Plasma Bile Acid Profiling – Application for Drug-Induced Liver Injury

Drug-induced hepatotoxicity, or drug-induced liver injury (DILI), is still a major concern for drug developers despite several significant recent developments in drug safety testing. Due to its central role in drug metabolism and detoxification, the liver is particularly vulnerable to off-target toxicity. DILI is still the leading cause of acute liver failure in the United States (13-16% of all cases)¹, and has also been the most cited toxicity reason for the withdrawal of medications from the market². To date, more than 50 drugs have faced market withdrawal due to hepatotoxicity, which includes numerous products carrying black box warnings.

Why is DILI a unique challenge?

Although important steps have been made to reduce the risk of DILI and improve patient safety, the predictive power of most safety toxicology approaches is lacking. DILI in humans is often poorly predicted from preclinical studies, with approximately 45% of known DILI effects going undetected and 67% of clinical stage terminations not predicted in animal studies³.

From a clinical perspective, DILI can be classified as direct, indirect, or idiosyncratic. Direct injuries arise from compounds that are inherently toxic to the liver, leading to dose-dependent damage that is significantly easier to predict than other types of DILI. In contrast, idiosyncratic DILI is rare, unpredictable, and can seldom be replicated in animal models. Idiosyncratic DILI is also often associated with less favorable prognosis and more severe outcomes than direct DILI, responsible for 11% of total acute liver failure cases in the US⁴. Its unpredictable nature, combined with severity, makes idiosyncratic DILI an especially pressing concern for drug developers. A third type, indirect DILI, has recently been described as a type of drug-induced liver toxicity that occurs through an indirect action on the liver or immune system.

To address liver toxicity in preclinical and clinical experiments, a standardized set of liver function markers is usually monitored. These mainly include hepatic enzyme levels in the plasma, such as alanine and aspartate aminotransferases (ALT and AST), and alkaline phosphatase (ALP). Additionally, total bilirubin (TBL) is often assessed.

On its own however, this panel has poor translatability between preclinical species and humans and shows limited success in predicting idiosyncratic DILI. The emerging application of novel hepatotoxicity biomarkers is thus crucial for better DILI identification and de-risking. Among emerging DILI biomarkers, profiling bile acids from the systemic circulation has been shown to be a selective and sensitive tool for assessing DILI occurrence and severity.

eGuide: Biomarker Solutions from Discovery to Clinical Trials
Learn how strategic biomarker integration can accelerate research & de-risk pipelines to bring life-saving therapies to patients faster. This guide helps you uncover smarter ways to de-risk studies and accelerate timelines, without compromising scientific rigor.
Read the Guide

Bile Acids as DILI Biomarkers

Bile acids (BAs), derived from cholesterol, are the major constituents of bile. They are primarily synthesized in the liver (as cholic acid [CA] and chenodeoxycholic acid [CDCA] in humans) and subsequently conjugated with amino acids (glycine mainly in humans, taurine mainly in rodents). While primary bile acids are produced directly by the liver, secondary bile acids are formed through the action of gut microbiota once the bile is released into the small intestine, contributing to the high diversity of BAs present in the gastrointestinal tract. Bile acids undergo enterohepatic recirculation: from the intestines, they return to the liver through the portal circulation, and only a small fraction enters the systemic blood flow, from where they may also be taken up into hepatocytes.

The physiological role of BAs goes beyond facilitating the absorption of lipids and fat-soluble vitamins and the elimination of cholesterol; they also function as signaling molecules and regulate glucose and lipid metabolism. However, under certain pathological conditions, including hepatobiliary diseases, bile acids can become cytotoxic, especially when present in elevated concentrations.

BA toxicity most often manifests as disruption of mitochondrial and ER functions, leading to apoptosis and cell necrosis. In case of hepatobiliary diseases and conditions, this further contributes to liver toxicity and potential carcinogenicity. While the mechanisms underlying DILI are not fully understood, one established mechanism is the alteration of BA homeostasis, and about 50% of registered DILI cases are the result of cholestatic injury associated with impairment of bile formation or flow. Cholestatic DILI has also been the main mechanism of toxicity in the case of a third of the drugs withdrawn due to hepatotoxic adverse effects⁵.

Recent studies utilizing targeted metabolomics have revealed changes in plasma bile acid profiles associated with various forms of hepatotoxicity in both humans and preclinical animals⁶. Notably, specific bile acids, including glycocholic acid (GCDCA), taurocholic acid (TCDCA), and deoxycholic acid (DCA), have been identified as key indicators of disease severity in humans. While GCDCA and TCDCA levels rise with increasing DILI severity, DCA levels decrease, highlighting the nuanced nature of bile acid response in liver injury scenarios. Further research has reinforced the potential for using plasma bile acid composition as biomarkers for DILI^7-8.

The ability to predict DILI in humans based on circulating bile acids in preclinical species remains a significant challenge in toxicology. Studies have demonstrated marked species differences in toxic effects, with safety assessments conducted in both rodent and non-rodent animals often failing to predict DILI outcomes in humans accurately. Differences in bile acid composition and homeostasis between species contribute to the variation in susceptibility to bile acid-induced hepatotoxicity. These differences arise from distinct metabolic pathways for bile acid elimination, resulting in a considerable variation in BA composition, which results in significant variability in DILI predictions⁹.

Development of a Bile Acid Quantification Assay

Charles River has developed an assay designed to quantify 20 different bile acids (BAs) utilizing liquid chromatography-tandem mass spectrometry (LC/MS-MS). This panel was carefully chosen based on existing literature and focuses on BAs known to have associations with liver injury. The validated methods require only 10 microliters of plasma, making it a practical tool for drug developers.

We also evaluated 16 bile acids across samples from 12 species, including five common rodent models, four large animal models, and human samples. A heat map was generated to visualize species signatures, highlighting the differential abundance of BAs across the various species. We also sought to identify which species exhibited BA profiles most similar to that of humans and found that species that are taxonomically closer to humans demonstrate higher similarity in their bile acid profiles. As we move further away from humans in the taxonomic tree, this correlation becomes less stringent, particularly with rodents, which cluster separately from larger mammals and humans.  Additionally, we observed that the primary to secondary bile acid ratios vary among species, following discernible patterns. These insights can significantly influence the selection of toxicological species when necessary.

As we move further away from humans in the taxonomic tree, this correlation becomes less stringent, particularly with rodents, which cluster separately from larger mammals and humans.  Additionally, we observed that the primary to secondary bile acid ratios vary among species, following discernible patterns. These insights can significantly influence the selection of toxicological species when necessary.

Comprehensively profiling bile acids in plasma not only aids in understanding the mechanisms of drug-induced liver injury but also holds significant potential as a diagnostic tool, providing a more complex diagnostic output and toxicity signals than conventional hepatotoxicity markers. Understanding the BA metabolism and incorporation of serum individual BA analysis may enhance conventional safety assessment in preclinical drug development studies. BA profiles could also be valuable in situations where traditional clinical chemistry measurements are ambiguous.

Discuss Your Project