Improving the prediction of drug disposition in the brain

Ensemble modeling with machine learning and deep learning to provide interpretable generalized rules for classifying CNS drugs with high prediction power

The trade-off between a machine learning (ML) and deep learning (DL) model's predictability and its interpretability has been a rising concern in central nervous system-related quantitative structure-activity relationship (CNS-QSAR) analysis. Many state-of-the-art predictive modeling failed to provide structural insights due to their black box-like nature. Lack of interpretability and further to provide easy simple rules would be challenging for CNS-QSAR models. To address these issues, we develop a protocol to combine the power of ML and DL to generate a set of simple rules that are easy to interpret with high prediction power. A data set of 940 market drugs (315 CNS-active, 625 CNS-inactive) with support vector machine and graph convolutional network algorithms were used. Individual ML/DL modeling methods were also constructed for comparison. The performance of these models was evaluated using an additional external dataset of 117 market drugs (42 CNS-active, 75 CNS-inactive). Fingerprint-split validation was adopted to ensure model stringency and generalizability. The resulting novel hybrid ensemble model outperformed other constituent traditional QSAR models with an accuracy of 0.96 and an F1 score of 0.95. With the power of the interpretability provided with this protocol, our model laid down a set of simple physicochemical rules to determine whether a compound can be a CNS drug using six sub-structural features. These rules displayed higher classification ability than classical guidelines, with higher specificity and more mechanistic insights than just for blood-brain barrier permeability. This hybrid protocol can potentially be used for other drug property predictions.

Prediction of the Blood-Brain Barrier (BBB) Permeability of Chemicals Based on Machine-Learning and Ensemble Methods

The ability of chemicals to enter the blood-brain barrier (BBB) is a key factor for central nervous system (CNS) drug development. Although many models for BBB permeability prediction have been developed, they have insufficient accuracy (ACC) and sensitivity (SEN). To improve performance, ensemble models were built to predict the BBB permeability of compounds. In this study, in silico ensemble-learning models were developed using 3 machine-learning algorithms and 9 molecular fingerprints from 1757 chemicals (integrated from 2 published data sets) to predict BBB permeability. The best prediction performance of the base classifier models was achieved by a prediction model based on an random forest (RF) and a MACCS molecular fingerprint with an ACC of 0.910, an area under the receiver-operating characteristic (ROC) curve (AUC) of 0.957, a SEN of 0.927, and a specificity of 0.867 in 5-fold cross-validation. The prediction performance of the ensemble models is better than that of most of the base classifiers. The final ensemble model has also demonstrated good accuracy for an external validation and can be used for the early screening of CNS drugs.

Adult precision medicine: learning from the past to enhance the future

Despite therapeutic advances for other malignancies, gliomas remain challenging solid tumors to treat. Complete surgical resection is nearly impossible due to gliomas’ diffuse infiltrative nature, and treatment is hampered by restricted access to the tumors due to limited transport across the blood–brain barrier. Recent advances in genomic studies and next-generation sequencing techniques have led to a better understanding of gliomas and identification of potential aberrant signaling pathways. Targeting the specific genomic abnormalities via novel molecular therapies has opened a new avenue in the management of gliomas, with encouraging results in preclinical studies and early clinical trials. However, molecular characterization of gliomas revealed significant heterogeneity, which poses a challenge for targeted therapeutic approaches. In this context, leading neuro-oncology researchers and clinicians, industry innovators, and patient advocates convened at the inaugural annual Remission Summit held in Orlando, FL in February 2019 to discuss the latest advances in immunotherapy and precision medicine approaches for the treatment of adult and pediatric brain tumors and outline the unanswered questions, challenges, and opportunities that lay ahead for advancing the duration and quality of life for patients with brain tumors. Here, we provide historical context for precision medicine in other cancers, present emerging approaches for gliomas, discuss their limitations, and outline the steps necessary for future success. We focus on the advances in small molecule targeted therapy, as the use of immunotherapy as an emerging precision medicine modality for glioma treatment has recently been reviewed by our colleagues.

Crocetin promotes clearance of amyloid-β by inducing autophagy via the STK11/LKB1-mediated AMPK pathway

Alzheimer disease (AD) is usually accompanied by two prominent pathological features, cerebral accumulation of amyloid-β (Aβ) plaques and presence of MAPT/tau neurofibrillary tangles. Dysregulated clearance of Aβ largely contributes to its accumulation and plaque formation in the brain. Macroautophagy/autophagy is a lysosomal degradative process, which plays an important role in the clearance of Aβ. Failure of autophagic clearance of Aβ is currently acknowledged as a contributing factor to increased accumulation of Aβ in AD brains. In this study, we have identified crocetin, a pharmacologically active constituent from the flower stigmas of Crocus sativus, as a potential inducer of autophagy in AD. In the cellular model, crocetin induced autophagy in N9 microglial and primary neuron cells through STK11/LKB1 (serine/threonine kinase 11)-mediated AMP-activated protein kinase (AMPK) pathway activation. Autophagy induction by crocetin significantly increased Aβ clearance in N9 cells. Moreover, crocetin crossed the blood-brain barrier and induced autophagy in the brains' hippocampi of wild-type male C57BL/6 mice. Further studies in transgenic male 5XFAD mice, as a model of AD, revealed that one-month treatment with crocetin significantly reduced Aβ levels and neuroinflammation in the mice brains and improved memory function by inducing autophagy that was mediated by AMPK pathway activation. Our findings support further development of crocetin as a pharmacological inducer of autophagy to prevent, slow down progression, and/or treat AD.

Towards Deep Neural Network Models for the Prediction of the Blood-Brain Barrier Permeability for Diverse Organic Compounds

Permeation through the blood–brain barrier (BBB) is among the most important processes controlling the pharmacokinetic properties of drugs and other bioactive compounds. Using the fragmental (substructural) descriptors representing the occurrence number of various substructures, as well as the artificial neural network approach and the double cross-validation procedure, we have developed a predictive in silico LogBB model based on an extensive and verified dataset (529 compounds), which is applicable to diverse drugs and drug-like compounds. The model has good predictivity parameters (Q2=0.815, RMSEcv=0.318) that are similar to or better than those of the most reliable models available in the literature. Larger datasets, and perhaps more sophisticated network architectures, are required to realize the full potential of deep neural networks. The analysis of fragment contributions reveals patterns of influence consistent with the known concepts of structural characteristics that affect the BBB permeability of organic compounds. The external validation of the model confirms good agreement between the predicted and experimental LogBB values for most of the compounds. The model enables the evaluation and optimization of the BBB permeability of potential neuroactive agents and other drug compounds.

The Brain Exposure Efficiency (BEE) Score

The blood-brain barrier (BBB), composed of microvascular tight junctions and glial cell sheathing, selectively controls drug permeation into the central nervous system (CNS) by either passive diffusion or active transport. Computational techniques capable of predicting molecular brain penetration are important to neurological drug design. A novel prediction algorithm, termed the Brain Exposure Efficiency Score (BEE), is presented. BEE addresses the need to incorporate the role of trans-BBB influx and efflux active transporters by considering key brain penetrance parameters, namely, steady state unbound brain to plasma ratio of drug (K_p,uu) and dose normalized unbound concentration of drug in brain (C_u,b). BEE was devised using quantitative structure-activity relationships (QSARs) and molecular modeling studies on known transporter proteins and their ligands. The developed algorithms are provided as a user-friendly open source calculator to assist in optimizing a brain penetrance strategy during the early phases of small molecule molecular therapeutic design.

Examining pharmacodynamic and pharmacokinetic properties of eleven analogues of saquinavir for HIV protease inhibition

HIV is one of the most lethal viral diseases in the human population. Patients often suffer from drug resistance, which hampers HIV therapy. Eleven different structural analogues of saquinavir (SQV), designed using ChemSketch™ and named S1 through S11, were compared with SQV with respect to their pharmacodynamic and pharmacokinetic properties. Pharmacokinetic predictions were carried out using AutoDock, and molecular docking between macromolecule HIV protease (PDB ID: 3IXO) and analogues S1 - S11 as ligands was performed. Analogues S1, S3, S4, S9 and S11 had lower binding scores when compared with saquinavir, whereas that of analogue S5 was similar. Pharmacokinetic predictions made using ACDilab2, including the Lipinski profile, general physical features, absorption, distribution, metabolism and excretion parameters, and toxicity values, for the eleven analogues and SQV suggested that S1 and S5 are pharmacodynamically and pharmacokinetically robust molecules that could be developed and established as lead molecules after in vitro and in vivo studies.

Developments in Blood-Brain Barrier Penetrance and Drug Repurposing for Improved Treatment of Glioblastoma

Glioblastoma (GBM) is one of the most common, deadly, and difficult-to-treat adult brain tumors. Surgical removal of the tumor, followed by radiotherapy (RT) and temozolomide (TMZ) administration, is the current treatment modality, but this regimen only modestly improves overall patient survival. Invasion of cells into the surrounding healthy brain tissue prevents complete surgical resection and complicates treatment strategies with the goal of preserving neurological function. Despite significant efforts to increase our understanding of GBM, there have been relatively few therapeutic advances since 2005 and even fewer treatments designed to effectively treat recurrent tumors that are resistant to therapy. Thus, while there is a pressing need to move new treatments into the clinic, emerging evidence suggests that key features unique to GBM location and biology, the blood-brain barrier (BBB) and intratumoral molecular heterogeneity, respectively, stand as critical unresolved hurdles to effective therapy. Notably, genomic analyses of GBM tissues has led to the identification of numerous gene alterations that govern cell growth, invasion and survival signaling pathways; however, the drugs that show pre-clinical potential against signaling pathways mediated by these gene alterations cannot achieve effective concentrations at the tumor site. As a result, identifying BBB-penetrating drugs and utilizing new and safer methods to enhance drug delivery past the BBB has become an area of intensive research. Repurposing and combining FDA-approved drugs with evidence of penetration into the central nervous system (CNS) has also seen new interest for the treatment of both primary and recurrent GBM. In this review, we discuss emerging methods to strategically enhance drug delivery to GBM and repurpose currently-approved and previously-studied drugs using rational combination strategies.

Improved Prediction of Blood-Brain Barrier Permeability Through Machine Learning with Combined Use of Molecular Property-Based Descriptors and Fingerprints

Blood-brain barrier (BBB) permeability of a compound determines whether the compound can effectively enter the brain. It is an essential property which must be accounted for in drug discovery with a target in the brain. Several computational methods have been used to predict the BBB permeability. In particular, support vector machine (SVM), which is a kernel-based machine learning method, has been used popularly in this field. For SVM training and prediction, the compounds are characterized by molecular descriptors. Some SVM models were based on the use of molecular property-based descriptors (including 1D, 2D, and 3D descriptors) or fragment-based descriptors (known as the fingerprints of a molecule). The selection of descriptors is critical for the performance of a SVM model. In this study, we aimed to develop a generally applicable new SVM model by combining all of the features of the molecular property-based descriptors and fingerprints to improve the accuracy for the BBB permeability prediction. The results indicate that our SVM model has improved accuracy compared to the currently available models of the BBB permeability prediction.

The impact of in silico screening in the discovery of novel and safer drug candidates

Drug discovery is a multidisciplinary and multivariate optimization endeavor. As such, in silico screening tools have gained considerable importance to archive, analyze and exploit the vast and ever-increasing amount of experimental data generated throughout the process. The current review will focus on the computer-aided prediction of the numerous properties that need to be controlled during the discovery of a preliminary hit and its promotion to a viable clinical candidate. It does not pretend to the almost impossible task of an exhaustive report but will highlight a few key points that need to be collectively addressed both by chemists and biologists to fuel the drug discovery pipeline with innovative and safe drug candidates.

Blood-brain barrier, cytotoxic chemotherapies and glioblastoma

INTRODUCTION

Glioblastomas (GBM) are the most common and aggressive primary malignant brain tumors in adults. The blood brain barrier (BBB) is a major limitation reducing efficacy of anti-cancer drugs in the treatment of GBM patients. Areas covered: Virtually all GBM recur after the first-line treatment, at least partly, due to invasive tumor cells protected from chemotherapeutic agents by the intact BBB in the brain adjacent to tumor. The passage through the BBB, taken by antitumor drugs, is poorly and heterogeneously documented in the literature. In this review, we have focused our attention on: (i) the BBB, (ii) the passage of chemotherapeutic agents across the BBB and (iii) the strategies investigated to overcome this barrier. Expert commentary: A better preclinical knowledge of the crossing of the BBB by antitumor drugs will allow optimizing their clinical development, alone or combined with BBB bypassing strategies, towards an increased success rate of clinical trials.

A Genetic Algorithm Based Support Vector Machine Model for Blood-Brain Barrier Penetration Prediction

Blood-brain barrier (BBB) is a highly complex physical barrier determining what substances are allowed to enter the brain. Support vector machine (SVM) is a kernel-based machine learning method that is widely used in QSAR study. For a successful SVM model, the kernel parameters for SVM and feature subset selection are the most important factors affecting prediction accuracy. In most studies, they are treated as two independent problems, but it has been proven that they could affect each other. We designed and implemented genetic algorithm (GA) to optimize kernel parameters and feature subset selection for SVM regression and applied it to the BBB penetration prediction. The results show that our GA/SVM model is more accurate than other currently available log BB models. Therefore, to optimize both SVM parameters and feature subset simultaneously with genetic algorithm is a better approach than other methods that treat the two problems separately. Analysis of our log BB model suggests that carboxylic acid group, polar surface area (PSA)/hydrogen-bonding ability, lipophilicity, and molecular charge play important role in BBB penetration. Among those properties relevant to BBB penetration, lipophilicity could enhance the BBB penetration while all the others are negatively correlated with BBB penetration.

In silico ADME/T modelling for rational drug design

In recent decades, in silico absorption, distribution, metabolism, excretion (ADME), and toxicity (T) modelling as a tool for rational drug design has received considerable attention from pharmaceutical scientists, and various ADME/T-related prediction models have been reported. The high-throughput and low-cost nature of these models permits a more streamlined drug development process in which the identification of hits or their structural optimization can be guided based on a parallel investigation of bioavailability and safety, along with activity. However, the effectiveness of these tools is highly dependent on their capacity to cope with needs at different stages, e.g. their use in candidate selection has been limited due to their lack of the required predictability. For some events or endpoints involving more complex mechanisms, the current in silico approaches still need further improvement. In this review, we will briefly introduce the development of in silico models for some physicochemical parameters, ADME properties and toxicity evaluation, with an emphasis on the modelling approaches thereof, their application in drug discovery, and the potential merits or deficiencies of these models. Finally, the outlook for future ADME/T modelling based on big data analysis and systems sciences will be discussed.

Molecular determinants of blood-brain barrier permeation

The blood-brain barrier (BBB) is a microvascular unit which selectively regulates the permeability of drugs to the brain. With the rise in CNS drug targets and diseases, there is a need to be able to accurately predict a priori which compounds in a company database should be pursued for favorable properties. In this review, we will explore the different computational tools available today, as well as underpin these to the experimental methods used to determine BBB permeability. These include in vitro models and the in vivo models that yield the dataset we use to generate predictive models. Understanding of how these models were experimentally derived determines our accurate and predicted use for determining a balance between activity and BBB distribution.

Toward precision medicine in glioblastoma: the promise and the challenges

Integrated sequencing strategies have provided a broader understanding of the genomic landscape and molecular classifications of multiple cancer types and have identified various therapeutic opportunities across cancer subsets. Despite pivotal advances in the characterization of genomic alterations in glioblastoma, targeted agents have shown minimal efficacy in clinical trials to date, and patient survival remains poor. In this review, we highlight potential reasons why targeting single alterations has yielded limited clinical efficacy in glioblastoma, focusing on issues of tumor heterogeneity and pharmacokinetic failure. We outline strategies to address these challenges in applying precision medicine to glioblastoma and the rationale for applying targeted combination therapy approaches that match genomic alterations with compounds accessible to the central nervous system.

Blood-brain barrier models and their relevance for a successful development of CNS drug delivery systems: a review

During the research and development of new drugs directed at the central nervous system, there is a considerable attrition rate caused by their hampered access to the brain by the blood-brain barrier. Throughout the years, several in vitro models have been developed in an attempt to mimic critical functionalities of the blood-brain barrier and reliably predict the permeability of drug candidates. However, the current challenge lies in developing a model that retains fundamental blood-brain barrier characteristics and simultaneously remains compatible with the high throughput demands of pharmaceutical industries. This review firstly describes the roles of all elements of the neurovascular unit and their influence on drug brain penetration. In vitro models, including non-cell based and cell-based models, and in vivo models are herein presented, with a particular emphasis on their methodological aspects. Lastly, their contribution to the improvement of brain drug delivery strategies and drug transport across the blood-brain barrier is also discussed.