REGULATORY TOXICOLOGY AND PHARMACOLOGY 27, 119–130 (1998)
ARTICLE NO. RT981201
Serum Transaminase Elevations as Indicators of Hepatic Injury
Following the Administration of Drugs
David E. Amacher
Drug Safety Evaluation, Pfizer Central Research, Groton, Connecticut 06340
Received December 20, 1997
During the preclinical, early clinical, late-stage clini-
cal, and postmarketing phases of the pharmaceutical
discovery and development process, one important as-
pect of drug safety assessment involves monitoring for
possible drug-induced hepatic injury. Hepatic injuries
vary in nature from direct, intrinsic effects that are
observed in most recipients and more than one species
to rare idiosyncratic responses seen only in a few clini-
cal subjects. Histological types of injuries vary from
hepatocellular to hepatobiliary with multiple cellular
effects characteristic of each type. Of the various clini-
cal laboratory markers for hepatic injury, serum
transaminases, especially alanine aminotransferase
(ALT), are the most universally important indicators
for studies ranging from early preclinical animal test-
ing to postmarketing patient monitoring. This review
examines the characteristics of hepatic toxicity that
result in serum ALT changes, the differences in the
etiology of hepatic responses which govern when liver
injury is most likely to be detected during the four
phases of the drug discovery and development process,
and those modulating factors which affect the utility
of ALT as a dependable marker of hepatic injury in
clinical populations. The paper concludes with a sum-
mary of some ancillary methods for early preclinical
screening such as in vitro metabolism and toxicity
assays, gene and protein expression analysis, and
some strategies for enhancing the probability for the
early detection of idiosyncratic hepatotoxic responses
which are infrequent but significant factors in the
safety assessment process. q 1998 Academic Press
Key Words: review; serum transaminases; alanine
aminotransferase; liver injury.
1. DRUG HEPATOTOXICITY AND THE
PHARMACEUTICAL DEVELOPMENT PROCESS
Because of its central role in drug metabolism, the
liver is particularly susceptible to injury following sys-
temic exposure to xenobiotics. Accordingly, liver toxic-
ity as suggested by slightly elevated serum enzymes of
hepatic origin is one of the more frequently encoun-
119
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
tered adverse effects in early clinical trials of therapeu-
tic agents, although the actual incidence of hepatic in-
jury is extremely low. During the pharmaceutical dis-
covery, development, and marketing processes, drug-
associated hepatotoxicity can be encountered at any of
four stages.
The first stage involves the detection of hepatotoxic
phenomena in preclinical safety assessment studies
using rats, dogs, monkeys, or other mammals at dose
levels in excess of the intended therapeutic dose. A
significant incidence of drug-associated hepatocellular
or hepatobiliary injury detected by clinical chemistry
procedures and confirmed by histological methods in
one or more species generally precludes further devel-
opment unless there is clear evidence that a similar
effect is not anticipated in humans. An exception could
be the demonstration that animal toxicity is due to a
specific metabolite that would not be a major metabo-
lite in humans. Most intrinsically hepatotoxic (type A)
compounds are detected in preclinical animal trials
and are not advanced to clinical trials. The proportion
of compounds causing some degree of hepatic effects
such as hepatomegaly, serum enzyme elevations, he-
patobiliary effects, or necrosis at high doses in animals
can reach as high as 50% of drugs tested in preclinical
studies.
The next level where evidence of hepatotoxicity may
be evaluated is in early clinical development. At thera-
peutic doses, untoward findings could include, for ex-
ample, the reporting of serum transaminase elevations
that are 2 times the upper limit of normal, jaundice,
or 1.5 times the normal alkaline phosphatase (Alk P;
EC 3.1.3.1) activity in a small treated population of
healthy volunteers during early clinical trials (Batt and
Ferrari, 1995). While these clinical chemistry changes
are usually transient and not accompanied by clinical
evidence of hepatic toxicity, drugs eliciting these
changes are typically not continued in development un-
less there is a compelling reason to do so.
The third stage typically involves the reporting of a
few patients within a controlled clinical population who
are found to have transaminase levels two to three
times the upper limit of normal but who are otherwise
0273-2300/98 $25.00
Copyright q 1998 by Academic Press
All rights of reproduction in any form reserved.
rtpal AP: RTP
120 DAVID E. AMACHER
asymptomatic. A fairly large number of drugs fall
within this category. These are not typically discon-
tinued for this phenomenon alone. For example, the
U.S. Physicians’ Desk Reference indicates that liver en-
zyme elevations may be found in up to 15% of patients
taking nonsteroidal anti-inflammatory drugs (NSAIDs).
One NSAID in particular, sulindac, actually accounts
for most of the risk (Fry and Seeff, 1995; Garcia Rodri-
guez et al., 1994). Another example is the 12% inci-
dence of asymptomatic hepatic enzyme abnormalities
reported in ketoconazole recipients (Mosca et al., 1985).
The disease under treatment may also affect the liver
adversely leading to elevated hepatic enzymes. Most of
these reactions are host-dependent, idiosyncratic re-
sponses which are not easily predicted in preclinical
testing or observed in early clinical trials because of
the small number of subjects.
The final stage involves similar findings in a large
postmarketing population. These cases involve very-
low-frequency, idiosyncratic (type B) events that are
not detectable in controlled clinical populations of 500–
1000, for example, due to mathematical probability.
These result from a novel aspect of the patient, not the
drug, and are more uncertain in etiology because they
are typically reported in an uncontrolled nonclinical
setting. Thus, drug-induced hepatic injury ranges from
frequent, dose-related effects in multiple species to rare
but profound hepatic effects only detectable in suscepti-
ble individuals at therapeutic doses. Serum transami-
nase activities are the single diagnostic indicator for
all of these effects and are the subject of this review.
2. HISTOLOGICAL TYPES OF TOXIC HEPATIC INJURY
In general, the type and degree of acute toxic liver
injury following drug administration depend upon the
dose and duration of exposure. In addition, many chem-
icals preferentially affect certain cell populations but
not others in the same host. Major histologic lesions of
the liver due to chemical toxicity include fatty change,
hepatocellular death, and hepatobiliary lesions. Gran-
ulomatous and mixed histological changes may also be
seen (Pirmohamed et al., 1992) or venoclusive effects,
especially with some antineoplastic drugs (McDonald
and Tirumali, 1984). Steatosis or fatty change refers
to an increase in hepatic lipid content greater than 5%
by weight (Moslen, 1996). Lipid accumulates as intra-
cytoplasmic vacuoles in either a zonal or diffuse pat-
tern. Microvesicular and macrovesicular steatoses, dif-
fering in the lipid vacuole size and the mechanism for
their formation, are distinguishing characteristics of
specific toxicants. Macrovesicular steatosis is charac-
terized by a single large fat droplet which displaces
the nucleus to the cell periphery, but in microvesicular
steatosis there are a number of small lipid droplets
that do not displace the nucleus (Stricker, 1992a).
Steatosis is reversible unless it is persistent enough to
cause secondary changes or so severe that cell death
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
results. In humans, the incidence of fatty liver disease
noted at autopsy for accident victims ranges from 40
to 80% (Hodgson et al., 1989) suggesting a high inci-
dence from unknown causes in the general adult popu-
lation. Necrotic cell death is accompanied by a disrup-
tion of cell membrane integrity and the loss of liver-
specific enzymes which forms the basis of the serum
chemistry analyses used systematically for its detec-
tion. Clinical manifestations of hepatic necrosis and
degeneration range from jaundice to fulminant hepatic
failure. Cholestatic injury occurs when there is a dis-
ruption or impairment of bile flow. Canalicular chole-
stasis can be caused by several mechanisms including
leaky paracellular junctions, a diminished contractility
of canaliculus, diminished transcytosis, impaired
transporters, and toxin concentration in the paracani-
cular area (Moslen, 1996). Toxins or their metabolites
can also damage the intrahepatic bile ducts. Chlor-
promazine, oral contraceptives, and androgens used to
increase muscle mass can produce a cholestatic re-
sponse. Hepatobiliary lesions lead to elevated serum
levels of bile constituents or enzymes localized in the
hepatobiliary tree such as Alk P or gamma glutamyl-
transpeptidase (GGT; EC 2.3.2.2). In humans, most he-
patic drug reactions appear within 3 months of starting
therapy (Davis, 1989).
3. BIOCHEMICAL TESTS USED FOR THE DETECTION
OF HEPATIC INJURY
3.1. Categories of Laboratory Tests
Three categories of noninvasive laboratory tests are
used to identify the type and extent of drug hepatotox-
icity based on the presence or absence of specific mark-
ers in the blood of exposed individuals. The first cate-
gory of clinical assays is that used to assess hepatocel-
lular damage leading to liver cell necrosis. Evidence of
this type of injury is based on the detection the hepatic
transaminases, alanine aminotransferase (ALT), and
aspartate aminotransferase (AST). These are not nor-
mal components of blood and serve no known function
outside the organ of origin. A third specific marker of
hepatocellular damage, serum F protein, has been re-
cently described (Foster et al., 1989), although it is not
yet as widely used as the transaminases. Based on liver
biopsy specimens, chronic ALT elevations in asymp-
tomatic patients have also been associated with fatty
liver (Hulteranz et al., 1986). In fact, nonalcoholic
steatosis has been cited as a very common cause of
chronic ALT elevations in the general population
(Craxi and Almasio, 1996). Thus, transaminase eleva-
tions are indicative of hepatocellular death and fatty
degeneration in particular, but can also be used in con-
junction with other serum enzymes for distinguishing
non-hepatocellular injury as described below.
The second group of tests includes markers for hepa-
tobiliary (chlolestatic) effects. This type of injury in-
rtpal AP: RTP
 SERUM TRANSAMINASES AND HEPATIC INJURY
cludes biliary obstruction or hepatic infiltrative pro-
cesses resulting in the retention of bile acids in liver
and leading to drug-induced jaundice if severe. Serum
markers include Alk P from the cell canalicular mem-
brane, 5* nucleotidase (5*NT), and GGT. Neither Alk P
nor GGT, which is also inducible by alcohol and certain
drugs (see below), is liver-specific. However, marked
serum Alk P elevations, especially when accompanied
by 5*NT or GGT elevations, suggest mechanical bile
duct obstruction, primary sclerosing cholangitis, pri-
mary biliary cirrhosis, or drug-induced hepatitis
(Herlong, 1994). Thus, serum enzyme profiles as op-
posed to individual enzyme changes are used for diag-
nostic purposes, especially for monitoring hepatobiliary
effects.
The final category of diagnostic procedures is based
upon altered liver function. These are methods that
monitor serum albumin, cholesterol, prothrombin time,
or serum bilirubin as general indicators of the synthetic
and general metabolic capacity of the liver (Fregia and
Jensen, 1994) as opposed to marking some specific toxic
injury. The serum bilirubin assay will indicate liver
injury; however, elevated serum enzyme assays as de-
scribed above usually reflect hepatotoxicity earlier
(Herlong, 1994). Although drug-associated hepatic dys-
function is uncommon in general, severely altered liver
function, especially those processes leading to coagula-
tion disorders or bilirubin encephalopathy, are indica-
tive of severe hepatic injury.
3.2. Aminotransferases
Aminotransferases are a group of enzymes that cata-
lyze the reversible transfer of the amino group from
an a-amino acid to an oxo acid. ALT and AST shunt
their amino acid and oxo acid substrates into several
intermediate pathways. Cytosolic ALT is associated
with the utilization of pyruvate in glycolysis, mitochon-
drial ALT is involved in the conversion of alanine to
pyruvate for gluconeogenesis, and AST plays an im-
portant role in the transport of reducing equivalents
across the mitochondrial membrane (Sakagishi, 1995;
Rej, 1989). Hepatocellular damage with the subse-
quent disruption of the plasma membrane allows leak-
age of intracellular enzymes such as ALT or AST into
the bloodstream. With some hepatotoxicants, in-
creased hepatic synthesis of aminotransferases has
also been suggested as a source of increased serum
enzyme levels in hepatocellular injury (Pappas, 1986).
The range of normal is determined by either cutoff
values of {2 SD or 97.5 percentile cutoffs in a popula-
tion without known disease. Due to half-lives of ap-
proximately 17 h for AST and 47 h for ALT, the pres-
ence of these enzymes in serum is considered an indica-
tor of recent hepatocyte injury (Scheig, 1996). Some
drugs actually decrease the activities of serum ALT
and AST. Examples include oxodipine which causes
hepatic damage and yet decreases ALT activities in
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
121
dogs and rats (Waner and Nyska, 1991), cefazolin
which significantly depresses ALT and AST activities
in rat liver (Dhrami et al., 1979), and isoniazid which
significantly inhibits hepatic and serum AST in ro-
dents (Yamada et al., 1984a,b). In addition to inhibi-
tory effects by antivitamin B6 compounds, transami-
nase levels are effected by other nutritional events.
For example, healthy human volunteers ingesting a
choline-deficient diet for 4 weeks showed significantly
elevated levels of ALT compared their cohorts (Zeisel
et al., 1991). A discussion of the characteristics of each
of these important enzymes follows.
3.2.1. ALT. ALT (L-alanine:2-oxoglutalate amino-
transferase, EC 2.6.1.2) is a pyridoxal enzyme which
catalyzes the reversible interconversion of L-alanine /
a-ketoglutarate to pyruvate / L-glutamate with peri-
doxal phosphate as a coenzyme. The presence of low
levels of ALT in the peripheral circulation represents
normal cell turnover or release from nonvascular
sources. Drug-related increases in aminotransferase
activity are typically transitory with values returning
to within normal reference limits within a few weeks
(Rej, 1989).
ALT is widely distributed. Human isozymes are
found in the cytosol and mitochondria of liver, kidney,
and skeletal and cardiac muscles (Sakagishi, 1995). Mi-
tochondrial ALT comprises only a small portion of the
tissue activity and has not been demonstrated in nor-
mal human serum. The largest pool of ALT is in the
cytosol of hepatic parenchymal cells (Sherman, 1991).
Considerable differences in both the organ distribution
and intracellular compartmentalization of ALT have
been found among species (Hoffmann et al., 1989).
Some nonhuman primates, for example, show little or
no ALT organ specificity (Clampitt and Hunt, 1978).
However, overall, serum ALT is one of the most univer-
sal markers for hepatic injury across species.
In the clinical laboratory, the measurement of ALT
is a routine part of serum chemistry panels used to
assess hepatic injury. ALT values in healthy blood do-
nors, however, can be influenced by age, sex, dietary
change, geographical location, ethnicity, obesity, long-
term acetaminophen use, alcohol use, and marital sta-
tus (Sherman, 1991), a complicating factor when moni-
toring human populations for transaminase elevations
following drug exposure. Even the difference in mean
ALT values between males and females in a donor pop-
ulation can be significant (Mijovic et al., 1987). ALT
activities are elevated for a few days following major
abdominal or thoracic surgery (Stricker et al., 1992a)
and can also be elevated by the disease being treated.
The incidence of hepatic damage due to cotrimaoxazole,
for example, is around 20% higher in AIDS patients
(Westphal et al., 1994) compared to other diseases. Cog-
nizant of these modulating factors in clinical popula-
tions, ALT is the single most important indicator of
rtpal AP: RTP
122 DAVID E. AMACHER
hepatocellular injury for preclinical animal studies,
clinical trials, and postmarketing monitoring.
3.2.2. AST. AST (EC 2.6.1.1) is found in both the
cytosol and mitochondria of hepatocytes (Herlong,
1994), but high tissue levels are also found in heart,
skeletal muscle, kidney, brain, and pancreas (Rej,
1989). Accordingly, muscle trauma and surgery by it-
self can lead to AST serum elevations (Clermont and
Chalmers, 1967). An estimated 60–70% of AST activity
in human hepatocytes is localized within mitochondria
(Schmidt and Schmidt, 1990). Hence, when found in
blood, AST is considered to be a sensitive indicator of
mitochondrial damage, especially in the hepatic centri-
lobular regions which are particularly sensitive to toxic
and hypoxic liver injury (Schmidt and Schmidt, 1990).
It should be noted, however, that depending upon the
assay used a number of drugs can reportedly produce
spurious elevations in AST (Davis, 1989). Serum AST
is affected to a greater degree by alcohol consumption
than ALT (Lewis, 1984). Within these limitations, AST
in conjunction with ALT is a very important marker of
hepatic injury.
4. NON-DRUG-RELATED TRANSAMINASE ELEVATIONS
The term transaminitis has been applied to mild ele-
vations of serum aminotransaminases in the absence
of other clinical laboratory abnormalities in asymptom-
atic individuals (Hodgson et al., 1989; Rej, 1989). When
observed after drug exposure, these small elevations
are often considered as being uninterpretable. A second
difficulty in monitoring transaminase levels in patient
populations is the underlying incidence of mildly ab-
normal results in a classical panel of liver tests used
in routine monitoring of asymptomatic, unaffected in-
dividuals. These may be associated with population
characteristics previously discussed and are not usu-
ally encountered in preclinical animal toxicology stud-
ies. Liver injury tests are abnormal in up to 8% of a
healthy population (Hodgson et al., 1989). Liver en-
zyme elevations noted during hospital admission have
been estimated at 5.7–7.6% (Grohmann et al., 1984;
Van Dijke et al., 1986). In one study, 25–30% of asymp-
tomatic, normal workers had serum chemistries on a
five-test liver panel in excess of the established normal
range in the absence of risk factors for liver disease
(Wright et al., 1988). Of course, specific estimates of
the incidence of abnormal ALT values alone are lower
in donor populations. In one study, the frequency of
elevated ALT in a donor population was reported to
be 4.6% above 40 IU/liter in 1986 and appeared to be
increasing with time (Mijovic et al., 1987). In the ab-
sence of any intentional drug exposure, extremely high
transaminase levels (ú8- to 10-fold normal) in a patient
population may indicate acute viral hepatitis, drug- or
toxin-mediated liver necrosis, or other damage; persis-
tent mild elevations (2- to 8-fold) are characteristic of
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
chronic hepatitis, steatosis, metabolic diseases (Fregia
et al., 1994), alcoholic liver disease, and, much less fre-
quently, drug-induced liver disease and nonalcoholic
steatosis (Renner and Dallenbach, 1992). In addition,
some disease states, especially malignancies, produce
hepatic effects culminating in elevated transaminases.
5. GUIDELINES FOR SERUM TRANSAMINASE
ELEVATION AS AN INDICATOR OF LIVER INJURY
By international consensus, liver injury, as might be
detected by liver tests, has been defined as an increase
greater than two times the upper limit of normal range
(ULN) in ALT or conjugated bilirubin or a combined
increase in AST, Alk P, and total bilirubin provided
that one of them is greater than two times the ULN
(Benichou, 1990). This injury is further defined as he-
patocellular when (1) there is an increase of greater
than two times the ULN for ALT alone or (2) there is
a ratio of serum ALT activity/serum Alk P activity
greater than five or cholestatic when there is an in-
crease of greater than two times the ULN of Alk P
alone or a ratio equal or less than two. Acute liver
injury is distinguished from chronic liver injury when
the increases have lasted less or more than 3 months,
respectively. Severe liver injury and fulminant liver
injury are defined by (1) the presence of jaundice, pro-
thrombin õ50% (or equivalent), and hepatic encepha-
lopathy and (2) the rapid development of hepatic en-
cephalopathy and severe coagulation disorders, respec-
tively.
6. SERUM ENZYME ELEVATIONS DUE TO HEPATIC
ENZYME INDUCTION
Enzyme induction is one reported iatrogenic effect
leading to elevated hepatic serum enzyme levels in pa-
tient populations that are not directly indicative of he-
patic injury. This has been well documented, especially
for anticonvulsants, but also for some other drugs. In
one such study, 56 children of 63 receiving phenobarbi-
tal and/or diphenylhydantoin had elevated GGT; 6 had
elevated ALT and AST for more than 20 weeks in the
absence of specific histopathology (Aiges et al., 1980).
Further evidence of hepatic enzyme induction by anti-
convulsants in asymptomatic patients was cited by
Wall et al. (1992) in a study of 206 adults and children.
Of these, serum GGT was elevated in 74.6%, Alk P in
29.7%, and ALT in 25.2%. Of 242 patients administered
antiepileptic drugs, 40 exhibited high levels of serum
GGT and nearly all cases indicated hepatic microsomal
enzyme induction as measured by antipyrine half-life,
leading Hirayanagi et al. (1991) to conclude that, in
these patients, elevated serum GGT did not necessarily
indicate hepatocellular damage. Elevated circulating
levels of GGT are commonly observed in patients tak-
ing cytochrome P450 enzyme-inducing drugs such as
rifamicin (Davis, 1989). Similar studies with co-antiep-
rtpal AP: RTP
 SERUM TRANSAMINASES AND HEPATIC INJURY
ileptic drug therapy (Haidukewych and John, 1986) in-
dicated ALT elevations up to three times and AST ele-
vations up to two times the upper limit of normal in
more than one-quarter of the patient population. These
were not considered clinically significant but instead
were attributed to enzyme induction. Liver biopsies in
similar patients undergoing long-term anticonvulsant
therapy showed no signs of chronic liver damage (Ja-
cobsen et al., 1976). Both ethanol and glucocorticoids
have been shown to induce GGT activity in rat liver
(Nishimura et al., 1981; Billon et al., 1980). In studies
with both man and rat, the primary cause for increased
GGT associated with prolonged alcohol exposure has
been attributed to hepatic enzyme induction rather
than liver cell injury (Teschke et al., 1983), although
it is clear that ethanol is hepatotoxic in general. In
developing rats, cortisol enhances hepatic activities of
ALT (Herzfeld and Raper, 1979).
7. TYPES OF DRUG HEPATOTOXICTY
7.1. Intrinsic Hepatotoxicity
One of two major categories of chemicals that can
produce dose-dependent, hepatic injury, either directly
or indirectly via a metabolite, includes the intrinsic or
‘‘direct’’ (type A) hepatotoxicants. These responses of-
ten can be anticipated from the known pharmacology
of the drug, are generally detectable in animal models,
and occur more frequently or with greater severity
when exposure is increased, i.e., with increasing dose
levels or duration of dosing. Salicylates (Lewis, 1984)
and acetaminophen (Fry and Seeff, 1995) are examples
of analgesics that produce intrinsic liver toxicity. At
high blood levels, aspirin can produce hepatocellular
injury, confirmed by liver biopsy studies (focal necro-
sis), with 10- to 40-fold elevations for serum transami-
nases noted (Zimmerman, 1978). Chemotherapeutic
agents often possess predictable, dose-dependent hepa-
totoxicity (King and Perry, 1995), even though they do
not intentionally target slowly dividing hepatocytes.
Tetracycline, methotrexate, mercaptopurine (Finlay-
son, 1973), and possibly suldinac (Boelsterli et al.,
1995) are other examples of direct hepatotoxins. Intrin-
sic hepatotoxins produce injury in a large percentage
of exposed individuals after a short fixed latent period
either by direct, nonselective physicochemical distor-
tion or disruption of hepatocytes or by indirect interfer-
ence with specific metabolic processes leading to struc-
tural damage (Lewis, 1984). This type of toxicity can
be alleviated by dose reduction in patient populations.
7.2. Idiosyncratic Hepatotoxicity
Although a number of drugs, as for example some
antineoplastic agents (McDonald and Tirumali, 1984),
exhibit intrinsic toxicity in man or animals, most hepa-
totoxic drug reactions in humans are considered to be
idiosyncratic (King and Perry, 1995), i.e., due to un-
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
123
usual susceptibility of an individual. Occurring at ther-
apeutic doses after a variable latent period, these re-
sponses are characterized by an incidence of hepatic
injury that is very low in frequency within a popula-
tion, dose-independent, and not reproducible in experi-
mental animals (Zimmerman, 1978). Due to their low
incidence, idiosyncratic responses are generally not de-
tected until late in the drug development process after
a large number of patients have been treated. Although
rare, serious adverse liver responses may include ful-
minant hepatitis and cholestasis. These account for
many drug-induced deaths worldwide. Drugs with-
drawn from the market due to idiosyncratic drug reac-
tions, including or due solely to hepatotoxicity, include
benoxaprofen, ibufenac, temafloxacin, tenilic acid (Park
et al., 1992), nomifensine, and perhexilene (Brecken-
ridge, 1996).
Etiologically, these reactions are of two major types:
(1) aberrant metabolism-based responses leading to the
accumulation of toxic metabolites in susceptible indi-
viduals and (2) hypersensitivity or immune-based tox-
icity. Other mechanisms may include abnormal recep-
tor sensitivity, latent biochemical abnormalities, or
multifactorial causes (Park et al., 1992). Almost all hep-
atotoxic reactions to antibacterials (George and Craw-
ford, 1996) or NSAIDs (Boelsterli et al., 1995), espe-
cially of the indole, pyrazole, and propionic acid classes
(Lewis, 1984), are idiosyncratic. Mechanisms may be
primarily metabolite-dependent (isoniazid, diclofenac),
hypersensitivity-mediated (beta-lactams, sulindac), or
both (sulfonamides, erythromycin derivatives) (West-
phal et al., 1994; Miyamoto et al., 1997; Boelsterli et
al., 1995). Some notable histamine (H2)-receptor antag-
onists and a number of antidepressants are associated
with hepatic idiosyncratic reactions, presumably medi-
ated via chemically reactive metabolites (Black, 1987;
Pirmohamed et al., 1992).
7.2.1. Metabolism-based toxicity. Interindividual
variations of drug effects are most frequently explained
by genetic variation or polymorphism of drug metabo-
lism. Based on immunogenicity and catalytic activities,
approximately 20 human cytochrome P450 isozymes
have been identified (Larrey and Pageaux, 1997) com-
prising at least 10 distinct P450 gene families (Wat-
kins, 1990). Of these, cytochromes CYP1A2, CYP2C9,
CYP2C19, CYP2E1, and, in particular, CYP2D6 and
CYP3A are especially important for the hepatic metab-
olism of drugs (Brockmoller and Roots, 1994). While
bioactivation of a drug to a toxic metabolite may repre-
sent õ1% of the overall metabolism of that drug (Pir-
mohamed et al., 1996), specific reactive metabolites
may directly lead to hepatic injury or may act via the
immune system. Genetic deficiencies in hepatic cyto-
chrome CYP2D6, CYP2C19, N-acetyltransferase 2
(NAT2), and glutathione synthetase have been cited
to explain individual variations in drug hepatotoxicity
(Larrey and Pageaux, 1997; Meyer, 1996). Seven muta-
rtpal AP: RTP
124 DAVID E. AMACHER
tions of the NAT2 gene are known to define numerous
alleles associated with decreased function, and two mu-
tant alleles of CYP2C19 have been described in poor
metabolizers (Meyer and Zanger, 1997; Goldstein and
de Morias, 1994). Oxidative polymorphism for cyto-
chrome CYP2D6, which metabolizes drugs typically
having a positively charged nitrogen atom (Pirmo-
hamed et al., 1996), includes poor, intermediate, and
ultrarapid metabolizers (Meyer and Zanger, 1997)
leading to a wide range of interindividual variability
in CYP2D6-based response in a healthy population
(Spina et al., 1994). For example, patients receiving the
antianginal agent perhexiline maleate who are defi-
cient in CYP2D6 can develop hepatotoxicity due to per-
hexiline accumulation (Morgan et al., 1984). Although
it is not subject to genetic polymorphism (Ereshefsky,
1996), both genetic and nongenetic factors lead to
marked (5- to 20-fold) individual variability in meta-
bolic clearance via CYP3A (Wilkinson, 1996) which
comprises 25% of the total hepatic cytochrome system.
This leads to wide variations in individual exposure
levels to drugs such as cyclosporine (Watkins, 1990).
Polymorphism has also been described for CYP2C9 and
CYP2E1 (Meyer and Zanger, 1997). The latter is im-
portant for the metabolism of general anesthetics such
as halothane which causes hepatitis in 1 in 35,000 pa-
tients on first exposure and in 1 in 3700 patients upon
multiple exposures (Pirmohamed et al., 1996).
7.2.2. Immune system-based toxicity. Immune-
based injuries represent the other major mechanism
for idiosyncratic responses in humans. Autoimmunity
triggered by xenobiotics involves the modification of
host tissues or immune cells by the chemical so that
self-antigens are erroneously targeted; drug hypersen-
sitivity or allergy refers to a situation where the im-
mune system responds in an exaggerated or inappro-
priate manner (Burns et al., 1996) by one of four mecha-
nisms (Combs and Gell, 1975). Most autoimmune
diseases are associated with specific alleles of the major
histocompatibility complex (MHC) class II genes (Fro-
nek et al., 1991). The high level of polymorphism that
is characteristic of the MHC genes may lead to subpop-
ulations of individuals who are highly immune to re-
sponses toward drug-related antigens (Park et al.,
1992). Four major categories of drugs known to cause
autoimmune responses include hydrazines, derivatives
of aromatic amines, sulfhydryl-containing compounds,
and compounds with a phenol ring (Adams and Hess,
1991). Specific examples of drugs known or suspected
of causing hepatitis via an autoimmune mechanism
include methyldopa, oxypehnisatin, isoniazid, nitrofu-
rantoin, clometacin, fenofibrate, papaverine, and tie-
nilic acid (Bigazzi, 1988). Although other nongenetic
factors may be involved in the etiology of altered im-
mune responses, the role of drug metabolism polymor-
phism leading to bioactivation is an important one. Ex-
trahepatic metabolism of some drugs associated with
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
a generalized type of idiosyncratic drug reaction medi-
ated through the immune system may also involve the
liver (Uetrecht, 1988).
7.3. Mixed or Species-Relevant Toxicity
Some drugs appear to cause idiosyncratic hepatic in-
jury in humans and yet can also produce hepatotoxic-
ity, perhaps by a different mechanism, in animal mod-
els at high doses. Thus, the same drug can be an intrin-
sic hepatotoxicant in one or more animal models and
yet cause an idiosyncratic liver response in some hu-
mans. Some of the angiotensin-converting enzyme in-
hibitors used for the treatment of hypertension and
congestive heart disease fall into this category. Capto-
pril, for example, has been associated with the rare
incidence of hepatotoxicity in humans (Rahmat et al.,
1985; Tabibian et al., 1987). In mice, acute high doses
of captopril cause moderate increases in ALT, de-
creases in hepatic GSH, and histological evidence of
hepatic necrosis with 24 h (Helliwell et al., 1985). Ena-
lapril maleate which has been reported to cause a rare
but potentially serious hepatotoxicity in humans, was
demonstrated by Jurima-Romet and Huang (1992) to
produce centrilobular necrosis and significant moder-
ate increases in ALT and AST 24 h after acute exposure
in Fisher 344 rats.
8. THERAPEUTIC CLASS AND DRUG-INDUCED
HEPATIC INJURY
The frequencies of drug-induced hepatitis at thera-
peutic doses vary among drug classes with some classes
more consistently demonstrating low levels of hepatic
injury than others. These can vary from 1/100 for a few
compounds, to 1/10,000 for some intermediate com-
pounds, to 1/100,000 for others (Larrey and Pageaux,
1997). For one antitumor antibiotic, elevations of ami-
notransferases can occur in up to 100% of patients
(King and Perry, 1995). In one large study (Sameshima
et al., 1984) based on 8156 case histories of drug-in-
duced hepatic injury over a 70-year period in Japan,
the greatest number of cases was due to antibiotics
(34%), followed by CNS drugs (15%), chemotherapeutic
drugs (14%), cardiovascular drugs (11%), carcinostatic
drugs (7%), hormones and hormone antagonists (6%),
diagnostic aids (4%), and other drugs (9%). A smaller
study based on 46 patients representing 20% of pa-
tients admitted for acute or chronic hepatitis during
a 3-year period cited anti-inflammatories, antibiotics,
antihypertensives, and diuretic drugs as being most
frequently implicated (Grippon et al., 1985). In the U.S.
Physicians Desk Reference, transaminase elevations,
albeit usually minor and infrequent, are not uncommon
for drugs classified as antineoplastic compounds, neu-
rotrophic drugs, and antibiotics/antibacterials. Table 1
summarizes available information for some of these
drug categories.
rtpal AP: RTP
 SERUM TRANSAMINASES AND HEPATIC INJURY 125
TABLE 1
The Relative Incidence of Liver Injury and Elevated Transaminase Levels for Various Drug Classes

Class Hepatic effects Mechanism of injury References

Nonsteroidal Incidence of acute liver injury is about 3.7
anti-inflammatory per 100,000 NSAID users. Transaminase
drugs levels abnormal in 5–15% of patients
taking a particular drug.
Anticancer drugs Hepatotoxic manifestations include
venoclusive disease, hepatocellular
necrosis, and fatty change. Incidence of
1.3% in 980 cases of hepatic injury in one
study.
Antibacterials Estimated frequency of hepatotoxic
reactions is between 1 and 10 per 100,000
drug prescriptions. Antibiotics accounted
for one-third of all suspected drug-
induced hepatic injury in one study. Mild
transient transminase elevations reported
in up to 30–50% of recipients for some
antibiotics.
Cardiovascular In one study, cardiovascular agents were
implicated in 27.3% of 980 cases of liver
injury. Diuretics accounted for 13.9% in
another study. Mild ALT/AST elevations
are common at the onset of treatment.
Mostly idiosyncratic responses.
Diverse injuries, both intrinsic
and idiosyncratic.
Injuries include necrosis,
cholestasis, cholangitis, and
other. Mostly idiosyncratic
except intrinsic for
tetracyclines and some
macrolides.
Autoimmune or hypersensitivity
reactions with antiarrhythmic
agents; heptocellular pattern
with antihypertensive agents.
Garcia Rodriguez et al. (1994),
Boelsterli et al. (1995)
King and Pery (1995),
McDonald and Tirumali
(1984), Jean-Paston and
Jouaglard (1984)
George and Crawford (1996),
Stricker et al. (1992b),
Stricker (1986)
Jean-Pastor and Jouglard
(1984), Stricker et al.
(1992b), Stricker (1986)

9. DRUGS FREQUENTLY ASSOCIATED
WITH HEPATOTOXICITY
Some marketed drugs have been associated with a
high, predictable incidence of suspected hepatotoxicity
in human populations, as, for example, they may cause
asymptomatic, transient elevations in serum transami-
nase levels in ¢10% of the patient population. It should
be noted that the actual frequency of isolated, asymp-
tomatic elevations of transaminases in patient popula-
tions may actually be understated because liver func-
tion tests are not routinely ordered during routine clini-
cal practice (Jick, 1995). Some examples are shown in
Table 2. Likewise, drug interaction, disease complica-
tions, viral infections, or combinations of drugs may be
responsible, in part, for some of these effects.
10. DISCUSSION
Drug-induced liver toxicity can be detected at various
stages of drug development depending upon the mecha-
nism responsible for hepatic injury. Intrinsic hepato-
toxicants are effectively identified in the preclinical
testing stage using animal models at doses much
higher than the intended human dose while, at the
other extreme, rare but sometimes fatal idiosyncratic
responses can only be discovered after the exposure of
thousands of patients to therapeutic levels of the drug.
Elevated serum transaminases, especially ALT, are the
single most important laboratory indicators of hepatic
effects from early preclinical testing to the postmarket
monitoring of patients responding to a variety of histo-
logic effects. Baseline serum transaminase levels in hu-
man populations are affected by a variety of both ge-
netic and environmental factors. In addition, some se-
rum enzymes of hepatic origin used for assessment of
liver injury or disease can be increased by hepatic en-
zyme induction instead of or in addition to membrane
leakage caused by hepatic damage. Within these limi-
tations, guidelines for the detection of hepatic injury
via these biochemical markers have been established.
The hepatic metabolism of drugs to reactive toxicants
or protein-binding moieties is the single most im-
portant, but by no means the only mechanism leading
to hepatic injury. Unlike preclinical laboratory animals
which are relatively homogenous in terms of their con-
stitutive hepatic cytochrome P450 isozymes, human
populations are remarkably diverse with cytochrome
P450 enzyme polymorphisms and other individual dif-
ferences in drug-metabolizing enzyme systems being
prevalent. Subtle individual differences in the hepatic
detoxification processes or the formation of minor but
highly reactive metabolites in susceptible individuals
can lead to rare but unpredictable consequences as an
increasing proportion of the population is exposed to a
drug at therapeutic doses. Frequently, these idiosyn-
cratic responses are caused by an aberrant immune
response to the drug or a drug-altered self-antigen. For
both direct, intrinsic toxic responses in humans or labo-
ratory animals and isolated idiosyncratic responses in
a clinical setting, early evidence of hepatic injury is
obtained via biochemical methods including the moni-
toring of serum transaminases.
While knowledge of the typical human metabolism
of experimental drugs will not predict idiosyncratic he-
patic toxicity, it can be useful in determining similari-

AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03 rtpal AP: RTP

126
TABLE 2
Drugs Frequently Associated with Elevated Transaminase Levels
Description of
Drug name transminase response
Amiodarone Approximately 25% of users
(antiarrhythmic) develop transminase
elevations ú2 times normal
during treatment.
Chlorpromazine Mild liver enzyme elevations
(anticonvulsant) in 10–42% of recipients;
serum transaminases
typically elevated first week
of treatment.
Cognex (tacrine–HCl) Elevated transaminases (ú3
cholinesterase times upper limit of normal)
inhibitor (cognition in 25% of patients.
activator)
Ibufenac Elevated transaminase levels
(nonsteroidal anti- in ú30%; hepatocellular
inflammatory) jaundice in 5% of
individuals.
Iproniazid Abnormal transaminases in up
(antidepressant) to 20% of recipients.
Isoniazid Serum transaminases elevated
(antibacterial) in 10–20% of patients.
a-Methyldopa Transient increases in
(antihypertensive) transaminases, Alk P,
bilirubin in 2.8% of patients.
Naltrexone–HCl Peak ALT levels 3–19 times
(Revia) (opioid baseline values in some
antagonist) patients after 3–8 weeks of
treatment.
Precose (acarbose) ALT and/or ALT elevated ú3
a-glucosidase times upper limit of normal
inhibitor (diabetes) in 15% of patients.
a-Methyldopa Transient increases in
(antihypertensive) transaminases, Alk P, and
bilirubin in 2.8% of patients.
Naltrexone–HCl Peak ALT levels 3–19 times
(Revia) (opioid baseline values in some
antagonist) patients after 3–8 weeks of
treatment.
Precose (acarbose) ALT and/or ALT elevated ú3
a-glucosidase times upper limit of normal
inhibitor (diabetes) in 15% of patients.
Proleukin Elevated in 56% of patients.
(aldesleukin;
interleukin-2
product)
(anticancer)
Riluzole (Rilutek) Elevated ALT in 10% and AST
(amyotrophic in 14% of patients at 100
lateral sclerosis mg/kg.
treatment)
Tienilic acid AST and ALT elevated.
(uricosuric diuretic)
Valproic acid Frequent, minor, dose-related
(anticonvulsant) elevations of ALT, AST, and
LDH reported.
AID RTP 1201 / 6e19$$1201 06-19-98 12:50:03
DAVID E. AMACHER
Hepatic histological
responses Mechanisms Reference
Phospholipidosis. Metabolic idiosyncratic. Lewis et al. (1989),
Sticker et al. (1992)
Cholestatic or mixed. Reactive metabolite Davis (1989), Pessayre
responsible, (1992), De Mol et al.
imunoallergy. (1984), Stricker et al.
(1992b)
Consistent with Mechanism of Watkins et al. (1994)
hepatocellular injury. hepatotoxicity is
unclear.
Hepatocellular necrosis. Intrinsic toxicity. Boelsterli et al. (1995),
Removed from clinical Fry and Seeff (1995)
use in 1967.
Hepatocellular necrosis. Reactive metabolite Sticker et al. (1992b)
Withdrawn from most
markets.
Liver injury reported in Reactive metabolite Davis (1989), Timbrell
0.5% of adults; higher responsible for et al. (1980), Stricker
in children metabolic (1986), Stricker et al.
idiosyncratic (1992b), PDR
response.
Acute hepatitis in about Reactive metabolite Davis (1989), Finlayson
6% of patients within responsible, (1973), Dybing et al.
the first few weeks. hypersensitivity. (1976)
No cases of hepatic Direct hepatotoxin. PDR
failure reported.
Hepatic abnormalities Not reported. PDR
improved or resolved
upon discontinuation.
Acute hepatitis in about Reactive metabolite Davis (1989), Finlayson
6% of patients within responsible, (1973), Dybing et al.
the first few weeks. hypersensitivity. (1976)
No cases of hepatic Direct hepatotoxin. PDR
failure reported.
Hepatic abnormalities Not reported. PDR
improved or resolved
upon discontinuation.
Elevated bilirubin, Not reported. PDR
alkaline phosphatase,
jaundice.
Dose-related (intrinsic). Bryson et al. (1996)
1 in 10,000 patients Reactive metabolite Dansette et al. (1991),
develop hepatitis of responsible, Stricker (1986)
the immunoallergic immunoallergy.
type. Withdrawn from
the market.
Microvesicular steatosis Reactive metabolite. Koch-Weser and Browne
(1980), Pessayre (1992),
Porubek et al. (1984),
PDR
rtpal AP: RTP
 SERUM TRANSAMINASES AND HEPATIC INJURY
ties between species that may lead to intrinsic toxicity
or to explain tolerance or sensitivity to the hepatic tox-
icity of a compound in one species versus another. Cur-
rent in vitro laboratory procedures utilize isolated mi-
crosomal preparations or specific recombinant human
cytochrome products in conjunction with chromato-
graphic procedures for the detection and identification
of drug metabolites. In vitro methods have been used
to predict the in vivo drug metabolism of verapamil,
loxidine, diazepam, lidocaine, phenacetin, and some
other compounds in the human liver (Iwatsubo et al.,
1997). In a similar manner, in vitro comparisons of
drug toxicity in metabolizing rat, dog, monkey, and,
where possible, human hepatocyte cultures can be used
for the preclinical assessment of drug hepatic toxicity
using the same clinical chemistry end points, namely
the transaminases, that are used in vivo. Primary he-
patocytes cultured on an extracellular collagen matrix
in time form bile canaliculi (LeCluyse et al., 1994) pro-
viding the opportunity to study the effects of drugs
on the hepatobiliary system in vitro. Examples of the
successful use of hepatocyte cultures from humans and
other species to study hepatic effects include the me-
tabolism of adinazolam, the cytotoxicity of trospecto-
mycin, and the human-specific hepatotoxicity of panad-
iplon (Ulrich et al., 1995).
Recent developments that may have application in
preclinical testing for potential liver toxicity include
procedures that have evolved in conjunction with com-
binatorial chemistry and pharmacological screening
methods (see Lam, 1997). Screening methods of partic-
ular interest include genomic and proteonomic assays
that are sensitive and plenary by nature, although they
may be of limited value in revealing specific toxicities.
These methods depend heavily upon automation, the
availability of specific probes or libraries for pattern
recognition, and advanced computer systems. The ulti-
mate advantage of these techniques would be the detec-
tion of early cellular events, as, for example, markers
of oxidative damage, at therapeutic dose levels that
may predispose the organism to frank toxicity with sus-
tained exposure. Examples of gene expression in liver
after toxic injury include heat shock and oxidative
stress-inducible genes (Schiaffonati and Tiberio, 1997).
Both induction and suppression of gene expression can
occur with some gene families (Runge-Morris, 1997).
Reactive intermediates of xenobiotics often bind cova-
lently to proteins as conjugates or adducts leading to
either direct toxicity or immune-mediated toxicity
(Pumford et al., 1997). In addition to the modification
of cellular proteins, xenobiotics may modify protein ex-
pression (Richardson et al., 1994). Two-dimensional gel
electrophoresis techniques have been described that
allow the resolution and quantitation of hundreds of
liver proteins that may be altered by the toxic effects
of xenobiotics (Anderson et al., 1996). If a clear associa-
tion between either altered gene expression or specific
changes in protein structure and abundance can be
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
127
established with specific liver injury, these techniques
may be useful in detecting hepatotoxic potential pre-
clinically, even at therapeutic doses.
Idiosyncratic responses remain difficult to assess at
a preclinical or early clinical level. Using separate cul-
tures from several donors, it may be possible to expose
human hepatocytes to higher concentrations of a drug
than would be found in clinical practice in order to
forcibly detect idiosyncratic responses due to minor re-
active metabolites that do not involve the immune sys-
tem. Because most compounds that have been tested
do cause observable changes in liver gene expression
(Anderson et al., 1996) and idiosyncratic hepatotoxins
such as diclofenac, halothane, tienilic acid, and valproic
among others are known to form protein adducts (Pum-
ford and Halmes, 1997), protein analysis by two-dimen-
sional gel electrophoresis may provide a useful preclini-
cal screening too. Future developments in assays may
include determinations of drug effects on transgenic
animals expressing human MHC class I and II and
associated genes (Campbell and Milner, 1993), utiliza-
tion of the sensitive reporter antigen/popliteal lymph
node assay (Albers et al., 1997), the comparative testing
of a compound at high doses or concentrations in
transgenic animals or cell lines expressing various con-
stitutive human cytochrome P450s, or the testing of
preclinical candidates in rodent strains genetically di-
verse in terms of the hepatic cytochrome P-450 system.
11. CONCLUSIONS
• The incidence of drug-induced hepatic injury var-
ies among the different drug classes as well as within
a drug class. Intrinsic responses are generally detected
early in the discovery and development process, while
host-dependent, idiosyncratic responses are usually
not detectable during the early phases of development
due to their rarity.
• Of the various clinical laboratory markers for he-
patic injury, serum transaminases, especially ALT, are
the most universally important indicators for studies
ranging from early preclinical animal testing to post-
marketing patient monitoring. Serum ALT is elevated
in response to several categories of hepatic lesions in
multiple species. However, slight transaminase eleva-
tions following drug exposure in humans must be inter-
preted with care due to the many extrinsic factors
known to elevate transaminases.
• The hepatic metabolism of drugs to reactive toxi-
cants or protein-binding moieties is a frequently en-
countered mechanism leading to hepatic injury. Subtle
individual differences in the hepatic detoxification pro-
cesses or the formation of minor but highly reactive
metabolites in susceptible individuals can lead to rare
but unpredictable consequences as an increasing pro-
portion of the population is exposed to a drug at thera-
peutic doses. Improvements in the metabolic profiling
of drugs using in vitro microsomal systems expressing
rtpal AP: RTP
128 DAVID E. AMACHER
multiple human cytochromes during the preclinical
testing phase may enhance our ability to anticipate
both intrinsic and idiosyncratic toxic responses.
• Although the monitoring of serum transaminases
will continue to be an important clinical laboratory pro-
cedure for the assessment of preclinical hepatotoxicity,
new developments involving comparative toxicity
assays in human and animal hepatocyte cultures, tech-
niques to detect altered gene expression or changes in
protein structure and abundance, or improved methods
for the detection of immunogenicity will augment the
safety assessment process.
REFERENCES
Adams, L. E., and Hess, E. V. (1991). Drug-related lupus. Incidence,
mechanisms and clinical implications. Drug Saf. 6, 431–449.
Aiges, H. W., Daum, F., Olson, M., Kahn, E., and Teichberg, S. (1980).
The effects of phenobarbital and diphenylhydantoin on liver func-
tion and morphology. J. Pediatr. 97, 22–26.
Albers, R., Broeders, A., Van der Pijl, A., Seinen, W., and Pieters,
R. (1997). The use of reporter antigens in the popliteal lymph node
assay to assess immunomodulation by chemicals. Toxicol. Appl.
Pharmacol. 143, 102–109.
Anderson, N. L., Taylor, J., Hoffman, J. P., Esquer-Blasco, R., Swift,
S., and Anderson, N. G. (1996). Simultaneous measurement of
hundreds of liver proteins: Application in assessment of liver func-
tion. Toxicol. Pathol. 24, 72–76.
Batt, A.-M., and Ferrari, L. (1995). Manifestations of chemically in-
duced liver damage. Clin. Chem. 41(12), 1882–1887.
Benichou, C. (1990). Criteria of drug-induced liver disorders. Report
of an international consensus meeting. J. Hepatol. 11, 272–276.
Bigazzi, P. E. (1988). Autoimmunity induced by chemicals. Clin. Tox-
icol. 26, 125–156.
Billon, M. C., Dupre, G., and Hanoune, J. (1980). In vivo modulation
of rat hepatic g-glutamyltransferase activity by glucocorticoids.
Mol. Cell. Endocrinol. 18, 99–108.
Black, M. (1987). Hepatotoxic and hepatoprotective potential of his-
tamine (H2)-receptor antagonists. Am. J. Med. 83(Suppl. 6A), 68–
75.
Boelsterli, U. A., Zimmerman, H. J., and Kretz-Rommel, A. (1995).
Idiosyncratic liver toxicity of nonsteroidal anti-inflammatory
drugs: Molecular mechanisms and pathology. Crit. Rev. Toxicol.
25, 207–235.
Breckenridge, A. (1996). A clinical pharmacologist’s view of drug
toxicity. Br. J. Clin. Pharmacol. 42, 53–58.
Brockmoller, J., and Roots, I. (1994). Assessment of liver metabolic
function. Clinical implications. Clin. Pharmacokinet. 27, 216–248.
Bryson, H. M., Fulton, B., and Benfield, P. (1996). Riluzole. A review
of its pharmacodynamic and pharmacokinetic properties and ther-
apeutic potential in amyotrophic lateral sclerosis. Drugs 52, 549–
563.
Burns, L. A., Meade, B. J., and Munson, A. E. (1996). Toxic responses
of the immune system. In Casarett and Doull’s Toxicology. The
Basic Science of Poisons (C. D. Klaassen, Ed.), 5th ed., pp. 355–
402. McGraw-Hill, New York.
Campbell, R. D., and Milner, C. M. (1993). MHC genes in autoimmu-
nity. Curr. Opin. Immunol. 5, 887–993.
Clampitt, R. B., and Hart, R. J. (1978). The tissue activities of some
diagnostic enzymes in ten mammalian species. J. Comp. Pathol.
88, 607–621.
Clermont, R. J., and Chalmers, T. C. (1967). The transaminase tests
in liver disease. Medicine 46, 197–207.
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
Coombs, R. R. A., and Gell, P. G. H. (1975). Classification of allergic
reactions responsible for clinical hypersensitivity and disease. In
Clinical Aspects of Immunology (P. G. H. Gell, R. R. A. Coombs,
and P. G. Lachmann, Eds.), pp. 761–781. Oxford Univ. Press, Ox-
ford.
Craxi, A., and Almasio, P. (1996). Diagnostic approach to liver en-
zyme elevation. J. Hepatol. 25(Suppl. 1), 47–51.
Dansette, P. M., Amar, C., Valadon, P., Pons, C., Beaune, P. H., and
Mansuy, D. (1991). Hydroxylation and formation of electrophilic
metabolites of tienilic acid and its isomer by human liver micro-
somes. Biochem. Pharmacol. 41, 553–560.
Davis, M. (1989). Drugs and abnormal ‘liver function tests.’ Adverse
Drug React. Bull. 139, 520–523.
De Mol, N. J., and Busker, R. W. (1984). Irreversible binding of the
chlorpromazine radical cation and of photoactivated chlorproma-
zine to biological macromolecules. Chem.–Biol. Interact. 52, 79–
92.
Dhami, M. S. I., Drangova, R., Farkas, R., Balazs, T., and Feuer, G.
(1979). Decreased aminotransferase activity of serum and various
tissues in the rat after cefazolin treatment. Clin. Chem. 25, 1263–
1266.
Dybing, E., Nelson, S. D., Mitchell, J. R., Sasame, H. A., and Gillette,
J. R. (1976). Oxidation of a-methyldopa and other catechols by
cytochrome P-450-generated superoxide anion: Possible mecha-
nism of methyldopa hepatitis. Mol. Pharmacol. 12, 911–920.
Ereshefsky, L. (1996). Drug–drug interactions involving antidepres-
sants: Focus on venlafaxine. J. Clin. Psychopharmacol. 16, 37S–
53S.
Finlayson, N. D. (1973). The adverse effects of drugs on the liver. Br.
J. Clin. Pract. 27, 286–290.
Foster, G. R., Goldin, R. D., and Oliveira, D. B. (1989). Serum F pro-
tein: A new sensitive and specific test of hepatocellular damage.
Clin. Chim. Acta 184, 85–92.
Fregia, A., and Jensen, D. M. (1994). Evaluation of abnormal liver
tests. Comp. Ther. 20, 50–54.
Fry, S. W., and Seeff, L. B. (1995). Hepatotoxicity of analgesics and
anti-inflammatory agents. Gastroenterol. Clin. N. Am. 24, 875–
905.
Fronek, Z., Timmerman, L. A., and McDevitt, H. O. (1991). A rare
HLA DQB allele sequenced from patients with systemic lupus ery-
thematosus. Hum. Immunol. 30, 77–84.
Garcia Rodriguez, L. A., Williams, R., Derby, L. E., Dean, A. D., and
Hershel, J. (1994). Acute liver injury associated with nonsteroidal
anti-inflammatory drugs and the role of risk factors. Arch. Intern.
Med. 154, 311–316.
George, D. K., and Crawford, D. H. G. (1996). Antibacterial-induced
hepatotoxicity. Incidence, prevention, and management. Drug Saf.
15, 79–85.
Goldstein, J. A., and de Morais, S. M. F. (1994). Biochemistry and
molecular biology of the human CYP2C subfamily. Pharmacoge-
netics 4, 285–299.
Grippon, P., Hirsch-Marrie, I., Biour, M., and Levy, V. G. (1984).
Drug induced liver disease in a liver unit. A retrospective study
of 46 patients in 3 years. Therapie 39, 503–507.
Grohmann, R., Hippius, H., Muller-Oerlinghausen, B., Ruther, E.,
Scherer, J., Schmidt L. G., Strauss, A., and Wolf, B. (1984). Assess-
ment of adverse drug reactions in psychiatric hospitals. Eur. J.
Clin. Pharmacol. 26, 727–734.
Haidukewych, D., and John, G. (1986). Chronic valproic acid and
coantiepileptic drug therapy and incidence of increases in serum
liver enzymes. Ther. Drug Monit. 8, 407–410.
Helliwell, T. R., Yeung, J. H. K., and Park, B. K. (1985). Hepatic ne-
crosis and glutathione depletion in captopril-treated mice. Br. J.
Exp. Pathol. 66, 67–78.
Herlong, H. F. (1994). Approach to the patient with abnormal liver
enzymes. Hosp. Pract. 29, 32–38.
rtpal AP: RTP
 SERUM TRANSAMINASES AND HEPATIC INJURY
Herzfeld, A., and Raper, S. M. (1979). Effects of cortisol or starvation
on the activities of four enzymes in small intestine and liver of the
rat during development. J. Dev. Physiol. 1, 315–327.
Hirayanagi, N., Fujii, E., and Teshirogi, T. (1991). The significance
of serum g-glutamyl transpeptidase (g-GTP) elevation caused by
antiepileptic drug(s). Using the antipyrine metabolic capacity as
a parameter of microsomal enzyme activity. Nippon Ika Daigaku
Zasshi 58, 547–560.
Hodgson, M. J., Van Thiel, D. H., Lauschus, K., and Karpf, M. (1989).
Liver injury tests in hazardous waste workers: The role of obesity.
J. Occup. Med. 31, 238–242.
Hoffmann, W. E., Kramer, J., Main, A. R., and Torres, J. L. (1989).
Clinical enzymology. In The Clinical Chemistry of Laboratory Ani-
mals (W. F. Loeb and F. W. Quimby, Eds.), pp. 237–278. Pergamon
Press, New York.
Hultcrantz, R., Glaumann, H., Lindberg, G., and Son Nilsson, L.
(1986). Liver investigation in 149 asymptomatic patients with
moderately elevated activities of serum aminotransferases. Scand.
J. Gastroenterol. 21, 109–113.
Iwatsubo, T., Hirota, N., Ooie, T., Suzuki, H., Shimada, N., Chiba, K.,
Ishizaki, T., Green, C. E., Tyson, C. A., and Sugiyama, Y. (1997).
Prediction of in vivo drug metabolism in the human liver from in
vitro metabolism data. Pharmacol. Ther. 73, 147–171.
Jacobsen, N. O., Mosekilde, L., Myhre-Jensen, O., Pedersen, E., and
Wildenhoff, K. E. (1976). Liver biopsies in epileptics during anti-
convulsant therapy. Acta Med. Scand. 199, 345–348.
Jean-Pastor, M. J., and Jouglard, J. (1984). Bilan des accidents hepa-
tiques medicamenteux recueillis par la pharmacovigilance fran-
caise. Therapie 39, 493–500.
Jick, H. (1995). Drug-associated asymptomatic elevations of trans-
aminase in drug safety assessments. Pharmacotherapy 15, 23–25.
Jurima-Romet, M., and Huang, H. S. (1992). Enalapril hepatotoxicity
in the rat. Effects of modulators of cytochrome P450 and glutathi-
one. Biochem. Pharmacol. 44, 1803–1810.
King, P. D., and Perry, M. C. (1995). Hepatotoxicity of chemothera-
peutic and oncologic agents. Gastroenterol. Clin. N. Am. 24, 969–
990.
Koch-Weser, J., and Browne, T. R. (1980). Drug therapy: Valproic
acid. New Engl. J. Med. 302, 661–666.
Lam, K. S. (1997). Application of combinatorial library methods in
cancer research and drug discovery. Anticancer Drug Des. 12, 145–
167.
Larrey, D., and Pageaux, G. P. (1997). Genetic predisposition to drug-
induced hepatotoxicity. J. Hepatol. 26(Suppl. 2), 12–21.
LeCluyse, E. L., Audus, K. L., and Hochman, J. H. (1994). Formation
of extensive canalicular networks by rat hepatocytes cultured in
collagen-sandwich configuration. Am. J. Physiol. 266, C1764–
C1774.
Lewis, J. H. (1984). Hepatic toxicity of nonsteroidal anti-inflamma-
tory drugs. Clin. Pharm. 3, 128–138.
Lewis, J. H., Ranard, R. C., Caruso, A., Jackson, L. K., Mullick, F.,
Ishak, K. G., Seeff, L. B., and Zimmerman, H. J. (1989). Amiodar-
one hepatotoxicity: Prevalence and clinicopathologic correlations
among 104 patients. Hepatology 9, 679–685.
McDonald, G. B., and Tirumali, N. (1984). Intestinal and liver toxic-
ity of antineoplastic drugs. West. J. Med. 140, 250–259.
Meyer, U. A. (1996). Overview of enzymes of drug metabolism. J.
Pharmacokinet. Biopharm. 24, 449–459.
Meyer, U. A., and Zanger, U. M. (1997). Molecular mechanisms of
genetic polymorphisms of drug metabolism. Annu. Rev. Pharma-
col. Toxicol. 37, 269–296.
Mijovic, V., Contreras, M., and Barbara, J. (1987). Serum alanine
aminotransferase (ALT) and g-glutamyltransferase (g-GT) activi-
ties in north London blood donors. J. Clin. Pathol. 40, 1340–1344.
Miyamoto, G., Zahid, N., and Uetrecht, J. P. (1997). Oxidation of
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
129
diclofenac to reactive intermediates by neutrophils, myeloperoxi-
dase, and hypochlorous acid. Chem. Res. Toxicol. 10, 414–419.
Morgan, M. Y., Reshef, R., Shah, R. R., Oates, N. S., Smith, R. L., and
Sherlock, S. (1984). Impaired oxidation of debrisoquine in patients
with perhexiline liver injury. Gut 25, 1057–1064.
Mosca, P., Bonazzi, P., Novelli, G., Jezequel, A. M., and Orlandi, F.
(1985). In vivo and in vitro inhibition of hepatic microsomal drug
metabolism by ketoconazole. Br. J. Exp. Pathol. 66, 737–742.
Moslen, M. T. (1996). Toxic responses of the liver. In Casarett and
Doull’s Toxicology. The Basic Science of Poisons (C. D. Klaassen,
Ed.), 5th ed., pp. 403–416. McGraw-Hill, New York.
Nishimura, M., Stein, H., Berges, W., and Teschke, R. (1981).
Gamma-glutamyltransferase activity in the liver plasma mem-
brane: Induction following chronic alcohol consumption. Biochem.
Biophys. Res. Commun. 99, 141–148.
Pappas, N. J., Jr. (1986). Source of increased serum aspartate and
alanine aminotransferase: Cycloheximide effect on carbon tetra-
chloride hepatotoxicity. Clin. Chim. Acta 154, 181–190.
Park, B. K., Pirmohamed, M., and Kitteringham, N. R. (1992). Idio-
syncratic drug reactions: A mechanistic evaluation of risk factors.
Br. J. Clin. Pharmacol. 34, 377–395.
Pessayre, D. (1992). III. Mechanisms of drug-induced hepatic injury.
In Drug Induced Hepatic Injury (B. H. Ch. Stricker, Ed.), 2nd ed.,
pp. 23–49. Elsevier, New York.
Pirmohamed, M., Kitteringham, N. R., and Park, B. K. (1992). Idio-
syncratric reactions to antidepressants: A review of possible mech-
anisms and predisposing factors. Pharmacol. Ther. 53, 105–125.
Pirmohamed, M., Madden, S., and Park, B. K. (1996). Idiosyncratic
drug reactions. Metabolic bioactivation as a pathogenic mecha-
nism. Clin. Pharmacokinet. 31, 215–230.
Porubek, D. J., Grillo, M. P., and Baillie, T. A. (1989). The covalent
binding to protein of valproic acid and its hepatotoxic metabolite,
2-n-propyl-4-pentenoic acid, in rats and in isolated rat hepatocytes.
Drug Metab. Dispos. 17, 123–130.
Pumford, N. R., and Halmes, N. C. (1997). Protein targets of xenobi-
otic reactive intermediates. Annu. Rev. Pharmacol. Toxicol. 37,
91–117.
Pumford, N. R., Halmes, N. C., and Hinson, J. A. (1997). Covalent
binding of xenobiotics to specific proteins in the liver. Drug Metab.
Rev. 29, 39–57.
Rahmat, J., Gelfand, R. L., Gelfand, M. C., Winchester, J. F.,
Schreiner, G. E., and Zimmerman, H. J. (1985). Captopril-associ-
ated cholestatic jaundice. Ann. Intern. Med. 102, 56–58.
Rej, R. (1989). Aminotransferases in disease. Diagn. Enzymol. 9,
667–687.
Renner, E. L., and Dallenbach, A. (1992). Erhohte leberenzyme: Was
nun? Ther. Umsch. 49, 281–286.
Richardson, F. C., Horn, D. M., and Anderson, N. L. (1994). Dose-
responses in rat hepatic protein modification and expression fol-
lowing exposure to the rat hepatocarcinogen methapyrilene. Carci-
nogenesis 15, 325–329.
Runge-Morris, M. A. (1997). Regulation of expression of the rodent
cytosolic sulfotransferases. FASEB J. 11, 109–117.
Sakagishi, Y. (1995). Alanine aminotransferase (ALT). Nippon Rin-
sho 53, 1146–1150.
Sameshima, Y., Shiozaki, Y., Mizuno, T., and Sasakawa, M. (1984).
Drug-induced hepatic injury. Jpn. J. Gastroenterol. 81, 37–45.
Scheig, R. (1996). Evaluation of test used to screen patients with
liver disorders. Primary care. Clin. Office Prac. 23, 551–560.
Schiaffonati, L., and Tiberio, L. (1997). Gene expression in liver after
toxic injury: Analysis of heat shock response and oxidative stress-
inducible genes. Liver 17, 183–191.
Schmidt, E., and Schmidt, F. W. (1990). Progress in the enzyme diag-
nosis of liver disease: Reality or illusion? Clin. Biochem. 23, 375–
382.
rtpal AP: RTP
130 DAVID E. AMACHER
Sherman, K. E. (1991). Alanine aminotransferase in clinical practice.
A review. Arch. Intern. Med. 151, 260–265.
Spina, E., Campo, G. M., Avenoso, A., Caputi, A. P., Zuccaro, P., Pa-
cifici, R., Gatti, G., Strada, G., Bartoli, A., and Perucca, E. (1994).
CYP2D6-related oxidation polymorphism in Italy. Pharmacol. Res.
29, 281–289.
Stricker, B. H. Ch. (1986). Hepatic injury by drugs and environmen-
tal agents. In The Liver Annual (I. M. Arias, M. Frenkel, and
J. H. P. Wilson, Eds.), Vol. 5, pp. 419–482. Elsevier, New York.
Stricker, B. H. Ch., Blok, A. P. R., and Desmot, V. J. (1992a). IV.
Pathology of drug-induced hepatic injury. In Drug-Induced Hepatic
Injury (B. H. Ch. Stricker, Ed.), 2nd ed., pp. 51–70. Elsevier, New
York.
Stricker, B. H. Ch., Biour, M., and Wilson, J. H. P. (1992b) V. Individ-
ual agents: Drugs. In Drug-Induced Hepatic Injury (B. H. Ch.
Stricker, Ed.), 2nd ed., pp. 71–524. Elsevier, New York.
Tabibian, N., Alpert, L., and Alpert, E. (1987). Captopril-induced
liver dysfunction. South. Med. J. 80, 1173–1175.
Teschke, R., Neuefeind, M., Nishimura, M., and Strohmeyer, G.
(1983). Hepatic gamma-glutamyltransferase activity in alcoholic
fatty liver: Comparison with other liver enzymes in man and rats.
Gut 24, 625–630.
Timbrell, J. A., Mitchell, J. R., Snodgrass, W. R., and Nelson, S. D.
(1980). Isoniazid hepatotoxicity: The relationship between covalent
binding and metabolism in vivo. J. Pharmacol. Exp. Ther. 213,
364–369.
Uetrecht, J. P. (1988). Mechanism of drug-induced lupus. Chem. Res.
Toxicol. 1, 133–143.
Ulrich, R. G., Bacon, J. A., Cramer, C. T., Peng, G. W., Petrella, D. K.,
Stryd, R. P., and Sun, E. L. (1995). Cultured hepatocytes as investi-
gational models for hepatic toxicity: Practical applications in drug
discovery and development. Toxicol. Lett. 82/83, 107–115.
Van Dijke, C. P. H., and Mattie, E. N, H. (1986). Bijwerkingen van
geneesmiddelen op een afdeling inwendige geneeskunde. Ned.
Tijdschr. Geneeskd. 130, 1889–1893.
AID RTP 1201 / 6e19$$$$21 06-19-98 12:50:03
Wall, M., Baird-Lambert, J., Buchanan, N., and Farrell, G. (1992).
Liver function tests in persons receiving anticonvulsant medica-
tions. Seizure 1, 187–190.
Waner, T., and Nyska, A. (1991). The toxicological significance of
decreased activities of blood alanine and aspartate aminotransfer-
ase. Vet. Res. Commun. 15, 73–78.
Watkins, P. B. (1990). Role of cytochromes P450 in drug metabolism
and hepatotoxicity. Semin. Liver Dis. 10, 235–250.
Watkins, P. B., Zimmerman, H. J., Knapp, M. J., Gracon, S. I., and
Lewis, K. W. (1994). Hepatotoxic effects of tacrine administration
in patients with Alzheimer’s disease. JAMA 271, 992–998.
Westphal, J. F., Vetter, D., and Brogard, J. M. (1994). Hepatic side
effects of antibiotics. J. Antimicrob. Chemother. 33, 387–401.
Wilkinson, G. R. (1996). Cytochrome P4503A (CYP3A) metabolism:
Prediction of in vivo activity in humans. J. Pharmacokinet. Bio-
pharm. 24, 475–490.
Wright, C., Rivera, J. C., and Baetz, J. H. (1988). Liver function test-
ing in a working population: Three strategies to reduce false-posi-
tive results. J. Occup. Med. 30, 693–697.
Yamada, R. H., Wakabayahsi, Y., and Iwashima, A. (1984a). Inhibi-
tion of serum aspartate aminotransferase induced by isoniazid ad-
ministration in mice. Acta Vitaminol. Enzymol. 6, 289–294.
Yamada, R., Wakabayahsi, Y., and Iwashima, A. (1984b). ‘‘In vivo’’
inhibition of murine liver aspartate aminotransferase by isoniazid.
Acta Vitaminol. Enzymol. 6, 29–39.
Zeisel, S. H., Da Costa, K.-A., Franklin, P. D., Alexander, E. A., La-
mont, J. T., Sheard, N. F., and Beiser, A. (1991). Choline, an essen-
tial nutrient for humans. FASEB J. 5, 2093–2098.
Zimmerman, H. J. (1984). Function and integrity of the liver. In
Clinical Diagnosis and Management by Laboratory Methods (J. B.
Henry, Ed.), pp. 217–250. Saunders, Philadelphia.
Zimmerman, H. J. (1974). Editorial: Aspirin-induced hepatic injury.
Ann. Intern. Med. 80, 103–105.
Zimmerman, H. J. (1978). Hepatotoxicity. The Adverse Effects of
Drugs and Other Chemicals on the Liver. Appleton-Century-Crofts,
New York.
rtpal AP: RTP