Drug-induced liver injury severity and toxicity (DILIst): binary classification of 1279 drugs by human hepatotoxicity

Identification of reversible OATP1B1 and time-dependent CYP3A4 inhibition as the major risk factors for drug-induced cholestasis (DIC)

Hepatic bile acid regulation is a multifaceted process modulated by several hepatic transporters and enzymes. Drug-induced cholestasis (DIC), a main type of drug-induced liver injury (DILI), denotes any drug-mediated condition in which hepatic bile flow is impaired. Our ability in translating preclinical toxicological findings to human DIC risk is currently very limited, mainly due to important interspecies differences. Accordingly, the anticipation of clinical DIC with available in vitro or in silico models is also challenging, due to the complexity of the bile acid homeostasis. Herein, we assessed the in vitro inhibition potential of 47 marketed drugs with various degrees of reported DILI severity towards all metabolic and transport mechanisms currently known to be involved in the hepatic regulation of bile acids. The reported DILI concern and/or cholestatic annotation correlated with the number of investigated processes being inhibited. Furthermore, we employed univariate and multivariate statistical methods to determine the important processes for DILI discrimination. We identified time-dependent inhibition (TDI) of cytochrome P450 (CYP) 3A4 and reversible inhibition of the organic anion transporting polypeptide (OATP) 1B1 as the major risk factors for DIC among the tested mechanisms related to bile acid transport and metabolism. These results were consistent across multiple statistical methods and DILI classification systems applied in our dataset. We anticipate that our assessment of the two most important processes in the development of cholestasis will enable a risk assessment for DIC to be efficiently integrated into the preclinical development process.

Improved Detection of Drug-Induced Liver Injury by Integrating Predicted In Vivo and In Vitro Data

Drug-induced liver injury (DILI) has been a significant challenge in drug discovery, often leading to clinical trial failures and necessitating drug withdrawals. Over the last decade, the existing suite of in vitro proxy-DILI assays has generally improved at identifying compounds with hepatotoxicity. However, there is considerable interest in enhancing the in silico prediction of DILI because it allows for evaluating large sets of compounds more quickly and cost-effectively, particularly in the early stages of projects. In this study, we aim to study ML models for DILI prediction that first predict nine proxy-DILI labels and then use them as features in addition to chemical structural features to predict DILI. The features include in vitro (e.g., mitochondrial toxicity, bile salt export pump inhibition) data, in vivo (e.g., preclinical rat hepatotoxicity studies) data, pharmacokinetic parameters of maximum concentration, structural fingerprints, and physicochemical parameters. We trained DILI-prediction models on 888 compounds from the DILI data set (composed of DILIst and DILIrank) and tested them on a held-out external test set of 223 compounds from the DILI data set. The best model, DILIPredictor, attained an AUC-PR of 0.79. This model enabled the detection of the top 25 toxic compounds (2.68 LR+, positive likelihood ratio) compared to models using only structural features (1.65 LR+ score). Using feature interpretation from DILIPredictor, we identified the chemical substructures causing DILI and differentiated cases of DILI caused by compounds in animals but not in humans. For example, DILIPredictor correctly recognized 2-butoxyethanol as nontoxic in humans despite its hepatotoxicity in mice models. Overall, the DILIPredictor model improves the detection of compounds causing DILI with an improved differentiation between animal and human sensitivity and the potential for mechanism evaluation. DILIPredictor required only chemical structures as input for prediction and is publicly available at https://broad.io/DILIPredictor for use via web interface and with all code available for download.

Beyond the hype and toward application: liver complex in vitro models in preclinical drug safety

INTRODUCTION

Drug induced Liver-Injury (DILI) is a leading cause of drug attrition and complex in vitro models (CIVMs), including three dimensional (3D) spheroids, 3D bio printed tissues and flow-based systems, could improve preclinical prediction. Although CIVMs have demonstrated good sensitivity and specificity in DILI detection their adoption remains limited.

AREAS COVERED

This article describes DILI, the challenges with its prediction and the current strategies and models that are being used. It reviews data from industry-FDA collaborations and strategic partnerships and finishes with an outlook of CIVMs in preclinical toxicity testing. Literature searches were performed using PubMed and Google Scholar while product information was collected from manufacturer websites.

EXPERT OPINION

Liver CIVMs are promising models for predicting DILI although, a decade after their introduction, routine use by the pharmaceutical industry is limited. To accelerate their adoption, several industry-regulator-developer partnerships or consortia have been established to guide the development and qualification. Beyond this, liver CIVMs should continue evolving to capture greater immunological mimicry while partnering with computational approaches to deliver systems that change the paradigm of predicting DILI.

Improved Detection of Drug-Induced Liver Injury by Integrating Predicted in vivo and in vitro Data

Drug-induced liver injury (DILI) has been significant challenge in drug discovery, often leading to clinical trial failures and necessitating drug withdrawals. The existing suite of in vitro proxy-DILI assays is generally effective at identifying compounds with hepatotoxicity. However, there is considerable interest in enhancing in silico prediction of DILI because it allows for the evaluation of large sets of compounds more quickly and cost-effectively, particularly in the early stages of projects. In this study, we aim to study ML models for DILI prediction that first predicts nine proxy-DILI labels and then uses them as features in addition to chemical structural features to predict DILI. The features include in vitro (e.g., mitochondrial toxicity, bile salt export pump inhibition) data, in vivo (e.g., preclinical rat hepatotoxicity studies) data, pharmacokinetic parameters of maximum concentration, structural fingerprints, and physicochemical parameters. We trained DILI-prediction models on 888 compounds from the DILIst dataset and tested on a held-out external test set of 223 compounds from DILIst dataset. The best model, DILIPredictor, attained an AUC-ROC of 0.79. This model enabled the detection of top 25 toxic compounds compared to models using only structural features (2.68 LR+ score). Using feature interpretation from DILIPredictor, we were able to identify the chemical substructures causing DILI as well as differentiate cases DILI is caused by compounds in animals but not in humans. For example, DILIPredictor correctly recognized 2-butoxyethanol as non-toxic in humans despite its hepatotoxicity in mice models. Overall, the DILIPredictor model improves the detection of compounds causing DILI with an improved differentiation between animal and human sensitivity as well as the potential for mechanism evaluation. DILIPredictor is publicly available at https://broad.io/DILIPredictor for use via web interface and with all code available for download and local implementation via https://pypi.org/project/dilipred/.

Clinical Features and Risk Factors for Drug-Induced Liver Injury: A Retrospective Study From China

PURPOSE

With the prolongation of human life expectancy and the outbreak of COVID-19, antineoplastic agents, anti-infective drugs, and cardiovascular system drugs have been widely applied, resulting in a growing incidence of drug-induced liver injury (DILI) year by year. This study aimed to investigate signals, clinical characteristics, and risk factors in patients with liver injury.

METHODS

A retrospective analysis was conducted on inpatients clinically diagnosed with DILI from 2019 to 2021 in one tertiary hospital in mainland China. The hepatic biochemical indices, clinical manifestations and suspected drugs of the patients were counted. We determined causality assessed by the Roussel Uclaf Causality Assessment Method in patients that the biochemistry met the diagnostic criteria recommended by the International Serious Adverse Events Consortium and compared them with contemporaneous patients diagnosed as DILI but with hepatic biochemical abnormalities only to identify the injure types and risk factors for DILI.

FINDINGS

A total of 1167 patients from 2639 initial participants with DILI were included. According to the injured target cells, it can be divided into hepatocellular injury type 351 cases (30.08%), cholestatic injury type 97 cases (8.31%), mixed injury type 27 cases (2.31%), and biochemical abnormal only type 692 cases (59.30%). It involved 1738 cases of suspected drugs, 349 drugs, and the top 3 drug categories were antineoplastic agents, anti-infectives, and traditional Chinese medicines, with Cyclophosphamide, Atorvastatin, and Liuzasulfapyridine as the top 3 in order of ranking. The main symptoms of patients were darker urine, decreased appetite, and yellow sclera. The overall prognosis of patients with DILI was favorable, with 280 recovered cases (23.99%), 691 improved cases (59.21%), 189 not improved cases (16.20%), and 7 deaths (0.60%). There were significant differences in gender, age, malignancy, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, total bilirubin, gamma-glutamyltransferase, albumin, international normalized ratio, and prognosis among patients with different injury types (P < 0.05). Multiple logistic regression analysis showed that female (odds ratio [OR] = 1.897, P < 0.001), alcohol use (OR = 1.905, P = 0.001), malignancy (OR = 0.417, P < 0.001), and pregnancy (OR = 0.201, P = 0.011) were independent factors influencing DILI.

IMPLICATIONS

For most patients with liver injury, the manifestations are mild elevation of liver biochemistry without other symptoms (biochemical abnormal only type). The rest of the patients are predominantly of the hepatocellular injury type. Female and alcohol abuse patients are the risk factors of DILI, reminding clinicians to strengthen education on safe drug use, give individualized treatment, and regularly monitor liver function indexes in the patients.

Image-based molecular representation learning for drug development: a survey

Artificial intelligence (AI) powered drug development has received remarkable attention in recent years. It addresses the limitations of traditional experimental methods that are costly and time-consuming. While there have been many surveys attempting to summarize related research, they only focus on general AI or specific aspects such as natural language processing and graph neural network. Considering the rapid advance on computer vision, using the molecular image to enable AI appears to be a more intuitive and effective approach since each chemical substance has a unique visual representation. In this paper, we provide the first survey on image-based molecular representation for drug development. The survey proposes a taxonomy based on the learning paradigms in computer vision and reviews a large number of corresponding papers, highlighting the contributions of molecular visual representation in drug development. Besides, we discuss the applications, limitations and future directions in the field. We hope this survey could offer valuable insight into the use of image-based molecular representation learning in the context of drug development.

Interplay Between Drug-Induced Liver Injury and Gut Microbiota: A Comprehensive Overview

Drug-induced liver injury is a prevalent severe adverse event in clinical settings, leading to increased medical burdens for patients and presenting challenges for the development and commercialization of novel pharmaceuticals. Research has revealed a close association between gut microbiota and drug-induced liver injury in recent years. However, there has yet to be a consensus on the specific mechanism by which gut microbiota is involved in drug-induced liver injury. Gut microbiota may contribute to drug-induced liver injury by increasing intestinal permeability, disrupting intestinal metabolite homeostasis, and promoting inflammation and oxidative stress. Alterations in gut microbiota were found in drug-induced liver injury caused by antibiotics, psychotropic drugs, acetaminophen, antituberculosis drugs, and antithyroid drugs. Specific gut microbiota and their abundance are associated closely with the severity of drug-induced liver injury. Therefore, gut microbiota is expected to be a new target for the treatment of drug-induced liver injury. This review focuses on the association of gut microbiota with common hepatotoxic drugs and the potential mechanisms by which gut microbiota may contribute to the pathogenesis of drug-induced liver injury, providing a more comprehensive reference for the interaction between drug-induced liver injury and gut microbiota.

Control compounds for preclinical drug-induced liver injury assessment: Consensus-driven systematic review by the ProEuroDILI network

BACKGROUND & AIMS

Idiosyncratic drug-induced liver injury (DILI) is a complex and unpredictable event caused by drugs, and herbal or dietary supplements. Early identification of human hepatotoxicity at preclinical stages remains a major challenge, in which the selection of validated in vitro systems and test drugs has a significant impact. In this systematic review, we analyzed the compounds used in hepatotoxicity assays and established a list of DILI-positive and -negative control drugs for validation of in vitro models of DILI, supported by literature and clinical evidence and endorsed by an expert committee from the COST Action ProEuroDILI Network (CA17112).

METHODS

Following 2020 PRISMA guidelines, original research articles focusing on DILI which used in vitro human models and performed at least one hepatotoxicity assay with positive and negative control compounds, were included. Bias of the studies was assessed by a modified 'Toxicological Data Reliability Assessment Tool'.

RESULTS

A total of 51 studies (out of 2,936) met the inclusion criteria, with 30 categorized as reliable without restrictions. Although there was a broad consensus on positive compounds, the selection of negative compounds lacked clarity. 2D monoculture, short exposure times and cytotoxicity endpoints were the most tested, although there was no consensus on drug concentrations.

CONCLUSIONS

Extensive analysis highlighted the lack of agreement on control compounds for in vitro DILI assessment. Following comprehensive in vitro and clinical data analysis together with input from the expert committee, an evidence-based consensus-driven list of 10 positive and negative control drugs for validation of in vitro models of DILI is proposed.

IMPACT AND IMPLICATIONS

Prediction of human toxicity early in the drug development process remains a major challenge, necessitating the development of more physiologically relevant liver models and careful selection of drug-induced liver injury (DILI)-positive and -negative control drugs to better predict the risk of DILI associated with new drug candidates. Thus, this systematic study has crucial implications for standardizing the validation of new in vitro models of DILI. By establishing a consensus-driven list of positive and negative control drugs, the study provides a scientifically justified framework for enhancing the consistency of preclinical testing, thereby addressing a significant challenge in early hepatotoxicity identification. Practically, these findings can guide researchers in evaluating safety profiles of new drugs, refining in vitro models, and informing regulatory agencies on potential improvements to regulatory guidelines, ensuring a more systematic and efficient approach to drug safety assessment.

Pyruvate dehydrogenase is a potential mitochondrial off-target for gentamicin based on in silico predictions and in vitro inhibition studies

During the drug development process, organ toxicity leads to an estimated failure of one-third of novel chemical entities. Drug-induced toxicity is increasingly associated with mitochondrial dysfunction, but identifying the underlying molecular mechanisms remains a challenge. Computational modeling techniques have proven to be a good tool in searching for drug off-targets. Here, we aimed to identify mitochondrial off-targets of the nephrotoxic drugs tenofovir and gentamicin using different in silico approaches (KRIPO, ProBis and PDID). Dihydroorotate dehydrogenase (DHODH) and pyruvate dehydrogenase (PDH) were predicted as potential novel off-target sites for tenofovir and gentamicin, respectively. The predicted targets were evaluated in vitro, using (colorimetric) enzymatic activity measurements. Tenofovir did not inhibit DHODH activity, while gentamicin potently reduced PDH activity. In conclusion, the use of in silico methods appeared a valuable approach in predicting PDH as a mitochondrial off-target of gentamicin. Further research is required to investigate the contribution of PDH inhibition to overall renal toxicity of gentamicin.

Understanding exosomes: Part 2-Emerging leaders in regenerative medicine

Exosomes are the smallest subset of extracellular signaling vesicles secreted by most cells with the ability to communicate with other tissues and cell types over long distances. Their use in regenerative medicine has gained tremendous momentum recently due to their ability to be utilized as therapeutic options for a wide array of diseases/conditions. Over 5000 publications are currently being published yearly on this topic, and this number is only expected to dramatically increase as novel therapeutic strategies continue to be developed. Today exosomes have been applied in numerous contexts including neurodegenerative disorders (Alzheimer's disease, central nervous system, depression, multiple sclerosis, Parkinson's disease, post-traumatic stress disorders, traumatic brain injury, peripheral nerve injury), damaged organs (heart, kidney, liver, stroke, myocardial infarctions, myocardial infarctions, ovaries), degenerative processes (atherosclerosis, diabetes, hematology disorders, musculoskeletal degeneration, osteoradionecrosis, respiratory disease), infectious diseases (COVID-19, hepatitis), regenerative procedures (antiaging, bone regeneration, cartilage/joint regeneration, osteoarthritis, cutaneous wounds, dental regeneration, dermatology/skin regeneration, erectile dysfunction, hair regrowth, intervertebral disc repair, spinal cord injury, vascular regeneration), and cancer therapy (breast, colorectal, gastric cancer and osteosarcomas), immune function (allergy, autoimmune disorders, immune regulation, inflammatory diseases, lupus, rheumatoid arthritis). This scoping review is a first of its kind aimed at summarizing the extensive regenerative potential of exosomes over a broad range of diseases and disorders.

Deep Learning Algorithm Based on Molecular Fingerprint for Prediction of Drug-Induced Liver Injury

Drug-induced liver injury (DILI) is one the rare adverse drug reaction (ADR) and multifactorial endpoints. Current preclinical animal models struggle to anticipate it, and in silico methods have emerged as a way with significant potential for doing so. In this study, a high-quality dataset of 1573 compounds was assembled. The 48 classification models, which depended on six different molecular fingerprints, were built via deep neural network (DNN) and seven machine learning algorithms. Comparing the results of the DNN and machine learning models, the optional performing model was found as the one developed based on the DNN with ECFP_6 as input, which achieved the area under the receiver operating characteristic curve (AUC) of 0.713, balanced accuracy (BA) of 0.680, and F1 of 0.753. In addition, we used the SHapley Additive exPlanations (SHAP) algorithm to interpret the models, identified the crucial structural fragments related to DILI risk, and selected the top ten substructures with the highest contribution rankings to serve as warning indicators for subsequent drug hepatotoxicity screening studies. The study demonstrates that the DNN models developed based on molecular fingerprints can be a trustworthy and efficient tool for determining the risk of DILI during the pre-development of novel medications.

Synergistic effect of potential alpha-amylase inhibitors from Egyptian propolis with acarbose using in silico and in vitro combination analysis

BACKGROUND

Type 2 Diabetes mellitus (DM) is an affliction impacting the quality of life of millions of people worldwide. An approach used in the management of Type 2 DM involves the use of the carbohydrate-hydrolyzing enzyme inhibitor, acarbose. Although acarbose has long been the go-to drug in this key approach, it has become apparent that its side effects negatively impact patient adherence and subsequently, therapeutic outcomes. Similar to acarbose in its mechanism of action, bee propolis, a unique natural adhesive biomass consisting of biologically active metabolites, has been found to have antidiabetic potential through its inhibition of α-amylase. To minimize the need for ultimately novel agents while simultaneously aiming to decrease the side effects of acarbose and enhance its efficacy, combination drug therapy has become a promising pharmacotherapeutic strategy and a focal point of this study.

METHODS

Computer-aided molecular docking and molecular dynamics (MD) simulations accompanied by in vitro testing were used to mine novel, pharmacologically active chemical entities from Egyptian propolis to combat Type 2 DM. Glide docking was utilized for a structure-based virtual screening of the largest in-house library of Egyptian propolis metabolites gathered from literature, in addition to GC-MS analysis of the propolis sample under investigation. Thereafter, combination analysis by means of fixed-ratio combinations of acarbose with propolis and the top chosen propolis-derived phytoligand was implemented.

RESULTS

Aucubin, identified for the first time in propolis worldwide and kaempferol were the most promising virtual hits. Subsequent in vitro α-amylase inhibitory assay demonstrated the ability of these hits to significantly inhibit the enzyme in a dose-dependent manner with an IC50 of 2.37 ± 0.02 mM and 4.84 ± 0.14 mM, respectively. The binary combination of acarbose with each of propolis and kaempferol displayed maximal synergy at lower effect levels. Molecular docking and MD simulations revealed a cooperative binding mode between kaempferol and acarbose within the active site.

CONCLUSION

The suggested strategy seems imperative to ensure a steady supply of new therapeutic entities sourced from Egyptian propolis to regress the development of DM. Further pharmacological in vivo investigations are required to confirm the potent antidiabetic potential of the studied combination.

Developing a GNN-based AI model to predict mitochondrial toxicity using the bagging method

Mitochondrial toxicity has been implicated in the development of various toxicities, including hepatotoxicity. Therefore, mitochondrial toxicity has become a major screening factor in the early discovery phase of drug development. Several models have been developed to predict mitochondrial toxicity based on chemical structures. However, they only provide a binary classification of positive or negative results and do not provide the substructures that contribute to a positive decision. Therefore, we developed an artificial intelligence (AI) model to predict mitochondrial toxicity and visualize structural alerts. To construct the model, we used the open-source software library kMoL, which employs a graph neural network approach that allows learning from chemical structure data. We also utilized the integrated gradient method, which enables the visualization of substructures that contribute to positive results. The dataset used to construct the AI model exhibited a significant imbalance, with significantly more negative than positive data. To address this, we employed the bagging method, which resulted in a model with high predictive performance, as evidenced by an F1 score of 0.839. This model can also be used to visualize substructures that contribute to mitochondrial toxicity using the integrated gradient method. Our AI model predicts mitochondrial toxicity based on chemical structures and may contribute to screening mitochondrial toxicity in the early stages of drug discovery.

Predicting drug-induced liver injury using graph attention mechanism and molecular fingerprints

Drug-induced liver injury (DILI) is a significant issue in drug development and clinical treatment due to its potential to cause liver dysfunction or damage, which, in severe cases, can lead to liver failure or even fatality. DILI has numerous pathogenic factors, many of which remain incompletely understood. Consequently, it is imperative to devise methodologies and tools for anticipatory assessment of DILI risk in the initial phases of drug development. In this study, we present DMFPGA, a novel deep learning predictive model designed to predict DILI. To provide a comprehensive description of molecular properties, we employ a multi-head graph attention mechanism to extract features from the molecular graphs, representing characteristics at the level of compound nodes. Additionally, we combine multiple fingerprints of molecules to capture features at the molecular level of compounds. The fusion of molecular fingerprints and graph features can more fully express the properties of compounds. Subsequently, we employ a fully connected neural network to classify compounds as either DILI-positive or DILI-negative. To rigorously evaluate DMFPGA's performance, we conduct a 5-fold cross-validation experiment. The obtained results demonstrate the superiority of our method over four existing state-of-the-art computational approaches, exhibiting an average AUC of 0.935 and an average ACC of 0.934. We believe that DMFPGA is helpful for early-stage DILI prediction and assessment in drug development.

Guided discovery of hepatoprotective polyhydroxy cembrane-type diterpenoids from the gum resin of Boswellia carterii by MS/MS molecular networking

Seven previously undescribed polyhydroxy cembrane-type diterpenoids, olibanols A-G (1-7) were obtained from the gum resin of Boswellia carterii by means of MS/MS molecular networking. Compound 2 possessed four hydroxy groups, 1, 3, 4, 5, and 6 had three hydroxy groups, 7 with one hydroxy group, among which 1 and 4 were a pair of epimers with double bond at C-3 and hydroxy at C-8. Structures of these previously undescribed compounds were determined by NMR analysis and ECD calculations. All the polyhydroxy cembrane-type diterpenoids obtained were assayed for their hepatoprotective effects against the anti-tuberculosis drug-induced hepatic damage to the HRZ-induced HepG2 cells. As results indicated, compounds 3, 4, and 6 showed significant hepatoprotective effects against the hepatic damage via the Nrf2 signal pathway, which could be developed as potential hepatoprotective agents against the anti-tuberculosis drug-induced hepatic damage.

A comprehensive review of Dynamic Chemical Labelling on Luminex xMAP technology: a journey towards Drug-Induced Liver Injury testing

Drug-Induced Liver Injury (DILI) is a grave global adverse event that can result in fatal consequences, causing drug failures, market withdrawals, and regulatory warnings, leading to substantial financial losses. The early detection of DILI remains a significant challenge in global healthcare. Although circulating microRNAs (miRs) show promise as clinical biomarkers for DILI, the current analytical methods for their measurement are insufficient. There is a pressing need for rapid and reliable miR detection methods that eliminate the need for nucleic acid extraction and PCR-based amplification. This review highlights recent advancements achieved by integrating Dynamic Chemical Labelling (DCL) with Luminex xMAP technology. This powerful combination has resulted in groundbreaking bead-based assays that allow (1) the direct, multiplex detection of miRs, and (2) the simultaneous testing of miR and protein biomarkers. This triple capability enables a comprehensive assessment that significantly enhances the detection and analysis of crucial biomarkers, thus improving the understanding and diagnosis of DILI. In conclusion, this review offers valuable insights into the capabilities and potential applications of these groundbreaking assays in DILI research, as well as their potential use in other diagnostic and research domains that require direct or multiplex analysis of miRs or analysis of miRs in combination with proteins.

GeoDILI: A Robust and Interpretable Model for Drug-Induced Liver Injury Prediction Using Graph Neural Network-Based Molecular Geometric Representation

Drug-induced liver injury (DILI) is a significant cause of drug failure and withdrawal due to liver damage. Accurate prediction of hepatotoxic compounds is crucial for safe drug development. Several DILI prediction models have been published, but they are built on different data sets, making it difficult to compare model performance. Moreover, most existing models are based on molecular fingerprints or descriptors, neglecting molecular geometric properties and lacking interpretability. To address these limitations, we developed GeoDILI, an interpretable graph neural network that uses a molecular geometric representation. First, we utilized a geometry-based pretrained molecular representation and optimized it on the DILI data set to improve predictive performance. Second, we leveraged gradient information to obtain high-precision atomic-level weights and deduce the dominant substructure. We benchmarked GeoDILI against recently published DILI prediction models, as well as popular GNN models and fingerprint-based machine learning models using the same data set, showing superior predictive performance of our proposed model. We applied the interpretable method in the DILI data set and derived seven precise and mechanistically elucidated structural alerts. Overall, GeoDILI provides a promising approach for accurate and interpretable DILI prediction with potential applications in drug discovery and safety assessment. The data and source code are available at GitHub repository (https://github.com/CSU-QJY/GeoDILI).

DICTrank: The largest reference list of 1318 human drugs ranked by risk of drug-induced cardiotoxicity using FDA labeling

Drug-induced cardiotoxicity (DICT) is a leading cause of drug trial failure and discontinuation. Current drug annotations for cardiotoxicity largely focus on individual outcomes or mechanisms. Considering the broad spectrum of adverse cardiac events, we developed Drug-Induced Cardiotoxicity Rank (DICTrank) using FDA labeling and comprehensively classified 1318 human drugs into four categories: Most-DICT-Concern (n = 341), Less-DICT-Concern (n = 528), No-DICT-Concern (n = 343), and Ambiguous-DICT-Concern (n = 106). Notably, DICTrank covers diverse therapeutic categories, of which several were enriched with Most-DICT-Concern drugs, such as antineoplastic agents, sex hormones, anti-inflammatory drugs, beta-blockers, and cardiac therapy. DICTrank currently presents the largest drug list of DICT annotation, and it could contribute to the development of new approach methods, including AI models for early identification of DICT risk during drug development and beyond.

Translational Relevance of Rodent Models to Predict Human Liver Disease

Animals models are essential to understand the complex pathobiology of human diseases. George Box's aphorism based on statistics "All models are wrong, but some are useful" certainly applies to animal models of disease. In this session, the translational relevance of various animal models applicable to human liver disease was explored starting with a historic overview of the rodent cancer bioassay with emphasis on hepatocarcinogenesis from early work at the National Cancer Institute, refinement by the National Toxicology Program and contemporary efforts to identify potential mechanisms and their relevance to human cancer risk. Subsequently, recently elucidated understanding of the molecular drivers and signaling mechanisms of liver pathophysiology and liver cancer, including factors associated with liver regeneration, metabolic hepatocellular zonation, and the role of macrophages and their crosstalk with stellate cells in understanding human liver disease was discussed. Next, our contemporary understanding of the role of nuclear receptors in hepatic homeostasis and drug response highlighting nuclear receptor activation and crosstalk in modulating biological responses associated with liver damage and neoplastic response were discussed. Finally, an overview and translational relevance of different drug-induced liver injury (DILI) rodent model systems focused on pathology and mechanisms with commentary on current relevant Food and Drug Administration (FDA) perspective were summarized with closing remarks.

Revolutionizing Medicinal Chemistry: The Application of Artificial Intelligence (AI) in Early Drug Discovery

Artificial intelligence (AI) has permeated various sectors, including the pharmaceutical industry and research, where it has been utilized to efficiently identify new chemical entities with desirable properties. The application of AI algorithms to drug discovery presents both remarkable opportunities and challenges. This review article focuses on the transformative role of AI in medicinal chemistry. We delve into the applications of machine learning and deep learning techniques in drug screening and design, discussing their potential to expedite the early drug discovery process. In particular, we provide a comprehensive overview of the use of AI algorithms in predicting protein structures, drug-target interactions, and molecular properties such as drug toxicity. While AI has accelerated the drug discovery process, data quality issues and technological constraints remain challenges. Nonetheless, new relationships and methods have been unveiled, demonstrating AI's expanding potential in predicting and understanding drug interactions and properties. For its full potential to be realized, interdisciplinary collaboration is essential. This review underscores AI's growing influence on the future trajectory of medicinal chemistry and stresses the importance of ongoing synergies between computational and domain experts.

In silico modeling-based new alternative methods to predict drug and herb-induced liver injury: A review

New approach methods (NAMs) have been developed to predict a wide range of toxicities through innovative technologies. Liver injury is one of the most extensively studied endpoints due to its severity and frequency, occurring among populations that consume drugs or dietary supplements. In this review, we focus on recent developments of in silico modeling for liver injury prediction using deep learning and in vitro data based on adverse outcome pathways (AOPs). Despite these models being mainly developed using datasets generated from drug-like molecules, they were also applied to the prediction of hepatotoxicity caused by herbal products. As deep learning has achieved great success in many different fields, advanced machine learning algorithms have been actively applied to improve the accuracy of in silico models. Additionally, the development of liver AOPs, combined with big data in toxicology, has been valuable in developing in silico models with enhanced predictive performance and interpretability. Specifically, one approach involves developing structure-based models for predicting molecular initiating events of liver AOPs, while others use in vitro data with structure information as model inputs for making predictions. Even though liver injury remains a difficult endpoint to predict, advancements in machine learning algorithms and the expansion of in vitro databases with relevant biological knowledge have made a huge impact on improving in silico modeling for drug-induced liver injury prediction.

Chemistry-Based Modeling on Phenotype-Based Drug-Induced Liver Injury Annotation: From Public to Proprietary Data

Drug-induced liver injury (DILI) is an important safety concern and a major reason to remove a drug from the market. Advancements in recent machine learning methods have led to a wide range of in silico models for DILI predictive methods based on molecule chemical structures (fingerprints). Existing publicly available DILI data sets used for model building are based on the interpretation of drug labels or patient case reports, resulting in a typical binary clinical DILI annotation. We developed a novel phenotype-based annotation to process hepatotoxicity information extracted from repeated dose in vivo preclinical toxicology studies using INHAND annotation to provide a more informative and reliable data set for machine learning algorithms. This work resulted in a data set of 430 unique compounds covering diverse liver pathology findings which were utilized to develop multiple DILI prediction models trained on the publicly available data (TG-GATEs) using the compound's fingerprint. We demonstrate that the TG-GATEs compounds DILI labels can be predicted well and how the differences between TG-GATEs and the external test compounds (Johnson & Johnson) impact the model generalization performance.

Recent Advances in Toxicity Prediction: Applications of Deep Graph Learning

The development of new drugs is time-consuming and expensive, and as such, accurately predicting the potential toxicity of a drug candidate is crucial in ensuring its safety and efficacy. Recently, deep graph learning has become prevalent in this field due to its computational power and cost efficiency. Many novel deep graph learning methods aid toxicity prediction and further prompt drug development. This review aims to connect fundamental knowledge with burgeoning deep graph learning methods. We first summarize the essential components of deep graph learning models for toxicity prediction, including molecular descriptors, molecular representations, evaluation metrics, validation methods, and data sets. Furthermore, based on various graph-related representations of molecules, we introduce several representative studies and methods for toxicity prediction from the perspective of GNN architectures and graph pretrained models. Compared to other types of models, deep graph models not only advance in higher accuracy and efficiency but also provide more intuitive insights, which is significant in the development of model interpretation and generalization ability. The graph pretrained models are emerging as they can extract prominent features from large-scale unlabeled molecular graph data and improve the performance of downstream toxicity prediction tasks. We hope this survey can serve as a handbook for individuals interested in exploring deep graph learning for toxicity prediction.

Navigating the complex landscape of benzodiazepine- and Z-drug diversity: insights from comprehensive FDA adverse event reporting system analysis and beyond

Introduction

Medications which target benzodiazepine (BZD) binding sites of GABAA receptors (GABAARs) have been in widespread use since the nineteen-sixties. They carry labels as anxiolytics, hypnotics or antiepileptics. All benzodiazepines and several nonbenzodiazepine Z-drugs share high affinity binding sites on certain subtypes of GABAA receptors, from which they can be displaced by the clinically used antagonist flumazenil. Additional binding sites exist and overlap in part with sites used by some general anaesthetics and barbiturates. Despite substantial preclinical efforts, it remains unclear which receptor subtypes and ligand features mediate individual drug effects. There is a paucity of literature comparing clinically observed adverse effect liabilities across substances in methodologically coherent ways.

Methods

In order to examine heterogeneity in clinical outcome, we screened the publicly available U.S. FDA adverse event reporting system (FAERS) database for reports of individual compounds and analyzed them for each sex individually with the use of disproportionality analysis. The complementary use of physico-chemical descriptors provides a molecular basis for the analysis of clinical observations of wanted and unwanted drug effects.

Results and Discussion

We found a multifaceted FAERS picture, and suggest that more thorough clinical and pharmacoepidemiologic investigations of the heterogenous side effect profiles for benzodiazepines and Z-drugs are needed. This may lead to more differentiated safety profiles and prescription practice for particular compounds, which in turn could potentially ease side effect burden in everyday clinical practice considerably. From both preclinical literature and pharmacovigilance data, there is converging evidence that this very large class of psychoactive molecules displays a broad range of distinctive unwanted effect profiles - too broad to be explained by the four canonical, so-called "diazepam-sensitive high-affinity interaction sites". The substance-specific signatures of compound effects may partly be mediated by phenomena such as occupancy of additional binding sites, and/or synergistic interactions with endogenous substances like steroids and endocannabinoids. These in turn drive the wanted and unwanted effects and sex differences of individual compounds.

Identifying chemicals based on receptor binding/bioactivation/mechanistic explanation associated with potential to elicit hepatotoxicity and to support structure activity relationship-based read-across

The liver is the most common target organ in toxicology studies. The development of chemical structural alerts for identifying hepatotoxicity will play an important role in in silico model prediction and help strengthen the identification of analogs used in structure activity relationship (SAR)- based read-across. The aim of the current study is development of an SAR-based expert-system decision tree for screening of hepatotoxicants across a wide range of chemistry space and proposed modes of action for clustering of chemicals using defined core chemical categories based on receptor-binding or bioactivation. The decision tree is based on ∼ 1180 different chemicals that were reviewed for hepatotoxicity information. Knowledge of chemical receptor binding, metabolism and mechanistic information were used to group these chemicals into 16 different categories and 102 subcategories: four categories describe binders to 9 different receptors, 11 categories are associated with possible reactive metabolites (RMs) and there is one miscellaneous category. Each chemical subcategory has been associated with possible modes of action (MOAs) or similar key structural features. This decision tree can help to screen potential liver toxicants associated with core structural alerts of receptor binding and/or RMs and be used as a component of weight of evidence decisions based on SAR read-across, and to fill data gaps.

Artificial Intelligence in Drug Toxicity Prediction: Recent Advances, Challenges, and Future Perspectives

Toxicity prediction is a critical step in the drug discovery process that helps identify and prioritize compounds with the greatest potential for safe and effective use in humans, while also reducing the risk of costly late-stage failures. It is estimated that over 30% of drug candidates are discarded owing to toxicity. Recently, artificial intelligence (AI) has been used to improve drug toxicity prediction as it provides more accurate and efficient methods for identifying the potentially toxic effects of new compounds before they are tested in human clinical trials, thus saving time and money. In this review, we present an overview of recent advances in AI-based drug toxicity prediction, including the use of various machine learning algorithms and deep learning architectures, of six major toxicity properties and Tox21 assay end points. Additionally, we provide a list of public data sources and useful toxicity prediction tools for the research community and highlight the challenges that must be addressed to enhance model performance. Finally, we discuss future perspectives for AI-based drug toxicity prediction. This review can aid researchers in understanding toxicity prediction and pave the way for new methods of drug discovery.

Administration of Secretome Derived from Human Mesenchymal Stem Cells Induces Hepatoprotective Effects in Models of Idiosyncratic Drug-Induced Liver Injury Caused by Amiodarone or Tamoxifen

Drug-induced liver injury (DILI) is one of the leading causes of acute liver injury. While many factors may contribute to the susceptibility to DILI, obese patients with hepatic steatosis are particularly prone to suffer DILI. The secretome derived from mesenchymal stem cell has been shown to have hepatoprotective effects in diverse in vitro and in vivo models. In this study, we evaluate whether MSC secretome could improve DILI mediated by amiodarone (AMI) or tamoxifen (TMX). Hepatic HepG2 and HepaRG cells were incubated with AMI or TMX, alone or with the secretome of MSCs obtained from human adipose tissue. These studies demonstrate that coincubation of AMI or TMX with MSC secretome increases cell viability, prevents the activation of apoptosis pathways, and stimulates the expression of priming phase genes, leading to higher proliferation rates. As proof of concept, in a C57BL/6 mouse model of hepatic steatosis and chronic exposure to AMI, the MSC secretome was administered endovenously. In this study, liver injury was significantly attenuated, with a decrease in cell infiltration and stimulation of the regenerative response. The present results indicate that MSC secretome administration has the potential to be an adjunctive cell-free therapy to prevent liver failure derived from DILI caused by TMX or AMI.

Molecular mechanism for the involvement of CYP2E1/NF-κB axis in bedaquiline-induced hepatotoxicity

Bedaquiline (BDQ) is a new class of anti-tubercular (anti-TB) drugs and is currently reserved for multiple drug resistance (MDR-TB). However, after receiving fast-track approval, its clinical studies demonstrate that its treatment is associated with hepatotoxicity and labeled as 'boxed warning' by the USFDA. No data is available on BDQ to understand the mechanism for drug-induced liver injury (DILI), a severe concern for therapeutic failure/unbearable tolerated toxicities leading to drug resistance. Therefore, we performed mechanistic studies to decipher the potential of BDQ at three dose levels (80 to 320 mg/kg) upon the repeated dose administration orally using a widely used mice model for TB. Results of BDQ treatment at the highest dose level showed that substantial increase of hepatic marker enzymes (SGPT and SGOT) in serum, oxidative stress marker levels (MDA and GSH) in hepatic tissue, and pro-inflammatory cytokine levels (TNF-α, IL-6, and IL-1β) in serum compared to control animals. Induction of liver injury situation was further evaluated by Western blotting for various protein expressions linked to oxidative stress (SOD, Nrf2, and Keap1), inflammation (NF-ĸB and IKKβ), apoptosis (BAX, Bcl-2, and Caspase-3) and drug metabolism enzymes (CYP3A4 and CYP2E1). The elevated plasma level of BDQ and its metabolite (N-desmethyl BDQ) were observed, corresponding to BDQ doses. Histopathological examination and SEM analysis of the liver tissue corroborate the above-mentioned findings. Overall results suggest that BDQ treatment-associated generation of its cytotoxic metabolite could act on CYP2E1/NF-kB pathway to aggravate the condition of oxidative stress, inflammation, and apoptosis in the liver and precipitating hepatotoxicity.

Towards Decoding Hepatotoxicity of Approved Drugs through Navigation of Multiverse and Consensus Chemical Spaces

Drug-induced liver injury (DILI) is the principal reason for failure in developing drug candidates. It is the most common reason to withdraw from the market after a drug has been approved for clinical use. In this context, data from animal models, liver function tests, and chemical properties could complement each other to understand DILI events better and prevent them. Since the chemical space concept improves decision-making drug design related to the prediction of structure-property relationships, side effects, and polypharmacology drug activity (uniquely mentioning the most recent advances), it is an attractive approach to combining different phenomena influencing DILI events (e.g., individual "chemical spaces") and exploring all events simultaneously in an integrated analysis of the DILI-relevant chemical space. However, currently, no systematic methods allow the fusion of a collection of different chemical spaces to collect different types of data on a unique chemical space representation, namely "consensus chemical space." This study is the first report that implements data fusion to consider different criteria simultaneously to facilitate the analysis of DILI-related events. In particular, the study highlights the importance of analyzing together in vitro and chemical data (e.g., topology, bond order, atom types, presence of rings, ring sizes, and aromaticity of compounds encoded on RDKit fingerprints). These properties could be aimed at improving the understanding of DILI events.

Using chemical and biological data to predict drug toxicity

Various sources of information can be used to better understand and predict compound activity and safety-related endpoints, including biological data such as gene expression and cell morphology. In this review, we first introduce types of chemical, in vitro and in vivo information that can be used to describe compounds and adverse effects. We then explore how compound descriptors based on chemical structure or biological perturbation response can be used to predict safety-related endpoints, and how especially biological data can help us to better understand adverse effects mechanistically. Overall, the described applications demonstrate how large-scale biological information presents new opportunities to anticipate and understand the biological effects of compounds, and how this can support predictive toxicology and drug discovery projects.

Supervised chemical graph mining improves drug-induced liver injury prediction

Drug-induced liver injury (DILI) is the main cause of drug failure in clinical trials. The characterization of toxic compounds in terms of chemical structure is important because compounds can be metabolized to toxic substances in the liver. Traditional machine learning approaches have had limited success in predicting DILI, and emerging deep graph neural network (GNN) models are yet powerful enough to predict DILI. In this study, we developed a completely different approach, supervised subgraph mining (SSM), a strategy to mine explicit subgraph features by iteratively updating individual graph transitions to maximize DILI fidelity. Our method outperformed previous methods including state-of-the-art GNN tools in classifying DILI on two different datasets: DILIst and TDC-benchmark. We also combined the subgraph features by using SMARTS-based frequent structural pattern matching and associated them with drugs' ATC code.

Applications of In Silico Models to Predict Drug-Induced Liver Injury

Drug-induced liver injury (DILI) is a major cause of the withdrawal of pre-marketed drugs, typically attributed to oxidative stress, mitochondrial damage, disrupted bile acid homeostasis, and innate immune-related inflammation. DILI can be divided into intrinsic and idiosyncratic DILI with cholestatic liver injury as an important manifestation. The diagnosis of DILI remains a challenge today and relies on clinical judgment and knowledge of the insulting agent. Early prediction of hepatotoxicity is an important but still unfulfilled component of drug development. In response, in silico modeling has shown good potential to fill the missing puzzle. Computer algorithms, with machine learning and artificial intelligence as a representative, can be established to initiate a reaction on the given condition to predict DILI. DILIsym is a mechanistic approach that integrates physiologically based pharmacokinetic modeling with the mechanisms of hepatoxicity and has gained increasing popularity for DILI prediction. This article reviews existing in silico approaches utilized to predict DILI risks in clinical medication and provides an overview of the underlying principles and related practical applications.

Adaptability of AI for safety evaluation in regulatory science: A case study of drug-induced liver injury

Artificial intelligence (AI) has played a crucial role in advancing biomedical sciences but has yet to have the impact it merits in regulatory science. As the field advances, in silico and in vitro approaches have been evaluated as alternatives to animal studies, in a drive to identify and mitigate safety concerns earlier in the drug development process. Although many AI tools are available, their acceptance in regulatory decision-making for drug efficacy and safety evaluation is still a challenge. It is a common perception that an AI model improves with more data, but does reality reflect this perception in drug safety assessments? Importantly, a model aiming at regulatory application needs to take a broad range of model characteristics into consideration. Among them is adaptability, defined as the adaptive behavior of a model as it is retrained on unseen data. This is an important model characteristic which should be considered in regulatory applications. In this study, we set up a comprehensive study to assess adaptability in AI by mimicking the real-world scenario of the annual addition of new drugs to the market, using a model we previously developed known as DeepDILI for predicting drug-induced liver injury (DILI) with a novel Deep Learning method. We found that the target test set plays a major role in assessing the adaptive behavior of our model. Our findings also indicated that adding more drugs to the training set does not significantly affect the predictive performance of our adaptive model. We concluded that the proposed adaptability assessment framework has utility in the evaluation of the performance of a model over time.

Impact of pharmacist active consultation on clinical outcomes and quality of medical care in drug-induced liver injury inpatients in general hospital wards: A retrospective cohort study

The utility of pharmacist consultation for drug-induced liver injury (DILI) management has not been explored. This retrospective cohort study evaluated the impact of a pharmacist active consultation (PAC) service on the management and outcome in patients with DILI. Consecutive patients meeting clinical biochemical criteria for DILI were enrolled at a tertiary teaching hospital between 1 January 2020 and 30 April 2022. The Roussel Uclaf Causality Assessment Method was used to assess causality between drug use and liver injury for each suspected DILI patient. Included patients were grouped according to whether they received PAC, and a proportional hazard model with multivariate risk adjustment, inverse probability of treatment weighting (IPTW), and propensity score matching (PSM) was used to assess DILI recovery. In the PSM cohort, the quality of medical care was compared between PAC and no PAC groups. A total of 224 patients with DILI (108 who received PAC and 116 who did not) were included in the analysis. Of these patients, 11 (10%) were classified as highly probable, 58 (54%) as probable, and 39 (36%) as possible DILI in the PAC group, while six patients (5%) were classified as highly probable, 53 (46%) as probable, and 57 (49%) as possible DILI in the no PAC group (p = 0.089). During patient recovery, PAC was associated with a ∼10% increase in the cumulative 180-day recovery rate. The PAC group had a crude hazard ratio (HR) of 1.73 [95% confidence interval (CI): 1.23–2.43, p = 0.001] for DILI 180-day recovery, which remained stable after multivariate risk adjustment (HR = 1.74, 95% CI: 1.21–2.49, p = 0.003), IPTW (HR = 1.72, 95% CI: 1.19–2.47, p = 0.003), and PSM (HR = 1.49, 95% CI: 1.01–2.23, p = 0.046). In the PSM cohort, PAC was more likely to identify suspect drugs (90% vs. 60%, p < 0.001) and lead to timely withdrawal of the medication (89% vs. 57%, p < 0.001). Thus, PAC is associated with a better quality of medical care for patients with DILI and can improve patient outcomes.

An Engineered Protein-Based Building Block (Albumin Methacryloyl) for Fabrication of a 3D In Vitro Cryogel Model

Drug-induced liver injury (DILI) is a leading cause of attrition in drug development or withdrawal; current animal experiments and traditional 2D cell culture systems fail to precisely predict the liver toxicity of drug candidates. Hence, there is an urgent need for an alternative in vitro model that can mimic the liver microenvironments and accurately detect human-specific drug hepatotoxicity. Here, for the first time we propose the fabrication of an albumin methacryloyl cryogel platform inspired by the liver’s microarchitecture via emulating the mechanical properties and extracellular matrix (ECM) cues of liver. Engineered crosslinkable albumin methacryloyl is used as a protein-based building block for fabrication of albumin cryogel in vitro models that can have potential applications in 3D cell culture and drug screening. In this work, protein modification, cryogelation, and liver ECM coating were employed to engineer highly porous three-dimensional cryogels with high interconnectivity, liver-like stiffness, and liver ECM as artificial liver constructs. The resulting albumin-based cryogel in vitro model provided improved cell–cell and cell–material interactions and consequently displayed excellent liver functional gene expression, being conducive to detection of fialuridine (FIAU) hepatotoxicity.

Deriving time-concordant event cascades from gene expression data: A case study for Drug-Induced Liver Injury (DILI)

Adverse event pathogenesis is often a complex process which compromises multiple events ranging from the molecular to the phenotypic level. In toxicology, Adverse Outcome Pathways (AOPs) aim to formalize this as temporal sequences of events, in which event relationships should be supported by causal evidence according to the tailored Bradford-Hill criteria. One of the criteria is whether events are consistently observed in a certain temporal order and, in this work, we study this time concordance using the concept of “first activation” as data-driven means to generate hypotheses on potentially causal mechanisms. As a case study, we analysed liver data from repeat-dose studies in rats from the TG-GATEs database which comprises measurements across eight timepoints, ranging from 3 hours to 4 weeks post-treatment. We identified time-concordant gene expression-derived events preceding adverse histopathology, which serves as surrogate readout for Drug-Induced Liver Injury (DILI). We find known mechanisms in DILI to be time-concordant, and show further that significance, frequency and log fold change (logFC) of differential expression are metrics which can additionally prioritize events although not necessary to be mechanistically relevant. Moreover, we used the temporal order of transcription factor (TF) expression and regulon activity to identify transcriptionally regulated TFs and subsequently combined this with prior knowledge on functional interactions to derive detailed gene-regulatory mechanisms, such as reduced Hnf4a activity leading to decreased expression and activity of Cebpa. At the same time, also potentially novel events are identified such as Sox13 which is highly significantly time-concordant and shows sustained activation over time. Overall, we demonstrate how time-resolved transcriptomics can derive and support mechanistic hypotheses by quantifying time concordance and how this can be combined with prior causal knowledge, with the aim of both understanding mechanisms of toxicity, as well as potential applications to the AOP framework. We make our results available in the form of a Shiny app (https://anikaliu.shinyapps.io/dili_cascades), which allows users to query events of interest in more detail.

Understanding mechanisms from systems-scale biological data is of great relevance in toxicology as well as drug discovery; however how to generate causal hypotheses instead of correlations is by no means clear. In this work, we study the conserved temporal order of events and present an automatable framework to quantify and characterize time concordance across a large set of time-series. We apply this concept to events derived from time-resolved gene expression and histopathology from the TG-GATEs in vivo liver data as a case study. We were able to recover known events involved in the pathogenesis of Drug-Induced Liver Injury (DILI), and identify potentially novel pathway and transcription factors (TFs) which precede adverse histopathology. As complementary sources of evidence for causality, we additionally show how time concordance and prior knowledge on plausible interactions between TFs can be combined to derive causal hypotheses on the TFs’ mode of regulation and interaction partners. Overall, the results derived in our case study can serve as valuable hypothesis-free starting points for the development of Adverse Outcome Pathways for DILI, and demonstrate that our approach provides a novel angle to prioritize mechanistically relevant events.

The prediction approach of drug-induced liver injury: response to the issues of reproducible science of artificial intelligence in real-world applications

In the previous study, we developed the generalized drug-induced liver injury (DILI) prediction model—ResNet18DNN to predict DILI based on multi-source combined DILI dataset and achieved better performance than that of previously published described DILI prediction models. Recently, we were honored to receive the invitation from the editor to response the Letter to Editor by Liu Zhichao, et al. We were glad that our research has attracted the attention of Liu’s team and they has put forward their opinions on our research. In this response to Letter to the Editor, we will respond to these comments.

Development of binary classification models for assessment of drug-induced liver injury in humans using a large set of FDA-approved drugs

Drug-induced liver injury (DILI) has been identified as one of the major causes for drugs withdrawn from the market, and even termination during the late stages of development. Therefore, it is imperative to evaluate the DILI potential of lead compounds during the research and development process. Although various computational models have been developed to predict DILI, most of which applied the DILI data were extracted from preclinical sources. In this investigation, the in silico prediction models for DILI were constructed based on 1140 FDA-approved drugs by using naïve Bayes classifier approach. The genetic algorithm method was applied for the molecular descriptors selection. Among these established prediction models, the NB-11 model based on eight molecular descriptors combined with ECFP_18 showed the best prediction performance for DILI, which gave 91.7% overall prediction accuracy for the training set, and 68.9% concordance for the external test set. Therefore, the established NB-11 prediction model can be used as a reliable virtual screening tool to predict DILI adverse effect in the early stages of drug design. In addition, some new structural alters for DILI were identified, which could be used for structural optimization in the future drug design by medicinal chemists.

Three-Dimensional-Bioprinted Liver Chips and Challenges

Drug testing, either on animals or on 2D cell cultures, has its limitations due to inaccurate mimicking of human pathophysiology. The liver, as one of the key organs that filters and detoxifies the blood, is susceptible to drug-induced injuries. Integrating 3D bioprinting with microfluidic chips to fabricate organ-on-chip platforms for 3D liver cell cultures with continuous perfusion can offer a more physiologically relevant liver-mimetic platform for screening drugs and studying liver function. The development of organ-on-chip platforms may ultimately contribute to personalized medicine as well as body-on-chip technology that can test drug responses and organ–organ interactions on a single or linked chip model.

In vitro cell-based models of drug-induced hepatotoxicity screening: progress and limitation

Drug-induced liver injury (DILI) is one of the major causes of post-approval withdrawal of therapeutics. As a result, there is an increasing need for accurate predictive in vitro assays that reliably detect hepatotoxic drug candidates while reducing drug discovery time, costs, and the number of animal experiments. In vitro hepatocyte-based research has led to an improved comprehension of the underlying mechanisms of chemical toxicity and can assist the prioritization of therapeutic choices with low hepatotoxicity risk. Therefore, several in vitro systems have been generated over the last few decades. This review aims to comprehensively present the development and validation of 2D (two-dimensional) and 3D (three-dimensional) culture approaches on hepatotoxicity screening of compounds and highlight the main factors affecting predictive power of experiments. To this end, we first summarize some of the recognized hepatotoxicity mechanisms and related assays used to appraise DILI mechanisms and then discuss the challenges and limitations of in vitro models.

Evaluation of Toxicity and Oxidative Stress of 2-Acetylpyridine-N(4)-orthochlorophenyl Thiosemicarbazone

Thiosemicarbazones are well known for their broad spectrum of action, including antitumoral and antiparasitic activities. Thiosemicarbazones work as chelating binders, reacting with metal ions. The objective of this work was to investigate the in silico, in vitro, and in vivo toxicity and oxidative stress of 2-acetylpyridine-N(4)-orthochlorophenyl thiosemicarbazone (TSC01). The in silico prediction showed good absorption by biological membranes and no theoretical toxicity. Also, the compound did not show cytotoxicity against Hep-G2 and HT-29 cells. In the acute nonclinical toxicological test, the animals treated with TSC01 showed behavioral changes of stimulus of the central nervous system (CNS) at 300 mg/kg. One hour after administration, a dose of 2000 mg/kg caused depressive signs. All changes disappeared after 24 h, with no deaths, which suggest an estimated LD50 of 5000 mg/kg and GSH 5. The group treated with 2000 mg/kg had an increase of water consumption and weight gain in the second week. The biochemical parameters presented no toxicity relevance, and the analysis of oxidative stress in the liver found an increase of lipid peroxidation and nitric oxide. However, histopathological analysis showed organ integrity was maintained without any changes. In conclusion, the results show the low toxicological potential of thiosemicarbazone derivative, indicating future safe use.

Farnesoid X receptor (FXR) agonists induce hepatocellular apoptosis and impair hepatic functions via FXR/SHP pathway

Farnesoid X receptor (FXR) plays an indispensable role in liver homeostasis and has been a promising drug target for hepatic diseases. However, the concerns of undesired biological actions limit the clinical applications of FXR agonists. To reveal the intrinsic mechanism of FXR agonist-induce hepatotoxicity, two typical FXR agonists with different structures (obeticholic acid (OCA) and Px-102) were investigated in the present study. By detecting MMP, ROS, and ATP and analyzing the fate of cells, we found that both OCA and Px-102 reduced the mitochondrial function of hepatocytes and promoted cell apoptosis. Gene ablation or inhibition of FXR or SHP ameliorated the cytotoxicities of OCA and Px-102, which indicated the adverse actions of FXR/SHP activation including down-regulation of phosphorylation of PI3K/AKT and functional hepatic genes. The dose-related injurious effects of OCA (10 mg/kg and 30 mg/kg) and Px-102 (5 mg/kg and 15 mg/kg) on the liver were confirmed on a high-fat diet mouse model. The decrease of hepatocyte-specific genes and augmenter of liver regeneration in the liver caused by OCA or Px-102 suggested an imbalance of liver regeneration and a disruption of hepatic functions. Exploration of intestinally biased FXR agonists or combination of FXR agonist with apoptosis inhibitor may be more beneficial strategies for liver diseases.

New Perspectives to Improve Mesenchymal Stem Cell Therapies for Drug-Induced Liver Injury

Drug-induced liver injury (DILI) is one of the leading causes of acute liver injury. Many factors may contribute to the susceptibility of patients to this condition, making DILI a global medical problem that has an impact on public health and the pharmaceutical industry. The use of mesenchymal stem cells (MSCs) has been at the forefront of regenerative medicine therapies for many years, including MSCs for the treatment of liver diseases. However, there is currently a huge gap between these experimental approaches and their application in clinical practice. In this concise review, we focus on the pathophysiology of DILI and highlight new experimental approaches conceived to improve cell-based therapy by the in vitro preconditioning of MSCs and/or the use of cell-free products as treatment for this liver condition. Finally, we discuss the advantages of new approaches, but also the current challenges that must be addressed in order to develop safer and more effective procedures that will allow cell-based therapies to reach clinical practice, enhancing the quality of life and prolonging the survival time of patients with DILI.

Predictive Model for Drug-Induced Liver Injury Using Deep Neural Networks Based on Substructure Space

Drug-induced liver injury (DILI) is a major concern for drug developers, regulators, and clinicians. However, there is no adequate model system to assess drug-associated DILI risk in humans. In the big data era, computational models are expected to play a revolutionary role in this field. This study aimed to develop a deep neural network (DNN)-based model using extended connectivity fingerprints of diameter 4 (ECFP4) to predict DILI risk. Each data set for the predictive model was retrieved and curated from DILIrank, LiverTox, and other literature. The best model was constructed through ten iterations of stratified 10-fold cross-validation, and the applicability domain was defined based on integer ECFP4 bits of the training set which represented substructures. For the robustness test, we employed the concept of the endurance level. The best model showed an accuracy of 0.731, a sensitivity of 0.714, and a specificity of 0.750 on the validation data set in the complete applicability domain. The model was further evaluated with four external data sets and attained an accuracy of 0.867 on 15 drugs with DILI cases reported since 2019. Overall, the results suggested that the ECFP4-based DNN model represents a new tool to identify DILI risk for the evaluation of drug safety.

In silico approaches in organ toxicity hazard assessment: current status and future needs in predicting liver toxicity

Hepatotoxicity is one of the most frequently observed adverse effects resulting from exposure to a xenobiotic. For example, in pharmaceutical research and development it is one of the major reasons for drug withdrawals, clinical failures, and discontinuation of drug candidates. The development of faster and cheaper methods to assess hepatotoxicity that are both more sustainable and more informative is critically needed. The biological mechanisms and processes underpinning hepatotoxicity are summarized and experimental approaches to support the prediction of hepatotoxicity are described, including toxicokinetic considerations. The paper describes the increasingly important role of in silico approaches and highlights challenges to the adoption of these methods including the lack of a commonly agreed upon protocol for performing such an assessment and the need for in silico solutions that take dose into consideration. A proposed framework for the integration of in silico and experimental information is provided along with a case study describing how computational methods have been used to successfully respond to a regulatory question concerning non-genotoxic impurities in chemically synthesized pharmaceuticals.

Prediction of Alternative Drug-Induced Liver Injury Classifications Using Molecular Descriptors, Gene Expression Perturbation, and Toxicology Reports

Motivation: Drug-induced liver injury (DILI) is one of the primary problems in drug development. Early prediction of DILI, based on the chemical properties of substances and experiments performed on cell lines, would bring a significant reduction in the cost of clinical trials and faster development of drugs. The current study aims to build predictive models of risk of DILI for chemical compounds using multiple sources of information.

Methods: Using several supervised machine learning algorithms, we built predictive models for several alternative splits of compounds between DILI and non-DILI classes. To this end, we used chemical properties of the given compounds, their effects on gene expression levels in six human cell lines treated with them, as well as their toxicological profiles. First, we identified the most informative variables in all data sets. Then, these variables were used to build machine learning models. Finally, composite models were built with the Super Learner approach. All modeling was performed using multiple repeats of cross-validation for unbiased and precise estimates of performance.

Results: With one exception, gene expression profiles of human cell lines were non-informative and resulted in random models. Toxicological reports were not useful for prediction of DILI. The best results were obtained for models discerning between harmless compounds and those for which any level of DILI was observed (AUC = 0.75). These models were built with Random Forest algorithm that used molecular descriptors.

Molecular Initiating Events Associated with Drug-Induced Liver Malignant Tumors: An Integrated Study of the FDA Adverse Event Reporting System and Toxicity Predictions

Liver malignant tumors (LMTs) represent a serious adverse drug event associated with drug-induced liver injury. Increases in endocrine-disrupting chemicals (EDCs) have attracted attention in recent years, due to their liver function-inhibiting abilities. Exposure to EDCs can induce nonalcoholic fatty liver disease and nonalcoholic steatohepatitis, which are major etiologies of LMTs, through interaction with nuclear receptors (NR) and stress response pathways (SRs). Therefore, exposure to potential EDC drugs could be associated with drug-induced LMTs. However, the drug classes associated with LMTs and the molecular initiating events (MIEs) that are specific to these drugs are not well understood. In this study, using the Food and Drug Administration Adverse Event Reporting System, we detected LMT-inducing drug signals based on adjusted odds ratios. Furthermore, based on the hypothesis that drug-induced LMTs are triggered by NR and SR modulation of potential EDCs, we used the quantitative structure-activity relationship platform for toxicity prediction to identify potential MIEs that are specific to LMT-inducing drug classes. Events related to cell proliferation and apoptosis, DNA damage, and lipid accumulation were identified as potential MIEs, and their relevance to LMTs was supported by the literature. The findings of this study may contribute to drug development and research, as well as regulatory decision making.

The Promise of AI for DILI Prediction

Drug-induced liver injury (DILI) is a common reason for the withdrawal of a drug from the market. Early assessment of DILI risk is an essential part of drug development, but it is rendered challenging prior to clinical trials by the complex factors that give rise to liver damage. Artificial intelligence (AI) approaches, particularly those building on machine learning, range from random forests to more recent techniques such as deep learning, and provide tools that can analyze chemical compounds and accurately predict some of their properties based purely on their structure. This article reviews existing AI approaches to predicting DILI and elaborates on the challenges that arise from the as yet limited availability of data. Future directions are discussed focusing on rich data modalities, such as 3D spheroids, and the slow but steady increase in drugs annotated with DILI risk labels.