Clinical Radiology 72 (2017) 276e285

Contents lists available at ScienceDirect

Clinical Radiology
journal homepage: www.clinicalradiologyonline.net

Review

Hepatocellular adenoma: imaging review of the

various molecular subtypes
H. Dharmana a, *, S. Saravana-Bawan a, S. Girgis b, G. Low a
a
Department of Radiology & Diagnostic Imaging, University of Alberta Hospital, 2A2.41 WMC, 8440-112 Street,
Edmonton, AB T6G 2B7, Canada
b
Department of Laboratory Medicine & Pathology, University of Alberta Hospital, 5B2.36 WMC, 8440-112 Street,
Edmonton, AB T6G 2B7, Canada

article in formation
This educational review focuses on the epidemiology and radiological evaluation of the various
Article history: subtypes of hepatic adenomas (HCAs). It includes detailed discussion of the imaging appear-
Received 10 May 2016 ances of each HCA subtype and the clinical relevance of the new classification system. Each
Received in revised form HCA subtype has a unique biological behaviour. Imaging plays a central role in diagnosis,
30 November 2016 subtype characterisation, identification of complications, and follow-up assessment. Man-
Accepted 22 December 2016 agement of patients should vary according to the specific HCA subtype.
Crown Copyright Ó 2017 Published by Elsevier Ltd on behalf of The Royal College of
Radiologists. All rights reserved.

Introduction behaviour.4 Furthermore, recent studies have shown that

Hepatic adenoma (HCA) is a rare benign monoclonal
neoplasm, most commonly encountered in women with a
history of oral contraceptive use. Although benign, ade-
nomas may be complicated by haemorrhage or malignant
transformation into hepatocellular carcinomas (HCCs).1e3
Recently, HCA has been categorised into three distinct
subtypes based on genetic and histopathological features:
(1) inflammatory (I-HCA), (2) hepatocyte nuclear factor-1
alpha inactivated (HNF-1a) HCA, and (3) b-catenin-acti-
vated HCA.4 “Unclassified” HCAs are a small group that lacks
specific genetic abnormalities.4 Although the classification
system is new, knowledge of the different subtypes is
important because each entity has a different biological
* Guarantor and correspondent: H. Dharmana, Department of Radiology &
Diagnostic Imaging, University of Alberta Hospital, 2A2.41 WMC, 8440-112
Street, Edmonton, AB T6G 2B7, Canada. Tel.: þ1 780 407 6907; fax: þ1 780
4073853.
E-mail address: hdharmana@gmail.com (H. Dharmana).
certain subtypes have characteristic imaging features that
may permit a confident non-invasive diagnosis.1,5 In this
article, we discuss the epidemiology, pathology, imaging
features, differential diagnoses, and management options of
HCA and the clinical relevance of the new classification
system.
Epidemiology
HCAs are extremely rare. In North America and Europe,
the annual incidence is 1e1.3 million cases per year with
approximately 86% of HCAs occurring in females of child-
bearing age.6 The male-to-female ratio is approximately
1:10 and most patients are between 15 to 45 years of age at
presentation.7,8 A greater HCA incidence is recognised in the
last several decades due to the introduction of the oral
contraceptive pill (OCP).9 In addition, HCAs are increasingly
discovered as incidental findings in patients undergoing
imaging tests for non-specific complaints or unrelated
clinical indications. This is likely a reflection of an overall

http://dx.doi.org/10.1016/j.crad.2016.12.020
0009-9260/Crown Copyright Ó 2017 Published by Elsevier Ltd on behalf of The Royal College of Radiologists. All rights reserved.
 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285 277

increase in the volume of imaging performed on a per
population basis over time.
Associations/risk factors
OCP
HCA was rarely reported before the advent of the OCP in
the 1960s. A strong causal association between OCP and the
development of HCA was first recognised by Baum in
1973.10 Subsequent studies have confirmed this link and
demonstrated that the development of HCA is related to
higher doses and longer duration of OCP use.9 The incidence
of HCA in women on the long-term OCP (3e4 per 100,000)
is approximately 25 times greater than that of the general
population.10 A few studies have also suggested that OCP
discontinuation may lead to tumour regression, but this is
controversial.11 In China, a male predominance (62.3%) has
been reported for HCA, possibly due to limited OCP use in
this country.6 Although OCP is considered the most
important risk factor, HCAs can also develop de novo or in
association with other conditions.
increasing muscle mass.13 AAS are synthetically manufac-
tured hormones of testosterone. The liver is a hormone-
sensitive organ that expresses oestrogen and androgen re-
ceptors. It is thought that excess androgenic stimulation of
hepatocytes leads to abnormal cell proliferation and HCA
development.14 Some studies have documented tumour
regression following AAS discontinuation.15,16
Glycogen storage disorders
HCA has been described in patients with glycogen stor-
age disorders (GSDs), particularly type 1 (Von Gierke’s
disease) and type 3.17 GSDs are hereditary disorders with an
autosomal recessive transmission. They are characterised
by abnormal glycogen accumulation in the liver, which
causes chronic inflammation and predisposes to HCA
development. HCAs associated with GSDs commonly affect
males (male to female ratio of 2:1) and typically develop in
patients <20 years of age.18,19 These HCAs are more likely to
be multiple17 and have a higher risk of malignant trans-
formation into HCC.17
Hepatic adenomatosis

Anabolic steroids
Long-term misuse of anabolic androgenic steroids (AAS)
has been implicated in HCA development.12 Since the 1990s,
there have been a growing number of reports of AAS misuse
amongst athletes and bodybuilders for the purpose of
Hepatic adenomatosis refers to cases where there are >
10 HCAs in the liver (Fig. 1). The condition was first
described by Flejou et al. in 1985.20 Unlike solitary HCAs,
hepatic adenomatosis generally develops in the absence
of predisposing factors and may be associated with a
higher risk of complications.21 One study suggests that

Figure 1 (aee) A 34-year-old woman with histopathologically proven hepatic adenomatosis due to inflammatory adenomas (I-HCA) with the
index mass in segment 7 (asterisk). (a) An axial T2-weighted image shows multifocal well-marginated masses (>10) of intermediate to high
signal intensity in both lobes of the liver. Signal loss in the background liver but not in the masses is noted on the opposed-phase image (c)
compared with the in-phase (b) T1-weighted image, denoting hepatic steatosis. Axial gadolinium-enhanced T1-weighted fat-saturated MRI
images show that the masses exhibit diffuse hyperenhancement on the arterial (d) and portal (e) phases. (feg) A 34-year-old woman with
histopathologically proven hepatic adenomatosis due inflammatory adenomas (I-HCA). (f) A coronal T2-weighted image shows multifocal well-
marginated masses (>10) of intermediate to high signal intensity in both lobes of the liver. (g) A coronal gadolinium-enhanced T1-weighted fat-
saturated MRI image shows that the masses exhibit diffuse hyperenhancement on the portal phase.
278 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285
adenomatosis is associated with a 63% risk of haemor-
rhage and 10% risk of malignant transformation into
HCC.22 Once viewed as a separate entity, it is now
considered as part of the spectrum of the same disease
process as HCA. Although Flejou et al.20 reported an equal
gender incidence, larger studies have found a female
predilection. HCAs in hepatic adenomatosis are histolog-
ically similar to HCAs that occur sporadically and may be
of I-HCA, HNF-1a or the b-catenin activated subtype. HNF-
1a HCAs with TFC1 gene mutations tend to develop fa-
milial adenomatosis, which is associated with maturity-
onset diabetes of the young (MODY3).23
Obesity/metabolic syndrome
Studies have shown an increased HCA incidence in obese
patients (particularly males)24 and those with metabolic
syndrome.25 The metabolic syndrome consists of a cluster
of factors (dyslipidaemia, insulin resistance or type 2 dia-
betes, hypertension, and obesity) associated with an
increased risk of cardiovascular disease, chronic liver dis-
ease, and hepatic neoplasms such as HCC and HCA.26,27
Miscellaneous
Other less common conditions associated with HCA
include pregnancy, Fanconi’s anaemia due to steroid ther-
apy, familial adenomatosis polyposis, beta thalassemia,
tyrosinaemia, and type 1 diabetes mellitus.24 HCA in preg-
nant women requires special consideration because
elevated levels of steroid hormones in pregnancy may
induce HCA growth and rupture.28
Clinical and biochemical features
Approximately 50% of cases are clinically occult and
detected incidentally on imaging.1 The overall risk of hae-
morrhage is 20e25% with HCA >5 cm, a subcapsular loca-
tion and longstanding OCP use being the main risk factors
for tumour rupture and haemorrhage (Fig. 2).29,30 Such
patients may present with a palpable mass, acute abdom-
inal pain, elevated liver enzymes, and hypovolaemic
Figure 2 A 27-year-old woman who presented to the emergency room with acute right upper quadrant pain. (a) Axial unenhanced CT image and
(b) arterial phase contrast-enhanced CT image show an 11 cm subcapsular mass that exhibits high-density haemorrhage (arrow in a) and
heterogeneous arterial enhancement (arrow in b). Histopathology from the surgical specimen confirmed that this was a ruptured I-HCA.
shock.31 Of the various subtypes, I-HCA has the highest risk
of haemorrhage and may show elevated levels of acute-
phase reactants, such as C-reactive protein and serum am-
yloid associated proteins, elevated gamma-glutamyl trans-
ferase, and/or signs of chronic anaemia.4 The exact
mechanism for malignant transformation into HCCs is not
well understood. Rapid increase in the size of a previously
stable HCA raises the possibility of malignant degeneration
or haemorrhage. A recent systematic review by Stoot et al.32
found that the incidence of malignant transformation was
approximately 4.2%. The b-catenin activated subtype has
the highest predilection for malignant transformation of all
HCAs.1,2
Histopathology
Microscopic
HCAs are monoclonal neoplasms that are composed
histologically of sheet- or cord-like arrangements of benign
hepatocytes that lack portal venules and bile ductules d a
key differentiating feature from focal nodular hyperplasia
(FNH).33 The individual sheets are separated by dilated si-
nusoids consisting of thin-walled capillaries that are solely
supplied by arterial feeding vessels as HCAs lack a portal
venous supply.7,34 Individual hepatocytes may contain
varying amounts of intracellular fat or glycogen. Kupffer
cells may be found but are typically of reduced number and
non-functioning. The extensive arterial supply in HCAs ac-
counts for its hypervascular nature, while poor connective
tissue support and the absence of a true capsule make these
tumours susceptible to haemorrhage.
Macroscopic
HCAs are usually well-defined unencapsulated lesions
that can range from 1 to 20 cm.1 The tumours may be sol-
itary or multifocal. HCAs have a characteristic yellow
appearance on cut specimen owing to the accumulation of
intracellular lipid. Large peritumoural or intratumoural
vessels and areas of central necrosis or haemorrhage may be
 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285
detected.35 Intratumoural calcification is rarely encoun-
tered. A pseudocapsule may be seen due to compressed
hepatic parenchyma around the tumour.
Histopathological subtypes and magnetic
resonance imaging characteristics
HCAs may show a range of imaging appearances, which
can be complicated by fat, haemorrhage, or malignant
transformation. Thus, achieving an accurate diagnosis can
sometimes be challenging. In 2006, the French collabora-
tive group introduced a phenotypeegenotype classifica-
tion, which categorised HCAs into several subtypes
according to the genetic and histopathological character-
istics of the tumours: (i) HNF-1a HCA, (ii) I-HCA, (iii) b-
catenin activated HCA, and (iv) unclassified HCAs.2,4 Since
the introduction of the new phenotypeegenotype classi-
fication, studies have shown that magnetic resonance
imaging (MRI) is the technique of choice for subtype
characterisation.3 Laumonier et al.3 demonstrated that
HNF-1a HCA and I-HCA have specific features on MRI
(Table 1).
I-HCA
Pathology
I-HCA is the most common subtype comprising
40e50% of HCAs and includes lesions formerly termed as
“telangiectatic” FNHs.4,36 This subtype is associated with
mutations in the interleukin 6 signal transducer gene
(IL6ST), which is located on chromosome 5q11 and en-
codes for glycoprotein 130.37,38 Mutations in glycoprotein
130 lead to sustained activation of the Janus kinase signal
transducer and activation of the transcription pathway,
which causes abnormal hepatocyte proliferation involved
in I-HCA pathogenesis. I-HCA most frequently occurs in
young women on the OCP, obese patients, and those with
metabolic syndrome.4 Patients with I-HCA may have signs
of chronic anaemia or “systemic inflammatory syndrome”
comprising fever, leukocytosis, and elevated serum C-
reactive protein. These patients may also show elevation
of serum gamma-glutamyl transferase, alkaline phospha-
tase, and serum amyloid-associated proteins, especially in
cases associated with intratumoural haemorrhage or
multiple HCAs.30,39 Of all subtypes, I-HCA has the highest
Table 1
Summary of hepatic adenoma (HCA) subtypes, frequencies, molecular findings and key magnetic resonance imaging (MRI) features.
HCA Subtype Frequency Molecular findings
Inflammatory 40e50% Interleukin 6 signal transducer gene (IL6ST)
(I-HCA) mutations in 10%
HNF-1a inactivated 30e35% Mutations in transcription factor one (TCF1) gene
b-Catenin 10e15% b-catenin activating gene mutations
activated
Unclassified <5% No specific gene mutations
HNF-1a, hepatocyte nuclear factor-1 alpha inactivated; HCC, hepatocellular carcinoma.
279
risk of bleeding, which occurs in up to 30% owing to the
presence of marked sinusoidal dilatation, abnormal thick-
walled arteries, and peliotic areas on histopathology.29
Malignant degeneration into HCC occurs in 5e10% of I-
HCAs due to a separate mutation involving the b-catenin
gene.1,30
Imaging
On MRI, I-HCA characteristically shows heterogeneous
high T2 signal that is more intense at the periphery of the
lesion (termed the “atoll” sign40) and arterial enhancement
that persists into the portal phase (Fig. 3). Both these im-
aging features correlate with dilated sinusoids on histopa-
thology and the imaging findings together are 85.2%
sensitive and 87.5% specific for I-HCA.1,3 Intratumoural
steatosis is rare and when present is usually focal or patchy
and heterogeneous, rather than diffuse (Fig. 4). The back-
ground liver may be steatotic. I-HCAs have a 30% incidence
of haemorrhage and a 5e10% incidence of b-catenin muta-
tions.29,41 Intratumoural haemorrhage often leads to
increased lesion heterogeneity. On MRI, acute haemorrhage
shows high T1 signal due to methaemoglobin, while chronic
haemorrhage shows low T1 and T2 signal due to haemosi-
derin. On computed tomography (CT), tumour bleed often
manifests as heterogeneous high-density foci within the
HCA and surrounding liver parenchyma while intraperito-
neal bleed appears as high-density free fluid that may show
a haematocrit effect.
HNF-1a HCA
Pathology
HNF-1a HCA is the second most common subtype
comprising 30e35% of lesions and is almost exclusively
found in women, most of whom (>90%) have a history of
OCP use.29 As previously mentioned, this subtype may be
associated with MODY3 and familial adenomatosis.29 Of all
subtypes, HNF-1a HCA has the most indolent biological
behaviour and the lowest risk of malignant trans-
formation.42 These tumours are often asymptomatic and
detected incidentally on imaging. HNF-1a HCA is associated
with mutations in the transcription factor one (TCF1) gene
located on chromosome 12q24.43 TCF1 is a tumour-
suppressor gene involved in hepatocyte differentiation,
and its inactivation promotes hepatocellular proliferation.
Key MRI features
Hyperintense on T2-weighted images
Arterial enhancement that persists on portal and delayed phases
Diffuse intralesional fat
Arterial enhancement
Mimics HCC showing arterial enhancement and portal
phase washout
No specific imaging features
280 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285

Figure 3 A 30-year-old woman with an incidentally detected inflammatory adenoma (I-HCA). (a) An axial T2-weighted MRI image and (b)
magnified image show a well-marginated ovoid tumour with a characteristic high T2 signal peripheral rim (arrow) termed the “atoll sign.” This
as it resembles a carol atoll. The mass is isointense on the T1-weighted in-phase MRI image (arrow in c) with no signal loss on the corresponding
opposed-phase MRI image (arrow in d). Gadolinium-enhanced axial T1-weighted fat-saturated MRI images obtained during the arterial (e) and
portal phase (f). The mass shows intense uniform arterial enhancement (arrow in e) which persists into the portal phase (arrow in f). A
percutaneous biopsy confirmed the diagnosis of I-HCA.

The gene also inactivates a liver fatty acid binding protein
(L-FABP) resulting in impaired fatty acid transport in he-
patocytes and excessive intralesional fat deposition.44 The
presence of diffuse intratumoural microscopic fat accounts
for its typical appearance on MRI.
Imaging
On MRI, HNF-1a HCA characteristically shows diffuse
intratumoural signal loss on chemical-shift T1-weighted
imaging (e.g., signal loss on opposed-phased images

Figure 4 A 30-year-old woman with an incidentally detected histopathologically proven I-HCA. (a) Axial T1-weighted in-phase MRI image, (b)
T1-weighted opposed-phase MRI image, and (c) axial T2-weighted MRI image show an ovoid mass (arrows) centrally within the liver. (b) On the
opposed-phase MRI image, the mass shows punctate internal foci of signal loss due to non-uniform steatosis within the tumour; while the
background liver shows diffuse signal loss from hepatic steatosis. (c) On the axial T2-weighted MRI image, the tumour shows heterogeneous
intermediate to high signal intensity.
 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285 281

compared to in-phase images) due to microscopic fat con- Unclassified

tent (Fig. 5)1,3; this finding alone is 86.7% sensitive and 100%
specific for this subtype.3 HNF-1a HCA is predominantly
hyper- or isointense on T1-weighted images and iso- or
hyperintense on T2-weighted images and is not associated
with significantly restricted diffusion. Following gadolin-
ium administration, HNF-1a HCA typically shows hyper-
enhancement in the arterial phase, which does not persist
into the portal phase.29 In addition, the background liver
often shows diffuse steatosis.
Unclassified comprises a small subset (approximately
10%) of HCAs that show adenoma-like morphology but lack
distinctive genetic and/or pathological abnormalities.46
Knowledge of these lesions is limited. On imaging, these
tumours have no specific MRI features.
Gadoxetic acid-enhanced MRI

b-Catenin-activated HCA Gadoxetic acid is a hepatocyte-specific contrast agent

Histopathology
b-catenin activated HCAs constitute 10e15% of all ade-
nomas; these tumours are caused by mutations in the b-
catenin gene (CTNNB1) located on chromosome 3p21,
which leads to sustained activation of a b-catenin protein
resulting in uncontrolled hepatocyte proliferation.45 This
subtype is frequently found in males and is associated with
GSDs, AAS, and familial adenomatosis polyposis.29 Identi-
fying the b-catenin mutation is of major interest because of
its high association with malignant transformation.1,3 On
histology, b-catenin-activated HCAs can be difficult to
distinguish from well-differentiated HCCs; often showing
cytological abnormalities such as nuclear atypia and a high
nuclear-to-cytoplasm ratio.
Imaging
b-catenin-activated HCAs do not have characteristic MRI
features. The lesions generally lack intratumoural fat. Ma-
lignant transformation is suspected if the tumours show
significant interval growth, local invasion, or contrast me-
dium washout on the portal or delayed phase (Fig. 6).
that is often used clinically to differentiate HCA from FNH.
Gadoxetic acid is taken up by functioning hepatocytes and
excreted by the biliary system resulting in high signal in
the liver parenchyma on the hepatobiliary phase at 20
minutes. Uptake of gadoxetic acid into hepatocytes is
mediated by a cell membrane transporter termed organic
anion transporting polypeptide (OATP), while cellular
excretion is mediated by other transporters such as
multidrug resistance-associated protein-2 and -3 (MRP2
and MRP3).5,42 A systemic review by McInnes et al. found
that gadoxetic acid-enhanced MRI was able to discrimi-
nate between HCA and FNH with 91e100% sensitivity and
87e100% specificity based on lesion signal intensity on
the hepatobiliary phase.42 On this phase, most FNHs show
iso- or high signal while most HCAs show low signal
compared with the background liver parenchyma (Fig. 7).
The higher signal in FNHs on the hepatobiliary phase is
due to higher expression of OATP transporters in FNHs
compared with HCAs, while increased expression of MRP3
in HCAs also contributes to greater contrast medium
washout in these lesions. The study did, however, find
that some HCAs (0e67%; Fig. 8), mainly I-HCA, can show
iso- or hyperintensity on the hepatobiliary phase. More-
over, Ba-Ssalamah et al.5 found significant differences in

Figure 5 A 28-year-old man with glycogen storage disorder and a histopathologically proven HNF-1a inactivated HCA. (a) An axial fat-
suppressed T2-weighted MRI image shows a well-marginated and encapsulated heterogeneous T2 hyperintense mass (arrow). The mass (ar-
rows) shows signal loss on the opposed-phase T1-weighted MRI image (c) compared with the T1-weighted in-phase MRI image (b) due to
diffuse intratumoural steatosis. The background liver also shows hepatic steatosis. (dee) Gadolinium-enhanced T1-weighted fat-saturated MRI
images obtained during the arterial (d) and portal phase (e) show that the tumour exhibits non-uniform arterial enhancement (arrow in d) and
portal phase contrast medium washout (arrow in e).
282 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285

Figure 6 A 40-year-old man with no known risk factors presented with non-specific abdominal pain. (a) An axial T2-weighted MRI image shows
a heterogeneous tumour (arrow) with high signal intensity and a solid and cystic internal consistency. The tumour (arrows) shows low signal on
the corresponding T1-weighted in-phase (b) and opposed-phase (c) MRI images with no signal loss between phases to suggest intratumoural
steatosis. Gadolinium-enhanced fat saturated T1-weighted MRI images obtained during the arterial (d) and portal phases (e) show that the
tumour has heterogeneous arterial enhancement (arrow in d) and portal phase washout (arrow in e). Histopathology from the surgical specimen
confirmed a diagnosis of b-catenin-activated HCA with malignant transformation into HCC.

Figure 7 A 37-year-old woman with a histopathologically proven HCA (HNF-1 a). (a) An axial T2-weighted image shows a mass (arrow) of
intermediate to high signal intensity. Diffuse signal loss in the mass (arrows) on the opposed-phase (c) image compared to the in-phase (b) T1-
weighted image confirms the presence of intratumoural microscopic fat content. On the axial post-gadoxetic acid-enhanced fat-saturated T1-
weighted MRI images, the mass (arrows) shows mild arterial hyperenhancement (d), portal phase washout (e), and absence of contrast me-
dium uptake compared to the background liver on the hepatobiliary phase (f) at 20 minutes.

Figure 8 A 21-year-old woman with a histopathologically proven HCA. (a) An axial T2-weighted image shows a mass (arrow) of intermediate to
high signal intensity. The mass is not well visualised on the in-phase (b) or opposed-phase (c) T1-weighted images. On the axial post-gadoxetic
acid-enhanced fat-saturated T1-weighted MRI images, the mass (arrows) shows heterogeneous arterial (d) and portal (e) hyperenhancement
and mild contrast uptake on the hepatobiliary phase (f) at 20 minutes.
 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285
the uptake of gadoxetic acid amongst HCA subtypes
depending on the varying expression of OATP and MRP3.
On the hepatobiliary phase at 20 minutes, all steatotic
(presumably HNF-1a, n¼10) and unclassified HCAs (n¼6)
showed low signal while six of 21 (39%) I-HCAs and six of
eight (83%) b- catenin activated HCAs showed iso or high
signal.
Differential diagnoses
Several benign and malignant lesions may simulate HCA.
These include FNH, HCC, angiomyolipoma, focal fat, and
hypervascular metastases.
FNH
FNHs are typically “stealth” lesions with near iso-
intensity compared with the background liver on T1- and
T2-weighted MRI images. Larger lesions may show a char-
acteristic T2 high signal central scar. FNHs usually show
arterial hypervascularity, while the central scar typically
shows delayed enhancement. Gadoxetic acid-enhanced
MRI is able to differentiate FNHs from HCAs in most but
not all cases.47 As previously mentioned, I-HCAs were pre-
viously classified as telangiectatic FNHs.1,2,46 These lesions
lack the histological features of regular FNHs such as the
presence of a central scar and a central artery. On imaging,
telangiectatic FNHs show findings atypical for FNHs, such as
tumour heterogeneity, microscopic fat, a pseudocapsule, or
a non-enhancing scar.36
HCC and fibrolamellar HCC
Distinguishing HCA from HCC in a non-cirrhotic liver
can be challenging. The presence of arterial enhancement
and intralesional fat (40% of HCCs) may be seen in either
tumour.8 Similar to HCC, HNF-1a and b-catenin-activated
HCA can show contrast medium washout or a delayed
enhancing capsule. Correlation with clinical history
including patient demographics and the presence of risk
factors for HCC may be useful in discriminating between
the aetiologies. Fibrolamellar HCC and HCAs typically
affect young adults, but the two can often be differentiated
based on imaging appearances. Fibrolamellar HCC often
presents as a large irregular mass that may exhibit a T2 low
signal central scar and internal calcifications, while rec-
ognised ancillary features include metastases and
lymphadenopathy.48
Hypervascular metastases
Hypervascular metastases from breast, renal, thyroid,
islet cell tumour, or melanoma can be difficult to distinguish
from HCA. Metastases are usually multiple and show T1
hypointensity and T2 intermediate hyperintensity on MRI.
Intralesional fat and haemorrhage are uncommon; how-
ever, haemorrhagic metastases (e.g., melanoma, renal) can
demonstrate T1 hyperintensity.49 Moreover, perilesional fat
deposition has been described in hepatic metastases from a
283
primary pancreatic insulinoma.50 A careful search for pri-
mary cancer should be performed and a detailed clinical
history obtained.
Focal steatosis
Hepatic steatosis is a common imaging finding. It is
usually diffuse or geographic in distribution, but can be
focal or multifocal and often mimics other hepatic lesions
on ultrasound and CT. A nodular type of steatosis, in
particular, can cause diagnostic confusion with HCA. MRI
can be used to distinguish between the two. Areas of focal
fat often occur in characteristic locations such as the gall-
bladder fossa, adjacent to the falciform ligament and the
porta hepatic, and do not exert mass effect on adjacent
blood vessels. Focal fat typically shows signal loss on
opposed-phase compared with in-phase MRI images and
behaves like normal liver tissue on the remaining MRI se-
quences. The absence of signal alteration on diffusion-
weighted imaging also supports the diagnosis of this
“pseudolesion”.
Hepatic angiomyolipoma
Hepatic angiomyolipoma (AML) is a rare benign tumour,
which often contains macroscopic fat. Histopathologically,
they are composed of vascular tissue, smooth muscle, and
varying amounts of mature fat. Hepatic AMLs typically
occur in the right lobe and almost always coexist with renal
AMLs.
Management
The new HCA classification system has led to a better
understanding of the natural history of these tumours and a
more tailored approach to patient management. Tradition-
ally, surgical resection was considered the treatment of
choice for HCA, with liver transplantation reserved for he-
patic adenomatosis. Current clinical management revolves
around the prevention and treatment of complications,
such as haemorrhage and malignant transformation. It is
now clear that the risk of such complications is influenced
by the histological subtype and tumour size. Management
also depends on whether the patient is symptomatic or
asymptomatic.
Imaging surveillance
In asymptomatic patients with a HNF-1a HCA <5 cm,
Bioulac-Sage et al.4 recommend clinical and imaging sur-
veillance as this subtype has a low risk of malignant
transformation and haemorrhage. Genetic counselling
should also be considered in patients with HNF-1a muta-
tions due to the association with MODY3 and hepatic ade-
nomatosis. In HCA patients on the OCP or AAS, watchful
waiting with repeat imaging at 3e6 months following
cessation of these drugs is recommended to determine if
the tumours regress.
284 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285
Percutaneous biopsy
HCAs that demonstrate growth despite discontinuation
of causative drugs should be biopsied to exclude malig-
nancy. Biopsy is also recommended in all non-HNF-1a HCAs
to identify the b-catenin gene mutation associated with a
high risk of malignancy.
Surgical resection
Surgical resection is recommended for HCAs >5 cm as
these tumours have a higher risk of haemorrhage and
malignant transformation.51 Surgical resection is also
advocated for HCAs that demonstrate growth despite
discontinuation of causative drugs, and HCAs occurring in
males and those with GSDs.30 The majority of tumours are
resected locally or with partial lobectomy.52 The reported
morbidity associated with elective resection is approxi-
mately 13%. Complications associated with emergency
surgery are significantly higher with a mortality rate of
approximately 5e8%.53
Hepatic artery embolisation
Symptomatic HCAs usually present with intratumoural
haemorrhage or rupture leading to haemoperitoneum. In
recent years, selective hepatic artery embolisation has
gained favour as the treatment of choice for ruptured HCAs
due to higher risk of morbidity and mortality associated
with emergency surgery.54 Newer and less invasive tech-
niques, such as transarterial chemoembolisation and radi-
ofrequency ablation, have also been adopted for the
treatment of symptomatic tumours <5 cm.55
Liver transplantation
Liver transplantation is rarely performed and is usually
reserved for symptomatic patients with hepatic adenoma-
tosis or cases with progressive liver failure or malignant
transformation.56
Conclusion
HCAs are rare benign tumours that can be associated
with serious complications such as haemorrhage and ma-
lignant transformation. The imaging appearances of HCA
are varied and are determined by the specific subtype. The
new classification system has led to a better understanding
of the natural history of these tumours and a refined
approach to the clinical management of these patients.
References
1. Shanbhogue AK, Prasad SR, Takahashi N, et al. Recent advances in cy-
togenetics and molecular biology of adult hepatocellular tumors: im-
plications for imaging and management. Radiology 2011;258:673e93.
2. Zucman-Rossi J, Jeannot E, Nhieu JT, et al. Genotypeephenotype corre-
lation in hepatocellular adenoma: new classification and relationship
with HCC. Hepatology 2006;43:515e24.
3. Laumonier H, Bioulac-Sage P, Laurent C, et al. Hepatocellular adenomas:
magnetic resonance imaging features as a function of molecular path-
ological classification. Hepatology 2008;48:808e18.
4. Bioulac-Sage P, Laumonier H, Couchy G, et al. Hepatocellular adenoma
management and phenotypic classification: the Bordeaux experience.
Hepatology 2009;50:481e9.
5. Ba-Ssalamah A, Antunes C, Feier D, et al. Morphologic and molecular
features of hepatocellular adenoma with gadoxetic acid-enhanced MR
imaging. Radiology 2015:142366.
6. Lin H, van den Esschert J, Liu C, et al. Systematic review of hepatocellular
adenoma in China and other regions. J Gastroenterol Hepatol
2011;26:28e35.
7. Grazioli L, Federle MP, Brancatelli G, et al. Hepatic adenomas: imaging
and pathologic findings. RadioGraphics 2001;21:877e92. discussion
892e874.
8. Paulson EK, McClellan JS, Washington K, et al. Hepatic adenoma: MR
characteristics and correlation with pathologic findings. AJR Am J
Roentgenology 1994;163:113e6.
9. Rooks JB, Ory HW, Ishak KG, et al. Epidemiology of hepatocellular ade-
noma. The role of oral contraceptive use. JAMA 1979;242:644e8.
10. Baum JK, Bookstein JJ, Holtz F, et al. Possible association between benign
hepatomas and oral contraceptives. Lancet 1973;2:926e9.
11. Edmondson HA, Reynolds TB, Henderson B, et al. Regression of liver cell
adenomas associated with oral contraceptives. Ann Intern Med
1977;86:180e2.
12. Nakao A, Sakagami K, Nakata Y, et al. Multiple hepatic adenomas caused
by long-term administration of androgenic steroids for aplastic anemia
in association with familial adenomatous polyposis. J Gastroenterol
2000;35:557e62.
13. Socas L, Zumbado M, Perez-Luzardo O, et al. Hepatocellular adenomas
associated with anabolic androgenic steroid abuse in bodybuilders: a
report of two cases and a review of the literature. Br J Sport Med
2005;39:e27.
14. Wilson JD, Griffin JE. The use and misuse of androgens. Metab Clin Exper
1980;29:1278e95.
15. Stimac D, Milic S, Dintinjana RD, et al. Androgenic/anabolic steroid-
induced toxic hepatitis. J Clin Gastroenterol 2002;35:350e2.
16. Gonzalez A, Canga F, Cardenas F, et al. An unusual case of hepatic ade-
noma in a male. J Clin Gastroenterol 1994;19:179e81.
17. Labrune P, Trioche P, Duvaltier I, et al. Hepatocellular adenomas in
glycogen storage disease type I and III: a series of 43 patients and review
of the literature. J Pediatr Gastroenterol Nutr 1997;24:276e9.
18. Volmar KE, Burchette JL, Creager AJ. Hepatic adenomatosis in glycogen
storage disease type Ia: report of a case with unusual histology. Arch
Pathol Lab Med 2003;127:e402e405.
19. Grazioli L, Federle MP, Ichikawa T, et al. Liver adenomatosis: clinical,
histopathologic, and imaging findings in 15 patients. Radiology
2000;216:395e402.
20. Flejou JF, Barge J, Menu Y, et al. Liver adenomatosis. An entity distinct
from liver adenoma? Gastroenterology 1985;89:1132e8.
21. Donato M, Hamidian Jahromi A, Andrade AI, et al. Hepatic adenomatosis:
a rare but important liver disease with severe clinical implications. Int
Surg 2015;100:903e7.
22. Greaves WO, Bhattacharya B. Hepatic adenomatosis. Arch Pathol Lab
Med 2008;132:1951e5.
23. Lewin M, Handra-Luca A, Arrive L, et al. Liver adenomatosis: classifica-
tion of MR imaging features and comparison with pathologic findings.
Radiology 2006;241:433e40.
24. Bunchorntavakul C, Bahirwani R, Drazek D, et al. Clinical features and
natural history of hepatocellular adenomas: the impact of obesity.
Aliment Pharmacol Ther 2011;34:664e74.
25. Farges O, Ferreira N, Dokmak S, et al. Changing trends in malignant
transformation of hepatocellular adenoma. Gut 2011;60:85e9.
26. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet
2005;365:1415e28.
27. Paradis V, Zalinski S, Chelbi E, et al. Hepatocellular carcinomas in pa-
tients with metabolic syndrome often develop without significant liver
fibrosis: a pathological analysis. Hepatology 2009;49:851e9.
28. Cobey FC, Salem RR. A review of liver masses in pregnancy and a pro-
posed algorithm for their diagnosis and management. Am J Surg
2004;187:181e91.
 H. Dharmana et al. / Clinical Radiology 72 (2017) 276e285
29. Katabathina VS, Menias CO, Shanbhogue AK, et al. Genetics and
imaging of hepatocellular adenomas: 2011 update. RadioGraphics
2011;31:1529e43.
30. Dokmak S, Paradis V, Vilgrain V, et al. A single-center surgical experi-
ence of 122 patients with single and multiple hepatocellular adenomas.
Gastroenterology 2009;137:1698e705.
31. Ribeiro Junior MA, Chaib E, Saad WA, et al. Surgical management of
spontaneous ruptured hepatocellular adenoma. Clinics (Sao Paulo)
2009;64:775e9.
32. Stoot JH, Coelen RJ, De Jong MC, et al. Malignant transformation of
hepatocellular adenomas into hepatocellular carcinomas: a system-
atic review including more than 1600 adenoma cases. HPB
2010;12:509e22.
33. Vilgrain V. Focal nodular hyperplasia. Eur J Radiol 2006;58:236e45.
34. Jenkins RL, Johnson LB, Lewis WD. Surgical approach to benign liver
tumors. Sem Liver Dis 1994;14:178e89.
35. Rubin RA, Mitchell DG. Evaluation of the solid hepatic mass. Med Clin N
Am 1996;80:907e28.
36. Paradis V, Benzekri A, Dargere D, et al. Telangiectatic focal nodular hy-
perplasia: a variant of hepatocellular adenoma. Gastroenterology
2004;126:1323e9.
37. Rebouissou S, Amessou M, Couchy G, et al. Frequent in-frame somatic
deletions activate gp130 in inflammatory hepatocellular tumours. Na-
ture 2009;457:200e4.
38. Rodriguez C, Grosgeorge J, Nguyen VC, et al. Human gp130 transducer
chain gene (IL6ST) is localized to chromosome band 5q11 and possesses
a pseudogene on chromosome band 17p11. Cytogenet Cell Genet
1995;70:64e7.
39. Paradis V, Champault A, Ronot M, et al. Telangiectatic adenoma: an
entity associated with increased body mass index and inflammation.
Hepatology 2007;46:140e6.
40. van Aalten SM, Thomeer MG, Terkivatan T, et al. Hepatocellular ade-
nomas: correlation of MR imaging findings with pathologic subtype
classification. Radiology 2011;261:172e81.
41. Chang CY, Hernandez-Prera JC, Roayaie S, et al. Changing epidemiology
of hepatocellular adenoma in the United States: review of the literature.
Int J Hepatol 2013;2013:604860.
42. Ronot M, Bahrami S, Calderaro J, et al. Hepatocellular adenomas: accu-
racy of magnetic resonance imaging and liver biopsy in subtype clas-
sification. Hepatology 2011;53:1182e91.
285
43. Bluteau O, Jeannot E, Bioulac-Sage P, et al. Bi-allelic inactivation of TCF1
in hepatic adenomas. Nat Genet 2002;32:312e5.
44. Rebouissou S, Imbeaud S, Balabaud C, et al. HNF1alpha inactivation
promotes lipogenesis in human hepatocellular adenoma independently
of SREBP-1 and carbohydrate-response element-binding protein
(ChREBP) activation. J Biol Chem 2007;282:14437e46.
45. Chen YW, Jeng YM, Yeh SH, et al. P53 gene and Wnt signaling in benign
neoplasms: beta-catenin mutations in hepatic adenoma but not in focal
nodular hyperplasia. Hepatology 2002;36:927e35.
46. Dhingra S, Fiel MI. Update on the new classification of hepatic ade-
nomas: clinical, molecular, and pathologic characteristics. Arch Pathol
Lab Med 2014;138:1090e7.
47. Mohajer K, Frydrychowicz A, Robbins JB, et al. Characterization of he-
patic adenoma and focal nodular hyperplasia with gadoxetic acid. J
Magn Reson Imaging 2012;36:686e96.
48. Ganeshan D, Szklaruk J, Kundra V, et al. Imaging features of fibrolamellar
hepatocellular carcinoma. AJR Am J Roentgenol 2014;202:544e52.
49. Kelekis NL, Semelka RC, Woosley JT. Malignant lesions of the liver with
high signal intensity on T1-weighted MR images. J Magn Reson Imaging
1996;6:291e4.
50. Semelka RC, Custodio CM, Cem Balci N, et al. Neuroendocrine tumors of
the pancreas: spectrum of appearances on MRI. J Magn Reson Imaging
2000;11:141e8.
51. Shanbhogue A, Shah SN, Zaheer A, et al. Hepatocellular adenomas:
current update on genetics, taxonomy, and management. J Comput Assist
Tomogr 2011;35:159e66.
52. Herman P, Coelho FF, Perini MV, et al. Hepatocellular adenoma: an
excellent indication for laparoscopic liver resection. HPB
2012;14:390e5.
53. Terkivatan T, de Wilt JH, de Man RA, et al. Indications and long-term
outcome of treatment for benign hepatic tumors: a critical appraisal.
Arch Surg 2001;136:1033e8.
54. Erdogan D, Busch OR, van Delden OM, et al. Management of sponta-
neous haemorrhage and rupture of hepatocellular adenomas. A single
centre experience. Liver Int 2006;26:433e8.
55. Kim YI, Chung JW, Park JH. Feasibility of transcatheter arterial chemo-
embolization for hepatic adenoma. J Vasc Interv Radiol 2007;18:862e7.
56. Chiche L, Dao T, Salame E, et al. Liver adenomatosis: reappraisal, diag-
nosis, and surgical management: eight new cases and review of the
literature. Ann Surg 2000;231:74e81.