Drug-induced liver injury: mechanisms and test systems

Drug-induced liver disease

Although the year 2001 did not see any prescription drugs withdrawn because of drug-induced liver disease, the US Food and Drug Administration requested that dietary supplements containing comfrey be taken off the market because of the danger of hepatic injury. The Food and Drug Administration remains very involved in the process by which drug-induced liver disease can be detected early in drug development and in the determination of how best to prevent hepatotoxicity after drug approval. A workshop on drug-induced liver disease cosponsored by the Food and Drug Administration, the Pharmaceutical Research and Manufacturers Association, and the American Association for the Study of Liver Diseases was held in Washington, DC, in February 2001, and the resulting white paper outlined several areas for research. A number of agents were newly described as causing various forms of liver injury, and several others had drug-induced liver disease confirmed by additional reports. Several investigators dealt with the difficulties inherent in establishing causality of drug-induced liver disease and the potential negative consequences of wrongly attributing hepatotoxicity to a particular agent. In one recent series, more than half the instances of alleged drug-induced liver disease were found to have other causes, often leading to a delay in the actual diagnosis and appropriate management. Case reports in particular were often misleading. Although several drug assessment scales have been developed, none appears to be foolproof.

Hepatobiliary pathology

Technologic advances using cDNA microarray hybridization, liver diseases characterized by mitochondrial DNA depletion, and new work characterizing bile salt transport problems in familial intrahepatic cholestasis syndromes were some of the major highlights of this past year. Analysis of normal livers by cDNA microarrays disclosed 2418 unique gene transcripts encoding a host of cellular structural and functional proteins. This technique was also applied to hepatocellular carcinoma, where enhanced expression of a number of genes involved in antiapoptosis and cell transformation may shed additional light on the process of hepatocarcinogenesis. Mitochondrial DNA depletion seen in Navajo neurohepatopathy and in respiratory chain disorders of infancy was associated with cholestasis and cirrhosis in the former and microvesicular steatosis and oncocytic transformation (mitochondrial hyperplasia) in the latter. Pathologists who routinely examine liver biopsies after liver or bone marrow transplantation should be aware of unusual biopsy features that mimic other diseases, such as the autoimmune hepatitis-like syndrome that may follow liver transplantation and chronic graft-versus-host disease that clinically and pathologically resembles acute hepatitis.

Hepatotoxicity with thiazolidinediones: is it a class effect?

Decreased insulin sensitivity plays a major role in various human diseases. particularly type 2 diabetes mellitus, and is associated with a higher risk of atherosclerosis and cardiovascular complications. Thiazolidinediones, more commonly termed glitazones, are the first drugs to specifically target muscular insulin resistance. They have proven efficacy for reducing plasma glucose levels in patients with type 2 diabetes mellitus treated with diet alone, sulphonylureas, metformin or insulin. In addition, they are associated with some improvement of the cardiovascular risk profile. However, troglitazone, the first compound approved by the Food and Drug Administration in the US, proved to be hepatotoxic and was withdrawn from the market after the report of several dozen deaths or cases of severe hepatic failure requiring liver transplantation. It remains unclear whether or not hepatotoxicity is a class effect or is related to unique properties of troglitazone. Rosiglitazone and pioglitazone, two other glitazones, appear to have similar efficacy with regard to blood glucose control in patients with type 2 diabetes mellitus as compared with troglitazone. In controlled clinical trials, the incidence of significant (> or =3 x upper limit of normal) increases in liver enzyme levels (ALT in particular) was similar with rosiglitazone or pioglitazone as compared with placebo, whereas troglitazone was associated with a 3-fold greater incidence. In contrast to the numerous case reports of acute liver failure in patients receiving troglitzone, only a few case reports of hepatotoxicity have been reported in patients treated with rosiglitazone until now, with a causal relationship remaining uncertain. Furthermore, no single case of severe hepatotoxicity has been reported yet with pioglitazone. It should be mentioned that troglitazone, unlike pioglitazone and rosiglitazone, induces the cytochrome P450 isoform 3A4, which is partly responsible for its metabolism, and may be prone to drug interactions. Importantly enough, obesity, insulin resistance and type 2 diabetes mellitus are associated with liver abnormalities, especially non-alcoholic steatohepatitis, independent of any pharmacological treatment. This association obviously complicates the selection of patients who are good candidates for a treatment with glitazones as well as the monitoring of liver tests after initiation of therapy with any thiazolidinedione compound. While regular monitoring of liver enzymes is still recommended and more long term data are desirable, current evidence from clinical trials and postmarketing experience in the US supports the conclusion that rosiglitazone and pioglitazone do not share the hepatotoxic profile of troglitazone.

Polymorphism of the N-acetyltransferase 2 gene as a susceptibility risk factor for antituberculosis drug-induced hepatitis

Antituberculosis drug-induced hepatitis is one of the most prevalent drug-induced liver injuries. Isoniazid is the major drug incriminated in this hepatotoxicity. Isoniazid is mainly metabolized to hepatotoxic intermediates by N-acetyltransferase (NAT). However, the association of polymorphic NAT acetylator status and antituberculosis drug-induced hepatitis is debatable. To determine whether acetylator status is a risk factor for antituberculosis drug-induced hepatitis, we genotyped NAT2 in 224 incident tuberculosis patients who received antituberculosis treatment. Antituberculosis drug-induced hepatitis was diagnosed based on a positive isoniazid rechallenge test and exclusion of viral hepatitis. Acetylator status was determined by genotyping NAT2 in patients using a polymerase chain reaction with restriction fragment length polymorphism. Univariate analysis and logistic regression analysis were used to evaluate the risk factors of isoniazid-induced hepatitis. Thirty-three patients (14.7%) were diagnosed with antituberculosis drug-induced hepatitis. Slow acetylators had a higher risk of hepatotoxicity than rapid acetylators (26.4% vs. 11.1%, P =.013). Among patients with hepatotoxicity, slow acetylators had significantly higher serum aminotransferase levels than rapid acetylators. Logistic regression showed that slow-acetylator status (odds ratio [OR], 3.66; 95% CI, 1.58-8.49; P =.003) and age (OR, 1.09; 95% CI, 1.04-1.14; P <.001) were the only 2 independent risk factors for antituberculosis drug-induced hepatitis. In conclusion, slow-acetylator status of NAT2 is a significant susceptibility risk factor for antituberculosis drug-induced hepatitis. Additionally, slow acetylators are prone to develop more severe hepatotoxicity than rapid acetylators. Regular monitoring of serum aminotransferase levels is mandatory in patients receiving antituberculosis treatment, especially in slow acetylators.

Lack of hepatotoxicity associated with nonnucleoside reverse transcriptase inhibitors

Nonnucleoside reverse transcriptase inhibitors (NNRTIs), particularly nevirapine, have been associated with hepatotoxicity. We performed a retrospective study to determine the incidence of NNRTI hepatotoxicity in a group of HIV-infected patients from a New York City practice. These patients are predominantly homosexual white males. We also analyzed the effect of coinfection with hepatitis B (HBV) or hepatitis C (HCV) virus. In total, 272 patients received NNRTIS: 40 (15%) received delavirdine, 91 (33%) received efavirenz, and 141 (52%) received nevirapine. Of the patients with known hepatitis status, 18 of 190 (9%) were coinfected with HBV, and 24 of 205 were coinfected (12%) with HCV. The overall rate of grade 3 to 4 elevations in aspartate aminotransferase (AST) or alanine aminotransferase (ALT) was 3 of 272 (1.1%) and did not differ significantly among the three NNRTIs. HBV or HCV was not associated with a significant increase in AST or ALT elevations. We conclude that NNRTIs are relatively free from hepatotoxicity in this population, despite the presence of coinfection with HBV or HCV.

Hepatic sinusoidal obstruction after gemtuzumab ozogamicin (Mylotarg) therapy

Gemtuzumab ozogamicin (Mylotarg) targets leukemia cells expressing the CD33 receptor by means of a monoclonal antibody conjugated to a cytotoxic agent, calicheamicin. Treatment of acute myeloid leukemia (AML) with gemtuzumab ozogamicin may result in liver injury. We reviewed the course of 23 patients who were given gemtuzumab ozogamicin for AML that had relapsed after hematopoietic cell transplantation. Liver toxicity was assessed through physical examination, serum tests, histologic examination, and hepatic venous pressure measurements. Liver injury developed in 11 patients after gemtuzumab ozogamicin administration; it was manifested as weight gain, ascites, and jaundice in 7 patients. Seven patients died with persistent liver dysfunction and either multiorgan failure or sepsis at a median of 40 days after gemtuzumab ozogamicin infusion. Portal pressure measurements were elevated in 2 patients. Results of liver histologic examination in 5 patients showed sinusoidal injury with extensive sinusoidal fibrosis, centrilobular congestion, and hepatocyte necrosis. Six patients experienced AML remission that was sustained for at least 60 days after gemtuzumab ozogamicin infusion. In summary, hepatic sinusoidal liver injury developed after gemtuzumab ozogamicin infusion. Histology showed striking deposition of sinusoidal collagen, suggesting that gemtuzumab ozogamicin targets CD33(+) cells residing in hepatic sinusoids as the mechanism for its hepatic toxicity.

Hepatoimmunology: a perspective

Premises for the subspecialty of hepatoimmunology include the recognition that the liver is a lymphoid organ with unique immunological properties. These properties ensure efficient innate defence against intestinal microbes and toxins, confer a particular capacity for induction of tolerance, and provide for apoptotic disposal of redundant lymphocytes. Pathological responses within the liver are elicited when: (i) hepatotropic viruses (hepatitis virus B and C) escape immune elimination and reside in hepatocytes; (ii) the liver becomes the site of autoimmune responses directed against either hepatocytes (autoimmune hepatitis) or biliary ductules (primary biliary cirrhosis); or (iii) the liver in the course of disposal of drugs generates neoantigens that provoke adverse allergic responses. Recent advances in the understanding of the immunopathogenesis of these entities are reviewed.