Anda di halaman 1dari 8

J Gastrointest Canc (2011) 42:228–235

DOI 10.1007/s12029-010-9200-x


Comparison of Cancer Care for Hepatocellular Carcinoma

at Two Tertiary-Care Referral Centers from High and Low
Endemic Regions for Viral Hepatitis
Fahad Alsohaibani & Geoffrey Porter &
Hamad Al-Ashgar & Mark Walsh & Robert Berry &
Michele Molinari & Kevork Minas Peltekian

Published online: 1 September 2010

# Springer Science+Business Media, LLC 2010

Abstract Results Mean age (±SEM) of the two groups were similar
Background Risk factors for hepatocellular carcinoma (64+1 and 65+1 years) with 93% versus 75% males
(HCC) have geographic variability but differences in care amongst Canadian and Saudi patients, respectively. In
have not been described. We reviewed the presentation, Canada, underlying disease was alcohol-related cirrhosis
management, and outcomes of HCC patients from two (45%), cryptogenic cirrhosis (26%), or hepatitis C (13%).
tertiary-referral centers in Central Saudi Arabia and Atlantic For Saudis, HCC cases were attributed to hepatitis C (47%),
Canada during 1997–2002. cryptogenic cirrhosis (27%), and hepatitis B (21%). At
Methods Data were extracted from health records of 96 initial presentation, Saudi patients had more vascular
Saudi and 80 Canadian consecutive patients with HCC. invasion and distant metastases while Canadians had more
advanced liver disease. The tumor-specific prognostic
F. Alsohaibani : K. M. Peltekian classifications were comparable. Due to center-specific
Department of Medicine, Dalhousie University,
Halifax, Nova Scotia, Canada
expertise or preference, symptomatic treatment was more
common amongst Saudi patients (83% versus 42%) while
F. Alsohaibani : H. Al-Ashgar more Canadians underwent local palliative interventions
Department of Medicine, (52% versus 12%). Frequency of potentially curative
King Faisal Specialist Hospital and Research Center,
Riyadh, Saudi Arabia
therapies including resection and transplantation were
similar at both centers. There was no difference in overall
G. Porter : M. Walsh : M. Molinari : K. M. Peltekian median survival (14 versus 10 months) amongst Canadian
Department of Surgery, Dalhousie University, and Saudi patients.
Halifax, Nova Scotia, Canada
Conclusions This study validates divergence in HCC
G. Porter presentation between low and high endemic regions for
Cancer Care Nova Scotia, viral hepatitis. In addition, for the first time, differences in
Halifax, Nova Scotia, Canada cancer care of HCC are documented.
M. Walsh : M. Molinari : K. M. Peltekian
Atlantic Multi-Organ Transplantation Program, Keywords Cancer care . Cirrhosis . Viral hepatitis .
Queen Elizabeth II Health Sciences Centre, Canada . Saudi Arabia . Health service delivery
Halifax, Nova Scotia, Canada

R. Berry
Department of Diagnostic Imaging, Dalhousie University,
Halifax, Nova Scotia, Canada Abbreviations
HCC Hepatocellular carcinoma
K. M. Peltekian (*)
QEII Queen Elizabeth II Health Science Centre
Atlantic Hepatology Services,
P. O. Box 60010, B3H 4R7 Halifax, Nova Scotia, Canada KFSH King Faisal Specialist Hospital and Research Center
e-mail: ASR Age standardized rate
J Gastrointest Canc (2011) 42:228–235 229

ICD International Classification of Disease Methods

AFP α-Fetoprotein CTP, Child–Turcotte–Pugh
CLIP Cancer of the Liver Italian Program Study Population
BCLC Barcelona Clinic Liver Cancer
HBV Hepatitis B virus Nova Scotia is the most populated of the four provinces in
HCV Hepatitis C virus Atlantic Canada with an estimated population of 950,000.
TACE Trans-arterial chemo-embolization Queen Elizabeth II Health Science Center (QEII) is the
PILI Percutaneous intra-lesional injection main teaching hospital for Dalhousie University and the
IQR Inter-quartile range province’s largest healthcare center with 1,100 beds—175
dedicated to Veterans. It is also the tertiary-referral center
for cancer care in the Maritimes and hosts the Atlantic
Multi-Organ Transplantation Program, which provides
Introduction advanced liver care and transplantation to approximately
2,500,000 Atlantic Canadians. King Faisal Specialist
Hepatocellular carcinoma, the most common primary Hospital and Research Center (KFSH) is situated in
malignancy of the liver, is observed characteristically as Riyadh, the capital of Saudi Arabia. KFSH is a
a complication of chronic liver disease. It accounts for government-funded tertiary medical and research facility
6% of all cancers worldwide. Hepatocellular carcinoma operating 894 beds and serve mainly the central part of
(HCC) is considered to be the fifth most common Saudi Arabia and the surrounding regions (with population
cancer in the world and the third most common cause of 4,250,000 as of September 2004). Similar to QEII, it is
of cancer mortality with a 5-year survival rate of less one of the main teaching hospitals in the Central Region
than 5% without treatment (1). providing cancer care and advanced hepato-biliary surgery
The incidence of HCC is characterized by marked and liver transplantation.
geographic and demographic variations that correlate
with the prevalence of HBV and HCV. The great Inclusion Criteria
majority of HCC occurs in either Southeast Asia or in
sub-Saharan Africa (2). Epidemiologic and laboratory All patients diagnosed with HCC between 1997 and 2002 at
studies have firmly established a strong and specific QEII and KFH were included. Although in recent years, there
association between HBV and HCC. In addition, cirrhosis, has been emphasis for outpatient investigation and manage-
regardless of its etiology, is the most significant risk factor ment of patients with HCC and even home-based palliative
for HCC (3). In North America, although alcohol and care for those with advanced disease, during the study period,
persistent HCV infection are well-documented etiologic most subjects diagnosed with HCC were hospitalized in both
factors for HCC, recent epidemiologic observations medical (oncology or gastroenterology) and surgical services
indicate that non-alcoholic fatty liver disease leading to for investigations, interventions, and palliation. In both
cryptogenic cirrhosis also contributes to the cases of HCC centers, often consultation was provided by hepatologists to
(4). patients in both hospital and outpatient settings.
In spite of the well-described geographic variation in the We used established criteria for the diagnosis of HCC (7).
global epidemiology of HCC, differences in cancer care HCC is a highly vascular neoplasm that usually arises in a
between regions are unknown. Atlantic Canada is considered cirrhotic liver. Based on this concept, we used noninvasive
to have low incidence for HCC, however, over the last 25 years diagnostic criteria whenever biopsy was not available. The
the incidence of HCC has increased significantly, from 0.20 diagnostic imaging criteria required that the liver mass be
per 100.000 (1974–1978) to 1.13 per 100.000 (1994–1998) characterized by at least two imaging modalities with contrast
(5). In Saudi Arabia, the overall age standardized rate (ASR) enhancement on computerized tomography, magnetic reso-
for HCC between January 1999 and December 2000 was nance imaging, or angiography. The presence of elevated
estimated to be 4.5/100.000 (6.5/100.000 in males and 2.6/ serum α-fetoprotein (AFP) greater than 400 ng/dl in cirrhotic
100.000 in females) with the highest ASR in the Central patients who had a liver mass was also considered diagnostic.
Region at 10.4/100.000 (6). If the purposes of this study when lesion mass size was less
We reviewed and compared the clinical presentation, than 2 cm in diameter, to be included in the study, histological
management and outcomes of patients referred from confirmation was necessary (7). There were six out of 86
Atlantic Canada (low endemic area for viral hepatitis) and patients from QEII and four out of 100 patients from KFSH
from Central Saudi Arabian Region (high endemic area for who were excluded because of insufficient data or non-
viral hepatitis). conclusive results.
230 J Gastrointest Canc (2011) 42:228–235

Data Source were considered potentially curative and all subjects under-
going hepatic resection or transplantation were grouped as
In this retrospective analysis, health records of 80 consecutive receiving curative therapy even if they received local
patients diagnosed and hospitalized with HCC between 1997 palliative therapies such as trans-arterial chemo-embolization
and 2002 at QEII were reviewed and compared with 96 (TACE) prior to liver transplantation.
consecutive patients hospitalized during the same time period The main approach for managing non-resectable focal
at KFSH with the diagnosis of HCC. One of the authors (FA) lesions less than 3 cm differed between QEII and KFSH
performed all the chart reviews and data extraction in during the study period. In spite of the availability of
standardized fashion to ensure comparable interpretation of radiofrequency ablation, in Atlantic Canada the primary
results among the two distant study centers. modality of intervention was percutaneous 99% ethanol or
50% acetic acid injection; radiofrequency ablation (RFA) at
Diagnostic Codes QEII was being utilized mainly for metastatic cancers
involving the liver rather than primary HCC. On the other
Our data included patients diagnosed with primary liver hand, in Central Saudi Arabia, RFA was used primarily to
cancer, using the International Classification of Disease control focal HCC. In addition, in Atlantic Canada because
(ICD-9 code 155.0 and ICD-10 code C22) for HCC. of early successful experience, chemo-embolization via
The main database included age, gender, performance hepatic artery was often used in subjects with non-
status, clinical and laboratory including serum AFP as a tumor resectable large tumors or while on the wait list for liver
marker for HCC, and platelet counts as a reflection on the transplantation. This intervention was not available at
presence of portal hypertension. Since the Child–Turcotte– KFSH during the study period. These three interventions:
Pugh (CTP) classification is one of the most commonly used percutaneous intra-lesional injection (PILI) with ethanol or
systems to evaluate liver function in cirrhotic patients we acetic acid, RFA whether percutaneous or intra-operative,
made every effort to extract the presentation data from charts and TACE were all considered local intervention modes for
to be able to calculate it. We then reviewed all the relevant management of HCC. For the purposes of this study,
radiological and histological data at presentation to determine patients undergoing these procedures were grouped as
tumor size, presence of vascular invasion or metastasis. Data receiving palliative therapy.
relating to various treatment modalities administered to each In addition, during the study period, at QEII subjects
patient and the final outcomes, when available, including with non-resectable and advanced HCC were often tried on
death were also extracted from patient charts and entered into tamoxifen. However, large controlled-trial (13) and later a
data for later analysis. systematic review with meta-analysis confirmed the lack of
Currently, there are many available prognostic staging or efficacy for tamoxifen in HCC (14). For the purpose of this
classification systems for HCC including (8). In our study, study, all patients referred for hospice or comfort care, those
we extracted data to determine each patient’s Okuda staging treated with systemic chemotherapy or hormonal therapy
(9) and Cancer of the Liver Italian Program (CLIP) score such as tamoxifen were grouped together as having only
(10) at the time of initial HCC diagnosis. CLIP score is symptomatic therapy.
easily computed and consisted of four variables: CTP stage,
tumor morphology and extension, serum AFP levels, and Statistical Analysis
presence of portal vein thrombosis. Since one of the main
objectives of our study was to compare the type of Descriptive statistics (mean, standard deviation, range, and
intervention in the management of HCC, we also gathered proportions) were used to describe each variable. Comparisons
data to categorize the patient in the appropriate Barcelona between groups were made with Student t-test for continuous
Clinic Liver Cancer (BCLC) staging classification (11). The variables and χ2 test for frequencies of categorical variables.
latter has been incorporated into clinical practice guidelines A Kaplan–Meier plot was constructed for visual representa-
since it stratifies patients according to HCC staging and tion of the deaths. Statistical significance was demonstrated
degree of liver disease in a process leading to a specific by a p-value<0.05. Statistical analysis was performed with
treatment; BCLC staging classification has shown the best Minitab version 14 (Minitab, State College, PA).
results in terms of survival (12).

Center-Specific Interventions Results

During study period, both centers provided hepatic resection At QEII the mean (±SEM) age of patients was 64+1 years
and liver transplantation. These two surgical interventions with 93% males compared to 75% males, and 65+1 years
J Gastrointest Canc (2011) 42:228–235 231

at KFSH. The mean performance status according to Table 1 Characteristics of HCC patients from Saudi Arabia and
Atlantic Canada
Eastern Cooperative Oncology Group classification was
comparable at time of presentation for KFSH versus Variable KFSH QEII p-value
QEII groups (Table 1). Underlying etiology for HCC at (n=96) (n=80)
QEII was alcohol-related cirrhosis (45%), cryptogenic
Agea, year 65±1 64±1 0.384
cirrhosis (26%), hereditary hemochromatosis (13%),
Male sex 75% 93% 0.002
chronic hepatitis C virus (13%), chronic hepatitis B virus
Primary etiology
(5%), primary biliary cirrhosis (2%), and chronic autoim-
HCV 47% 13% <0.001
mune hepatitis (2%). For KFSH group, most HCC cases
HBV 21% 6% 0.006
were attributed to chronic hepatitis C virus (47%),
Alcohol 2% 45% <0.001
cryptogenic cirrhosis (27%), chronic hepatitis B virus
Cryptogenic 27% 26% 0.901
(21%), schistosomiasis (5%) and alcohol-related cirrhosis
Hemochromatosis 0% 13% <0.001
Schistosomiasis 5% 0% 0.038
Fifty percent of QEII patients and 40% of patients at
Others 1% 2% 0.457
KFSH had histological confirmation of tumor based on
sampling of liver lesion either by percutaneous biopsy
Albumin, g/L 30±1 29±1 0.228
or surgical resection (p = 0.166). Partial portal vein
Bilirubin, μmol/L 48±10 42±6 0.654
thrombosis was observed in 12% at QEII versus 16% in
INR 1.10±0.03 1.32±0.04 <0.001
KFSH, while complete portal vein thrombosis was similar
at 14%. More vascular invasion was documented in Platelet count, × 109/L 193±11 177±12 0.315
patients from KFSH than from QEII with 29% versus Child–Turcotte–Pugh classification 0.021
8% (p<0.001) indicating more advanced presentation for A 59% 39%
HCC among Saudi patients. B 28% 39%
C 13% 22%
Morphological Assessment of HCC Performance statusb <0.001
0 18% 11%
Patients with HCC hospitalized at QEII at presentation 1-2 60% 60%
2% had tumor <2 cm in diameter, 50% between 2-5 cm 3 4% 23%
and in 48% the tumor was greater than 5 cm. In KFSH 4 18% 6%
the respective percentages were 4%, 38% and 58% AFP, ng/mL 0.625
(Table 1). Most of QEII patients had unifocal tumor <20 35% 42%
(51% versus 41% in KFSH), while larger number of 20 – 400 26% 23%
KFSH patients had multifocal tumor (42% versus 36% in > 400 39% 35%
QEII). Eight percent versus 5% had diffuse tumor and 5% Tumor diameter, cm 0.235
versus 12% had metastatic tumor in QEII and KFSH, <2 4% 2%
respectively. 2-5 38% 50%
>5 58% 48%
Liver Disease Staging and HCC Classification Tumor distribution 0.214
Unifocal 41% 51%
Patients at QEII had more advanced liver disease with CTP Multifocal 42% 36%
class B–C 61% versus 41% at KFSH (p=0.021). Half of the Diffuse 5% 8%
patients at QEII presented with ascites while only 28% of Metastatic 12% 5%
KFSH patients ascites was noted at presentation (p=0.009). Portal vein thrombosis
The mean tumor-specific staging scores in patient Complete 14% 14% 0.968
populations at both Centers were comparable with CLIP Partial 16% 12% 0.554
score, and Okuda Staging (Table 1). The latter would be CLIP score 0.23
expected since both CLIP score and Okuda staging 0 19% 14%
combine the stage of liver disease and the stage of HCC 1 21% 29%
to predict prognosis. At QEII 87% of patients were 2 32% 27%
classified at BCLC stages C–D versus 80% of patients at 3 19% 11%
KFSH were at stage C–D. The majority of patients were 4 7% 10%
presenting at stages where no efficacious interventions 5 1% 6%
were available that would prolong survival. 6 1% 3%
232 J Gastrointest Canc (2011) 42:228–235

Table 1 (continued)
On the other hand, the forms of palliative treatments
Variable KFSH QEII p-value available at QEII included TACE which was delivered to
(n=96) (n=80) 44% of patients and PILI with either alcohol or acetic
acid was performed in 9%. At KFSH, PILI with alcohol
Okuda staging 0.215
was performed in 5% of patients and percutaneous or
0 68% 53%
intra-operative RFA in 7%. The RFA technology was
1-2 16% 22%
available at QEII at the time of the study but its use was
3-4 17% 25%
limited to small metastatic tumors in the liver. Although
BCLC classification 0.143
there was an early trend for better outcomes amongst
A 11% 11%
Atlantic Canadians undergoing local palliative therapies
B 9% 2%
including TACE, the long-term outcomes were similar
C 53% 48%
between the two Centers. Finally potentially curative
D 27% 39%
treatment with surgical resection was achieved in 10% of
Mean+SEM QEII group versus 7% of KFSH group. Liver transplantation
Assessed by the Eastern Cooperative Oncology Group criteria was performed successful for 10% of QEII patients versus 3%
of KFSH patients with HCC.

Median Survival
Treatment Modalities
Survival was measured from first day of diagnosis to the
Symptomatic treatment was the main modality of therapy at day of last follow up and compared between the two
KFSH (83% versus 42% at QEII). The latter group at QEII groups. The median survival for patients at QEII and KFSH
included patients who received hormonal therapy with is shown in Fig. 1. This was predictable outcome based on
tamoxifen and those who received systemic chemotherapy the CLIP score, the Okuda Stage and BCLC classification.
at KFSH; both these interventions were proven subsequent The patients that underwent curative interventions (either
to study period not to impact survival and hence were tumor resection or orthotopic liver transplantation) had the
combined in the same treatment category (Table 2). Unlike best survival 78% 5-year survival (Fig. 2). Those patients
patients at QEII, patients with multifocal HCC who were with palliative therapies such as TACE, PILI, or RFA had
not candidates for resection or transplantation were only median survival of 24 months (inter-quartile range (IQR)
offered symptomatic treatments since TACE was not 14 months) while those who received symptomatic therapy
available during the study period at KFSH. In spite of the median survival of 6 months (IQR 14 months; Fig. 2). In
more advanced stage of HCC amongst the Saudi patients at our study, only 10% to 20% of patients diagnosed with
presentation, there was tendency for better survival HCC were eligible for curative treatments, which included
amongst the Saudi patients (median survival 8.3 months surgical resection or orthotopic liver transplantation. Of these
versus 0.9 months amongst Atlantic Canadians—presenting two therapies, liver transplantation has the best long-term
with more advanced liver disease) but this did not translate outcomes 1, 3, and 5-year survival rates of 78%, 63%, and
to significant overall differences. 57%, respectively.

Table 2 Initial treatment modal-

ities for HCC patients from Saudi Modalitya Intervention KFSH (n=96) QEII (n=80) p-value
Arabia and Atlantic Canada
Curative 10 (10%) 16 (20%)
Surgical resection 7 8 0.522
Liver transplantation 3 8 0.061
Palliative 12 (12%) 42 (52%)
RFAb 7 0 0.014
PILI 5 7 0.353
Patients could have received
TACEb 0 35 <0.001
more than one modality (e.g.,
patient receiving TACE while Symptomatic 80 (83%) 34 (42%)
awaiting liver transplantation) Hormonalb 0 33 <0.001
Difference due to local preferences Systemic chemotherapy 6 1 0.091
or availability of center-specific Supportive 79 27 <0.001
J Gastrointest Canc (2011) 42:228–235 233

affected than Asians, Pacific Islanders, or Native Americans

(15). It is clear that HCC will become a significant cause of
morbidity and mortality in the near future (16).
The most effective way to decrease the incidence of HCC
is to prevent development of cirrhosis. Although there is no
clear evidence to support the benefit of periodic screening for
detection of HCC using ultrasonography and serum AFP;
consensus guidelines based on expert opinions recommend
ultrasonography every 6 to 12 months in high-risk patients
who are candidates for liver transplantation, surgical resection
or percutaneous therapy (12).
Accurate clinical staging of HCC is important to
Fig. 1 Kaplan–Meier survival curve for patients with HCC referred to determine the prognosis and provide the best treatment for
KFSH in Central Saudi Arabia (n=96) and QEII in Atlantic Canada (n= each patient. Unlike for the majority of solid tumors, the
80). The median survival at KFSH and QEII were 10.1 and 13.6 months, outcomes in liver cancer are highly dependent on the
respectively (Log rank, p-value=0.126)
residual liver function and general health of the patient (17).
This study also provides a glimpse on how the choice of
interventions depends on center-specific preferences or
Discussion availability of interventions. Life expectancy of patient
with HCC is hard to predict and despite a statistically
Viral hepatitis is the primary underlying etiology of cirrhosis significant improvement in the survival rates over the last
with subsequent HCC worldwide. Just as expected, in our 20 years, patients with HCC continue to have poor
study, viral hepatitis comprises significant portion of etiologic prognosis with a median survival of approximately
factor for HCC but cryptogenic cirrhosis (presumably 7 months (16). When symptomatic, HCC is usually non-
secondary to fatty liver disease) is the second most common resectable and associated with median survival of less than
underlying cause for HCC in both regions. Hemochromatosis 6 months (18). Despite many available treatment options,
is a significant etiologic factor among Atlantic Canadians who unfortunately the prognosis remains poor.
are largely descendents of Celtics from Scotland, Ireland, In our study, overall patients hospitalized at KFSH had
Wales, and Normandy. more advanced pathology for HCC with larger tumor load
The incidence of HCC increases with age and occurs and multifocality; however, patients hospitalized at QEII
four to nine times more frequently in males than in females had more advanced liver disease. Presentation with either
except in the group without preexistent liver disease, in advanced HCC pathology or advanced liver disease is not
whom the ratio is 1:1. In the United States, Caucasians are surprising since both processes remain asymptomatic until
two to three times less often affected than African end-stages. This is one of the rationales for supporting
Americans, who in turn are two to three times less often programs for HCC surveillance. However, the difference in
presentation may be explained by the fact that in Atlantic
Canada liver disease and cirrhosis is usually multi-factorial
with many having two or more risks for cirrhosis such as
HCV, alcohol, and obesity. In these patients, HCC is being
identified because patients have declared themselves as
having advanced liver disease. On the other hand, in Saudi
Arabia where the majority of HCC patients had viral
etiology acquiring it at earlier age with increasing propor-
tions of female patients where HCC becomes symptomatic
on a background of more indolent chronic liver disease the
latter being identified only after HCC has been docu-
mented. These differences in presentation, if proven to be
valid in other studies, will have major impact regional
policies for liver disease screening or HCC surveillance.
We were able to document differences not only in the
Fig. 2 Kaplan–Meier survival curve for patients with all HCC by
treatment categories. The median survival rates for curative interventions,
initial presentation of patients hospitalized with HCC but
local palliative interventions, and symptomatic treatments were 25.1, also in the type of intervention between low and high-
11.1, and 5.8 months, respectively (Log rank, p-value<0.001) intermediate endemic regions for viral hepatitis. The main
234 J Gastrointest Canc (2011) 42:228–235

discrepancy in the approach was due to the more advanced References

HCC in the high-intermediate endemic region with often
only symptomatic therapy. Some of the differences were
also related to disparities in acceptance and uptake by 1. Parkin DM, Bray F, Ferlay J, Pisani P. Estimating the world
providers for novel interventions. The variations and cancer burden: Globocan 2000. Int J Cancer. 2001;94:153–6.
inequities between the different countries are being recognized 2. McGlynn KA, London WT. Epidemiology and natural history of
hepatocellular carcinoma. Best Pract Res Clin Gastroenterol.
globally in the inability of patients to access new innovative 2005;19:3–23.
cancer drug treatments (19). At KFSH, intra-operative radio- 3. Zaman SN, Melia WM, Johnson RD, Portmann BC, Johnson PJ,
frequency ablation had been available for many years and Williams R. Risk factors in development of hepatocellular
well-accepted procedure. On the other hand, because of carcinoma in cirrhosis: prospective study of 613 patients. Lancet.
strong interventional radiology group at QEII, both PILI and 4. Caldwell SH, Oelsner DH, Iezzoni JC, Hespenheide EE, Battle
TACE of HCC via selective hepatic artery injection were EH, Driscoll CJ. Cryptogenic cirrhosis: clinical characterization
utilized frequently. and risk factors for underlying disease. Hepatology. 1999;29:664–
In our study, the best outcomes were those patients 9.
5. Farina DM, Reaume N, McAlister VC, Colwell BD, Peltekian
eligible for curative treatments which comprised only 10% KM. Rising incidence in hepatocellular carcinoma and cholangio-
to 20% of patients diagnosed with HCC. With wider use of carcinoma in Atlantic Canada over a 25-year period. Gastroenter-
HCC surveillance programs the proportion of patients ology. 2000;118:A1437.
eligible for curative treatments will increase. Once HCC is 6. Cancer Incidence Report in Saudi Arabia. 1999–2000:54–56.
7. Talwalkar JA, Gores GJ. Diagnosis and staging of hepatocellular
determined not to be resectable or transplantable, no carcinoma. Gastroenterology. 2004;127:S126–32.
standard treatment has been shown to be effective; 8. Marrero JA, Fontana RJ, Barrat A, Askari F, Conjeevaram HS, Su GL,
currently, there is no proven treatment modality that et al. Prognosis of hepatocellular carcinoma: comparison of 7 staging
prolongs survival in those patients (20). One exception to systems in an American cohort. Hepatology. 2005;41:707–16.
9. Okuda K, Ohtsuki T, Obata H, Tomimatsu M, Okazaki N,
this is TACE that has been shown to improve survival of Hasegawa H, et al. Natural history of hepatocellular carcinoma
highly selected (Child class A–B) patients with unrespectable and prognosis in relation to treatment: study of 850 patients.
HCC. Survival probabilities at 1 and 2 years were 82% and Cancer. 1985;56:918–28.
63% for TACE, and 63% and 27% for control (21). PILI has 10. The Cancer of the Liver Italian Program (CLIP) Investigators.
Prospective validation of the CLIP score: a new prognostic
been most widely performed as an attractive alternative to system for patients with cirrhosis and hepatocellular carcinoma.
surgery in patients with small HCC with similar outcomes Hepatology. 2000;31:840–5.
(22). In recent years, however, in many centers percutaneous 11. Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma:
and intra-operative RFA has been steadily replacing PILI. the BCLC staging classification. Semin Liver Dis. 1999;19:329–
For small HCC, outcomes following RFA and hepatic 12. Bruix J, Sherman M, Practice Guidelines Committee, American
resection are comparable (23). Although sorafenib has been Association for the Study of Liver Diseases. Management of
shown to be effective in non-resectable HCC, its exact role hepatocellular carcinoma. Hepatology. 2005;42:1208–36.
in the algorithm of HCC management needs further studies 13. CLIP Group (Cancer of the Liver Italian Programme). Tamoxifen
in treatment of hepatocellular carcinoma: a randomised controlled
(24). The overall median survival between the two groups trial. Lancet. 1998;352:17–20.
showed minor advantage for patients in Atlantic Canada. 14. Nowak AK, Stockler MR, Chow PK, Findlay M. Use of tamoxifen in
This difference was mainly related to the utilization of TACE advanced-stage hepatocellular carcinoma: a systematic review.
at QEII. The outcomes of all the other interventions were Cancer. 2005;103:1408–14.
15. El-Serag HB, Mason AC. Rising incidence of hepatocellular
similar individually or when grouped into curative, palliative carcinoma in the United States. N Engl J Med. 1999;340:745–
or symptomatic therapy. 50.
This study emphasizes the differences not only in the 16. El-Serag HB, Mason AC, Key C. Trends in survival of patients
initial presentation of subjects with HCC but also in the with hepatocellular carcinoma between 1977 and 1996 in the
United States. Hepatology. 2001;33:62–5.
type of intervention between low and high endemic 17. Huo TL, Lin HC, Lee FY, Hou MC, Lee PC, Wu JC, et al.
regions for viral hepatitis. The differences attributed to Occurrence of cirrhosis-related complications is a time-dependent
cancer care are based on availability of local expertise prognostic predictor independent of baseline model for end-stage
but also due to divergence in health care cultures. liver disease score. Liver Int. 2006;26:55–61.
18. Sasaki Y, Imaoka S, Nakano H, Ishikawa O, Ohigashi H, Yasuda T,
Increasingly, investigators are relying on multicenter or et al. Indications for hepatectomy for hepatocellular carcinoma—
multi-group studies to demonstrate effectiveness and what stage of the disease is the best indication for surgery. J
generalizability (25). Comparative studies such as this Hepatobiliary Pancreat Surg. 1998;5:14–7.
one highlight the necessity for global multicenter trials 19. Jonsson B, Wilking N. A global comparison regarding patient
access to cancer drugs: summary. Ann Oncol. 2007;18:iii2–7.
because of the existence of local variations in expertise 20. Simonetti RG, Liberati A, Angiolini C, Pagliaro L. Treatment of
and delivery of cancer care on a background of differences hepatocellular carcinoma: a systematic review of randomized
in the presentation of patients with HCC. controlled trials. Ann Oncol. 1997;8:117–36.
J Gastrointest Canc (2011) 42:228–235 235

21. Llovet JM, Real MI, Montana X, Planas R, Coll S, Aponte J, et al. 23. Liu JG, Wang YJ, Du Z. Radiofrequency ablation in the treatment
Arterial embolisation or chemoembolisation versus symptomatic of small hepatocellular carcinoma: a meta analysis. World J
treatment in patients with unresectable hepatocellular carcinoma: a Gastroenterol. 2010;16:3450–6.
randomised controlled trial. Lancet. 2002;359:1734–9. 24. Lee JM, Han KH. Positioning and indication of sorafenib in the
22. Livraghi T, Bolondi L, Buscarini L, Cottone M, Mazziotti A, treatment algorithm and real practice setting: Western and eastern
Morabito A, et al. No treatment, resection and ethanol injection in approach—Asian perspective. Oncology. 2010;78 Suppl 1:167–71.
hepatocellular carcinoma: a retrospective analysis of survival in 25. Localio AR, Berlin JA, Ten Have TR, Kimmel SE. Adjustments
391 patients with cirrhosis. Italian Cooperative HCC Study for Center in Multicenter Studies: an overview. Ann Intern Med.
Group. J Hepatol. 1995;22:522–6. 2001;135:112–23.