ReviewDiclofenac-induced liver injury: a paradigm of idiosyncratic drug toxicity
Introduction
Diclofenac (Voltaren and a number of generics) is a nonsteroidal antiinflammatory drug (NSAID) widely used for treatment of a variety of rheumatoid disorders, including osteoarthritis, rheumatoid arthritis, ankylosing spondylitis, and acute muscle pain. Similar to other NSAIDs, diclofenac causes, in rare instances, hepatic injury in patients. Although the apparent incidence rate of severe diclofenac-induced hepatic adverse drug reactions (ADRs) is very low (estimates range from 1 to 2 cases per million prescriptions (Purcell et al., 1991) to 6 to 18 cases/100,000 person-years (Walker, 1997), it is the large number of patients treated worldwide with diclofenac that makes the absolute number of hepatic cases appear impressive. In fact, published cases of severe diclofenac hepatotoxicity amount to approximately 250 reports, with a case fatality rate of approximately 10% (Lewis, 2003), but the number of total serious hepatic adverse reactions (mostly unpublished) reported to the health authorities is severalfold higher. Worse even, the real incidence has been estimated to be 10- to 20-fold higher than all reported cases, due to notorious underreporting (Sgro et al., 2002). These numbers, however, have to be seen as always against the background of millions of people successfully treated with this NSAID.
The pathogenesis of low-incidence/high-severity diclofenac hepatotoxicity is largely unknown. One reason for this is the apparent lack of animal models available to date that mimic the human situation and with which one could investigate the mechanisms of the toxic response. Clearly, because liver injury from diclofenac is not a reproducible effect and lacks a simple dose–response relationship, it is generally accepted that individual patient-specific susceptibility factors eventually determine whether a patient will tolerate the drug (as in the vast majority of cases) or whether an individual in rare cases will develop a toxic response. Thus, diclofenac-induced liver injury is a typical example of idiosyncratic drug toxicity (see below). This notion alone, however, will not tell us anything about the mechanism.
There are two major and plausible reasons why it is not only desirable but indispensable to better understand the mechanisms of diclofenac injury at the molecular level. Insight in these underlying mechanisms will improve prediction, in particular
- (1)
to help identify populations (subsets) at increased risk, as well as identify biomarkers that would point to an increased risk for an individual; and
- (2)
to assist in candidate selection during the development of new drugs (let us recall that a significant number of other, related NSAIDs had to be removed from the market because of an unacceptably high risk of hepatic toxicity and that the development of certain new promising drugs had to be discontinued, sometimes at a late stage of progression, due to unpredictable liver liability).
A number of reviews on NSAID-induced liver injury are available Tolman, 1990, Rabinovitz and VanThiel, 1992, Carson et al., 1993, Farrell, 1994, Garcia Rodriguez et al., 1994, Boelsterli et al., 1995, Fry and Seeff, 1995, Manoukian and Carson, 1996, Boelsterli, 2002a, Boelsterli, 2002c, Lewis, 2003. Many NSAIDs share among themselves certain critical pathways of bioactivation, disposition, and toxic responses in the liver. The aim of this review, however, is to be topical rather than comprehensive, and to summarize the diclofenac-specific recent advances in the field. The past years have seen major progress in the understanding of certain diclofenac-specific mechanisms of liver cell injury at the molecular level, and a number of crucial pathways have been identified, while others still remain in the dark. These new insights into some of the underlying mechanisms can hopefully be translated into predictive tools for risk evaluation both at the patient level and at the drug development level.
Section snippets
Idiosyncratic drug hepatotoxicity
More than 50% of all cases of acute liver failure can be attributed to drugs (Gill and Sterling, 2001). Among these, acetaminophen is the most common cause. While acetaminophen produces predictable and dose-dependent acute toxicity (Cohen et al., 1998), 14% of all cases of acute liver failure encompass unexpected and unpredictable drug reactions from a large variety of other drugs. Thus, it is now generally accepted that drugs are the single major cause of acute liver failure (in the Western
Pathology and clinical phenotype of diclofenac hepatotoxicity: a clue to understanding mechanisms?
Histopathological analysis of liver specimens (upon biopsy or autopsy) of patients with diclofenac-induced liver injury alone cannot reveal probable modes of diclofenac hepatotoxicity. This is because in the majority of cases such liver samples are not available and because in those very few available they represent a terminal stage of injury that is commonly found as a result of drug-induced liver failure. Specifically, the typical pathologic picture is hepatic necrosis (usually diffuse but
Putative mechanisms of diclofenac hepatotoxicity
In idiosyncratic drug toxicity, mechanisms pertaining to the drug (molecule), the individual (patient), and the underlying disease (therapeutic indication) all contribute to the overall pathogenesis. Here, new pieces of experimental evidence that have given us more insights into some of the putative mechanisms (viewed from both the toxicokinetic and toxicodynamic perspective) are discussed, with a focus on the most recent advances.
The molecule
Diclofenac (sodium salt, Mr = 318.1) is a lipophilic and weakly acidic compound (Poct = 28.4; pKa = 3.9) (Sengupta et al., 1985) that features two twisted aromatic rings (Sallmann, 1985). Important for its behavior in hepatocytes are two structural components (Fig. 1). First, it is a phenylacetic acid derivative, with the carboxylic acid moiety being the substrate for the production of reactive glucuronide metabolites (see below). Second, diclofenac has a diphenylamine backbone, with the
The patient
On the search of genetic or acquired patient factors, polymorphisms and other, more rarely (<1%) occurring, abnormalities in the expression of drug-metabolizing enzymes or other key proteins have increasingly been scrutinized. However, the link between such individual variations and the occurrence of diclofenac-induced liver injury has proven extremely difficult so far. For example, because the formation of an electrophilic intermediate (the 2,5-quinone imine, as inferred from the urinary
The underlying disease
Factors of the preexisting disease (pathophysiological alterations, altered gene expression, and altered toxic response) can greatly modulate the susceptibility to drug-induced hepatotoxicity (Boelsterli, 2003b). For example, the condition of rheumatoid arthritis (as opposed to other therapeutic indications that are treated with NSAIDs) can increase the risk for NSAID-induced liver injury by 10-fold (Rodriquez et al., 1994). Specifically for diclofenac, osteoarthritis is a risk factor. This can
New mechanistic insights from transcriptomics and proteomics
The advent of the novel toxicogenomics techniques made possible for the first time comparative expression profiles for thousands of genes. While the validation of these techniques (for prediction) is still ongoing, it has become clear (perhaps not too surprisingly) that for many drugs that are potentially hepatotoxic the gene expression profiles correlate well with traditional histopathological and biochemical analyses (Waring et al., 2001). No data are available for diclofenac or related
Conclusions
Based on the numerous reports on diclofenac-induced liver injury (both clinical reports and experimental studies), diclofenac has developed into a paradigm of NSAID-induced idiosyncratic hepatotoxicity (although the estimated incidence rate is not higher than that for some other NSAIDs). Thus, on the one hand, our insights into many of the underlying molecular pathways has considerably increased. A number of diclofenac-related hazards have been identified and studied in detail. These include
- (1)
Conflict of interest statement
There are no conflicts of interest.
References (106)
Serum transaminase elevations as indicators of hepatic injury following the administration of drugs
Reg. Toxicol. Pharmacol.
(1998)- et al.
Drug enterocyte adductspossible causal factors for diclofenac enteropathy in rats
Gastroenterology
(2000) - et al.
Acyl glucuronide reactivity in perspectivebiological consequences
Chem.–Biol. Interact.
(2003) - et al.
Diclofenac-associated hepatotoxicityanalysis of 180 cases reported to the Food and Drug Administration as adverse reactions
Hepatology
(1995) - et al.
Possible role of HLA in hepatotoxicityan exploratory study in 71 patients with drug-induced idiosyncratic hepatitis
J. Hepatol.
(1994) - et al.
Characterization of a novel diclofenac metabolite in human urine by capillary gas chromatography-negative chemicalionization mass spectrometry
J. Chromatogr. B
(1996) - et al.
Hepatic metabolism of diclofenacrole of human CYP in the minor oxidative pathways
Biochem. Pharmacol.
(1999) - et al.
Sensitivity to a metabolite of diclofenac as a cause of acute immune hemolytic anemia
Blood
(1997) - et al.
Induction of hepatic heme oxygenase-1 by diclofenac in rodentsrole of oxidative stress and cytochrome P-450 activity
J. Hepatol.
(2003) - et al.
Modulation of the mitochondrial permeability transition pore by pyridine nucleotides and dithiol oxidation at two separate sites
J. Biol. Chem.
(1996)