| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Diagnostic and Interventional Radiology, University of Mainz,
Langenbeckstr. 1, 55124 Mainz, Germany.
2 Department of Hepatobiliary Surgery, University of Mainz, Mainz,
Germany.
3 I. Medical Clinic, University of Mainz, Mainz, Germany.
Received June 20, 2007;
accepted after revision October 24, 2007.
Address correspondence to S. C. A. Herber
(s.herber{at}kk-koblenz.de).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSIONS. Prognosis depends on local response, Okuda score,
-fetoprotein level, and tumor size and is independent of the presence
of portal venous thrombosis.
Keywords: hepatocellular carcinoma • Kaplan-Meier • mitomycin C • transarterial chemoembolization
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We report our results with patients treated over a 5-year period who were not eligible for liver transplantation. Almost one third of the patients were found to have a diffusely infiltrating tumor, and almost one third had partial or complete portal venous thrombosis at the initial CT evaluation. The aim of this study was to evaluate our experience and results with a heterogeneous cohort of patients undergoing palliative therapy. This cohort encompasses patients with a more favorable prognosis because of limited disease and patients with a so-called poor prognosis because of tumors in advanced stages.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-fetoprotein level (> 400 µg/L).
|
The treatment protocol consisted of repeated TACE treatments with a mixture of mitomycin C and iodized oil (Lipiodol Ultra Fluid, Guerbet) at an interval of 8 ± 2 weeks between procedures. A total of 688 TACE procedures (mean, 4.6 ± 3.55 [SD] per patient; 95% CI, 3.9–4.9; range, 1–17) were performed during the study period. A one-time TACE procedure was performed on 34 (22.8%) of the patients, two or three procedures on 38 (25.5%) of the patients, and more than three procedures on 77 (51.7%) patients.
Patients were included who had untreated HCC not suitable for curative
treatment. The patients were not eligible for liver transplantation because of
advanced age (> 65 years), tumor size or growth exceeding the Milan
criteria (less than three tumors with a maximum diameter 
3.0 cm or one
tumor with a maximum diameter 5.0 cm), or the presence of a severe
comorbid condition.
Patients were excluded from TACE if they had advanced liver disease (Child-Pugh class C; Okuda score III), a serum bilirubin concentration greater than 3.0 mg/dL, bacterial infection, extrahepatic tumor spread, encephalopathy, portosystemic shunt, or any contraindication to an arterial procedure (clotting test result showing impairment, platelet count < 50,000/mm3, prothrombin activity < 50%). They were excluded in the case of renal failure (serum creatinine concentration > 1.8 mg/dL), leukocyte count less than 3 x 109, and refractory ascites not controlled with diuretics. In patients with portal venous thrombosis, an individual decision was made for or against TACE on the basis of liver function (absence of Child-Pugh class B or C disease), performance status, and the vote of an interdisciplinary tumor board. Approval for TACE as palliative therapy for inoperable liver cancer was granted by the institutional investigation and ethics committee.
TACE was stopped if one of the following conditions was present: progressive tumor disease, occlusion of the common hepatic artery or of the tumor-feeding arteries, extrahepatic tumor spread, or progressive hepatic decompensation (Child-Pugh class C; Okuda score III).
Imaging
CT images were acquired within 24 hours before each TACE treatment. The
investigation protocol included dynamic contrast-enhanced helical CT with a
4-MDCT scanner (Volume Zoom, Siemens Medical Solutions). An unenhanced CT scan
was followed by arterial and portal venous phase acquisitions. Investigation
parameters for the three phases were as follows: slice thickness, 3 mm;
increment, 2.5 mm; collimation, 2 mm; 150 mL IV iomeprol (Imeron 300, Bracco);
50 mL saline bolus; flow, 5 mL/s; 150 mAs at 120 kV. After each TACE session,
unenhanced CT was performed on the same or the next day to document the
iodized oil distribution and to exclude displacement of the embolizing
agent.
Technical Procedure
Transfemoral access with a 5-French catheter sheath was selected for all
patients, and a 5-French standard catheter (Sidewinder or Cobra, Terumo) was
used for angiography. The first step was to document the arterial supply to
the liver. The mesenteric artery was examined to visualize any atypical blood
supply to the liver. The common hepatic artery then was cannulated, and
according to the individual anatomic disposition, the catheter was placed in a
position distal to the origin of the gastroduodenal artery in the common
hepatic artery or in the right or left lobar artery. If a stable catheter
position was not achieved at this level, a 2.2-French microcatheter was used
for superselective intubation of the segmental vessels.
TACE was performed with a suspension of a maximum of 20 mL iodized oil (Lipiodol Ultra Fluid, Guerbet) and mitomycin C (10 mg) mixed immediately before administration. In the case of bilateral tumor spread, the total dose was distributed under continuous fluoroscopic guidance to both liver lobes (two thirds of the total dose to the right lobe, one third to the left); in cases of unilateral tumor growth, the dose was administered only to the right or the left lobar artery. The treatment was terminated after the total dose was administered or stasis or reflux was attained in the appropriate section. The patients were usually hospitalized for one night and were dismissed the next day if no complications developed. Nausea, pain, or fever that occurred in association with the TACE treatment was managed symptomatically with antiemetics (e.g., tropisetron 5–15 mg IV), analgesics (e.g., metamizole IV), or antipyretics (e.g., paracetamol 500–1,000 mg orally). No antibiotic prophylaxis was given.
Laboratory Studies and Clinical Findings
Biochemical and tumor markers were checked every 8 ± 2 weeks. In all
patients, serum -fetoprotein, transaminase (aspartate amino
transferase, alanine amino transferase), and serum cholinesterase levels,
blood count, and coagulation values (partial thromboplastin time, thrombin
time, Quick's test result or international normalized ratio) were assessed.
The degree of cirrhosis was classified according to Child-Pugh class
(A–C) and Okuda score (I–III). Liver function was impaired in
23.5% of the patients (Child-Pugh class B). An Okuda score of II was
documented in 50.3% of the patients and an Okuda score of I in 49.7%
(Table 1).
CT Evaluation
The number of tumors and the maximal transverse and longitudinal diameters
of the marker lesion in the disease process were assessed through synopsis of
the contrast-enhanced CT images. For patients who received TACE more than six
times, an interim analysis was performed after three procedures. Tumor
response was assessed on the basis of the cross-sectional CT findings, and
interim best response to therapy was assessed according to the World Health
Organization classification
[8]. The maximum diameters of
the largest lesion were recorded and compared with the measurements on the
last available follow-up CT scans. Progressive disease was considered more
than a 25% increase in maximum tumor diameter or the presence of new tumor;
stable disease, less than 25% decrease or an increase in tumor size; minor
response, 26–50% decrease in maximum tumor size; partial response, more
than 50% decrease in maximum tumor size; complete remission, no tumor found.
Iodized oil accumulation was classified as 0–25%, 26–50%,
51–75%, and 76–100% on a visual ana log scale on postintervention
CT scans. In addi tion, correct distribution of embolization agent was
confirmed on a postintervention unenhanced CT scan.
Tumor attenuation was analyzed with a visual analog score in the arterial phase and classified as high, low, or mixed. Patency of the portal vein and hepatic arteries and veins was determined on the preinterventional CT scan and during catheter angiography. Infiltration of the liver capsule was suggested if the tumor had contact with the liver capsule over a longer distance [9]. On the follow-up CT scans, presence or absence of pathologic lymph nodes was checked, as was extrahepatic tumor spread. Lymphatic metastasis was suggested if lymph nodes in the hepatoduodenal ligament were not suspected at previous evaluation and had grown to a diameter of more than 2.0 cm in the course of disease without evidence of regression in the subsequent course.
Statistical Analysis
The database was analyzed with the statistical program SPSS version 12.0
for Windows (SPSS). Continuous ordinal data were expressed as mean ±
SD, and qualitative data as frequency and rate. The statistical significance
of quantitative data was determined with the Mann-Whitney test, and categoric
data were compared with use of the chi-square test. Univariate analysis to
identify predictors of survival was performed with the Kaplan-Meier method for
the baseline variables. The log-rank test was used to describe significant
differences. On the basis of the results of the univariate analysis, factors
found significant or considered important were subjected to multivariate
analysis with a Cox proportional hazards model. Two-tailed p <
0.05 was considered statistically significant. All statistical tests were
performed bilaterally.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Course of Disease
At the end of the investigation period, 16.8% (n = 25) of the
patients were alive, and 83.2% (n = 124) had died
(Table 3). Follow-up CT scans
were available for 91.9% (137/149) of the patients. Stable disease,
regression, or progression was found in 65.9%, 14.1%, and 18.8% of the
patients, respectively. The mean size of the marker lesion was 6.3 ±
3.8 cm (95% CI, 5.3–6.4 cm; range, 0.8–17.0 cm) before TACE and
6.5 ± 4.8 cm (95% CI, 5.3–6.7 cm; range, 0.8–20.9 cm) after
the last TACE procedure. After the last TACE session, extrahepatic tumor
manifestation was suspected in 48.3% of the patients. Presumably malignant
intraabdominal lymph nodes were found in 23.5% and lung metastasis in 15.4% of
the patients. The cause of the deaths of 6.5% of the patients remained
unclear. Approximately three fourths (74.2%) of the patients probably died of
tumor disease, but in 13.7% of the cases, progressive liver decompensation was
considered the cause of death. Three patients died of esophageal variceal
bleeding.
|
Survival and Prognosis
The overall 1-, 2-, and 3-year survival rates for the total cohort were
53.6%, 24.6%, and 13.2%, respectively, and the mean survival time was 18.3
months (95% CI, 15.0–21.4 months)
(Fig. 1). Results of univariate
and multivariate analyses (Tables
4,
5,
6) showed a significant
correlation between death and response to TACE (response or stable disease vs
tumor progression, p = 0.013)
(Fig. 2). Response depended on
tumor size and the number of tumors. The response was significantly better for
small tumors than for larger tumor nodules (p < 0.0001). In
addition, response was significantly better in hypervascular lesions than in
lesions with mixed or low attenuation (p = 0.007). A significant
benefit was found among patients with limited disease (p = 0.02). The
1-, 2- and 3-year survival rates for patients with unifocal tumors were
calculated to be 68.4%, 46.1%, and 25.6%. Patients with multifocal diffusely
infiltrating tumors had overall survival rates of only 37.5%, 16.6%, and
9.2%.
|
|
|
|
|
Initial serum -fetoprotein level was related to extension of the
tumor disease (p = 0.024). Univariate and multivariate analyses
showed a significant correlation between elevation of the serum
-fetoprotein level and course of disease. Patients with elevated (>
10 µg/L) -fetoprotein levels had a significantly worse outcome.
Furthermore, functional liver status had a major influence on long-term
survival. Results of univariate and multivariate analyses showed that Okuda
score had a significant influence on survival. Patients with an Okuda score of
II at the first CT evaluation had 2- and 3-year survival rates of only 16.1%
and 8.5% (Fig. 3).
|
-fetoprotein and serum bilirubin levels (p < 0.05) also
were observed. Serum cholinesterase levels were not significantly different
before and after TACE.
|
Clinically apparent complications developed in 35 (23.5%) of the patients (Table 8). Even after repeated treatments, permanent occlusion of the common hepatic artery or right or left lobar artery forcing interruption of the TACE procedure occurred in only 2.7% of the patients. Distinct hepatic arterial spasm was found in 6.8% of the patients. In 4.7% of patients who underwent repeated treatments, early stasis or reflux developed after injection of the embolic agent. One case of dissection of the common hepatic artery necessitating interruption of treatment was documented. On control angiography, however, the situation had improved, so the treatment was resumed. Severe vomiting necessitating prolonged hospitalization occurred in 7.4% of the patients. Displacement of iodized oil into the gastroduodenal artery or gastric branches with subsequent severe pancreatitis in one case and gastric ulcer in another was found in 1.3% of the patients. No deaths associated with the TACE treatment occurred.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most (86.6%) patients have multifocal tumor disease and are not eligible for liver transplantation. Therefore the aim of our embolization protocol, which is similar to that of other authors [2], was not superselective treatment of only tumor-feeding vessels. In the first treatment session, we routinely administered some of the embolization material to presumably uninvolved regions. This procedure was based on histopathologic observations and experience that in as many as one third of patients, additional tumor nodules are found in surgical specimens that are not recognized on CT or MRI [12–14].
Even though many authors have proposed using embolizing agents, such as absorbable gelatin sponge, polyvinyl alcohol particles, and microspheres, after administering the iodized oil–chemotherapeutic suspension, there are concerns about provoking early occlusion of the hepatic arteries [9, 15, 16]. Because our treatment protocol provided for regular sequential treatment, which is considered superior to a one-time procedure [11, 17–19], we abstained from using particles or microspheres.
The probability of survival for the total cohort was even worse than that reported in previous studies [1, 2, 9, 20, 21]. In 2002 Llovet and colleagues [1] conducted a randomized controlled trial in which results among patients treated with TACE were compared with those among patients who received only the best supportive care. The authors reported favorable cumulative 1-, 2-, and 3-year survival rates of 82%, 63%, and 29% among patients treated with chemoembolization. Another randomized trial in 2002, by Lo et al. [2], had 1-, 2-, and 3-year survival rates of 57%, 31%, and 26%. In our cohort, the 1-year survival was 53.6%, and the 2-year and 3-year survival rates were only 24.6% and 13.2%. Llovet et al. explained their very positive findings by the use of restrictive selection criteria. They included only patients with predominantly preserved liver function and a limited tumor load. They excluded patients with factors considered less favorable for outcome, such as age older than 75 years and the presence of diffusely infiltrating neoplasms, portal venous thrombosis, or vascular invasion. Unlike Llovet and colleagues, we did not formulate such restrictive selection criteria from the start. We did not even exclude elderly patients or patients with portal venous thrombosis or large tumors.
In line with the findings of Tsukioka et al. [22], patient age did not have a negative influence on outcome after TACE in our cohort. However, as in another study [9], univariate analysis revealed a significantly worse survival rate among patients with portal venous thrombosis. As in other studies [23, 24], portal venous thrombosis was present in almost one third of the patients in our cohort. As in still other studies [20, 25, 26], the mean 3-year survival rate was poor. None of the patients with portal venous thrombosis survived 3 years, and only 12% survived 2 years or more. In our cohort, however, the 1-year survival rate among patients with portal venous thrombosis was comparable with that in a study by Uraki et al. [27] in 2004. Those investigators found a mean survival time of 15 months among patients presenting with portal venous thrombosis; the 1-year survival rate was 42%.
In contrast to studies conducted by others [15, 28, 29], multivariate analysis in our study did not reveal a statistically significant difference in survival among patients with and those without portal venous thrombosis. One explanation may be the circumstance that patients who died very early of advanced tumor disease were included in the total cohort, so the influence of portal venous thrombosis was masked.
Even though the danger of provoking acute liver failure is estimated to be quite small if liver function is well preserved [30], there are concerns about the long-term consequences for hepatic function. As others [20, 31, 32] have found, TACE proved safe for our patients, even those with complete or partial thrombosis of the portal vein. Frequency of death within 30 days after TACE did not reach statistical significance. We found no consistent data in the literature about the frequency of acute or chronic liver failure after TACE.
Although we excluded patients with disease in Child-Pugh class C or with an Okuda score of III, in more than one half of our patients, disease was classified as having an Okuda score of II. We found distinct changes between the first and last control laboratory evaluations. Liver function, characterized by bilirubin level, serum protein level, and prothrombin ratio, worsened significantly in our cohort. As a sign of progressive liver decompensation, ascites was observed significantly more often on the last control CT scan than on the initial scan. Retrospective analysis showed that death was related to liver decompensation and not to tumor progression in only 13.5% of the cases. This finding is almost identical to the results presented by Kiely and colleagues [21] in 2006. Those authors proposed that TACE may be safe even for patients with limited liver volume. Although survival tended to be shorter among patients with Child-Pugh class B disease, statistical analysis showed no significant difference. As in a study by Lo and colleagues [2], however, univariate and multivariate analyses showed Okuda score to be a relevant predictor of the course of disease. The probability of survival was significantly better among patients with Okuda I disease than those with Okuda II disease.
In addition to Okuda score at the first treatment session, individual response to the procedure had a highly significant effect on patient survival, as has been found by other authors [1, 27, 33]. Adequate local tumor control and stable or regressive tumor disease were associated with a favorable long-term prognosis significantly more often than in patients with tumor progression while being treated with TACE. In addition, tumor control in the whole liver, particularly suppression of new tumor nodules, was a crucial criterion in prognosis. Patients with no evidence of new tumors in addition to the initially documented tumor nodules had a clearly better outcome. Cox regression analysis showed that response to TACE and lack of new tumors are key to a more favorable outcome. As have other study groups, we found a close correlation between response or treatment failure and the extent and number of tumors, vascularization of the nodule, and uptake of iodized oil [1–2, 9, 21, 33]. TACE was less promising in the management of large tumors, poorly vascularized lesions, multifocal disease, and poor iodized oil accumulation within the tumor (p < 0.05).
As found by Huppert et al.
[9], local tumor progression
was associated significantly more often with massive (> 50%) diffusely
infiltrating tumors. The threshold for serum -feto protein
concentration in our analysis was 100 µg/L. When the -fetoprotein
level exceeded this limit, we found a significantly worse outcome than among
patients with -fetoprotein levels that did not exceed this limit in the
univariate analysis. Multivariate analysis with the Cox model revealed
-fetoprotein value was an independent prognostic factor, as found in
other studies [15,
28,
34,
35].
We are aware of the limitations of our study. First, there might have been
selection bias in that only patients not eligible for other treatments were
included and thus represented a negative proportion of all the HCC patients
treated during the last 5 years. Second, there was no randomized control group
for comparison. Nevertheless, our findings from a single university hospital
emphasize the value of palliative TACE treatment of these patients. Major
benefit was found particularly among patients who responded to treatment.
Initial -fetoprotein level and Okuda score of liver function were
considered further independent prognostic factors for survival. We found that
even with portal venous thrombosis, patients may profit from TACE and should
not be routinely excluded from this treatment.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
