Original Article

The Outcomes of Elderly Patients

With Hepatocellular Carcinoma
Treated With Transarterial
Chemoembolization
Thomas Yau, MD1; Tzy Jyun Yao, PhD2; Pierre Chan, MD1; Richard John Epstein, MD1; Kelvin K. Ng, MD3;
Siu Ho Chok, MD3; Tan Teo Cheung, MD3; Sheung Tat Fan, MD3; and Ronnie Tung Ping Poon, MD3

BACKGROUND: The authors evaluated and compared the treatment outcomes of transarterial chemoem-
bolization (TACE) between young (
70 years) and elderly (>70 years) patients at their institute over an
18-year period. METHODS: Advanced hepatocellular carcinoma (HCC) patients who received TACE at the
authors’ center were analyzed retrospectively. The demographic data, TACE-related morbidities, and sur-
vival outcome were compared between these 2 age groups. RESULTS: Between 1989 and 2006, 843
patients who were
70 years old and 197 patients who were >70 years old received TACE treatment for
advanced HCC. There were significantly more comorbid illnesses associated with the elderly patients than
the young patients (64 % vs 33%, P < .01). Moreover, elderly patients who received TACE treatment for
HCC were at earlier stages of disease (P < .01). Both the overall median survival (14.0 months vs 8.1 months,
P < .003) and disease-specific survival (15.2 months vs 8.7 months, P < .001) were significantly higher in el-
derly than young patients. The most commonly encountered TACE-related morbidity in both age groups
was liver function derangement. Young patients had a significantly higher rate of developing liver derange-
ment after TACE than elderly patients (21% vs 11%, P < .01). Conversely, the elderly patients had a signifi-
cantly higher rate of developing peptic ulcer disease with TACE treatment than young patients (2.5% vs
0.5%, P ¼.01). Overall, there was no significant difference in TACE-related mortality between the young and
elderly patients (3% vs 4%, P ¼.49). CONCLUSIONS: This study has confirmed the comparable efficacy and
tolerability in using TACE for the treatment of advanced HCC in young and elderly patient populations.
Cancer 2009;115:5507–15. V C 2009 American Cancer Society.

KEY WORDS: transarterial chemoembolization, hepatocellular carcinoma, elderly, hepatitis.

Hepatocellular carcinoma (HCC) is the 5th most common cancer worldwide.1 The majority of
cases occur in developing countries, especially in Asia, because of the high prevalence of endemic hepatitis
B virus (HBV). Recently, there has been a rising number of HCC cases in Western countries, because of
the sequel of hepatitis C virus (HCV) and alcoholic cirrhosis.2

Corresponding author: Ronnie Tung Ping Poon, MD, Department of Surgery, The University of Hong Kong, 102 Pokfulam Road, Hong Kong;
Fax: (011) 852 2816 2863; [email protected]
1
Department of Medicine, University of Hong Kong Medical Center, Queen Mary Hospital, Hong Kong, China; 2Clinical Trials Center, University of
Hong Kong Medical Center, Queen Mary Hospital, Hong Kong, China; 3Department of Surgery, University of Hong Kong Medical Center, Queen
Mary Hospital, Hong Kong, China
Received: November 24, 2008; Revised: February 13, 2009; Accepted: March 24, 2009
Published online August 21, 2009 in Wiley InterScience (www.interscience.wiley.com)
DOI: 10.1002/cncr.24636, www.interscience.wiley.com

Cancer December 1, 2009 5507

Original Article

The incidence of cancer increases greatly with age, HCC and cirrhosis would be considered for liver trans-
with >60% of all cancers occurring in those aged 65 years plantation if they were within the Milan15 or expanded
or older.3 In fact, the risk of developing HCC is known to University of California San Francisco criteria.16 If the
be age dependent,4 and thus there will be an increasing patients were not surgical candidates, local ablation proce-
number of elderly patients diagnosed with HCC in the dures such as percutaneous ethanol injection and radiofre-
coming years because of the increase in the longevity of quency ablation or TACE were offered, depending on
the population. Compared with the younger age group, tumor size, number, and position. The remaining patients
elderly patients have significantly more comorbidity, par- who were not candidates for surgery, transplantation, or
ticularly from pulmonary and cardiovascular diseases. any locoregional therapies were offered either systemic
Moreover, the therapeutic benefit and the toxicities of therapy or best supportive care.
cancer treatment in the elderly population is largely At our center, TACE was considered as the treat-
unknown, as they have generally been underrepresented ment of choice for locally advanced HCC that is not ame-
in clinical trials.5 Many elderly patients are not receiving nable to resection, transplantation, or local ablation. In
optimal cancer treatment, as it is often withheld from patients with anatomically resectable tumors that are con-
them because of the fear of potential toxicities and per- sidered inoperable for other reasons, such as comorbidities
ceived minimal survival advantage.6 or suboptimal liver reserve, and patients who refused sur-
Transarterial chemoembolization (TACE) is com- gical interventions were also managed with TACE if pos-
monly used for the treatment of locally advanced HCC. It sible. However, main portal vein thrombosis,
was found to prolong the overall survival of patients with arteriovenous shunting, Child’s C cirrhosis, and extrahe-
unresectable HCC compared with conservative treatment patic metastases were generally regarded as major contra-
in 2 recent randomized trials.7,8 Furthermore, a meta- indications to TACE. Cisplatin was used as the
analysis of published randomized trials also concluded chemotherapeutic agent and delivered with Lipiodol, fol-
that TACE is an effective palliative treatment for HCC.9 lowed by Gelfoam particle embolization.17 Selective can-
Previously, advanced age was regarded by some clinicians nulation and embolization of the feeding arteries of the
as a contraindication for TACE.10 Moreover, there are tumors were performed whenever possible. Technical
few data in the literature regarding the efficacy and toler- details of the TACE procedure have been described in our
ability of TACE in elderly patients aged 70 years.11,12 previous papers.7,18 TACE was usually repeated every 2 to
In this study, we evaluated and compared the treatment 3 months until tumor progression, liver intolerance, or
outcomes of TACE between young (
70 years old) and other severe complications. Response to TACE was
elderly (>70 years old) patients based on a large cohort of assessed every 2 to 3 months with serial AFP level and CT
patients treated at our institute over an 18-year period. scan. In the current study, TACE-related morbidity was
defined as any complications within 30 days of treatment,
MATERIALS AND METHODS and TACE-related mortality was defined as death from a
Between January 1989 and December 2006, there were complication within 30 days of treatment. Common
4074 patients diagnosed with HCC who presented to the minor complaints of fever, abdominal pain, and vomiting
Division of Hepatobiliary Surgery in the Department of after TACE were excluded from the analyses.
Surgery, Queen Mary Hospital, Hong Kong. The diagno- All patients’’ data were retrieved and analyzed from
sis of HCC was confirmed either by histology or cytology, a prospectively collected HCC database. Demographic
by elevated serum a-fetoprotein (AFP) (400 ng/mL), or data, details of tumor characteristics, details of TACE
by typical radiological appearance. Patients were staged treatment and complications, treatment duration,
mostly by computed tomography (CT) scan. Patients response to TACE, last follow-up dates, and survival sta-
aged >70 years were defined as elderly in the current tus were reviewed and analyzed retrospectively.
study according to previous studies.13,14
All patients with adequate liver function and Statistical Analysis
radiologically resectable tumor were initially evaluated for Chi-square or Mann-Whitney tests were used to compare
partial hepatectomy. Patients of aged
65 years with early the nominal and continuous variables, respectively,

5508 Cancer December 1, 2009

 TACE in Elderly Patients/Yau et al

between young and elderly patients. Overall survival was Table 1. Treatment Modalities of Hepatocellular Carcinoma
in Young and Elderly Patients
calculated from the date of commencement of TACE to
the date of death or last follow-up. In the present study, Treatment Modality Young Elderly P
we defined malignant cachexia, bleeding, hepatic failure, Patients, Patients,
n52452 n5550
multiorgan failure, and rupture of HCC as disease-related
death, and disease-specific survival was calculated by treat- Locoregional therapies
TACE 843 197
ing nondisease-related death as censored cases.19 Both RFA 201 34
overall survival and disease-specific survival were com- PEIT 14 3
Cryotherapy 9 2
puted by the Kaplan-Meier method. All risk factors were 1067 (44%) 236 (43%) .46
Total
tested univariately using the log-rank test. The
Systemic treatment
log(log[estimated disease-free survival]) was plotted ver- Tamoxifen 360 86
sus log(time) to check the proportional hazard assump- Chemotherapy 108 2
Others 115 15
tion. All factors with P values <.1 were included in the Total 583 (23%) 103 (19%) <.001
Cox proportional hazard regression analysis through Best supportive care 802 (33%) 211 (38%) .01
backward selection to study the effect of age adjusting for
other risk factors. All statistical analyses were performed TACE indicates transarterial chemoembolization; RFA, radiofrequency
ablation; PEIT, percutaneous ethanol injection therapy.
using the SAS 9.1 (SAS Institute Inc., Cary, NC; 2003).

elderly HCC patients were carriers of HCV infection,

RESULTS whereas only 10% of young HCC patients had HCV
Among the 4074 HCC patients presenting at our center infection. Nevertheless, there was no significant difference
during the study period, 1072 patients received surgical in the Child-Pugh liver function status between young
treatment for HCC, whereas the other 3002 patients were and elderly patients (P ¼ .2). In the present cohort, 10
not surgical candidates. Among these nonsurgical young patients had Child-Pugh class C liver function sta-
patients, 2452 patients were
70 years old, and 550 tus and had received TACE treatment. Although there
patients were >70 years of age. Table 1 shows the treat- was no significant difference in tumor size between the
ment modalities for these patients. About 44% of patients young and elderly HCC patients (P ¼ .15), elderly
received various locoregional therapies for the treatment patients who received TACE treatment for HCC were in
of locally advanced HCC. Among these locoregional earlier stages of disease than young patients (P < .01). In
therapies, TACE was the most commonly used procedure. particular, a higher proportion of elderly patients with
Notably, only about 1=5 of patients received systemic treat- stage 1 disease were offered TACE instead of surgical
ment for the treatment of advanced HCC. The young resection than young patients. Furthermore, there was a
patients more frequently received systemic therapy for the lower frequency of raised serum AFP level in elderly
treatment of advanced HCC than the elderly (23% vs patients than young patients (P ¼ .03). Conversely, there
19%, P < .001). The remaining 1=3 of the nonsurgical were significantly more comorbid illnesses in the elderly
patients received best supportive care only. There was a patients than in young patients (64% vs 33%, P < .01),
significantly higher proportion of elderly patients who especially underlying cardiovascular diseases, respiratory
received best supportive care than young patients (33% vs problems, and diabetes mellitus.
38%, P ¼ .01) Overall, there was no significant difference in the
Table 2 shows the details of the 1040 patients who number of courses of TACE between young and elderly
received TACE treatment for locally advanced HCC. patients (P ¼ .19). Both groups received a median of 2
Among these 1040 patients, 843 patients were 70 years or courses. The median overall survival of all patients with
younger and 197 patients were older than 70 years. There TACE treatment was 8.7 months. However, elderly
were a significantly higher proportion of female HCC patients who received TACE treatment for HCC had a
patients in the elderly group. Moreover, among patients significantly higher overall median survival than young
with data available for hepatitis virus, around 18% of patients (14.0 months vs 8.1 months, P < .01) (Fig. 1).

Cancer December 1, 2009 5509

Original Article

Table 2. Demographic Characteristics of Young and Elderly Hepatocellular Carcinoma

Patients Who Received Transarterial Chemoembolization as Primary Treatment

Patient Characteristics Overall, Age £70 Age >70 P

n51040 years, n5843 years, n5197
Age, median y (range) 61 (17-88) 58 (17-70) 75 (71-88)

Gender
Men 858 (83%) 715 (85%) 143 (73%)
Women 182 (18%) 128 (15%) 54 (27%) <.01

Hepatitis Virus Status

HBV 1 ve, n5989 815 (82%) 703 (87%) 112 (63%) <.01
HCV 1 ve, n5539 64 (12%) 43 (10%) 21 (18%) <.01

AFP (ng/mL)
£400 546 (53%) 429 (51%) 117 (59%)
>400 494 (47%) 414 (49%) 80 (41%) .03

Child-Pugh
A 839 (81%) 674 (80%) 165 (84%)
B 191 (18%) 159 (19%) 32 (16%)
C 10 (1%) 10 (1%) 0 (0%) .2

Tumor size
5 cm 305 (29%) 240 (29%) 65 (33%)
>5 cm 715 (69%) 584 (69%) 131 (66%)
Diffuse 20 (2%) 19 (2%) 1 (1%) .15

New AJCC TNM Staging

Stage I 242 (23%) 175 (21%) 67 (34%)
Stage II 161 (16%) 128 (15%) 33 (17%)
Stage IIIA 559 (54%) 470 (56%) 89 (45%)
Stage IIIB 49 (5%) 41 (5%) 8 (4%)
Stage IIIC 15 (1%) 15 (2%) 0 (0%)
Stage IV 14 (1%) 14 (2%) 0 (0%) <.01
Comorbid condition 408 (39%) 282 (33%) 126 (64%) <.01
Cardiovascular disease 264 (25%) 175 (21%) 89 (45%) <.01
Respiratory disease 137 (13%) 98 (12%) 39 (20%) <.01
Diabetes mellitus 156 (15%) 104 (12%) 52 (26%) <.01

AFP indicates a-fetoprotein; AJCC, American Joint Committee on Cancer.

Table 3. Overall Survival of Patients Who Received TACE

Patients Age£70, Age>70, P

n5843 n5197
Median courses of TACE, No. 2 (1-27) 2 (1-22) .19
6-month survival 58.3% 68.4%
1-year survival 39.2% 54.4%
3-year survival 14.9% 23.2%
5-year survival 8.4% 10.6%
Median overall survival, mo 8.1 14.0 <.003

TACE indicates transarterial chemoembolization.

Also, the elderly patients had a higher survival rate than

the young patients at 1-year, 3-year, and 5-year follow-up
FIGURE 1. A Kaplan-Meier survival curves represent hepato- (Table 3). The main causes of death in patients who
cellular carcinoma patients who received transarterial chemo-
embolization (TACE), stratified by age groups (in years). received TACE are listed in Table 4 and stratified accord-
ing to age group. The disease-specific survival of the 2 age

5510 Cancer December 1, 2009

 TACE in Elderly Patients/Yau et al

Table 4. The Main Causes of Death in Both the Young and Table 5. The Disease-Specific Survival of Patients Who
Elderly Patients Who Received Transarterial Received Transarterial Chemoembolization
Chemoembolization
Age £70 Age >70 P
Cause of Death Age£70 Age>70 years, years,
years, years, n5840 n5196
n5843 n5197
6-month survival 59.8% 71.0%
Alive 120 (14.2%) 34 (17.3%) 1-year survival 40.6% 58.4%
3-year survival 15.9% 27.7%
Disease related 5-year survival 9.5% 13.1%
Malignant cachexia 581 (68.9%) 130 (66.0%) Median overall survival, mo 8.7 15.2 <.001
Bleeding 58 (6.9%) 11 (5.6%)
Hepatic failure 33 (3.9%) 4 (2.0%)
Multiorgan failure 8 (1.0%) 0 (0.0%)
Rupture of HCC 11 (1.3%) 0 (0.0%)

Disease unrelated Table 7 shows TACE-related complications and

Cardiac problem 2 (0.2%) 2 (1.0%) mortality. There was no significant difference in TACE-
Pulmonary problem 5 (0.6%) 3 (1.5%)
2 (0.2%) 2 (1.0%)
related complications between young and elderly patients
Renal failure
Sepsis 6 (0.7%) 3 (1.5%) (27% vs 24%, P ¼ .5). In particular, there were no signifi-
Suicide 1 (0.1%) 0 (0.0%) cant differences in the incidence of liver abscess, acute
Other 13 (1.5%) 7 (3.6%)
Unknown 3 (0.4%) 1 (0.5%) cholecystitis, acute pancreatitis, pancytopenia, renal func-
tion impairment, rupture HCC, and hepatic artery dissec-
HCC indicates hepatocellular carcinoma. tion between these 2 age groups. The most commonly
encountered TACE-related morbidity in both age groups
was liver function derangement. Notably, young patients
groups at 0.5, 1, 3, and 5 years are listed in Table 5. At all had a significant higher rate of liver derangement after
time points, elderly patients had a higher disease-specific TACE than elderly patients (21% vs 12%, P < .01). Fur-
survival than young patients (log-rank test, P < .01). thermore, a significantly higher proportion of young
Moreover, the univariate and multivariate Cox propor- patients than elderly patients had to stop TACE because
tional hazard regression analysis and the stratified analysis of TACE-related liver derangement (16% vs 10%,
of disease-specific survival demonstrated very similar P ¼ .02). Conversely, the elderly patients had a significant
results to those of the overall survival (Table 6). This higher rate of developing peptic ulcer disease with TACE
shows that older age still had a significantly smaller hazard treatment than young patients (2.5% vs 0.5%, P ¼ .01),
for overall survival after adjusting for other risk factors. albeit there was no significant difference in the rate of
The hazard ratio was 0.988 for each additional year. A developing perforation of peptic ulcer (0.2% vs 0.5%,
stratified analysis by American Joint Committee on Can- P ¼ .47). Overall, 34 (3.7% of 919 patients with available
cer (AJCC) stage indicated that the hazard ratio for age data) patients died of TACE complications, and the ma-
was 0.989 (95% confidence interval [CI], 0.979-1.000, jority of TACE-related death was related to hepatic fail-
P ¼ .043) and 0.988 (95% CI, 0.982-0.995, P ¼ .001) ure. There was no significant difference in TACE-related
per additional year for stage I/II and stage III A/B, respec- mortality rate between the young and elderly patients
tively, after adjusting for other risk factors. Conversely, (3.5% vs 4.7%, P ¼ .50).
Figure 2 shows the Kaplan-Meier estimates of overall sur-
vival for younger (
70) and elderly (>70) patients within
the 2 AJCC stage groups. For AJCC stage I/II patients, DISCUSSION
the medium overall survival was 19.9 months and 21.1 Globally, the optimal management of elderly patients
months for younger and elderly patients, respectively with advanced HCC is a pressing issue because of the age-
(P ¼ .93, log-rank test); for AJCC stage IIIA/B patients, ing population and the increasing incidence of HCC. It
the medium overall survival was 5.3 months and 9.0 has been previously shown that elderly patients with
months for younger and elderly patients, respectively HCC were more likely to receive less standard investiga-
(P ¼ .01, log-rank test). tion and more conservative treatment than young patients

Cancer December 1, 2009 5511

Original Article

Table 6. Cox Proportional Hazard Regression Analysis for Overall Survival

Univariate Multivariate Analysisy

Analysis* P
P Hazard Ratio 95.0% CI

Age <.0001 <.0001 0.988 0.982-0.993

Sex .123
Tumor size >5 cm/diffuse <.0001 .003 1.383 1.123-1.702
Child-Pugh B/C .039 <.0001 1.439 1.216-1.702

Tumor staging (AJCC)

Stage I/II 1
Stage IIIA/B <.0001 <.0001 1.723 1.418-2.095
Stage IIIC/IV <.0001 <.0001 3.202 2.107-4.866
Serum AFP >400 ng/mL <.0001 <.0001 1.737 1.510-1.998
Comorbid condition .0002
Diabetes Mellitus .009
CVD .001

CI indicates confidence interval; AJCC, American Joint Committee on Cancer; CVD, cardiovascular disease.
* Sex and ll factors with P < .1 under univariate analysis are shown.
y
The final model by backward selection is shown.

despite similar disease staging.20 In particular, a study There are few data in the literature regarding the
conducted in Japan showed that TACE was less frequently treatment outcome of elderly HCC patients, in particular
performed in elderly patients because of the perceived those patients treated with TACE. Most of these studies
higher treatment-related toxicities.21 As shown in this were conducted about a decade ago with a relatively small
study, TACE is the most commonly used locoregional number of patients included. Moreover, the definition of
therapy for the treatment of locally advanced HCC elderly in these studies is heterogeneous and ranges from
patients at our institution. There was no difference in the 65 to 80 years. Our previous study12 and a Japanese
proportion of patients who received TACE among the 2 study11 suggested comparable efficacy of TACE treatment
age groups at our center. However, similar to the afore- in the young and elderly HCC patient populations. In
mentioned studies, elderly patients were more likely to contrast, a small Italian study showed that the elderly
receive best supportive care than young patients. Of note, patients treated with TACE had poorer outcomes than
there were more female patients in the elderly cohort. young patients.22 Therefore, there is no concrete evidence
This may be related to a larger female elderly population yet in the literature to suggest the efficacy and safety of
in our society because of the longer life expectancy of TACE treatment in elderly patients. In the current study,
women. Furthermore, elderly patients were more likely to 1040 HCC patients treated with TACE were included,
be carriers of HCV infection. This finding may be with 197 patients of age 70 years. On the basis of this
explained by the finding that most hepatitis B carriers ac- large HCC patient population, our results have indicated
quire HBV infection via vertical transmission in the peri- favorable outcomes of elderly HCC patients treated with
natal period, whereas most HCV carriers are infected at a TACE. Although about half and 1=5 of the elderly patients
later stage of life. Thus HCC is manifested as 1 of the have underlying cardiovascular and respiratory problems,
complications of HCV carriers much later than in HBV respectively, they still have similar treatment efficacy and
carriers. Men are far more prone to become the chronic morbidity compared with the young patients. Overall, the
carrier of HBV infection than their female counterparts. median survival of patients treated with TACE at our cen-
Conversely, most of the female patients with HCV- ter was only 8.7 months. This result is apparently lower
related HCC contracted the virus because of blood trans- than the results in the literature.7,8,23 This is mainly
fusion during pregnancy or delivery before the era of attributed to the finding that in the present analysis,
screening for HCV among blood donors in Hong Kong. TACE was used in relatively unselected advanced HCC

5512 Cancer December 1, 2009

 TACE in Elderly Patients/Yau et al

FIGURE 2. Kaplan-Meier estimates are shown of overall survival for younger (
70 years) and elderly (>70 years) patients within
the 2 American Joint Committee on Cancer stage groups.

patients. Thus, the results of TACE are less promising finding that the elderly patients treated with TACE were
when compared with data from prospective clinical trials, in significantly earlier stages of disease than the young
which mostly enrolled a highly selected patient popula- patients. Notably, in our patient cohort, 34% of the el-
tion. Interestingly, in contrast to our previous results12 derly patients had stage I disease. Elderly patients are gen-
and other studies,11,21 we have demonstrated that the erally considered at a higher risk for surgical
overall survival of the elderly patients who underwent interventions, and thus TACE was used as an alternative
TACE was seemingly better than the young patients. for the treatment of elderly patients with early stage dis-
There are 2 main reasons to account for this interesting ease. In contrast, only 21% of young patients had received
result. First, the better survival is accounted for by the TACE for the treatment of stage I disease, as young HCC

Cancer December 1, 2009 5513

Original Article

Table 7. TACE-Related Morbidity and Mortality related morbidity. In the literature, TACE is known to be
associated with peptic ulcers,24 albeit advanced age is not
Age £70, Age >70, P
n5843 n5197 a known risk. The results from our study show that the
overall incidence of peptic ulcers after TACE is low. How-
Median courses of TACE, No. 2 (1-27) 2 (1-22) .19
ever, the elderly patients were more prone to develop pep-
Complications
Overall 227 (26.9%) 48 (24.4%) .53
tic ulcers than the young patients. This finding may
Liver abscess 6 (0.7%) 4 (2.0%) .10 suggest the benefits in prescribing prophylactic proton
Acute cholecystitis 1 (0.1%) 0 (0%) 1.0
pump inhibitor to prevent the occurrence of ulcers in el-
Acute pancreatitis 1 (0.1%) 0 (0%) 1.0
Peptic ulcers 4 (0.5%) 5 (2.5%) .01 derly HCC patients receiving TACE treatment. Con-
Perforation of peptic ulcer 2 (0.2%) 1 (0.5%) .47 versely, a significantly higher proportion of young HCC
Liver function derangement 181 (22%) 23 (12%) <.01
Liver function derangement 136 (16%) 19 (10%) .02
patients had liver function derangement after TACE than
led to termination of TACE elderly patients, and a great majority of them had to stop
Renal impairment 8 (1.0%) 5 (2.5%) .08
TACE because of deteriorating liver function. This phe-
Rupture HCC 10 (1.2%) 3 (1.5%) .72
Pancytopenia 2 (0.2%) 0 (0%) 1.0 nomenon may be partly because of the finding that
Hepatic artery dissection 0 (0%) 1 (0.5%) .19 younger patients usually have poorer liver function before
TACE-related mortality TACE. Thus, this finding highlights the importance of
Total, n5919 26 (3.5%) 8 (4.7%) .50
vigilant monitoring of liver function during TACE, par-
Hepatic failure 22 3
Rupture HCC 3 1 ticularly in the young patient population.
Bleeding from varices/ulcers 5 3 In conclusion, the current study has indicated the
comparable efficacy and tolerability of TACE for the
TACE indicates transarterial chemoembolization; HCC, hepatocellular treatment of advanced HCC patients in both young age
carcinoma.
and elderly HCC patient populations under current prac-
tice in determination of the optimal treatments. Advanced
patients are more aggressively treated by surgical interven- age is not a contraindication for TACE treatment. With
tions. Even in stage IV disease or in Child-Pugh C cirrho- future incorporation of targeted therapy with TACE, the
sis, a small proportion of the young patients still received survival of elderly patients with locally advanced HCC
TACE treatment at our center. Second, our findings may may incrementally improve in the near future.
also suggest that elderly HCC patients have distinct de-
mographic and clinical characteristics when compared Conflict of Interest Disclosures
with the young patients. Their disease is biologically less Dr. Thomas Yau has participated on advisory boards for Bayer-
aggressive than that of young patients, and thus they have Schering and GlaxoSmithKline. Professor Ronnie T. P. Poon
has participated on advisory boards for GlaxoSmithKline and
better treatment outcomes. Indeed, studies12,13 have sug- Bayer-Schering.
gested a different mechanism of hepatocarcinogenesis in
these 2 age groups; elderly HCC patients have been asso-
ciated with more clinicopathologically favorable factors,
References
such as smaller size of the tumors, higher incidence of
encapsulated tumors, and lower incidence of satellite nod- 1. Parkin DM, Bray F, Ferlay J, Pisani P. Estimating the
world cancer burden: Globocan 2000. Int J Cancer. 2001;
ules. However, it is important to note that the elderly 94:153-156.
cohort patients included in the analysis are highly selected, 2. El-Serag HB, Davila JA, Petersen NJ, McGlynn KA. The
as they are likely to have an earlier stage of disease and bet- continuing increase in the incidence of hepatocellular carci-
ter liver function. noma in the United States: an update. Ann Intern Med.
2003;139:817-823.
Despite the finding that the elderly patients have sig-
3. Ries LAG, Kosary CL, Hankey BF, et al. SEER. Cancer
nificantly more underlying medical conditions, there is no
Statistics Review. Bethesda, MD: National Cancer Institute;
difference in TACE-related mortality between these 2 age 2000:1973-1997.
groups. Nonetheless, in contrast to our previous report,12 4. Cho SJ, Yoon JH, Hwang SS, Lee HS. Do young hepato-
these 2 age groups have a significant difference in TACE- cellular carcinoma patients with relatively good liver

5514 Cancer December 1, 2009

 TACE in Elderly Patients/Yau et al

function have poorer outcomes than elderly patients? J Gas- 15. Mazzaferro V, Regalia E, Doci R, et al. Liver transplanta-
troenterol Hepatol. 2007;22:1226-1231. tion for the treatment of small hepatocellular carcinomas
5. Hutchins LF, Unger JM, Crowley JJ, Coltman CA Jr, in patients with cirrhosis. N Engl J Med. 1996;334:693-
Albain KS. Underrepresentation of patients 65 years of age 699.
or older in cancer-treatment trials. N Engl J Med. 1999; 16. Yao FY, Ferrell L, Bass NM, et al. Liver transplantation
341:2061-2067. for hepatocellular carcinoma: expansion of the tumor size
6. Earle CC, Venditti LN, Neumann PJ, et al. Who gets limits does not adversely impact survival. Hepatology.
chemotherapy for metastatic lung cancer? Chest. 2000; 2001;33:1394-1403.
117:1239-1246. 17. Kasugai H, Kojima J, Tatsuta M, et al. Treatment of he-
7. Lo CM, Ngan H, Tso WK, et al. Randomized controlled patocellular carcinoma by transcatheter arterial embolization
trial of transarterial lipiodol chemoembolization for unre- combined with intraarterial infusion of a mixture of cispla-
sectable hepatocellular carcinoma. Hepatology. 2002; tin and ethiodized oil. Gastroenterology. 1989;97:965-971.
35:1164-1171. 18. Poon RT, Ngan H, Lo CM, Liu CL, Fan ST, Wong J.
8. Llovet JM, Real MI, Montana X, et al. Arterial embolisa- Transarterial chemoembolization for inoperable hepatocellu-
tion or chemoembolisation versus symptomatic treatment lar carcinoma and postresection intrahepatic recurrence. J
in patients with unresectable hepatocellular carcinoma: a Surg Oncol. 2000;73:109-114.
randomised controlled trial. Lancet. 2002;359:1734-1739. 19. Kalbfleisch JD, Prentice RL. The Statistical Analysis of
9. Llovet JM, Bruix J. Systematic review of randomized trials Failure Time Data. New York, NY: John Wiley & Sons;
for unresectable hepatocellular carcinoma: chemoemboliza- 1980.
tion improves survival. Hepatology. 2003;37:429-442. 20. Collier JD, Curless R, Bassendine MF, James OF. Clinical
10. Mondazzi L, Bottelli R, Brambilla G, et al. Transarterial features and prognosis of hepatocellular carcinoma in Brit-
oily chemoembolization for the treatment of hepatocellular ain in relation to age. Age Ageing. 1994;23:22-27.
carcinoma: a multivariate analysis of prognostic factors. 21. Dohmen K, Shirahama M, Shigematsu H, Irie K, Ishibashi
Hepatology. 1994;19:1115-1123. H. Optimal treatment strategy for elderly patients with
11. Hoshida Y, Ikeda K, Saito S, et al. The efficacy and prog- hepatocellular carcinoma. J Gastroenterol Hepatol. 2004;
nosis of transcatheter chemoembolization for hepatocellular 19:859-865.
carcinoma in the elderly [in Japanese]. Nippon Shokakibyo 22. Biselli M, Forti P, Mucci F, et al. Chemoembolization
Gakkai Zasshi. 1999;96:142-146. versus chemotherapy in elderly patients with unresectable
12. Poon RT, Fan ST, Lo CM, et al. Hepatocellular carci- hepatocellular carcinoma and contrast uptake as prognostic
noma in the elderly: results of surgical and nonsurgical factor. J Gerontol A Biol Sci Med Sci. 1997;52:M305-
management. Am J Gastroenterol. 1999;94:2460-2466. M309.
13. Nomura F, Ohnishi K, Honda M, Satomura Y, Nakai T, 23. Takayasu K, Arii S, Ikai I, et al. Prospective cohort study
Okuda K. Clinical features of hepatocellular carcinoma in of transarterial chemoembolization for unresectable hepato-
the elderly: a study of 91 patients older than 70 years. Br J cellular carcinoma in 8510 patients. Gastroenterology. 2006;
Cancer. 1994;70:690-693. 131:461-469.
14. Nagasue N, Chang YC, Takemoto Y, Taniura H, Kohno 24. Su YC, Chen LT, Jan CM, et al. Reappraisal of gastroduo-
H, Nakamura T. Liver resection in the aged (seventy years denal lesions after transcatheter arterial chemoembolization
or older) with hepatocellular carcinoma. Surgery. 1993; of liver neoplasms—selective versus superselective method.
113:148-154. J Clin Gastroenterol. 1994;18:118-121.

Cancer December 1, 2009 5515