Gastrointestinal Imaging • Original Research

Lewis et al.
HCC in Patients With HCV Without Fibrosis or Cirrhosis

Gastrointestinal Imaging
Original Research
Downloaded from www.ajronline.org by 182.253.105.42 on 02/13/18 from IP address 182.253.105.42. Copyright ARRS. For personal use only; all rights reserved

Hepatocellular Carcinoma in
Chronic Hepatitis C in the Absence
of Advanced Fibrosis or Cirrhosis
Sara Lewis1 OBJECTIVE. The objective of our study was to describe the cross-sectional imaging
Sasan Roayaie2 appearance of hepatocellular carcinoma (HCC) in patients with chronic hepatitis C virus
Stephen C. Ward 3 (HCV) infection in the absence of advanced fibrosis and cirrhosis.
Inna Shyknevsky 1 MATERIALS AND METHODS. This study is a retrospective review of our surgical
Ghalib Jibara2 database to identify patients with chronic HCV infection and HCC who underwent hepatec-
tomy and who had undergone preoperative CT or MRI. Only patients with a Metavir fibro-
Bachir Taouli1
sis score of F0, F1, or F2 on pathology were included. Patients with hepatitis B virus coin-
Lewis S, Roayaie S, Ward SC, Shyknevsky I, Jibara fection or other causes of chronic liver disease and patients with histopathologic evidence of
G, Taouli B advanced fibrosis or cirrhosis (Metavir scores F3 and F4) were excluded. Contrast-enhanced
CT or MRI examinations performed within 2 months before surgery were reviewed for the
number, size, and location of tumors; tumor enhancement characteristics; and presence of
macrovascular invasion.
RESULTS. Two hundred forty-five resections of HCC in patients with HCV were per-
formed in our institution from 1987 to 2012. Of this group, 26 patients (10.6%) had a Metavir
fibrosis score of F0, F1, or F2; of those patients, 19 (18 men and one woman; 18 non-Asian
patients and one Asian patient; mean age, 64 years) had imaging studies available for review.
Twenty-one HCCs (mean size, 4.5 cm; range, 0.9–14.8 cm) were evaluated at imaging. Typi-
cal wash-in and washout characteristics were seen in 16 of 19 viable lesions (84.2%). The re-
maining two HCCs were completely necrotic after transarterial chemoembolization. Eighteen
patients had a solitary tumor. Most tumors (15/21, 71.4%) developed in the right hepatic lobe.
CONCLUSION. HCC can develop in patients with chronic HCV without advanced fi-
Keywords: chronic viral hepatitis C, cirrhosis, fibrosis, brosis or cirrhosis, most frequently in older non-Asian men, and usually appears as a large
hepatocellular carcinoma solitary tumor with a typical wash-in–washout enhancement pattern.
DOI:10.2214/AJR.12.9151

Received April 24, 2012; accepted after revision
October 16, 2012.
1
Department of Radiology/Body MRI and Translational
and Molecular Imaging Institute, Ichan School of
Medicine at Mount Sinai, One Gustave Levy Pl, Box 1234,
New York, NY 10029. Address correspondence to
B. Taouli (bachir.taouli@mountsinai.org).
2 tion, and the current guidelines proposed by
Mount Sinai Liver Cancer Program, Ichan School of
Medicine at Mount Sinai, New York, NY.
3
Department of Pathology, Ichan School of Medicine at
Mount Sinai, New York, NY.
WEB
This is a Web exclusive article.
AJR 2013; 200:W610–W616
0361–803X/13/2006–W610
© American Roentgen Ray Society
C
hronic hepatitis C virus (HCV)
infection is the leading indica-
tion for liver transplantation and
the leading cause of death due to
liver disease in the United States [1]. Cirrhosis
is believed to be an underlying risk factor for
the development of hepatocellular carcinoma
(HCC) in patients with chronic HCV infec-
the American Association for the Study of
Liver Diseases (AASLD) recommend routine
HCC screening for HCV patients only once
the presence of cirrhosis has been established
[2]. However, clinical and pathologic reports
note that up to 20% of HCCs may develop in
noncirrhotic liver in patients with HCV and
that the exact mechanism of hepatocarcino-
genesis remains unclear in these cases [3, 4].
Several published surgical series have report-
ed the outcomes of resection for HCC in non-
cirrhotic patients, some of which include HCV
patients [5–13]. Knowledge about the imaging
appearance of HCC developing in noncirrhot-
ic liver is, however, limited [14, 15], and to
our knowledge there is no published study that
specifically focuses on the imaging appear-
ance of HCC in noncirrhotic HCV patients.
The purpose of our study was to describe the
cross-sectional imaging appearance of HCC
with histopathologic correlation in HCV pa-
tients without advanced fibrosis or cirrhosis.
Materials and Methods
Institutional review board approval was ob-
tained for this retrospective study, and patient in-
formed consent was not required. A retrospective
review of the hepatobiliary surgical and patho-
logic database in a single tertiary care center was
performed to identify patients with chronic HCV
infection and HCC who underwent liver resec-

W610 AJR:200, June 2013

 HCC in Patients With HCV Without Fibrosis or Cirrhosis
tion from January 1987 to May 2012. The inclu-
sion criteria for resection at our institution require
Child class A liver function, a single tumor on im-
aging without evidence of extrahepatic spread,
no clinical evidence of portal hypertension, and
a platelet count of 100,000/µL or greater. The
Downloaded from www.ajronline.org by 182.253.105.42 on 02/13/18 from IP address 182.253.105.42. Copyright ARRS. For personal use only; all rights reserved
presence of gross vascular invasion was not con-
sidered a contraindication to surgery. Additional
inclusion criteria for this study were as follows:
adult patients with HCC and chronic HCV infec-
tion who underwent hepatectomy and had a histo-
pathologic Metavir fibrosis score of F0¸–F2 and
had preoperative contrast-enhanced CT or MRI
within 2 months before surgery.
The exclusion criteria were chronic hepatitis B
virus (HBV) coinfection based on serologic mark-
ers; other causes of chronic liver disease such as, but
not limited to, alcoholic liver disease or nonalcohol-
ic steatohepatitis; and histopathologic evidence of
advanced fibrosis or cirrhosis (i.e., Metavir scores
F3 and F4). The clinical practice of the surgical on-
cologists at our institution is to perform segmen-
tal hepatic resection in patients with HCC with or
without cirrhosis and preserved liver function, so no
transplant cases were included in this series. Liver
biopsy is not routinely performed for patients who
undergo locoregional treatment of HCC, so histo-
pathologic correlation cannot be performed in these
cases; therefore, these cases were excluded.
A total of 245 liver resections for HCC were
performed at our institution from January 1987 to
May 2012. Of these, 40 and 179 patients with a
score of F3 and F4, respectively, were excluded.
Patients with a score of F3 were excluded from the
study to avoid potentially including cirrhosis due
to sampling error or observer variability [16, 17].
Patients with coexistent HBV based on serolog-
ic markers were excluded, and patients with coex-
istent alcohol abuse according to clinical history
were excluded [18]. Histopathologic results were
used to exclude patients with chronic liver disease
from other causes. Laboratory values (liver func-
tion test results and α-fetoprotein levels) were re-
corded at the time of surgery when available.
Twenty-six patients with HCV and a histopath-
ologic Metavir score of F0, F1, or F2 were identi-
fied (10.6% of total resections of HCV). Informa-
tion about this cohort was published in a separate
recent article [13] written by the surgical group at
our institution, but they did not specifically report
imaging findings. Seven patients were excluded
because an imaging study or pathologic specimen
obtained at an outside institution was not available
for review. Our final study population consisted of
19 patients with HCV and HCC (18 men and one
woman) with a mean age of 64 years (range, 51–75
years) who were evaluated from August 2004 to
May 2012. The distribution of patients by ethnic-
AJR:200, June 2013 W611
ity is as follows: African American (n = 10), white
(n = 7), Hispanic (n = 1), and Asian (n = 1).
Pathologic Assessment
The pathologic specimens were retrospectively
reviewed by a pathologist with 5 years of experience
in liver pathology for tumor size, location, and dif-
ferentiation [19] and for the presence of micro- and
macrovascular invasion. Background liver fibrosis
score and grade were assessed using the Metavir
staging system [20]. All histologic specimens were
deemed sufficient to confirm or exclude the pres-
ence of cirrhosis.
Imaging Technique
All patients underwent contrast-enhanced MDCT
(n = 14) or MRI (n = 5) examinations. The mean
delay between imaging and surgery was 19 days
(range, 5–65 days).
CT technique—CT scans were obtained with
either a 16- or 64-MDCT scanner (LightSpeed
Ultra or HiSpeed [GE Healthcare]; or Defini-
tion AS or Emotion 16 [Siemens Healthcare]).
CT examinations included unenhanced and
contrast-enhanced imaging through the liver, in-
cluding hepatic arterial phase and portal venous
phase imaging (60–70 seconds), after the initia-
tion of a bolus of IV contrast material (100 mL
of iopamidol [Isovue 370, Bracco Diagnostics]).
A bolus-tracking technique was used to acquire
arterial phase images. The CT parameters were
120–130 kVp, 170–240 mAs, a 5-mm collima-
tion, and a 1.5-mm reconstruction interval.
MRI technique—MRI was performed at 1.5
T (n = 3) or 3 T (n = 2) (Signa or Discovery
750, GE Healthcare; or Magnetom Avanto,
Siemens Healthcare). The following sequences
were performed: coronal and axial T2-weight-
ed HASTE, axial fat-suppressed fast spin-echo
T2-weighted, T1-weighted gradient-recalled
echo (GRE) in- and opposed phase, and ax-
ial dynamic contrast-enhanced 3D GRE T1-
weighted sequences (liver acceleration volume
acquisition [LAVA] or volumetric interpolated
breath-hold examination [VIBE]) before and
after gadolinium contrast administration. Gad-
opentetate dimeglumine ([Magnevist, Bay-
er HealthCare], n = 3) or gadobenate dimeg-
lumine ([MultiHance, Bracco Diagnostics],
n = 1) was administered at a standard dose of
0.1 mmol/kg or gadoxetate disodium ([Eovist,
Bayer HealthCare], n = 1) was administered at
a fixed dose of 10 mL. The arterial phase delay
was determined using a test bolus with 1 mL of
contrast material. Portal venous and equilibri-
um phase images were obtained at 1 and 3 min-
utes after contrast injection, respectively. Diffu-
sion-weighted imaging (DWI) was performed
using breath-hold single-shot echo-planar im-
aging sequences of three patients who under-
went imaging after 2009.
Image Analysis
CT and MR images were reviewed on a
PACS (Centricity PACS, GE Healthcare) by
two observers with 9 and 2 years of experi-
ence in abdominal imaging in consensus. The
observers were aware of the diagnosis of HCC.
Images were reviewed for the following: num-
ber of lesions; lesion size; lesion location; lesion
enhancement characteristics; and presence of
macrovascular invasion, which was defined as
invasion involving the portal vein and hepatic
vein branches. The lesions were measured us-
ing the largest single measurement on axial con-
trast-enhanced arterial phase images. The le-
sions with well-circumscribed margins were
described as discrete, and poorly delineated le-
sions were described as infiltrative. The dynam-
ic enhancement characteristics of the lesions on
contrast-enhanced imaging were categorized
as follows: typical wash-in (during the arte-
rial phase) and washout (during the portal ve-
nous phase, equilibrium phase, or both phases),
wash-in only, and hypovascular. The presence
of tumor necrosis was noted. The T1, T2, and
DWI characteristics of the lesions in patients
who underwent MRI were recorded when avail-
able. The presence of intravoxel fat, as deter-
mined by signal loss on T1 out-of-phase images
relative to T1 in-phase images, was also noted.
Results
Clinical and Laboratory Findings
The clinical and laboratory findings are
shown in Table 1. Serologic evidence of chron-
ic HCV was confirmed in all patients. Two
patients had HIV coinfection. The clinical
circumstances of the HCC diagnoses were
as follows: elevated α-fetoprotein (n = 9), clini-
cal symptoms (abdominal pain, n = 6; or pal-
pable mass, n = 1), incidental detection during
imaging for suspected pancreatitis (n = 1) dur-
ing abdominal ultrasound (n = 1), or inciden-
tal detection at routine follow-up imaging for
colorectal cancer (n = 1).
Laboratory analysis revealed normal liv-
er function test results as follows: aspartate
aminotransferase (n = 8), alanine aminotrans-
ferase (n = 10), alkaline phosphatase (n =
12), γ-glutamyltransferase (n = 3), and to-
tal bilirubin (n = 18). Six patients had normal
α-fetoprotein values (< 9 ng/mL), whereas 10
had mildly to markedly elevated α-fetoprotein
values. For three patients, α-fetoprotein values
were not available for review.
 Lewis et al.

TABLE 1: Clinical, Pathologic, and Imaging Findings in 19 Patients With Chronic Hepatitis C Virus Infection Without
Advanced Fibrosis and Cirrhosis and Hepatocellular Carcinoma
Patient Tumor Invasiona
Metavir
Age AFP Fibrosis Inflammation
No. (y) Sex Ethnicity (ng/mL) Score Grade No. Sizeb (cm) Grade Micro Macro Enhancement Characteristics
Downloaded from www.ajronline.org by 182.253.105.42 on 02/13/18 from IP address 182.253.105.42. Copyright ARRS. For personal use only; all rights reserved

1 67 M AA 13,625 F2 A2 1 8.0 PD + + Infiltrative, mildly hypovascular

2 55 M AA 107.6 F2 A2 1 2.9 PD + + WI-WO
3 65 M AA 86,136 F2 A1 1 14.8 PD + – WI-WO
4 66 M White 80.7 F2 A2 3 4.4, 2.7, 2.0 MD, N, N +, −, − −, −, − WI-WO, N, N
5 58 M AA NA F2 A2 1 6.6 MD + – WI-WO
6 66 M AA 23.8 F2 A2 1 4.1 MD + – WI-WO
7 55 M White NA F1 A2 1 2.6 MD + – WI-WO
8 75 F White 4 F0 A0 1 3.8 MD + + WI-WO
9 61 M AA 9.8 F2 A2 1 4.6 MD + – WI-WO
10 61 M Hispanic 21.9 F2 A2 1 2.6 WD + + WI-WO
11 73 M White NA F2 A2 1 6.5 MD + – WI-WO
12 63 M AA 13.2 F2 A2 1 1.3 MD – – WI, hypervascular on PV phase
13 61 M AA 3.7 F1 A1 1 4.6 PD + – WI-WO
14 67 M White 2.8 F2 A2 1 5.3 WD – – WI-WO
15 64 M Asian 29 F2 A2 1 2.6 MD + – WI-WO
16 51 M White 6.7 F2 A2 1 3.6 PD + + WI-WO
17 67 M AA 11.3 F2 A2 1 6.2 MD + – WI-WO
18 72 M AA 3.2 F1 A2 1 4.6 PD + – WI-WO
19 70 M White 4.4 F2 A2 1 0.9 WD – – Hypovascular
Note—AFP = α-fetoprotein, Micro = microvascular, Macro = macrovascular, AA = African American, PD = poorly differentiated, plus sign (+) = present, WI = wash-in,
WO = washout, MD = moderately differentiated, N = necrotic, minus sign (–) = absent, NA = data not available, PV = portal venous, WD = well differentiated.
a At pathology.
bTumor size determined at imaging.

Pathologic Findings
Twenty-one HCCs were diagnosed on par-
tial hepatectomy specimens. Mean tumor size
at pathology was 4.8 cm (range, 1.3–15.9 cm).
Tumors were well differentiated (n = 3, 14.3%),
moderately differentiated (n = 10, 47.6%), or
poorly differentiated (n = 6, 28.6%). Differ-
entiation could not be assessed for two com-
pletely necrotic lesions that had undergone
presurgical transarterial chemoembolization
(TACE). Microvascular invasion was ob-
served in 16 of 21 (76.2%) lesions, whereas
macrovascular invasion was observed in five
of 21 (23.8%) lesions. Pathologic assessment
of the background liver showed Metavir fi-
brosis scores of F0 (n = 1, 5.3%), F1 (n = 3,
15.8%), and F2 (n = 15, 78.9%) and inflam-
mation grades of A0 (n = 1, 5.3%), A1 (n = 2,
10.5%), and A2 (n = 16, 84.2%).
Imaging Findings
All 21 HCCs diagnosed at pathology were
identified on imaging. Tumors had an av-
erage size of 4.5 cm (range, 0.9–14.8 cm).
HCCs were solitary in 18 patients (94.7%).
Fifteen of 21 (71.4%) HCCs developed in the
right hepatic lobe. Sixteen of 19 (84.2%) vi-
able HCCs showed a typical wash-in–wash-
out enhancement pattern (Figs. 1 and 2). One
patient (patient 4 in Table 1) who underwent
CT 10 days after TACE had three HCCs
(measuring 2.0, 2.7, and 4.4 cm). The larg-
est lesion showed a typical wash-in–washout
enhancement pattern and the two smaller le-
sions did not enhance. The larger viable le-
sion revealed moderately differentiated HCC
with microvascular invasion and the two
smaller lesions were completely necrotic at
pathology. Review of CT performed 27 days
before TACE revealed the typical wash-in–
washout enhancement pattern in all three le-
sions. Approximately 16% of viable lesions
(3/19) did not show the typical imaging fea-
tures of HCC in our series (excluding the two
lesions that were completely necrotic after
TACE). The three atypical HCCs were vari-
able in size (range, 0.9–8.0 cm) and had the
following enhancement patterns: infiltrative
and hypovascular in patient 1 (Fig. 3); hyper-
vascular at the arterial phase and remaining
hyperdense at the portal venous phase in pa-
tient 12; and hypovascular in patient 19.
Five HCCs were identified in the five pa-
tients who underwent MRI. All lesions de-
tected on MRI showed hypointensity on un-
enhanced T1-weighted imaging, and four
lesions were hyperintense and one lesion was
isointense on T2-weighted imaging. Two of
three lesions showed diffusion restriction
when the DWI sequence was included. One
lesion (0.9 cm) was not identified on DWI.
Two of five HCCs showed intravoxel fat on
in-phase and opposed-phase imaging.
The margins of the tumor were well defined
in 20 lesions and poorly defined (infiltrative)
in one lesion. Macroscopic vascular invasion
was detected at imaging in two of five patients
with macrovascular invasion diagnosed at
histopathology: left portal vein invasion was
present in one patient with an 8.0-cm lesion in
the left hepatic lobe (Fig. 3) and right anterior
portal vein invasion was detected in the other

W612 AJR:200, June 2013

 HCC in Patients With HCV Without Fibrosis or Cirrhosis
Downloaded from www.ajronline.org by 182.253.105.42 on 02/13/18 from IP address 182.253.105.42. Copyright ARRS. For personal use only; all rights reserved
patient. Macroscopic vascular invasion detect-
ed at histopathologic analysis in the three re-
maining patients was not detected on imaging.
One patient (patient 3 in Table 1) present-
ing with a palpable mass underwent biopsy
at an outside institution. MRI performed after
the biopsy revealed a 15.9-cm partially necrot-
AJR:200, June 2013 W613
ic HCC exophytic arising from the left later-
al hepatic lobe with surrounding subcapsular
hemorrhage (Fig. 4). A partially necrotic and
ruptured left hepatic lobe HCC with intraperi-
toneal hemorrhage was confirmed at surgery.
Of the 16 of 21 HCCs with the typical im-
aging appearance, most lesions showed moder-
Fig. 1—61-year-old man (patient 9 in Table 1) with
chronic hepatitis C virus infection and hepatocellular
carcinoma (HCC) without cirrhosis.
A and B, Axial contrast-enhanced CT images acquired
at arterial phase (A) and portal venous phase (B) show
4.6-cm hypervascular lesion (arrows) in segments V
and VI with washout during portal venous phase (B) in
noncirrhotic liver.
C and D, Photomicrographs (H and E, ×40 [C] and
×100 [D]) of histopathologic specimen reveal
findings consistent with Metavir score of F2 in liver
parenchyma (C) and 4.5-cm moderately differentiated
HCC (D) with microvascular invasion.
Fig. 2—67-year-old man (patient 14 in Table 1) with
chronic hepatitis C virus infection and hepatocellular
carcinoma (HCC) without cirrhosis.
A and B, Axial contrast-enhanced CT images show
5.3-cm HCC (solid arrows) in segment VIII with
enhancement in arterial phase (A) and washout
in portal venous phase (B) in noncirrhotic liver.
Incidentally noted is 3-cm hemangioma (dashed
arrows) in left lateral segment.
C, Photomicrograph (trichrome stain, ×100) of
histopathologic specimen reveals Metavir score of F2
in liver parenchyma.
D, Photomicrograph (H and E, ×100) shows well-
differentiated HCC.
ate or poor differentiation. Of the three viable
lesions with atypical enhancement patterns,
one HCC was well differentiated, one HCC
was moderately differentiated, and one HCC
was poorly differentiated.
Discussion
In this study, we described the CT and
MRI appearances of HCC developing in a
predominantly non-Asian older male patient
population with HCV but without advanced
fibrosis or cirrhosis. We found that the most
common CT or MRI appearance of HCC
arising in this setting is a solitary large tu-
mor with a typical wash-in–washout pattern.
The current AASLD 2011 practice guide-
lines [2] suggest that screening for HCC is
most cost-effective in HCV patients when the
 Lewis et al.

risk of HCC is greater than 1.5% and that dy-
namic imaging should be used only once the
presence of cirrhosis has been established.
Furthermore, according to the Hepatitis C An-
Downloaded from www.ajronline.org by 182.253.105.42 on 02/13/18 from IP address 182.253.105.42. Copyright ARRS. For personal use only; all rights reserved
tiviral Long-Term Treatment Against Cirrho-
sis Study [21], the 5-year risk for noncirrhotic
HCV patients to develop HCC is 4.8%, which
falls below the 1.5%/y incidence screening
recommendation [2]. Thus, based on these
guidelines, HCC screening is not routinely
recommended for noncirrhotic patients with
HCV. Despite the AASLD recommendations,
it is extremely important to note that previous
work has shown that up to 20% of all HCCs

occur in noncirrhotic liver. HCC occurs more

commonly in noncirrhotic patients with HCV
compared with the background population due
to chance [3, 22, 23]. At our institution, we dis-
covered a prevalence of 10.6% (26/245 cases)
of HCC going on to resection that had devel-
oped in patients with chronic HCV infection
without histopathologic evidence of advanced
fibrosis or cirrhosis [13]. Review of serologic
results and history excluded concomitant risk
factors such as HBV or alcohol.
There are several published surgical se-
Fig. 3—67-year-old man (patient 1 in Table 1) with chronic hepatitis C virus infection and hepatocellular ries reporting the outcomes of resection for
carcinoma (HCC) without cirrhosis. HCC in noncirrhotic patients, some of which
A and B, Axial contrast-enhanced CT images show normal hepatic morphology and infiltrative HCC (long include HCV patients [5–13]. Some of these
arrows) in left hepatic lobe with mild enhancement in arterial phase (A), washout in portal venous phase (B),
and enhancing tumor thrombus in left portal vein (short arrow). Pathologic results confirmed diagnosis of series also reported male predominance in
infiltrative HCC with both macroscopic and microscopic invasion and Metavir fibrosis score of F2. the noncirrhotic population with HCC [7–10,

Fig. 4—65-year-old man (patient 3 in Table 1) with chronic hepatitis C virus infection and ruptured hepatocellular carcinoma (HCC) without cirrhosis assessed with 3-T
MRI.
A, Axial T2-weighted image shows noncirrhotic liver with large (14.8 cm) exophytic hyperintense mass arising from lateral left hepatic lobe (white arrow) with
surrounding hemorrhage (black arrows).
B, Axial unenhanced 3D gradient-recalled echo (GRE) T1-weighted image shows hyperintense signal corresponding to hemorrhage (black arrows) surrounding lesion
(white arrow).
C, Axial contrast-enhanced 3D GRE T1-weighted image obtained at arterial phase shows that tumor (arrow) is hypervascular.
D, Contrast-enhanced 3D T1-weighted image obtained at portal venous phase shows lesion (arrow) washout with central necrosis and pseudocapsule.
E, Axial diffusion-weighted image (b = 500 s/mm2) shows tumor hyperintensity. Partially necrotic and ruptured left hepatic lobe HCC (arrow) with intraperitoneal
hemorrhage was confirmed at surgery. Background liver shows findings consistent with Metavir score of F2 at pathology.
F, Photograph of surgical specimen shows tan mass with surrounding hemorrhage (arrow).

W614 AJR:200, June 2013

 HCC in Patients With HCV Without Fibrosis or Cirrhosis
12]. In a recent pathologic series, Yeh et al.
[23] described 18 HCCs developing in non-
cirrhotic HCV patients and reported that the
frequency of HCC in noncirrhotic HCV pa-
tients was statistically significantly higher
(16%) than the frequency of HCC develop-
Downloaded from www.ajronline.org by 182.253.105.42 on 02/13/18 from IP address 182.253.105.42. Copyright ARRS. For personal use only; all rights reserved
ing in the general population due to chance
(1.6%). In another pathologic series, De Mi-
tri et al. [24] assessed 19 patients with non-
cirrhotic HCV and HCC. They found that all
patients had a negative result for HBV sur-
face antigen and detected the HBV genome
in seven patients. All lesions were typical
HCCs with the exception of one fibrolamel-
lar HCC. Histopathologic analysis revealed
the presence and replication of HCV ge-
nomes in HCC, with genotype 1b being the
most prevalent. This finding suggests a possi-
ble role for HCV in the development of HCC.
The histologic features of the HCCs arising
in noncirrhotic HCV infection showed usu-
al and nondistinguishing features in both our
study and prior work.
There are limited published data on the
imaging appearance of HCC in noncirrhot-
ic liver [14, 15, 25]. Our findings are similar
to those of prior studies describing the im-
aging appearance of HCC in noncirrhotic pa-
tients with a variety of underlying risk fac-
tors. In a case control study, Winston et al.
[14] described the imaging appearance of
HCC in noncirrhotic and cirrhotic livers in
36 patients. However, none of the noncirrhotic
patients had HCV. They found that HCC le-
sions in noncirrhotic liver were significantly
larger than HCCs in cirrhotic liver (mean, 9.9
vs 2.2 cm, respectively) and were more fre-
quently solitary (72% vs 27%, p < 0.5) with a
central scar more frequently present (50% vs
6%, p < 0.05). The enhancement characteris-
tics were not assessed. The CT appearance of
HCC in 39 noncirrhotic patients was described
by Brancatelli et al. [15]. Risk factors for the
development of liver disease were detected in
15 patients and only two of those patients had
HCV infection. The typical HCC was solitary
or dominant (n = 32), was large (range, 2–23
cm), and occurred in the right hepatic lobe
(n = 21). Tumors were well (n = 6), moder-
ately (n = 32), or poorly (n = 1) differentiat-
ed. Imaging showed arterial phase enhance-
ment in 97% (n = 38) and venous washout in
90% (n = 35). Both Winston et al. and Bran-
catelli et al. reviewed histopathologic data for
the presence or absence of cirrhosis with no
mention of fibrosis score. A case series of six
patients with noncirrhotic chronic HCV and
HCC was also recently described by Nash et
AJR:200, June 2013 W615
al. [25]. Similar to our study, Nash et al. had
a predominantly male population (five men
and one woman). Five patients (83.3%) pre-
sented with a single lesion (mean size, 3.7
cm; range, 1.7–6.5 cm). The CT appearance
in four patients revealed three hypervascular
HCCs. Fibrosis was mild (n = 1), moderate
(n = 4), or bridging (n = 1).
The mechanism of HCC carcinogenesis in
noncirrhotic HCV patients remains to be elu-
cidated given that prior works in the surgi-
cal and pathologic literature have described
a higher-than-expected tumor incidence. The
sequence of carcinogenesis in chronic HCV
infection is classically described as occurring
in the cirrhotic liver as follows: fibrosis and
cirrhosis, development of regenerative nod-
ules and dysplastic nodules, and subsequent
development of HCC when cirrhosis is pres-
ent [26]. HCV is not historically believed to
be oncogenic; however, a few studies have
suggested the possibility that HCV viral pro-
teins have direct oncogenic properties [24,
27, 28]. HCV-induced cycles of chronic in-
flammation, necrosis, and regeneration and
the ensuing oxidative DNA damage may pre-
dispose to the development of HCC. It is also
possible that cirrhosis may have regressed in
noncirrhotic patients with HCC [23]. Anoth-
er consideration is the possible contribution
of occult HBV infection contributing to the
increased risk of developing HCC.
Our study has several limitations includ-
ing its retrospective and descriptive nature and
small sample size. The imaging methods were
therefore not homogeneous, resulting in a mix
of CT and MRI studies. We did not compare
imaging appearance of our study group with
a control group of HCV patients with cirrho-
sis. Last, although we excluded patients with
serologic evidence of HBV infection, patients
with clinical evidence of concomitant alco-
hol intoxication, and patients with histopath-
ologic evidence of other causes of chronic liv-
er disease, we did not formally exclude occult
HBV infection by assessing for HBV genome
in pathologic specimens. However, coexistent
HBV was considered unlikely especially be-
cause our population consisted primarily of
non-Asian patients (18/19).
In conclusion, our results have shown that
HCC can arise in patients with chronic HCV
without underlying advanced liver fibrosis and
cirrhosis; the affected patients are predominant-
ly older non-Asian men and tumors frequently
appear on imaging as large solitary lesions with
a typical wash-in–washout enhancement pat-
tern. The implications of these findings in terms
of expanding surveillance imaging to include
noncirrhotic HCV patients and the role of HCV
in the development of HCC in the noncirrhotic
liver need further study.
References
1. Ghany MG, Strader DB, Thomas DL, Seeff LB.
Diagnosis, management, and treatment of hepati-
tis C: an update. Hepatology 2009; 49:1335–1374
2. Bruix J, Sherman M. Management of hepatocel-
lular carcinoma: an update. Hepatology 2011;
53:1020–1022
3. Alkofer B, Lepennec V, Chiche L. Hepatocellular
cancer in the non-cirrhotic liver. J Visc Surg 2011;
148:3–11
4. Bralet MP, Regimbeau JM, Pineau P, et al. Hepa-
tocellular carcinoma occurring in nonfibrotic liv-
er: epidemiologic and histopathologic analysis of
80 French cases. Hepatology 2000; 32:200–204
5. Bismuth H, Chiche L, Castaing D. Surgical treat-
ment of hepatocellular carcinomas in noncirrhotic
liver: experience with 68 liver resections. World J
Surg 1995; 19:35–41
6. Fong Y, Sun RL, Jarnagin W, Blumgart LH. An
analysis of 412 cases of hepatocellular carcinoma
at a Western center. Ann Surg 1999; 229:790–799;
discussion, 799–800
7. Shimada M, Rikimaru T, Sugimachi K, et al. The
importance of hepatic resection for hepatocellular
carcinoma originating from nonfibrotic liver. J
Am Coll Surg 2000; 191:531–537
8. Nagasue N, Ono T, Yamanoi A, et al. Prognostic
factors and survival after hepatic resection for he-
patocellular carcinoma without cirrhosis. Br J
Surg 2001; 88:515–522
9. Chen MF, Tsai HP, Jeng LB, et al. Prognostic fac-
tors after resection for hepatocellular carcinoma
in noncirrhotic livers: univariate and multivariate
analysis. World J Surg 2003; 27:443–447
10. Chang CH, Chau GY, Lui WY, Tsay SH, King
KL, Wu CW. Long-term results of hepatic resec-
tion for hepatocellular carcinoma originating
from the noncirrhotic liver. Arch Surg 2004;
139:320–325; discussion, 326
11. Laurent C, Blanc JF, Nobili S, et al. Prognostic fac-
tors and longterm survival after hepatic resection
for hepatocellular carcinoma originating from non-
cirrhotic liver. J Am Coll Surg 2005; 201:656–662
12. Bège T, Le Treut YP, Hardwigsen J, et al. Prog-
nostic factors after resection for hepatocellular
carcinoma in nonfibrotic or moderately fibrotic
liver: a 116-case European series. J Gastrointest
Surg 2007; 11:619–625
13. Shrager B, Jibara G, Schwartz M, Roayaie S. Re-
section of hepatocellular carcinoma without cir-
rhosis. Ann Surg 2012; 255:1135–1143
14. Winston CB, Schwartz LH, Fong Y, Blumgart
LH, Panicek DM. Hepatocellular carcinoma: MR

imaging findings in cirrhotic livers and noncir-
rhotic livers. Radiology 1999; 210:75–79
15. Brancatelli G, Federle MP, Grazioli L, Carr BI.
Hepatocellular carcinoma in noncirrhotic liver:
CT, clinical, and pathologic findings in 39 U.S.
residents. Radiology 2002; 222:89–94
Downloaded from www.ajronline.org by 182.253.105.42 on 02/13/18 from IP address 182.253.105.42. Copyright ARRS. For personal use only; all rights reserved
16. Regev A, Berho M, Jeffers LJ, et al. Sampling er-
ror and intraobserver variation in liver biopsy in
patients with chronic HCV infection. Am J Gas-
troenterol 2002; 97:2614–2618
17. Rousselet MC, Michalak S, Dupre F, et al. Sourc-
es of variability in histological scoring of chronic
viral hepatitis. Hepatology 2005; 41:257–264
18. Becker U, Deis A, Sorensen TI, et al. Prediction of
risk of liver disease by alcohol intake, sex, and
age: a prospective population study. Hepatology
1996; 23:1025–1029
19. [No authors listed]. Terminology of nodular hepa-
W616 AJR:200, June 2013
Lewis et al.
tocellular lesions: International Working Party.
Hepatology 1995; 22:983–993
20. [No authors listed]. Intraobserver and interob-
server variations in liver biopsy interpretation in
patients with chronic hepatitis C: the French
METAVIR Cooperative Study Group. Hepatolo-
gy 1994; 20(1 Pt 1):15–20
21. Lok AS, Seeff LB, Morgan TR, et al.; HALT-C
Trial Group. Incidence of hepatocellular carcino-
ma and associated risk factors in hepatitis C–re-
lated advanced liver disease. Gastroenterology
2009; 136:138–148
22. Nzeako UC, Goodman ZD, Ishak KG. Hepatocel-
lular carcinoma in cirrhotic and noncirrhotic livers:
a clinico-histopathologic study of 804 North Amer-
ican patients. Am J Clin Pathol 1996; 105:65–75
23. Yeh MM, Daniel HD, Torbenson M. Hepatitis C–
associated hepatocellular carcinomas in non-cir-
rhotic livers. Mod Pathol 2010; 23:276–283
24. De Mitri MS, Poussin K, Baccarini P, et al. HCV-
associated liver cancer without cirrhosis. Lancet
1995; 345:413–415
25. Nash KL, Woodall T, Brown AS, Davies SE, Al-
exander GJ. Hepatocellular carcinoma in patients
with chronic hepatitis C virus infection without cir-
rhosis. World J Gastroenterol 2010; 16:4061–4065
26. Sherman M. Hepatocellular carcinoma: epidemi-
ology, risk factors, and screening. Semin Liver Dis
2005; 25:143–154
27. el-Refaie A, Savage K, Bhattacharya S, et al.
HCV-associated hepatocellular carcinoma with-
out cirrhosis. J Hepatol 1996; 24:277–285
28. Moriya K, Fujie H, Shintani Y, et al. The core pro-
tein of hepatitis C virus induces hepatocellular
carcinoma in transgenic mice. Nat Med 1998;
4:1065–1067