bond donors and acceptors, on the log BB values. The model has been cross-validated, and all statistics indicate that it is very good and has high predictive ability. The simplicity of the developed model, and its usefulness in screening studies of novel drug candidates that are able to cross the BBB by passive diffusion, are emphasized.

Molecular Design of Magnetic Resonance Imaging Agents Binding to Amyloid Deposits

The ability to detect and monitor amyloid deposition in the brain using non-invasive imaging techniques provides valuable insights into the early diagnosis and progression of Alzheimer's disease and helps to evaluate the efficacy of potential treatments. Magnetic resonance imaging (MRI) is a widely available technique offering high-spatial-resolution imaging. It can be used to visualize amyloid deposits with the help of amyloid-binding diagnostic agents injected into the body. In recent years, a number of amyloid-targeted MRI probes have been developed, but none of them has entered clinical practice. We review the advances in the field and deduce the requirements for the molecular structure and properties of a diagnostic probe candidate. These requirements make up the base for the rational design of MRI-active small molecules targeting amyloid deposits. Particular attention is paid to the novel cryo-EM structures of the fibril aggregates and their complexes, with known binders offering the possibility to use computational structure-based design methods. With continued research and development, MRI probes may revolutionize the diagnosis and treatment of neurodegenerative diseases, ultimately improving the lives of millions of people worldwide.

Chromatographic Data in Statistical Analysis of BBB Permeability Indices

Blood-brain barrier (BBB) permeability is an essential phenomena when considering the treatment of neurological disorders as well as in the case of central nervous system (CNS) adverse effects caused by peripherally acting drugs. The presented work contains statistical analyses and the correlation assessment of the analyzed group of active pharmaceutical ingredients (APIs) with their BBB-permeability data collected from the literature (such as computational log BB; Kp,uu,brain, and CNS+/- groups). A number of regression models were constructed in order to observe the connections between the APIs' physicochemical properties in combination with their retention data from the chromatographic experiments (TLC and HPLC) and the indices of bioavailability in the CNS. Conducted analyses confirm that descriptors significant in BBB permeability modeling are hydrogen bond acceptors and donors, physiological charge, or energy of the lowest unoccupied molecular orbital. These molecular descriptors were the basis, along with the chromatographic data from the TLC in log BB regression analyses. Normal-phase TLC data showed a significant contribution to the creation of the log BB regression model using the multiple linear regression method. The model using them showed a good predictive value at the level of R² = 0.87. Models for Kp,uu,brain resulted in lower statistics: R² = 0.56 for the group of 23 APIs with the participation of k IAM.

Computer-Aided Drug Design towards New Psychotropic and Neurological Drugs

Central nervous system (CNS) disorders are a therapeutic area in drug discovery where demand for new treatments greatly exceeds approved treatment options. This is complicated by the high failure rate in late-stage clinical trials, resulting in exorbitant costs associated with bringing new CNS drugs to market. Computer-aided drug design (CADD) techniques minimise the time and cost burdens associated with drug research and development by ensuring an advantageous starting point for pre-clinical and clinical assessments. The key elements of CADD are divided into ligand-based and structure-based methods. Ligand-based methods encompass techniques including pharmacophore modelling and quantitative structure activity relationships (QSARs), which use the relationship between biological activity and chemical structure to ascertain suitable lead molecules. In contrast, structure-based methods use information about the binding site architecture from an established protein structure to select suitable molecules for further investigation. In recent years, deep learning techniques have been applied in drug design and present an exciting addition to CADD workflows. Despite the difficulties associated with CNS drug discovery, advances towards new pharmaceutical treatments continue to be made, and CADD has supported these findings. This review explores various CADD techniques and discusses applications in CNS drug discovery from 2018 to November 2022.

New science, drug regulation, and emergent public health issues: The work of FDA's division of applied regulatory science

The U.S. Food and Drug Administration (FDA) Division of Applied Regulatory Science (DARS) moves new science into the drug review process and addresses emergent regulatory and public health questions for the Agency. By forming interdisciplinary teams, DARS conducts mission-critical research to provide answers to scientific questions and solutions to regulatory challenges. Staffed by experts across the translational research spectrum, DARS forms synergies by pulling together scientists and experts from diverse backgrounds to collaborate in tackling some of the most complex challenges facing FDA. This includes (but is not limited to) assessing the systemic absorption of sunscreens, evaluating whether certain drugs can convert to carcinogens in people, studying drug interactions with opioids, optimizing opioid antagonist dosing in community settings, removing barriers to biosimilar and generic drug development, and advancing therapeutic development for rare diseases. FDA tasks DARS with wide ranging issues that encompass regulatory science; DARS, in turn, helps the Agency solve these challenges. The impact of DARS research is felt by patients, the pharmaceutical industry, and fellow regulators. This article reviews applied research projects and initiatives led by DARS and conducts a deeper dive into select examples illustrating the impactful work of the Division.