Risk Of Drug-induced Liver Injury Research Articles

Machine Learning to Predict Drug-Induced Liver Injury and Its Validation on Failed Drug Candidates in Development

Drug-induced liver injury (DILI) poses a significant challenge for the pharmaceutical industry and regulatory bodies. Despite extensive toxicological research aimed at mitigating DILI risk, the effectiveness of these techniques in predicting DILI in humans remains limited. Consequently, researchers have explored novel approaches and procedures to enhance the accuracy of DILI risk prediction for drug candidates under development. In this study, we leveraged a large human dataset to develop machine learning models for assessing DILI risk. The performance of these prediction models was rigorously evaluated using a 10-fold cross-validation approach and an external test set. Notably, the random forest (RF) and multilayer perceptron (MLP) models emerged as the most effective in predicting DILI. During cross-validation, RF achieved an average prediction accuracy of 0.631, while MLP achieved the highest Matthews Correlation Coefficient (MCC) of 0.245. To validate the models externally, we applied them to a set of drug candidates that had failed in clinical development due to hepatotoxicity. Both RF and MLP accurately predicted the toxic drug candidates in this external validation. Our findings suggest that in silico machine learning approaches hold promise for identifying DILI liabilities associated with drug candidates during development.

Analysis of the Consumption of Medicinal Products Associated with a High Risk of Drug-Induced Liver Injury in Patients with COVID-19

INTRODUCTION. The risk of liver damage correlates with the severity of COVID-19. However, a growing number of studies have shown an association between liver function impairment and combinations of medicinal products used to treat COVID-19.AIM. The study aimed to analyse the annual consumption of medicinal products associated with a high risk of drug-induced liver injury (DILI) used as part of combination therapy in COVID-19 patients and to review a number of medication administration records in order to develop measures to prevent DILI.MATERIALS AND METHODS. The study used the ATC/DDD methodology to study consumption data for 2020, 2021, and 2022 and analysed a sample of 1250 inpatient medical records and medication administration records of COVID-19 patients treated in a Volgograd region hospital converted into a COVID-19 care centre. For genetically engineered biologicals and cyclophosphamide, which were lacking DDDs, the authors calculated the volume of consumption using the average dose per treatment course. The authors identified medicines capable of causing clinically apparent liver damage (according to the LiverTox database and Russian clinical practice guidelines) and/or elevated liver enzymes in ≥1% of patients (according to safety reports).RESULTS. The study found that 28% of the medicinal products used in combination for inpatient treatment of COVID-19 were associated with a high risk of DILI. In 2020, 2021, and 2022, the total consumption of medicinal products associated with a high risk of DILI was 342.3, 425.3, and 402.3 DDDs per 100 bed days, and the total consumption of genetically engineered biologicals (administered as a single dose) and cyclophosphamide was 3.5, 16.9, and 29.7 average course doses per 100 patients, respectively. According to the selective analysis of medical records, 19.8% (247/1250) reported concomitant use of 5 or more medicinal products associated with a high risk of DILI, which increased the risk of adverse drug interactions leading to DILI. In 2022, the most prescribed medicinal products with a high risk of DILI were omeprazole (188.7 DDDs per 100 bed days), non-steroidal anti-inflammatory drugs and paracetamol (54.4 DDDs per 100 bed days), atorvastatin (46.2 DDDs per 100 bed days), levofloxacin (26.4 DDDs per 100 bed days), ceftriaxone (20.5 DDDs per 100 bed days), favipiravir (17.3 DDDs per 100 bed days), and genetically engineered biologicals (24.0 DDDs per 100 bed days).CONCLUSIONS. To reduce the risk of DILI in COVID-19 patients admitted to infectious disease units, including the risk of DILI due to drug interactions, it is necessary to limit the use of hepatotoxic antibacterial agents, proton-pump inhibitors, and non-steroidal anti-inflammatory drugs, or consider alternative medicinal products with a lower risk of hepatotoxicity.

Microfluidic Human Physiomimetic Liver Model as a Screening Platform for Drug Induced Liver Injury

The pre-clinical animal models often fail to predict intrinsic and idiosyncratic drug induced liver injury (DILI), thus contributing to drug failures in clinical trials, black box warnings and withdrawal of marketed drugs. This suggests a critical need for human-relevant in vitro models to predict diverse DILI phenotypes. In this study, a porcine liver extracellular matrix (ECM) based biomaterial ink with high printing fidelity, biocompatibility and tunable rheological and mechanical properties is formulated for supporting both parenchymal and non-parenchymal cells. Further, we applied 3D printing and microfluidic technology to bioengineer a human physiomimetic liver acinus model (HPLAM), recapitulating the radial hepatic cord-like structure with functional sinusoidal microvasculature network, biochemical and biophysical properties of native liver acinus. Intriguingly, the human derived hepatic cells incorporated HPLAM cultured under physiologically relevant microenvironment, acts as metabolic biofactories manifesting enhanced hepatic functionality, secretome levels and biomarkers expression over several weeks. We also show that the matured HPLAM reproduces dose- and time-dependent hepatotoxic response of human clinical relevance to drugs typically recognized for inducing diverse DILI phenotypes as compared to conventional static culture. Overall, the developed HPLAM emulates in vivo like functions and may provide a useful platform for DILI risk assessment to better determine safety and human risk.

Pharmacogenomic Predictors of Antibiotic-Associated Drug-Induced Liver Injury in Critically Ill Children: Observational Study Results

INTRODUCTION. The pathogenesis of antibiotic-associated drug-induced liver injury (DILI) in children has not been fully elucidated to date. Certain genotypes in patients increase the probability of developing DILI. Therefore, the identification of pharmacogenetic markers associated with DILI in children is essential.АIM. This study aimed to identify pharmacogenetic biomarkers of new-onset DILI associated with tigecycline and meropenem in children.MATERIALS AND METHODS. This prospective observational study was conducted in the Morozov Children’s CityClinical Hospital from 1 February 2020 to 1 September 2021. The study analysed the incidence and types of antibiotic-associated adverse drug reactions (ADRs) in 100 critically ill children aged 0 to 17 years (44 boys and 56 girls). Pharmacogenetic testing was performed in children with ADRs (n=30) to identify potential mechanisms involved in the development of their ADRs. The authors isolated and tested DNA from buccal epithelium swabs using the Agena Bioscience iPLEX® PGx Pro-based VeriDose® Core Panel covering 68 single nucleotide polymorphisms (SNPs) or short insertions and deletions (INDELs) and 5 copy number variants (CNVs).RESULTS. The odds of developing DILI associated with meropenem and tigecycline were higher in carriers of the homozygous cytochrome genotype CYP3A5*3/*3 (OR: 12.6, 95% CI: 1.9–79.4, r=6.54, p=0.011) than in patients with the heterozygous genotype CYP3A5*1A/*3. The odds were even higher in patients not carrying the CYP3A5*1A/*3 genotype (OR: 17.14, 95% CI: 1.79–16.3, r=6.24, p=0.013). The detection of the CYP3A5*3/*3 genotype had a prognostic accuracy of 76.7%, a sensitivity of 82%, and a specificity of 74% in predicting the risk of DILI associated with meropenem and tigecycline. Moreover, children with DILI carried the heterozygous genotype SLCO1B1*1/*5 (rs4149056 polymorphism) more often than children with other adverse reactions associated with meropenem and tigecycline (r=9.8, p=0.002).CONCLUSION. The results of this study prove the prognostic significance of the homozygous cytochrome genotype CYP3A*3/*3 as an indicator of a potential risk for developing DILI associated with meropenem and tigecycline in children in critical conditions.The study was registered at ClinicalTrials.gov under No. NCT04141657 on 24 October 2019.

Pharmacogenetic variability of tuberculosis biomarkers in native and mestizo Peruvian populations.

In Peru, 29 292 people were diagnosed with tuberculosis in 2022. Although tuberculosis treatments are effective, 3.4%-13% are associated with significant adverse drug reactions, with drug-induced liver injury (DILI) considered the most predominant. Among the first-line antituberculosis drugs, isoniazid is the main drug responsible for the appearance of DILI. In liver, isoniazid (INH) is metabolized by N-acetyltransferase-2 (NAT2) and cytochrome P450 2E1 (CYP2E1). Limited information exists on genetic risk factors associated with the presence of DILI to antituberculosis drugs in Latin America, and even less is known about these factors in the native and mestizo Peruvian population. The aim of this study was to determine the prevalence of NAT2 and CYP2E1 genotypes in native and mestizo population. An analytical cross-sectional analysis was performed using genetic data from mestizo population in Lima and native participants from south of Peru. NAT2 metabolizer was determined as fast, intermediate and slow, and CYP2E1 genotypes were classified as c1/c1, c1/c2 and c2/c2, from molecular tests and bioinformatic analyses. Of the 472 participants, 36 and 6 NAT2 haplotypes were identified in the mestizo and native population, respectively. In mestizo population, the most frequent NAT2*5B and NAT2*7B haplotypes were associated with DILI risk; while in natives, NAT2*5G and NAT2*13A haplotypes were associated with decreased risk of DILI. For CYP2E1, c1/c1 and c1/c2 genotypes are the most frequent in natives and mestizos, respectively. The linkage disequilibrium of NAT2 single nucleotide polymorphisms (SNPs) was estimated, detecting a block between all SNPs natives. In addition, a block between rs1801280 and rs1799929 for NAT2 was detected in mestizos. Despite the limitations of a secondary study, it was possible to report associations between NAT2 and CYP2E alleles with Peruvian native and mestizo by prevalence ratios. The results of this study will help the development of new therapeutic strategies for a Tuberculosis efficient control between populations.

CDK4/6 inhibitors in drug-induced liver injury: a pharmacovigilance study of the FAERS database and analysis of the drug-gene interaction network.

The aim of this study was to investigate the potential risk of drug-induced liver injury (DILI) caused by the CDK4/6 inhibitors (CDK4/6is abemaciclib, ribociclib, and palbociclib by comprehensively analyzing the FDA Adverse Event Reporting System (FAERS) database. Moreover, potential toxicological mechanisms of CDK4/6is-related liver injury were explored via drug-gene network analysis. In this retrospective observational study, we collected reports of DILI associated with CDK4/6i use from the FAERS dated January 2014 to March 2023. We conducted disproportionality analyses using the reporting odds ratio (ROR) with a 95% confidence interval (CI). Pathway enrichment analysis and drug-gene network analyses were subsequently performed to determine the potential mechanisms underlying CDK4/6i-induced liver injury. We found positive signals for DILI with ribociclib (ROR = 2.60) and abemaciclib (ROR = 2.37). DILIs associated with liver-related investigations, signs, and symptoms were confirmed in all three reports of CDK4/6is. Moreover, ascites was identified as an unlisted hepatic adverse effect of palbociclib. We isolated 189 interactive target genes linking CDK4/6 inhibitors to hepatic injury. Several key genes, such as STAT3, HSP90AA1, and EP300, were revealed via protein-protein analysis, emphasizing their central roles within the network. KEGG pathway enrichment of these genes highlighted multiple pathways. Our study revealed variations in hepatobiliary toxicity among the different CDK4/6 inhibitors, with ribociclib showing the highest risk of liver injury, followed by abemaciclib, while palbociclib appeared relatively safe. Our findings emphasize the need for cautious use of CDK4/6 inhibitors, and regular liver function monitoring is recommended for long-term CDK4/6 inhibitor use.

The effect of statins on the risk of anti-tuberculosis drug-induced liver injury among patients with active tuberculosis: A cohort study

BackgroundTuberculosis (TB) remains prevalent worldwide, and anti-TB drugs are associated with drug-induced liver injury (DILI). Statins have pleiotropic effects which may decrease inflammation and achieve immunomodulation. However, few studies have investigated the pleiotropic effects of statins on the risk of DILI. The purpose of this study was to investigate whether statins prevent anti-tuberculosis DILI among active TB patients on standard anti-TB drug therapy. MethodsWe conducted a hospital-based retrospective cohort study using claims data from the Integrated Medical Database of National Taiwan University Hospital (NTUH-iMD). Patients with a positive TB culture were included. The use of statins was defined as a daily equivalent dose >0.5 mg of pitavastatin. Deterioration in liver function was evaluated according to elevated liver enzyme levels. The primary and secondary endpoints were the DILI and the severe DILI. The prognostic value of statins was evaluated by Kaplan–Meier analysis, and Cox proportional hazards models. ResultsA total of 1312 patients with a diagnosis of TB and receiving anti-TB treatment were included. During the study period, 193 patients had the DILI and 140 patients had the severe DILI. Kaplan–Meier analysis showed a significant difference between the usual statin users and controls in the DILI. In multivariable Cox proportional hazards analysis, statins showed a protective effect against the primary and secondary endpoints. In addition, the protective effect of statins showed a dose–response relationship against the DILI. ConclusionStatin treatment had a protective effect against the risk of anti-TB DILI with a positive dose–response relationship.

Drug Induced Liver Injury in Children: A Nationwide Cohort Study from China

Background & AimsCurrently, there is limited knowledge on the clinical profile of drug induced liver injury (DILI) in Chinese children. We aimed to assess the clinical characteristics, suspected drugs, and outcomes associated with pediatric DILI in China. MethodsThis nationwide, multi-center, retrospective study, conducted between 2012 and 2014, analyzed 25,927 cases of suspected DILI at 308 medical centers using the inpatient medical register system. Utilizing the Roussel Uclaf Causality Assessment Method (RUCAM) score, only patients with scores>6 or diagnosed with DILI by at least two experts after scoring <6 were included in the analysis. Among them, 460 cases met the European Association for the Study of Liver (EASL) biochemical criteria. The study categorized children into three age groups: toddlers (≥30 days to <6 years old), school-age children (6 to <12 years old), and adolescents (12 to <18 years old). ResultsHepatocellular injury was the predominant clinical classification, accounting for 63% of cases, with 34% of these cases meeting Hy's law criteria. Adolescents comprised the majority of children with moderate/severe DILI (65%). Similarly, adolescents faced a significantly higher risk of severe liver injury compared to younger children (adjusted odd ratio [aOR]=4.75, p =.002). The top three most frequently prescribed drug classes across all age groups were antineoplastic agents (25.9%), antimicrobials (21.5%), and traditional Chinese medicine (TCM, 13.7%). For adolescents, the most commonly suspected drugs were antituberculosis drugs (22%) and TCM (23%). ConclusionAdolescents are at a greater risk of severe and potentially fatal liver injury compared to younger children. Recognizing the risk of pediatric DILI is crucial for ensuring safe medical practices.

Angiotensin Receptor Blockers and the Risk of Suspected Drug-Induced Liver Injury: A Retrospective Cohort Study Using Electronic Health Record-Based Common Data Model in South Korea.

Angiotensin receptor blockers are widely used antihypertensive drugs in South Korea. In 2021, the Korea Ministry of Food and Drug Safety acknowledged the need for national compensation for a drug-induced liver injury (DILI) after azilsartan use. However, little is known regarding the association between angiotensin receptor blockers and DILI. We conducted a retrospective cohort study in incident users of angiotensin receptor blockers from a common data model database (1 January, 2017-31 December, 2021) to compare the risk of DILI among specific angiotensin receptor blockers against valsartan. Patients were assigned to treatment groups at cohort entry based on prescribed angiotensin receptor blockers. Drug-induced liver injury was operationally defined using the International DILI Expert Working Group criteria. Cox regression analyses were conducted to derive hazard ratios and the inverse probability of treatment weighting method was applied. All analyses were performed using R. In total, 229,881 angiotensin receptor blocker users from 20 university hospitals were included. Crude DILI incidence ranged from 15.6 to 82.8 per 1000 person-years in treatment groups, most were cholestatic and of mild severity. Overall, the risk of DILI was significantly lower in olmesartan users than in valsartan users (hazard ratio: 0.73 [95% confidence interval 0.55-0.96]). In monotherapy patients, the risk was significantly higher in azilsartan users than in valsartan users (hazard ratio: 6.55 [95% confidence interval 5.28-8.12]). We found a significantly higher risk of suspected DILI in patients receiving azilsartan monotherapy compared with valsartan monotherapy. Our findings emphasize the utility of real-world evidence in advancing our understanding of adverse drug reactions in clinical practice.

An analysis on clinical characteristics and prognosis-related risk factors in patients with drug-induced liver injury

Objective: To explore the drugs and clinical characteristics causing drug-induced liver injury (DILI) in recent years, as well as identify drug-induced liver failure, and chronic DILI risk factors, in order to better manage them timely. Methods: A retrospective investigation and analysis was conducted on 224 cases diagnosed with DILI and followed up for at least six months between January 2018 and December 2020. Univariate and multivariate logistic regression analyses were used to identify risk factors for drug-induced liver failure and chronic DILI. Results: Traditional Chinese medicine (accounting for 62.5%), herbal medicine (accounting for 84.3% of traditional Chinese medicine), and some Chinese patent medicines were the main causes of DILI found in this study. Severe and chronic DILI was associated with cholestatic type. Preexisting gallbladder disease, initial total bilirubin, initial prothrombin time, and initial antinuclear antibody titer were independent risk factors for DILI. Prolonged time interval between alkaline phosphatase (ALP) and alanine aminotransferase (ALT) falling from the peak to half of the peak (T(0.5ALP) and T(0.5ALT)) was an independent risk factor for chronic DILI [area under the receiver operating characteristic curve (AUC) = 0.787, 95%CI: 0.697~0.878, P < 0.001], with cutoff values of 12.5d and 9.5d, respectively. Conclusion: Traditional Chinese medicine is the main contributing cause of DILI. The occurrence risk of severe DILI is related to preexisting gallbladder disease, initial total bilirubin, prothrombin time, and antinuclear antibodies. T(0.5ALP) and T(0.5ALT) can be used as indicators to predict chronic DILI.

CYP7A1 Gene Induction via SHP-Dependent or Independent Mechanisms can Increase the Risk of Drug-Induced Liver Injury Independently or in Synergy with BSEP Inhibition.

Drug-induced liver injury (DILI) is a frequent cause of clinical trial failures during drug development. While inhibiting bile salt export pump (BSEP) is a well-documented DILI mechanism, interference with genes related to bile acid (BA) metabolism and transport can further complicate DILI development. Here, the effects of twenty-eight compounds on genes associated with BA metabolism and transport were evaluated, including those with discontinued development or use, boxed warnings, and clean labels for DILI. The study also included rifampicin and omeprazole, pregnane X receptor and aryl hydrocarbon receptor ligands, and four mitogen-activated protein kinase kinase (MEK1/2) inhibitors. BSEP inhibitors with more severe DILI, notably pazopanib and CP-724714, significantly upregulated the expression of 7 alpha-hydroxylase (CYP7A1), independent of small heterodimer partner (SHP) expression. CYP7A1 expression was marginally induced by omeprazole. In contrast, its expression was suppressed by mometasone (10-fold), vinblastine (18-fold), hexachlorophene (2-fold), bosentan (2.1-fold), and rifampin (2-fold). All four MEK1/2 inhibitors that show clinical DILI were not potent BSEP inhibitors but significantly induced CYP7A1 expression, accompanied by a significant SHP gene suppression. Sulfotransferase 2A1 and BSEP were marginally upregulated, but no other genes were altered by the drugs tested. Protein levels of CYP7A1 were increased with the treatment of CYP7A1 inducers and decreased with obeticholic acid, an farnesoid X receptor ligand. CYP7A1 inducers significantly increased bile acid (BA) production in hepatocytes, indicating the overall regulatory effects of BA metabolism. This study demonstrates that CYP7A1 induction via various mechanisms can pose a risk for DILI, independently or in synergy with BSEP inhibition, and it should be evaluated early in drug discovery. SIGNIFICANCE STATEMENT: Kinase inhibitors, pazopanib and CP-724714, inhibit BSEP and induce CYP7A1 expression independent of small heterodimer partner (SHP) expression, leading to increased bile acid (BA) production and demonstrating clinically elevated drug-induced liver toxicity. MEK1/2 inhibitors that show BSEP-independent drug-induced liver injury (DILI) induced the CYP7A1 gene accompanied by SHP suppression. CYP7A1 induction via SHP-dependent or independent mechanisms can pose a risk for DILI, independently or in synergy with BSEP inhibition. Monitoring BA production in hepatocytes can reliably detect the total effects of BA-related gene regulation for de-risking.

Risk of potential hepatotoxicity from pirfenidone or nintedanib in patients with idiopathic pulmonary fibrosis: results of a retrospective analysis of a large insurance database in Taiwan.

Background: A growing population of individuals diagnosed with idiopathic pulmonary fibrosis (IPF) are receiving treatment with nintedanib and pirfenidone. The aim of our study was to assess the incidence of drug-induced liver injury (DILI) associated with the use of pirfenidone and nintedanib in patients with IPF in Taiwan. Methods: We collected a cohort of adult patients diagnosed with IPF between 2017 and 2020. The research outcomes involved assessing the incidence of DILI in patients treated with nintedanib or pirfenidone. Poisson regression analysis was employed to estimate incidence rates, with and without adjustments for covariates, to calculate and present both unadjusted and adjusted incidence rate ratios (IRRs). Results: The risk of DILI was greater in patients who received nintedanib than in those who received pirfenidone during the 1-year follow-up. Patients treated with nintedanib exhibited a heightened risk of DILI based on inpatient diagnoses using specific codes after adjusting for variables such as gender, age group, comorbidities and concomitant medications, with an adjusted incidence rate ratio (aIRR) of 3.62 (95% confidence interval (CI) 1.11-11.78). Similarly, the risk of DILI was elevated in patients treated with nintedanib according to a per-protocol Poisson regression analysis of outcomes identified from inpatient diagnoses using specific codes. This was observed after adjusting for variables including gender, age group, comorbidities, and concomitant medications, with an aIRR of 3.60 (95% CI 1.11-11.72). Conclusion: Data from postmarketing surveillance in Taiwan indicate that patients who received nintedanib have a greater risk of DILI than do those who received pirfenidone.

Utility of human cytochrome P450 inhibition data in the assessment of drug-induced liver injury

1. Drug-induced liver injury (DILI) is a major cause of drug development discontinuation and drug withdrawal from the market, but there are no golden standard methods for DILI risk evaluation. Since we had found the association between DILI and CYP1A1 or CYP1B1 inhibition, we further evaluated the utility of cytochrome P450 (P450) inhibition assay data for DILI risk evaluation using decision tree analysis. 2. The inhibitory activity of drugs with DILI concern (DILI drugs) and no DILI concern (no-DILI drugs) against 10 human P450s was assessed using recombinant enzymes and luminescent substrates. The drugs were also subjected to cytotoxicity assays and high-content analysis using HepG2 cells. Molecular descriptors were calculated by alvaDesc. 3. Decision tree analysis was performed with the data obtained as variables with or without P450-inhibitory activity to discriminate between DILI drugs and no-DILI drugs. The accuracy was significantly higher when P450-inhibitory activity was included. After the decision tree discrimination, the drugs were further discriminated with the P450-inhibitory activity. The results demonstrated that many false-positive and false-negative drugs were correctly discriminated by using the P450 inhibition data. 4. These results suggest that P450 inhibition assay data are useful for DILI risk evaluation.

Artificial Intelligence in Liver Diseases: Recent Advances.

Liver diseases cause a significant burden on public health worldwide. In spite of great advances during recent years, there are still many challenges in the diagnosis and treatment of liver diseases. During recent years, artificial intelligence (AI) has been widely used for the diagnosis, risk stratification, and prognostic prediction of various diseases based on clinical datasets and medical images. Accumulative studies have shown its performance for diagnosing patients with nonalcoholic fatty liver disease and liver fibrosis and assessing their severity, and for predicting treatment response and recurrence of hepatocellular carcinoma, outcomes of liver transplantation recipients, and risk of drug-induced liver injury. Herein, we aim to comprehensively summarize the current evidence regarding diagnostic, prognostic, and/or therapeutic role of AI in these common liver diseases.

Computational models for predicting liver toxicity in the deep learning era.

Drug-induced liver injury (DILI) is a severe adverse reaction caused by drugs and may result in acute liver failure and even death. Many efforts have centered on mitigating risks associated with potential DILI in humans. Among these, quantitative structure-activity relationship (QSAR) was proven to be a valuable tool for early-stage hepatotoxicity screening. Its advantages include no requirement for physical substances and rapid delivery of results. Deep learning (DL) made rapid advancements recently and has been used for developing QSAR models. This review discusses the use of DL in predicting DILI, focusing on the development of QSAR models employing extensive chemical structure datasets alongside their corresponding DILI outcomes. We undertake a comprehensive evaluation of various DL methods, comparing with those of traditional machine learning (ML) approaches, and explore the strengths and limitations of DL techniques regarding their interpretability, scalability, and generalization. Overall, our review underscores the potential of DL methodologies to enhance DILI prediction and provides insights into future avenues for developing predictive models to mitigate DILI risk in humans.

Drug-Induced Liver Injury in the Elderly: Consensus Statements and Recommendations from the IQ-DILI Initiative.

The elderly demographic is the fastest-growing segment of the world's population and is projected to exceed 1.5 billion people by 2050. With multimorbidity, polypharmacy, susceptibility to drug-drug interactions, and frailty as distinct risk factors, elderly patients are especially vulnerable to developing potentially life-threatening safety events such as serious forms of drug-induced liver injury (DILI). It has been a longstanding shortcoming that elderly individuals are often a vulnerable population underrepresented in clinical trials. As such, an improved understanding of DILI in the elderly is a high-priority, unmet need. This challenge is underscored by recent documents put forward by the U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMA) that encourage data collection in the elderly and recommend improved practices that will facilitate a more inclusive approach. To establish what is already known about DILI in the elderly and pinpoint key gaps of knowledge in this arena, a working definition of "elderly" is required that accounts for both chronologic and biologic ages and varying states of frailty. In addition, it is critical to characterize the biological role of aging on liver function, as well as the different epidemiological factors such as polypharmacy and inappropriate prescribing that are common practices. While data may not show that elderly people are more susceptible to DILI, DILI due to specific drugs might be more common in this population. Improved characterization of DILI in the elderly may enhance diagnostic and prognostic capabilities and improve the way in which liver safety is monitored during clinical trials. This summary of the published literature provides a framework to understand and evaluate the risk of DILI in the elderly. Consensus statements and recommendations can help to optimize medical care and catalyze collaborations between academic clinicians, drug manufacturers, and regulatory scientists to enable the generation of high-quality research data relevant to the elderly population.

The integrated stress response-related expression of CHOP due to mitochondrial toxicity is a warning sign for DILI liability.

Drug-induced liver injury (DILI) is one of the most frequent reasons for failure of drugs in clinical trials or market withdrawal. Early assessment of DILI risk remains a major challenge during drug development. Here, we present a mechanism-based weight-of-evidence approach able to identify certain candidate compounds with DILI liabilities due to mitochondrial toxicity. A total of 1587 FDA-approved drugs and 378 kinase inhibitors were screened for cellular stress response activation associated with DILI using an imaging-based HepG2 BAC-GFP reporter platform including the integrated stress response (CHOP), DNA damage response (P21) and oxidative stress response (SRXN1). In total 389, 219 and 104 drugs were able to induce CHOP-GFP, P21-GFP and SRXN1-GFP expression at 50 μM respectively. Concentration response analysis identified 154 FDA-approved drugs as critical CHOP-GFP inducers. Based on predicted and observed (pre-)clinical DILI liabilities of these drugs, nine antimycotic drugs (e.g. butoconazole, miconazole, tioconazole) and 13 central nervous system (CNS) agents (e.g. duloxetine, fluoxetine) were selected for transcriptomic evaluation using whole-genome RNA-sequencing of primary human hepatocytes. Gene network analysis uncovered mitochondrial processes, NRF2 signalling and xenobiotic metabolism as most affected by the antimycotic drugs and CNS agents. Both the selected antimycotics and CNS agents caused impairment of mitochondrial oxygen consumption in both HepG2 and primary human hepatocytes. Together, the results suggest that early pre-clinical screening for CHOP expression could indicate liability of mitochondrial toxicity in the context of DILI, and, therefore, could serve as an important warning signal to consider during decision-making in drug development.

InterDILI: interpretable prediction of drug-induced liver injury through permutation feature importance and attention mechanism

Safety is one of the important factors constraining the distribution of clinical drugs on the market. Drug-induced liver injury (DILI) is the leading cause of safety problems produced by drug side effects. Therefore, the DILI risk of approved drugs and potential drug candidates should be assessed. Currently, in vivo and in vitro methods are used to test DILI risk, but both methods are labor-intensive, time-consuming, and expensive. To overcome these problems, many in silico methods for DILI prediction have been suggested. Previous studies have shown that DILI prediction models can be utilized as prescreening tools, and they achieved a good performance. However, there are still limitations in interpreting the prediction results. Therefore, this study focused on interpreting the model prediction to analyze which features could potentially cause DILI. For this, five publicly available datasets were collected to train and test the model. Then, various machine learning methods were applied using substructure and physicochemical descriptors as inputs and the DILI label as the output. The interpretation of feature importance was analyzed by recognizing the following general-to-specific patterns: (i) identifying general important features of the overall DILI predictions, and (ii) highlighting specific molecular substructures which were highly related to the DILI prediction for each compound. The results indicated that the model not only captured the previously known properties to be related to DILI but also proposed a new DILI potential substructural of physicochemical properties. The models for the DILI prediction achieved an area under the receiver operating characteristic (AUROC) of 0.88–0.97 and an area under the Precision-Recall curve (AUPRC) of 0.81–0.95. From this, we hope the proposed models can help identify the potential DILI risk of drug candidates at an early stage and offer valuable insights for drug development.

Drug-induced liver injury associated with elexacaftor/tezacaftor/ivacaftor: a pharmacovigilance analysis of the FDA Adverse Event Reporting System (FAERS)

Introduction: The efficacy and safety of elexacaftor/tezacaftor/ivacaftor (ETI) have been established in prospective clinical trials. Liver function test elevations were observed in a greater proportion of patients receiving ETI compared with placebo; however, the relatively small number of patients and short duration of study preclude detection of rare but clinically significant associations with drug-induced liver injury (DILI). To address this gap, we assessed the real-world risk of DILI associated with ETI through data mining of the FDA Adverse Event Reporting System (FAERS). Methods: Disproportionality analyses were conducted on FAERS data from the fourth quarter of 2019 through the third quarter of 2022. Comparative patient demographics, onset time and outcomes for ETI-DILI were also obtained. Results: 452 reports of DILI associated with ETI were found, representing 2.1% of all adverse event reports for ETI. All disproportionality measures were significant for ETI-DILI at p < 0.05; the reporting odds ratio (ROR) (2.82) was comparable to that of drugs classified by FDA as “Most-DILI concern”. The most notable demographic finding was a male majority (5:4 male to female ratio) for ETI-DILI compared to a female majority (4:5 male to female ratio) for non ETI-DILI. Median ETI-DILI onset time was 50.5 days, and hospitalization was the second most common complication. Conclusion: Using FAERS data, ETI was found to be disproportionately associated with DILI. Future research is needed to investigate the hepatotoxic mechanisms and assess potential mitigation strategies for ETI-induced hepatotoxicity.

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.