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

HepatologyVolume 60, Issue 3 p. 1015-1022 Liver Injury/RegenerationFree Access Human drug-induced liver injury severity is highly associated with dual inhibition of liver mitochondrial function and bile salt export pump Michael D. Aleo, Corresponding Author Michael D. Aleo Investigative Toxicology, Drug Safety Research and Development, Pfizer Inc, Groton, CTAddress reprint requests to: Dr. Michael D. Aleo, Drug Safety Research and Development, MS 8274-1229, Eastern Point Road, Groton, CT 06340. E-mail: michael.d.aleo@pfizer.com; fax: 860-715-8528.Search for more papers by this authorYi Luo, Yi Luo Compound Safety Prediction, Worldwide Medicinal Chemistry, Pfizer Inc., Groton, CTSearch for more papers by this authorRachel Swiss, Rachel Swiss Compound Safety Prediction, Worldwide Medicinal Chemistry, Pfizer Inc., Groton, CTSearch for more papers by this authorPaul D. Bonin, Paul D. Bonin Assay Development and Pharmacology, Primary Pharmacology Group, Pfizer Inc., Groton, CTSearch for more papers by this authorDavid M. Potter, David M. Potter Regulatory Strategy & Compliance, Drug Safety Research and Development, Pfizer Inc., Groton, CTSearch for more papers by this authorYvonne Will, Yvonne Will Compound Safety Prediction, Worldwide Medicinal Chemistry, Pfizer Inc., Groton, CTSearch for more papers by this author Michael D. Aleo, Corresponding Author Michael D. Aleo Investigative Toxicology, Drug Safety Research and Development, Pfizer Inc, Groton, CTAddress reprint requests to: Dr. Michael D. Aleo, Drug Safety Research and Development, MS 8274-1229, Eastern Point Road, Groton, CT 06340. E-mail: michael.d.aleo@pfizer.com; fax: 860-715-8528.Search for more papers by this authorYi Luo, Yi Luo Compound Safety Prediction, Worldwide Medicinal Chemistry, Pfizer Inc., Groton, CTSearch for more papers by this authorRachel Swiss, Rachel Swiss Compound Safety Prediction, Worldwide Medicinal Chemistry, Pfizer Inc., Groton, CTSearch for more papers by this authorPaul D. Bonin, Paul D. Bonin Assay Development and Pharmacology, Primary Pharmacology Group, Pfizer Inc., Groton, CTSearch for more papers by this authorDavid M. Potter, David M. Potter Regulatory Strategy & Compliance, Drug Safety Research and Development, Pfizer Inc., Groton, CTSearch for more papers by this authorYvonne Will, Yvonne Will Compound Safety Prediction, Worldwide Medicinal Chemistry, Pfizer Inc., Groton, CTSearch for more papers by this author First published: 06 May 2014 https://doi.org/10.1002/hep.27206Citations: 161 Potential conflict of interest: Dr. Aleo is employed by and owns stock in Pfizer. Dr. Bonin is employed by and owns stock in Pfizer. Dr. Potter is employed by and owns stock in Pfizer. Dr. Luo is employed by and owns stock in Pfizer and Bristol-Myers Squibb. Dr. Will is employed by and owns stock in Pfizer. Dr. Swiss is employed by and owns stock in Pfizer. This work did not receive external public or private foundation funding for this project. AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract 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. (Hepatology 2014;60:1015–1022) Abbreviations ATP adenosine triphosphate BBW black box warnings BSEP bile salt export pump Cmax highest plasma drug concentration CMH Cochran-Mantel-Haenszel statistical test DILI drug-induced liver injury FDA United States Food and Drug Administration IC50 apparent inhibitory concentration at 50% effect NCTR-LTKB National Center for Toxicological Research Liver Toxicity Knowledge Base NS nonsignificant The incidence of human drug-induced liver injury (DILI) has been reported in both retrospective and prospective studies.1 It accounts for 20-40% of all instances of hepatic failure, is an important cause of liver transplantation due to acute liver failure, and is still a leading reason for withdrawal of approved drugs.1-4 Human DILI can range in severity from minor, inconsequential elevations in liver transaminases to overwhelming damage leading to liver failure and death. For drugs that cause human DILI, nonclinical safety assessment studies do not always demonstrate appreciable signals of liver injury in animals (e.g., significant transaminase and/or bilirubin elevations in the presence of histological evidence of necrosis) that would highlight the liver as a potential target organ of concern in humans.5, 6 However, human variability in susceptibility to DILI is thought to be linked to various patient centric DILI risk factors that contribute to individual patient susceptibility and outcome.7-11 Given the numerous patient-centric factors that lead to DILI, it is difficult to understand which risk factor(s) are more important in patient-specific responses. However, a number of compound-specific liability factors have been determined through mechanistic evaluations and affirmed with further animal studies or human marketing experience. Some of these risk factors can be mitigated during the drug design/development process to identify drugs with better chemical attributes with reduced potential to cause human DILI. For example, Chen et al.12 recently reported that high lipophilicity in combination with high daily oral dose increases DILI risk potential in humans. Another drug design feature involved with increased DILI liability is the biotransformation of certain chemical structures that lead to the generation of reactive metabolites, redox cycling, and a high covalent body burdens.13, 14 Hepatobiliary transporters also appear to play a role in adverse drug reactions.15 One transporter of interest is the canalicular adenosine triphosphate (ATP)-dependent bile salt export pump (BSEP, ABCB11). This transporter belongs to the ATP-binding cassette superfamily of transporters,16-18 is predominantly expressed in liver,19 and is responsible for the excretion of bile salts, the major driving force for biliary flow. A number of drugs, such as bosentan, troglitazone, nefazodone, and cyclosporine A, inhibit BSEP activity and induce cholestasis in humans.20-25 A recent analysis of marketed drugs suggests that interference of BSEP activity is associated with clinical DILI.26 However, while some potent BSEP inhibitors have been withdrawn from the market (e.g., troglitazone and benzbromarone), others were associated with only a low risk for DILI (e.g., glimepiride, pioglitazone, simvastatin). It would therefore appear that BSEP inhibitory potency alone is insufficient for determining the severity of the DILI risk and that additional liability factors are involved.27 The involvement of mitochondrial dysfunction appears to be an additional mechanistic liability for DILI.27-29 Mitochondrial dysfunction can de-energize a cell and lead to oxidative stress, apoptosis, and hepatocellular injury. More important, reductions in hepatic ATP levels are associated with reduced biliary excretion in rats30, 31 and, given that several ATP-dependent canalicular transporters are responsible for hepatobiliary excretion (e.g., bile acids through BSEP and bilirubin through MRP2 transport) it is known that reductions in hepatic ATP levels are correlated with reductions in bile acid and indocyanine green (an MRP2 substrate) excretion in humans.32 Moreover, the accumulation of cytotoxic bile acids within hepatocytes, either through inhibition of BSEP or a reduction in transporter function, has been long known to disrupt mitochondrial function.33 It should also be noted that in certain disease states, like type 2 diabetes34, 35 and nonalcoholic steatohepatitis,36 there are significant deficits in normal mitochondrial function, which in turn may further predispose individual patients to DILI through this mechanism.37 During an internal review of in-house mitochondrial toxicity data accumulated over several years, it was noted that some of the same compounds reported to inhibit mitochondrial function were also potent BSEP inhibitors and that these agents were withdrawn from the market due to severe DILI (e.g., nefazodone, benzbromarone, troglitazone).38-40 For the reasons listed above, we hypothesized that the combined attributes of potent inhibition of mitochondrial function and BSEP transport may be more frequently associated with drugs that cause more severe forms of human DILI. Here, 72 compounds were investigated for their effects on mitochondrial respiration and apparent inhibition of human BSEP transport activity. Using annotations of human DILI concern collected by Chen et al.41, 42 or from internal efforts examining the currently approved label using the same annotation principles, differences in these intrinsic liability factors were correlated with varying degrees of clinical DILI concern. Materials and Methods Compounds were obtained from Sigma-Aldrich (St. Louis, MO), Toronto Research Chemicals (Toronto, Ontario), or from the Pfizer chemical bank (Groton, CT). The phosphorescent oxygen-sensitive probe, type A65N-1, was procured from Luxcel Biosciences (Cork, Ireland). For the BSEP vesicle assay [3H]taurocholic acid (2 Ci/mmol) was purchased from NEN Life Science Products (Boston, MA). Vesicles prepared from Sf9 insect cells transfected with human BSEP (SB-BSEP-Sf9-VT) were purchased from Solvo Biotechnology (Boston, MA). Animals Care and maintenance of animals used in this study were reviewed and approved by an in-house animal care and use committee and conducted in accordance with the principles described in the “Guide for Care and Use of Laboratory Animals,” Eighth Edition (Institute of Laboratory Animal Resources, National Research Council, National Academy Press, Washington, DC, 2011). Male Sprague-Dawley Rats (150-180 g) were purchased from Charles River (Wilmington, MA). The rats were housed in pairs in a controlled environment with constant temperature (21 ± 2°C) and a 12-hour light/dark cycle. Food and water were available ad libitum during the 2-week acclimatization period. Animals were euthanized with an overdose of carbon dioxide. The liver was rapidly excised and placed into ice-cold mitochondrial isolation buffer. Measurement of Mitochondrial Respiration Oxygen consumption in freshly isolated rat liver mitochondria was monitored in 96-well plate format using a phosphorescent oxygen-sensitive probe.38, 40, 43 Oxygen consumption rates were determined based on the known relationship between probe fluorescence and dissolved oxygen concentration43, 44 and were determined from the slopes of these concentration profiles using 100 nmol/mg mitochondrial protein as the highest tested concentration. The interassay variability has been published.45 Measurement of Bile Salt Transport Inhibition in Human BSEP Vesicle Assay Human BSEP activity was determined by 96-well plates measuring the transport of [3H]taurocholic acid in SB-BSEP-Sf9-VT vesicles. The assay methodology has been published,46 used glyburide as a positive control, and used 100 μM as the highest tested concentration. The interassay variability is presented in Supporting Table 1. Statistical Methods Drug names and data for this analysis are contained in Supporting Table 2. The Cochran-Mantel-Haenszel (CMH) test47 was used to determine if an overall relationship between: 1) category of liver injury and category of BSEP inhibition; 2) category of liver injury and mitochondrial inhibition; 3) category of liver injury and a combined categorization based on BSEP and mitochondrial inhibition; and 4) category of DILI Label Sections and a combined categorization based on BSEP and mitochondrial inhibition. The CMH test was used since it takes into account the ordering of both categories using a minimum value of P < 0.05. Results Pharmaceutical agents from three different categories of clinical liver injury, as annotated by Chen et al.,42 were investigated: 24 (Most-DILI-concern), 28 (Less-DILI-concern), and 20 (No-DILI-concern). The data for this analysis are found in online Supporting Table 2. Of the 24 Most-DILI-concern compounds investigated, 10 (42%) were highly active inhibitors of human BSEP (apparent inhibitory concentration at 50% effect [IC50] ≤10 μM), whereas the remaining 14 (58%) had intermediate activity (apparent IC50 value ranged between >10 to ≤100 μM) (Fig. 1). A similar distribution pattern for BSEP inhibition was observed for drugs with Less-DILI-concern, except for furosemide and buspirone (7%), which had no activity in the BSEP assay (IC50 >100 μM). In contrast, for drugs in the No-DILI-concern category, 45% had no activity in the BSEP assay while 10 and 45% of drugs had high or intermediate activity, respectively. There was an overall difference in the distribution of apparent BSEP inhibitory activity across the three DILI categories, with the most discrimination found between compounds in the Most- or Less- versus No-DILI-concern category. Figure 1Open in figure viewerPowerPoint Drug distribution of apparent inhibition of human bile salt export protein (BSEP) transport across Most-, Less-, and No-DILI-Concern categories. Black shading = BSEP IC50 values ≤10 μM. Gray shading = BSEP IC50 values >10 to ≤100 μM. No shading = BSEP IC50 values >100 μM. Overall significance was determined using Cochran-Mantel-Haenszel test (P < 0.001) with values of pairwise comparisons noted within the figure or nonsignificant (NS) (P > 0.05). A similar association was found for drugs that inhibited mitochondrial function across the Most-, Less-, and No-DILI-concern categories (Fig. 2). Highly active mitochondrial inhibitors (IC50 ≤100 nmol/mg mitochondrial protein) represented 38, 14, and 0% of compounds in the Most-, Less-, and No-DILI-concern categories, respectively. Compounds with intermediate inhibitory activity (IC50 >100 nmol/mg mitochondrial protein) also showed a similar distribution pattern. However, the majority of the compounds in the No-DILI-concern category (90%) had “No Effect” (no signal at the highest concentration tested, 100 nmol/mg mitochondrial protein) in the mitochondrial inhibition assay, compared to a distribution pattern of 64 and 33% of compounds in the Less- and Most-DILI-concern category, respectively. The overall difference in the distribution of mitochondrial inhibitory activity in drugs among the three DILI categories was also most significant between the Most- and No-DILI-concern categories. Figure 2Open in figure viewerPowerPoint Drug distribution of apparent inhibition of rat liver mitochondrial function across Most-, Less-, and No-DILI-Concern categories. Black shading = IC50 values ≤100 nmol/mg mitochondrial protein. Gray shading = IC50 values >100 nmol/mg mitochondrial protein. No shading = no effect (NE) in the assay up to the 100 nmol/mg mitochondrial protein testing concentration limit. Overall significance was determined using Cochran-Mantel-Haenszel test (P = 0.002) with values of pairwise comparisons noted within the figure or nonsignificant (NS) (P > 0.05). Pharmaceuticals with mechanistic liabilities as both BSEP and mitochondrial inhibitors showed a disproportionate distribution across DILI-concern categories (Fig. 3). The majority of dual potency drugs (potency defined here as BSEP IC50 <100 μM with mitochondrial inhibition IC50 ≤100 nmol/mg protein) were found in the Most-DILI-concern category (50%), while only 29 and 0% were found in the Less-DILI and No-DILI-concern categories, respectively. Dual moderate activity in both assays (BSEP IC50 >10 to ≤100 μM with mitochondrial inhibition IC50 >100 nmol/mg protein) represented 17, 7, and 10% of drugs within the Most-, Less-, and No-DILI-concern category, respectively. The presence of compounds with combinations of both “Dual Potency” and “Dual Moderate Activity” liabilities represented 67, 36, and only 10% of compounds in the Most-, Less-, and No-DILI-concern category, respectively. Compounds that were active BSEP inhibitors alone (BSEP IC50 ≤100 μM with “No Effect” in mitochondrial inhibition assay up to 100 nmole/mg mitochondrial protein) were uniformly represented across the three DILI concern categories. In this dataset, compounds with no intrinsic liability in either assay (BSEP IC50 >100 μM with “No Effect” in mitochondrial inhibition assay) were largely represented (45%) in the No-DILI-concern category and were not found in the Most-DILI-concern category. Furthermore, drugs with BSEP inhibition either had mitochondrial inhibitory liabilities or not. Overall, the combination of activity (potency) in these dual risk factors discriminated well between compounds in the Most- or Less- versus No-DILI-concern category. Figure 3Open in figure viewerPowerPoint Distribution of apparent inhibition human bile salt export protein (BSEP) transport and rat liver mitochondrial function across Most-, Less-, and No-DILI-Concern categories. Black shading = dual potency compounds where BSEP IC50 values ≤100 μM and mitochondrial IC50 values ≤100 nmol/mg mitochondrial protein. Dark gray shading = dual moderate activity compounds with BSEP IC50 values > 10-≤100 μM with mitochondrial IC50 values >100 nmol/mg mitochondrial protein. Light gray shading = compounds with BSEP IC50 values ≤100 μM alone with no effect in mitochondrial assay up to the 100 nmol/mg mitochondrial protein testing concentration limit. No shading = compounds with no effect (NE) in either assay (BSEP IC50 values >100 μM and no effect in mitochondrial assay). Overall significance was determined using Cochran-Mantel-Haenszel test (P < 0.001) with values of pairwise comparisons noted within the figure or nonsignificant (NS) (P > 0.05). This association of dual potency in these two mechanistic liabilities with DILI severity was further reinforced using the Food and Drug Administration (FDA) currently approved label categorization for these drugs (Fig. 4). Over half of the drugs that were “Withdrawn” from the marketplace/clinical development or carried black box warnings (BBW) for severe liver injury were potent dual inhibitors of BSEP and mitochondria (67%), compared to 31% of drugs with a “Warnings & Precautions” label, 35% with “Adverse Reactions,” and 0% with “No Mention” of liver injury in their current label. Drugs with dual moderate activity represented 22, 12, 6, and 10% of compounds with “Withdrawn or BBW,” “Warnings & Precautions,” “Adverse Reactions,” and “No Mention” of liver injury in their FDA label, respectively. Once again, drugs with activity in the BSEP assay alone (BSEP IC50 ≤100 μM with “No Effect” in mitochondrial inhibition assay) were represented across all four DILI label categories (Withdrawn or BBW [11%], Warnings & Precautions [58%], Adverse Reactions [47%] and No Mention [45%]), while drugs with “No Effect” in either assay represented 12 and 45% of the compounds with “Adverse Reactions” or “No Mention” of liver injury in their current label, respectively. Figure 4Open in figure viewerPowerPoint Breakdown of intrinsic BSEP and mitochondrial inhibition liability categories as a function of FDA label categories. Figure legend for potency categories as stated in Fig. 3. Overall significance was determined using Cochran-Mantel-Haenszel test (P < 0.001) with values of pairwise comparisons noted within the figure or nonsignificant (NS) (P > 0.05). The relationship between these dual risk factors and their approved labeling as a function of human exposure was also examined (Fig. 5). There is good representation of highest plasma drug concentration (Cmax) (total) across at least 3 orders of magnitude at approved clinical doses for compounds in each of the intrinsic liability factor categories (Dual Potency, Dual Moderate Activity, BSEP Alone, and No Effect). Compounds with “No Mention” or “Adverse Reactions” in their respective labels and have “No Activity” in either assay were not represented by compounds with low clinical exposure alone. Approximately half of the compounds have Cmax (total) values of greater than 1 μM (Fig. 5). In contrast, compounds with either “Dual Potency” or “Dual Moderate Activity” as intrinsic liability categories have “Withdrawn or BBW” labels starting at a Cmax (total) of ∼0.4 μM compared to ∼4 μM for compounds with a single liability (BSEP Alone), suggesting that dual potency not only drives differences in DILI severity categories but also sensitivity to DILI based on human exposure. Figure 5Open in figure viewerPowerPoint Distribution of FDA label categories for liver injury across categories of intrinsic BSEP and mitochondrial inhibition liability as a function of human Cmax (μM, total). Figure legend for potency categories as stated in Fig. 3. Discussion The convergence of multiple risk factors, as an important consideration for understanding the manifestation of DILI in humans, is not a new concept. Numerous hypotheses have been proposed by various researchers in this area as well as their relationship to highlighting a purported human DILI concern.48 However, the idea of integrating interdependent physiological processes and weighting to multiple risk factors, although the subject of modeling efforts, has not been investigated beyond individual risk factors alone.26, 49 There are many physiological reasons outlined in the introduction to suspect that drugs that are dual and potent BSEP and mitochondrial inhibitors may play an important role in the manifestation of clinical DILI such as: 1) BSEP is an ATP-dependent transport protein; 2) hepatobiliary transporter activity (BSEP and MRP2) and/or bile flow in rats and humans is directly proportional to intracellular ATP levels30-32; and 3) diminished activity of ATP-dependent transport pumps causes a cascade of toxic events due to the accumulation of cytotoxic bile acids within the liver.20, 46, 50 Furthermore, individuals with impaired mitochondrial function (e.g., diabetics) or dysfunctional BSEP (e.g., patients with progressive familial or benign recurrent intrahepatic cholestasis), patients taking concomitant medications that have combinations of these mechanistic liabilities (e.g., the combination interaction of bosentan with glyburide),20 may be more sensitive to drugs that are potent inhibitors of these important physiological processes. The relationship between concurrent inhibition of BSEP and mitochondrial function was demonstrated across the three DILI concern categories (NCTR-LTKB) and four FDA label categories for liver injury. Although as a single liability no activity in the BSEP assay was better at separating drugs represented from the Most- and No-DILI-concern categories (0% compared to 45%, respectively) (Fig. 1), mitochondrial inhibition was disproportionately represented in the three DILI categories as well (Fig. 2). Activity in the BSEP assay alone was represented fairly uniformly across the three DILI categories, suggesting that inhibition of BSEP alone is insufficient to differentiate across potential DILI risks (Fig. 3). When these intrinsic mechanistic risk liability factors are considered in combination, Dual Potency inhibitors were found to a greater extent in the Most- compared to Less-DILI-concern category (50% compared to 29%) with none present in the No-DILI-concern group (Fig. 3). In contrast, drugs with “No Effect” in the combined assays were primarily found in the No-DILI-concern group (45%). Therefore, the combination of these mechanistic liabilities appears to have more potential to discriminate drugs with DILI concern better than investigating these liabilities in isolation. Human exposure is an important factor in understanding how intrinsic liabilities and potencies translate into biological systems. Approximations of human Cmax at the highest approved clinical doses are a good surrogate for placing in vitro information into better biological context and perspective since actual measurements of drug exposure in patients affected with DILI are difficult to obtain. In Fig. 5, the pharmaceuticals in this analysis represent a wide range of human Cmax values. Most intriguing is that drugs with no activity in either assay had no or low concerns for human DILI across a wide range of human doses and in many cases drugs that exhibited high potency/activity in either assay but were of No-DILI-concern had low clinical exposure (e.g., beclomethasone, clobetasol, econazole, mometasone, nimodipine), thereby negating the intrinsic liability. Drugs with either dual potency/activity in these two assays appeared to have a lower threshold at which a more significant DILI concern was highlighted. In this dataset the lowest Cmax where this occurred was ∼0.4 μM (Cmax, total) compared to ∼4 μM (Cmax, total) for BSEP inhibitors alone (Fig. 5). Therefore, although from a chemical design perspective the absence of these combined risk factors increases the likelihood for a safer drug in terms of clinical liver injury, lower efficacious clinical exposures can help mitigate these risk factors when present. Recognizing the interplay between these two important mechanistic liability factors for human DILI is critical to current research. Pharmaceutical screening strategies can be improved, especially if structure toxicity relationships can be found. Besides obvious implications in pharmaceutical chemical design, there are also important elements in in silico modeling efforts like DILIsym. Two compounds of importance in this modeling effort are CP-724714 and etomoxir. As exemplar compounds, CP-724714 was modeled as a BSEP inhibitor,50 although it is also a potent mitochondrial inhibitor (IC50 ∼45 nmol/mg mitochondrial protein, Rachel Swiss, unpublished results), while etomoxir was modeled as a mitochondrial inhibitor, but is also a potent BSEP inhibitor (IC50 = 11.8 μM, Paul Bonin, unpublished results). This modeling effort should consider how these potent dual liabilities may affect physiological responses to generate more accurate simulations of the human clinical response. In summary, although at this time nothing can truly replace human clinical experience to understand the safety of new medicines, there appears to be a strong relationship between drugs that have dual and potent effects on mitochondrial respiration and BSEP inhibition that correlate proportionately with different DILI concern categories and labeling concerns. The duality of these potent effects in terms of associations with Most-DILI-concern compounds may well represent physiological interdependencies on hepatocellular mitochondrial ATP production and impaired ATP utilization by numerous cellular processes, including ATP-dependent transporters such as BSEP, in addition to an apparent inhibitory effect on BSEP transporter functionality. This analysis demonstrates that BSEP inhibition alone is not an absolute predictor of liver injury potential. Further investigations into the combined role these dual liabilities have in human DILI are warranted and an expanded analysis is currently under way with a larger compound dataset. Acknowledgment The authors thank Rouchelle Mireles for technical support. Supporting Information Additional Supporting Information may be found in the online version of this article at the publisher's website. Filename Description hep27206-sup-0001-suppinfo01.pdf80.5 KB Supporting Information Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. References 1 Bell LN, Chalasani N. Epidemiology of idiosyncratic drug-induced liver injury. Semin Liver Dis 2009; 29: 337- 347.