AI/ML Models to Predict the Severity of Drug-Induced Liver Injury for Small Molecules.

Drug-induced liver injury (DILI), believed to be a multifactorial toxicity, has been a leading cause of attrition of small molecules during discovery, clinical development, and postmarketing. Identification of DILI risk early reduces the costs and cycle times associated with drug development. In recent years, several groups have reported predictive models that use physicochemical properties or in vitro and in vivo assay endpoints; however, these approaches have not accounted for liver-expressed proteins and drug molecules. To address this gap, we have developed an integrated artificial intelligence/machine learning (AI/ML) model to predict DILI severity for small molecules using a combination of physicochemical properties and off-target interactions predicted in silico. We compiled a data set of 603 diverse compounds from public databases. Among them, 164 were categorized as Most DILI (M-DILI), 245 as Less DILI (L-DILI), and 194 as No DILI (N-DILI) by the FDA. Six machine learning methods were used to create a consensus model for predicting the DILI potential. These methods include k-nearest neighbor (k-NN), support vector machine (SVM), random forest (RF), Naïve Bayes (NB), artificial neural network (ANN), logistic regression (LR), weighted average ensemble learning (WA) and penalized logistic regression (PLR). Among the analyzed ML methods, SVM, RF, LR, WA, and PLR identified M-DILI and N-DILI compounds, achieving a receiver operating characteristic area under the curve of 0.88, sensitivity of 0.73, and specificity of 0.9. Approximately 43 off-targets, along with physicochemical properties (fsp3, log S, basicity, reactive functional groups, and predicted metabolites), were identified as significant factors in distinguishing between M-DILI and N-DILI compounds. The key off-targets that we identified include: PTGS1, PTGS2, SLC22A12, PPARγ, RXRA, CYP2C9, AKR1C3, MGLL, RET, AR, and ABCC4. The present AI/ML computational approach therefore demonstrates that the integration of physicochemical properties and predicted on- and off-target biological interactions can significantly improve DILI predictivity compared to chemical properties alone.

An in vitro coculture system of human peripheral blood mononuclear cells with hepatocellular carcinoma-derived cells for predicting drug-induced liver injury.

Preventing clinical drug-induced liver injury (DILI) remains a major challenge, because DILI develops via multifactorial mechanisms. Immune and inflammatory reactions are considered important mechanisms of DILI; however, biomarkers from in vitro systems using immune cells have not been comprehensively studied. The aims of this study were (1) to identify promising biomarker genes for predicting DILI in an in vitro coculture model of peripheral blood mononuclear cells (PBMCs) with a human liver cell line, and (2) to evaluate these genes as predictors of DILI using a panel of drugs with different clinical DILI risk. Transcriptome-wide analysis of PBMCs cocultured with HepG2 or differentiated HepaRG cells that were treated with several drugs revealed an appropriate separation of DILI-positive and DILI-negative drugs, from which 12 putative biomarker genes were selected. To evaluate the predictive performance of these genes, PBMCs cocultured with HepG2 cells were exposed to 77 different drugs, and gene expression levels in PBMCs were determined. The MET proto-oncogene receptor tyrosine kinase (MET) showed the highest area under the receiver-operating characteristic curve (AUC) value of 0.81 among the 12 genes with a high sensitivity/specificity (85/66%). However, a stepwise logistic regression model using the 12 identified genes showed the highest AUC value of 0.94 with a high sensitivity/specificity (93/86%). Taken together, we established a coculture system using PBMCs and HepG2 cells and selected biomarkers that can predict DILI risk. The established model would be useful in detecting the DILI potential of compounds, in particular those that involve an immune mechanism.

Most Influential Physicochemical and In Vitro Assay Descriptors for Hepatotoxicity and Nephrotoxicity Prediction.

Drug-induced organ injury is a major reason for drug candidate attrition in preclinical and clinical drug development. The liver, kidneys, and heart have been recognized as the most common organ systems affected in safety-related attrition or the subject of black box warnings and postmarket drug withdrawals. In silico physicochemical property calculations and in vitro assays have been utilized separately in the early stages of the drug discovery and development process to predict drug safety. In this study, we combined physicochemical properties and in vitro cytotoxicity assays including mitochondrial dysfunction to build organ-specific univariate and multivariable logistic regression models to achieve odds ratios for the prediction of clinical hepatotoxicity, nephrotoxicity, and cardiotoxicity using 215 marketed drugs. The multivariable hepatotoxic predictive model showed an odds ratio of 6.2 (95% confidence interval (CI) 1.7-22.8) or 7.5 (95% CI 3.2-17.8) for mitochondrial inhibition or drug plasma Cmax >1 µM for drugs associated with liver injury, respectively. The multivariable nephrotoxicity predictive model showed an odds ratio of 5.8 (95% CI 2.0-16.9), 6.4 (95% CI 1.1-39.3), or 15.9 (95% CI 2.8-89.0) for drug plasma Cmax >1 µM, mitochondrial inhibition, or hydrogen-bond-acceptor atoms >7 for drugs associated with kidney injury, respectively. Conversely, drugs with a total polar surface area ≥75 Å were 79% (odds ratio 0.21, 95% CI 0.061-0.74) less likely to be associated with kidney injury. Drugs belonging to the extended clearance classification system (ECCS) class 4, where renal secretion is the primary clearance mechanism (low permeability drugs that are bases/neutrals), were 4 (95% CI 1.8-9.5) times more likely to to be associated with kidney injury with this data set. Alternatively, ECCS class 2 drugs, where hepatic metabolism is the primary clearance (high permeability drugs that are bases/neutrals) were 77% less likely (odds ratio 0.23 95% CI 0.095-0.54) to to be associated with kidney injury. A cardiotoxicity model was poorly defined using any of these drug physicochemical attributes. Combining in silico physicochemical properties descriptors along with in vitro toxicity assays can be used to build predictive toxicity models to select small molecule therapeutics with less potential to cause liver and kidney organ toxicity.

A Novel Lipidomics-Based Approach to Evaluating the Risk of Clinical Hepatotoxicity Potential of Drugs in 3D Human Microtissues.

The importance of adsorption, distribution, metabolism, excretion, and toxicity (ADMET) analysis is expected to grow substantially due to recent failures in detecting severe toxicity issues of new chemical entities during preclinical/clinical development. Traditionally, safety risk assessment studies for humans have been conducted in animals during advanced preclinical or clinical phase of drug development. However, potential drug toxicity in humans now needs to be detected in the drug discovery process as soon as possible without reliance on animal studies. The "omics", such as genomics, proteomics, and metabolomics, have recently entered pharmaceutical research in both drug discovery and drug development, but to the best of our knowledge, no applications in high-throughput safety risk assessment have been attempted so far. This paper reports an innovative method to anticipate adverse drug effects in an early discovery phase based on lipid fingerprints using human three-dimensional microtissues. The risk of clinical hepatotoxicity potential was evaluated for a data set of 22 drugs belonging to five different therapeutic chemical classes and with various drug-induced liver injury effect. The treatment of microtissues with repeated doses of each drug allowed collecting lipid fingerprints for five time points (2, 4, 7, 9, and 11 days), and multivariate statistical analysis was applied to search for correlations with the hepatotoxic effect. The method allowed clustering of the drugs based on their hepatotoxic effect, and the observed lipid impairments for a number of drugs was confirmed by literature sources. Compared to traditional screening methods, here multiple interconnected variables (lipids) are measured simultaneously, providing a snapshot of the cellular status from the lipid perspective at a molecular level. Applied here to hepatotoxicity, the proposed workflow can be applied to several tissues, being tridimensional microtissues from various origins.

Moving beyond Binary Predictions of Human Drug-Induced Liver Injury (DILI) toward Contrasting Relative Risk Potential.

The hepatic risk matrix (HRM) was developed and used to differentiate lead clinical and back-up drug candidates against competitor/marketed drugs within the same pharmaceutical class for their potential to cause human drug-induced liver injury (DILI). The hybrid HRM scoring system blends physicochemical properties (Rule of Two Model: dose and lipophilicity or Partition Model: dose, ionization state, lipophilicity, and fractional carbon bond saturation) with common toxicity mechanisms (cytotoxicity, mitochondrial dysfunction, and bile salt export pump (BSEP) inhibition) that promote DILI. HRM scores are based on bracketed safety margins (<1, 1-10, 10-100, and >100× clinical Cmax,total). On the basis of well-established clinical safety experience of marketed/withdrawn drug candidates, the background analysis consists of 200 drugs from the Liver Toxicity Knowledge Base annotated as Most-DILI- (79), Less-DILI- (56), No-DILI- (47), and Ambiguous-DILI-concern (18) drugs. Scores were generated for over 21 internal and 7 external drug candidates discontinued for unacceptable incidence/magnitude of liver transaminase elevations during clinical trials or withdrawn for liver injury severity. Both hybrid scoring systems identified 70-80% Most-DILI-concern drugs, but more importantly, stratified successful/unsuccessful drug candidates for liver safety (incidence/severity of transaminase elevations and approved drug labels). Incorporating other mechanisms (reactive metabolite and cytotoxic metabolite generation and hepatic efflux transport inhibition, other than BSEP) to the HRM had minimal beneficial impact in DILI prediction/stratification. As is, the hybrid scoring system was positioned for portfolio assessments to contrast DILI risk potential of small molecule drug candidates in early clinical development. This stratified approach for DILI prediction aided decisions regarding drug candidate progression, follow-up mechanistic work, back-up selection, clinical dose selection, and due diligence assessments in favor of compounds with less implied clinical hepatotoxicity risk.

Evaluation of in Vitro Mitochondrial Toxicity Assays and Physicochemical Properties for Prediction of Organ Toxicity Using 228 Pharmaceutical Drugs.

Mitochondrial toxicity has been shown to contribute to a variety of organ toxicities such as liver, cardiac, and kidney. In the past decades, two high-throughput applicable screening assays (isolated rat liver mitochondria; glucose-galactose grown HepG2 cells) to assess mitochondrial toxicity have been deployed in many pharmaceutical companies, and numerous publications have demonstrated its usefulness for mechanistic investigations. However, only two publications have demonstrated the utility of these screens as a predictor of human drug-induced liver injury. In the present study, we screened 73 hepatotoxicants, 46 cardiotoxicants, 49 nephrotoxicants, and 60 compounds not known to cause human organ toxicity for their effects on mitochondrial function(s) in the assays mentioned above. Predictive performance was evaluated using specificity and sensitivity of the assays for predicting organ toxicity. Our results show that the predictive performance of the mitochondrial assays are superior for hepatotoxicity as compared to cardiotoxicity and nephrotoxicity (sensitivity 63% vs 33% and 28% with similar specificity of 93%), when the analysis was done at 100* Cmax (drug concentration in human plasma level). We further explored the association of mitochondrial toxicity with physicochemical properties such as calculated log partition coefficient (cLogP), topological polar surface area, ionization status, and molecular weight of the drugs and found that cLogP was most significantly associated mitochondrial toxicity. Since these assays are amenable to higher throughput, we recommend that chemists use these assays to perform structure activity relationship early in the drug discovery process, when chemical matter is abundant. This assures that compounds that lack the propensity to cause mitochondrial dysfunction (and associated organ toxicity) will move forward into animals and humans.

Liver safety evaluation of endothelin receptor antagonists using HepatoPac® : A single model impact assessment on hepatocellular health, function and bile acid disposition.

Marketed (bosentan, ambrisentan) and discontinued (sitaxsentan, CI-1034) endothelin receptor antagonists were examined in the human micropatterned hepatocyte co-culture (MPCC) model HepatoPac® . Differences across hepatocellular health (cellular adenosine triphosphate/glutathione content), function (urea production/albumin secretion) and taurocholic acid transport (biliary clearance/excretion index) were compared using amiodarone and ciclosporin A as positive controls. Ambrisentan had the weakest potency in all six endpoints, while sitaxsentan, bosentan and CI-1034 had more potent effects on hepatobiliary transport than health/function endpoints. Normalization to clinical Cmax gave the following relative rank order of safety based on margins for each endpoint: ambrisentan ≥ CI-1034 ~ bosentan > sitaxsentan. These data suggested impaired hepatobiliary disposition might contribute to a more prominent role in liver injury associated within sensitive human populations exposed to these compounds than direct hepatocellular toxicity. Rat, dog and monkey MPCCs also showed greater sensitivity potential to disrupted hepatobiliary disposition compared with hepatocellular health/functional endpoints. Drug metabolism competency was exhibited across all species. In vivo, rats and dogs appear more resistant to transaminase elevations and/or histological evidence of liver injury caused by these mechanisms even at exceedingly high systemic exposures relative to sensitive humans. Rats and dogs are resistant to hepatobiliary toxicants due to physiological differences in bile composition/handling. Although traditional animal testing provides adequate safety coverage for advancement of novel pharmaceuticals into clinical trials, supplemental assays employing human MPCCs may strengthen weight-of-evidence predictions for sensitive human populations. Proving the predictive value of this single impact assessment model in advance of clinical trial information for human liver injury risk is needed across more pharmaceuticals.

Lens cholesterol biosynthesis inhibition: A common mechanism of cataract formation in laboratory animals by pharmaceutical products.

CJ-12,918, a 5-lipoxygenase (5-LO) inhibitor, caused cataracts during a 1-month safety assessment studies in rats whereas the structurally similar ZD-2138 was without effect. For CJ-12,918 analogs, blocking different sites of metabolic liability reduced (CJ-13,454) and eliminated (CJ-13,610) cataract formation in both rats and dogs. Using this chemical series as a test set, models and mechanisms of toxicity were first explored by testing the utility of ex vivo rat lens explant cultures as a safety screen. This model overpredicted the cataractogenic potential of ZD-2138 due to appreciably high lens drug levels and was abandoned in favor of a mechanism-based screen. Perturbations in lens sterol content, from a decline in lathosterol content, preceded cataract formation suggesting CJ-12,918 inhibited lens cholesterol biosynthesis (LCB). A 2-day bioassay in rats using ex vivo LCB assessments showed that the level of LCB inhibition was correlated with incidence of cataract formation in animal studies by these 5-LO inhibitors. Thereafter, this 2-day bioassay was applied to other pharmaceutical programs (neuronal nitric oxide synthase, sorbitol dehydrogenase inhibitor, squalene synthetase inhibitor and stearoyl-CoA desaturase-1 inhibitors/D4 antagonists) that demonstrated cataract formation in either rats or dogs. LCB inhibition >40% was associated with a high incidence of cataract formation in both rats and dogs that was species specific. Bioassay sensitivity/specificity were further explored with positive (RGH-6201/ciglitazone/U18666A) and negative (tamoxifen/naphthalene/galactose) mechanistic controls. This body of work over two decades shows that LCB inhibition was a common mechanism of cataract formation by pharmaceutical agents and defined a level of inhibition >40% that was typically associated with causing cataracts in safety assessment studies typically ≥1 month.

Evaluating the Role of Multidrug Resistance Protein 3 (MDR3) Inhibition in Predicting Drug-Induced Liver Injury Using 125 Pharmaceuticals.

The role of bile salt export protein (BSEP) inhibition in drug-induced liver injury (DILI) has been investigated widely, while inhibition of the canalicular multidrug resistant protein 3 (MDR3) has received less attention. This transporter plays a pivotal role in secretion of phospholipids into bile and functions coordinately with BSEP to mediate the formation of bile acid-containing biliary micelles. Therefore, inhibition of MDR3 in human hepatocytes was examined across 125 drugs (70 of Most-DILI-concern and 55 of No-DILI-concern). Of these tested, 41% of Most-DILI-concern and 47% of No-DILI-concern drugs had MDR3 IC50 values of <50 µM. A better distinction across DILI classifications occurred when systemic exposure was considered where safety margins of 50-fold had low sensitivity (0.29), but high specificity (0.96). Analysis of physical chemical property space showed that basic compounds were twice as likely to be MDR3 inhibitors as acids, neutrals, and zwitterions and that inhibitors were more likely to have polar surface area (PSA) values of <100 Å2 and cPFLogD values between 1.5 and 5. These descriptors, with different cutoffs, also highlighted a group of compounds that shared dual potency as MDR3 and BSEP inhibitors. Nine drugs classified as Most-DILI-concern compounds (four withdrawn, four boxed warning, and one liver injury warning in their approved label) had intrinsic potency features of <20 µM in both assays, thereby reinforcing the notion that multiple inhibitory mechanisms governing bile formation (bile acid and phospholipid efflux) may confer additional risk factors that play into more severe forms of DILI as shown by others for BSEP inhibitors combined with multidrug resistance-associated protein (MRP2, MRP3, MRP4) inhibitory properties. Avoiding physical property descriptors that highlight dual BSEP and MDR3 inhibition or testing drug candidates for inhibition of multiple efflux transporters (e.g., BSEP, MDR3, and MRPs) may be an effective strategy for prioritizing drug candidates with less likelihood of causing clinical DILI.

Inhibition of Hepatobiliary Transport Activity by the Antibacterial Agent Fusidic Acid: Insights into Factors Contributing to Conjugated Hyperbilirubinemia/Cholestasis.

Conjugated hyperbilirubinemia accompanied by cholestasis is a frequent side effect during chronic treatment with the antimicrobial agent fusidic acid. Previous studies from our laboratory, addressing mechanisms of musculoskeletal toxicity arising from coadministration of fusidic acid with statins, demonstrated the ability of fusidic acid to potently inhibit human organic anion transporting polypeptides OATP1B1 (IC50 = 1.6 µM) and OATP1B3 (IC50 = 2.5 µM), which are responsible for the uptake-limited clearance of statins as well as bilirubin glucuronide conjugates. In the present work, inhibitory effects of fusidic acid were characterized against additional human hepatobiliary transporters [Na+/taurocholate cotransporting polypeptide (NTCP), bile salt export pump (BSEP), and multidrug resistance-associated proteins MRP2 and MRP3] as well as uridine glucuronosyl transferase (UGT1A1), which mediate the disposition of bile acids and bilirubin (and its conjugated metabolites). Fusidic acid demonstrated concentration-dependent inhibition of human NTCP- and BSEP-mediated taurocholic acid transport with IC50 values of 44 and 3.8 µM, respectively. Inhibition of BSEP activity by fusidic acid was also consistent with the potent disruption of cellular biliary flux (AC50 = 11 µM) in the hepatocyte imaging assay technology assay, with minimal impact on other toxicity end points (e.g., cytotoxicity, mitochondrial membrane potential, reactive oxygen species generation, glutathione depletion, etc.). Fusidic acid also inhibited UGT1A1-catalyzed ß-estradiol glucuronidation activity in human liver microsomes with an IC50 value of 16 µM. Fusidic acid did not demonstrate any significant inhibition of ATP-dependent LTC4 transport (IC50's > 300 µM) in human MRP2 or MRP3 vesicles. R values, which reflect maximal in vivo inhibition, were estimated from a static mathematical model by taking into consideration the IC50 values generated in the various in vitro assays and clinically efficacious unbound fusidic acid concentrations. The magnitudes of in vivo interaction (R values) resulting from the inhibition of OATP1B1, UGT1A1, NTCP, and BSEP transport were ∼1.9-2.6, 1.1-1.2, 1.0-1.1, and 1.4-1.7, respectively, which are indicative of some degree of inherent toxicity risk, particularly via inhibition of OATP and BSEP. Collectively, these observations indicate that inhibition of human BSEP by fusidic acid could affect bile acid homeostasis, resulting in cholestatic hepatotoxicity in the clinic. Lack of direct inhibitory effects on MRP2 transport by fusidic acid suggests that conjugated hyperbilirubinemia does not arise via interference in MRP2-mediated biliary disposition of bilirubin glucuronides. Instead, it is possible that elevation in the level of bilirubin conjugates in blood is mediated through inhibition of hepatic OATPs, which are responsible for their reuptake and/or downregulation of MRP2 transporter as a consequence of cholestatic injury.

Human drug-induced liver injury severity is highly associated with dual inhibition of liver mitochondrial function and bile salt export pump.

UNLABELLED: Drug-induced liver injury (DILI) accounts for 20-40% of all instances of clinical hepatic failure and is a common reason for withdrawal of an approved drug or discontinuation of a potentially new drug from clinical/nonclinical development. Numerous individual risk factors contribute to the susceptibility to human DILI and its severity that are either compound- and/or patient-specific. Compound-specific primary mechanisms linked to DILI include: cytotoxicity, reactive metabolite formation, inhibition of bile salt export pump (BSEP), and mitochondrial dysfunction. Since BSEP is an energy-dependent protein responsible for the efflux of bile acids from hepatocytes, it was hypothesized that humans exposed to drugs that impair both mitochondrial energetics and BSEP functional activity are more sensitive to more severe manifestations of DILI than drugs that only have a single liability factor. As annotated in the United States National Center for Toxicological Research Liver Toxicity Knowledge Base (NCTR-LTKB), the inhibitory properties of 24 Most-DILI-, 28 Less-DILI-, and 20 No-DILI-concern drugs were investigated. Drug potency for inhibiting BSEP or mitochondrial activity was generally correlated across human DILI concern categories. However, drugs with dual potency as mitochondrial and BSEP inhibitors were highly associated with more severe human DILI, more restrictive product safety labeling related to liver injury, and appear more sensitive to the drug exposure (Cmax) where more restrictive labeling occurs. CONCLUSION: These data affirm that severe manifestations of human DILI are multifactorial, highly associated with combinations of drug potency specifically related to known mechanisms of DILI (like mitochondrial and BSEP inhibition), and, along with patient-specific factors, lead to differences in the severity and exposure thresholds associated with clinical DILI.

Toxicity assessments of nonsteroidal anti-inflammatory drugs in isolated mitochondria, rat hepatocytes, and zebrafish show good concordance across chemical classes.

To reduce costly late-stage compound attrition, there has been an increased focus on assessing compounds in in vitro assays that predict attributes of human safety liabilities, before preclinical in vivo studies are done. Relevant questions when choosing a panel of assays for predicting toxicity are (a) whether there is general concordance in the data among the assays, and (b) whether, in a retrospective analysis, the rank order of toxicity of compounds in the assays correlates with the known safety profile of the drugs in humans. The aim of our study was to answer these questions using nonsteroidal anti-inflammatory drugs (NSAIDs) as a test set since NSAIDs are generally associated with gastrointestinal injury, hepatotoxicity, and/or cardiovascular risk, with mitochondrial impairment and endoplasmic reticulum stress being possible contributing factors. Eleven NSAIDs, flufenamic acid, tolfenamic acid, mefenamic acid, diclofenac, meloxicam, sudoxicam, piroxicam, diflunisal, acetylsalicylic acid, nimesulide, and sulindac (and its two metabolites, sulindac sulfide and sulindac sulfone), were tested for their effects on (a) the respiration of rat liver mitochondria, (b) a panel of mechanistic endpoints in rat hepatocytes, and (c) the viability and organ morphology of zebrafish. We show good concordance for distinguishing among/between NSAID chemical classes in the observations among the three approaches. Furthermore, the assays were complementary and able to correctly identify "toxic" and "non-toxic" drugs in accordance with their human safety profile, with emphasis on hepatic and gastrointestinal safety. We recommend implementing our multi-assay approach in the drug discovery process to reduce compound attrition.

Comparisons between in vitro whole cell imaging and in vivo zebrafish-based approaches for identifying potential human hepatotoxicants earlier in pharmaceutical development.

Drug-induced liver injury (DILI) is a major cause of attrition during both the early and later stages of the drug development and marketing process. Reducing or eliminating drug-induced severe liver injury, especially those that lead to liver transplants or death, would be tremendously beneficial for patients. Therefore, developing new pharmaceuticals that have the highest margins and attributes of hepatic safety would be a great accomplishment. Given the current low productivity of pharmaceutical companies and the high costs of bringing new medicines to market, any early screening assay(s) to identify and eliminate pharmaceuticals with the potential to cause severe liver injury in humans would be of economic value as well. The present review discusses the background, proof-of-concept, and validation studies associated with high-content screening (HCS) by two major pharmaceutical companies (Pfizer Inc and Jansen Pharmaceutical Companies of Johnson & Johnson) for detecting compounds with the potential to cause human DILI. These HCS assays use fluorescent-based markers of cell injury in either human hepatocytes or HepG2 cells. In collaboration with Evotec, an independent contract lab, these two companies also independently evaluated larval zebrafish as an early-stage in vivo screen for hepatotoxicity in independently conducted, blinded assessments. Details about this model species, the need for bioanalysis, and, specifically, the outcome of the phenotypic-based zebrafish screens are presented. Comparing outcomes in zebrafish against both HCS assays suggests an enhanced detection for hepatotoxicants of most DILI concern when used in combination with each other, based on the U.S. Food and Drug Administration DILI classification list.

Structural alert/reactive metabolite concept as applied in medicinal chemistry to mitigate the risk of idiosyncratic drug toxicity: a perspective based on the critical examination of trends in the top 200 drugs marketed in the United States.

Because of a preconceived notion that eliminating reactive metabolite (RM) formation with new drug candidates could mitigate the risk of idiosyncratic drug toxicity, the potential for RM formation is routinely examined as part of lead optimization efforts in drug discovery. Likewise, avoidance of "structural alerts" is almost a norm in drug design. However, there is a growing concern that the perceived safety hazards associated with structural alerts and/or RM screening tools as standalone predictors of toxicity risks may be over exaggerated. In addition, the multifactorial nature of idiosyncratic toxicity is now well recognized based upon observations that mechanisms other than RM formation (e.g., mitochondrial toxicity and inhibition of bile salt export pump (BSEP)) also can account for certain target organ toxicities. Hence, fundamental questions arise such as: When is a molecule that contains a structural alert (RM positive or negative) a cause for concern? Could the molecule in its parent form exert toxicity? Can a low dose drug candidate truly mitigate metabolism-dependent and -independent idiosyncratic toxicity risks? In an effort to address these questions, we have retrospectively examined 68 drugs (recalled or associated with a black box warning due to idiosyncratic toxicity) and the top 200 drugs (prescription and sales) in the United States in 2009 for trends in physiochemical characteristics, daily doses, presence of structural alerts, evidence for RM formation as well as toxicity mechanism(s) potentially mediated by parent drugs. Collectively, our analysis revealed that a significant proportion (∼78-86%) of drugs associated with toxicity contained structural alerts and evidence indicating that RM formation as a causative factor for toxicity has been presented in 62-69% of these molecules. In several cases, mitochondrial toxicity and BSEP inhibition mediated by parent drugs were also noted as potential causative factors. Most drugs were administered at daily doses exceeding several hundred milligrams. There was no obvious link between idiosyncratic toxicity and physicochemical properties such as molecular weight, lipophilicity, etc. Approximately half of the top 200 drugs for 2009 (prescription and sales) also contained one or more alerts in their chemical architecture, and many were found to be RM-positive. Several instances of BSEP and mitochondrial liabilities were also noted with agents in the top 200 category. However, with relatively few exceptions, the vast majority of these drugs are rarely associated with idiosyncratic toxicity, despite years of patient use. The major differentiating factor appeared to be the daily dose; most of the drugs in the top 200 list are administered at low daily doses. In addition, competing detoxication pathways and/or alternate nonmetabolic clearance routes provided suitable justifications for the safety records of RM-positive drugs in the top 200 category. Thus, while RM elimination may be a useful and pragmatic starting point in mitigating idiosyncratic toxicity risks, our analysis suggests a need for a more integrated screening paradigm for chemical hazard identification in drug discovery. Thus, in addition to a detailed assessment of RM formation potential (in relationship to the overall elimination mechanisms of the compound(s)) for lead compounds, effects on cellular health (e.g., cytotoxicity assays), BSEP inhibition, and mitochondrial toxicity are the recommended suite of assays to characterize compound liabilities. However, the prospective use of such data in compound selection will require further validation of the cellular assays using marketed agents. Until we gain a better understanding of the pathophysiological mechanisms associated with idiosyncratic toxicities, improving pharmacokinetics and intrinsic potency as means of decreasing the dose size and the associated "body burden" of the parent drug and its metabolites will remain an overarching goal in drug discovery.

Hepatotoxicity reports in the FDA adverse event reporting system database: A comparison of drugs that cause injury via mitochondrial or other mechanisms.

Drug-induced liver injury (DILI) is a leading reason for preclinical safety attrition and post-market drug withdrawals. Drug-induced mitochondrial toxicity has been shown to play an essential role in various forms of DILI, especially in idiosyncratic liver injury. This study examined liver injury reports submitted to the Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) for drugs associated with hepatotoxicity via mitochondrial mechanisms compared with non-mitochondrial mechanisms of toxicity. The frequency of hepatotoxicity was determined at a group level and individual drug level. A reporting odds ratio (ROR) was calculated as the measure of effect. Between the two DILI groups, reports for DILI involving mitochondrial mechanisms of toxicity had a 1.43 (95% CI 1.42-1.45; P < 0.0001) times higher odds compared to drugs associated with non-mitochondrial mechanisms of toxicity. Antineoplastic, antiviral, analgesic, antibiotic, and antimycobacterial drugs were the top five drug classes with the highest ROR values. Although the top 20 drugs with the highest ROR values included drugs with both mitochondrial and non-mitochondrial injury mechanisms, the top four drugs (ROR values > 18: benzbromarone, troglitazone, isoniazid, rifampin) were associated with mitochondrial mechanisms of toxicity. The major demographic influence for DILI risk was also examined. There was a higher mean patient age among reports for drugs that were associated with mitochondrial mechanisms of toxicity [56.1 ± 18.33 (SD)] compared to non-mitochondrial mechanisms [48 ± 19.53 (SD)] (P < 0.0001), suggesting that age may play a role in susceptibility to DILI via mitochondrial mechanisms of toxicity. Univariate logistic regression analysis showed that reports of liver injury were 2.2 (odds ratio: 2.2, 95% CI 2.12-2.26) times more likely to be associated with older patient age, as compared with reports involving patients less than 65 years of age. Compared to males, female patients were 37% less likely (odds ratio: 0.63, 95% CI 0.61-0.64) to be subjects of liver injury reports for drugs associated with mitochondrial toxicity mechanisms. Given the higher proportion of severe liver injury reports among drugs associated with mitochondrial mechanisms of toxicity, it is essential to understand if a drug causes mitochondrial toxicity during preclinical drug development when drug design alternatives, more clinically relevant animal models, and better clinical biomarkers may provide a better translation of drug-induced mitochondrial toxicity risk assessment from animals to humans. Our findings from this study align with mitochondrial mechanisms of toxicity being an important cause of DILI, and this should be further investigated in real-world studies with robust designs.

Using an in vitro cytotoxicity assay to aid in compound selection for in vivo safety studies.

Recent publications have demonstrated that using calculated physiochemical properties can help in the design of compounds that have a decreased risk of significant findings in rodent toxicology studies. In this Letter, we extend this concept and incorporate results from a high throughput cytotoxicity assay to help the drug discovery community select compounds for progression into in vivo studies. The results are presented in an easily interpretable odds ratio so that teams can readily compare compounds and progress potential clinical candidates to the necessary rodent in vivo studies.

A current practice for predicting ocular toxicity of systemically delivered drugs.

The ability to predict ocular side effects of systemically delivered drugs is an important issue for pharmaceutical companies. Although animal models involving standard clinical ophthalmic examinations and postmortem microscopic examinations of eyes are still used to identify ocular issues, these methods are being supplemented with additional in silico, in vitro, and in vivo techniques to identify potential safety issues and assess risk. The addition of these tests to a development plan for a potential new drug provides the opportunity to save time and money by detecting ocular issues earlier in the program. This review summarizes a current practice for minimizing the potential for systemically administered, new medicines to cause adverse effects in the eye.

Phase I study of PF­04895162, a Kv7 channel opener, reveals unexpected hepatotoxicity in healthy subjects, but not rats or monkeys: clinical evidence of disrupted bile acid homeostasis.

During a randomized Phase 1 clinical trial the drug candidate, PF-04895162 (ICA-105665), caused transaminase elevations (≥grade 1) in six of eight healthy subjects treated at 300 mg twice daily for 2-weeks (NCT01691274). This was unexpected since studies in rats (<6 months) and cynomolgus monkeys (<9 months) treated up to 100 mg/kg/day did not identify the liver as a target organ. Mechanistic studies showed PF-04895162 had low cytotoxic potential in human hepatocytes, but inhibited liver mitochondrial function and bile salt export protein (BSEP) transport. Clinical relevance of these postulated mechanisms of liver injury was explored in three treated subjects that consented to analysis of residual pharmacokinetic plasma samples. Compared to a nonresponder, two subjects with transaminase elevations displayed higher levels of miRNA122 and total/conjugated bile acid species, whereas one demonstrated impaired postprandial clearance of systemic bile acids. Elevated taurine and glycine conjugated to unconjugated bile acid ratios were observed in two subjects, one before the onset of elevated transaminases. Based on the affinity of conjugated bile acid species for transport by BSEP, the profile of plasma conjugated/unconjugated bile acid species was consistent with inhibition of BSEP. These data collectively suggest that the human liver injury by PF-04895162 was due to alterations in bile acid handling driven by dual BSEP/mitochondrial inhibition, two important risk factors associated with drug-induced liver injury in humans. Alterations in systemic bile acid composition were more important than total bile acids in the manifestation of clinical liver injury and may be a very early biomarker of BSEP inhibition.

Quantitative systems toxicology (QST) reproduces species differences in PF-04895162 liver safety due to combined mitochondrial and bile acid toxicity.

Many compounds that appear promising in preclinical species, fail in human clinical trials due to safety concerns. The FDA has strongly encouraged the application of modeling in drug development to improve product safety. This study illustrates how DILIsym, a computational representation of liver injury, was able to reproduce species differences in liver toxicity due to PF-04895162 (ICA-105665). PF-04895162, a drug in development for the treatment of epilepsy, was terminated after transaminase elevations were observed in healthy volunteers (NCT01691274). Liver safety concerns had not been raised in preclinical safety studies. DILIsym, which integrates in vitro data on mechanisms of hepatotoxicity with predicted in vivo liver exposure, reproduced clinical hepatotoxicity and the absence of hepatotoxicity observed in the rat. Simulated differences were multifactorial. Simulated liver exposure was greater in humans than rats. The simulated human hepatotoxicity was demonstrated to be due to the interaction between mitochondrial toxicity and bile acid transporter inhibition; elimination of either mechanism from the simulations abrogated injury. The bile acid contribution occurred despite the fact that the IC50 for bile salt export pump (BSEP) inhibition by PF-04895162 was higher (311 µmol/L) than that has been generally thought to contribute to hepatotoxicity. Modeling even higher PF-04895162 liver exposures than were measured in the rat safety studies aggravated mitochondrial toxicity but did not result in rat hepatotoxicity due to insufficient accumulation of cytotoxic bile acid species. This investigative study highlights the potential for combined in vitro and computational screening methods to identify latent hepatotoxic risks and paves the way for similar and prospective studies.