| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu
University, Higashi-ku, Fukuoka, Japan.
2 Present address: Department of Radiology, University of Iowa Hospitals and
Clinics, 200 Hawkins Dr., Iowa City, IA 52242.
3 Department of Anatomic Pathology, Graduate School of Medical Sciences, Kyushu
University, Higashi-ku, Fukuoka, Japan.
4 Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu
University, Higashi-ku, Fukuoka, Japan.
Received February 23, 2007;
accepted after revision August 7, 2007.
WEB This is a Web exclusive article.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Sixty HCC nodules were confirmed at pathology in 59 patients who had undergone CT hepatic arteriography and CT during arterioportography (CTAP). Lesions were classified into one of the four groups: group 1, nodules that appeared to show preserved portal perfusion on CTAP and showed hypo- or isoattenuation on CT hepatic arteriography; group 2, very hyperattenuating on CT hepatic arteriography; group 3, hyperattenuating on CT hepatic arteriography; and group 4, hypo- or isoattenuating on CT hepatic arteriography. Groups 2, 3, and 4 showed the portal perfusion defect on CTAP. These hemodynamic patterns were compared among different histologic grades. We also examined the number of unpaired arteries and the Ki-67 labeling index of the tumor pathologically.
RESULTS. The numbers of lesions in each group were as follows for groups 1, 2, 3, and 4, respectively: 17, one, two, and 0 well-differentiated HCCs; 0, 10, nine, and one moderately differentiated HCC; and 0, three, 12, and five poorly differentiated HCCs. The lesions showed significantly different hemodynamic patterns among different histologic grades (Cramer's V = 0.6919, p < 0.0001). The number of unpaired arteries showed a strong correlation with Ki-67 labeling index in well-differentiated HCC and moderately differentiated HCC (rho = 0.673, p < 0.0001) and a moderate inverse correlation with Ki-67 labeling index in poorly differentiated HCC (rho = -0.540, p = 0.0185).
CONCLUSION. In the late stage of HCC development, the arterial blood supply significantly decreases as the histologic grade progresses.
Keywords: CT • hemodynamics • hepatocellular carcinoma • liver disease • proliferative activity • unpaired artery • vascular imaging
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
During hepatocarcinogenesis, most hepatocellular nodules show a deterioration of arterial blood flow before the loss of portal blood flow. Subsequently, as a result of a marked increase in neovascularized arteries, arterial blood flow becomes dominant. The portal blood flow decreases with the advancement of tumor, and the tumor is eventually fed mainly by arterial flow [3]. The changes of intranodular blood supply in a regenerative nodule, dysplastic nodule, and well-differentiated HCC have been well described [2-4].
Advanced HCC usually indicates moderately differentiated HCC or poorly differentiated HCC, and these two have been studied inseparably in the radiology literature until now [3, 4]. The hemodynamic changes in moderately differentiated HCC and poorly differentiated HCC have not been well evaluated radiologically or pathologically in a previous study, to our knowledge. The survival of patients with poorly differentiated HCC, however, is reportedly worse than that of those with moderately differentiated HCC in surgically resected cases [5]. In particular, Jonas et al. [6] revealed that the survival of patients with poorly differentiated HCC was significantly worse than survival of patients with well- and moderately differentiated HCC after liver transplantation. Therefore, it would be of clinical benefit if these two groups could be differentiated clinically or radiologically. We have previously reported that some poorly differentiated HCCs showed a decreased arterial blood supply [7].
In terms of hepatocarcinogenesis, the number of unpaired arteries is substantially greater in HCCs than in cirrhotic nodules or dysplastic nodules and tends to correlate with the degree of arterial contrast enhancement [2, 3, 8].
The most widely used proliferation-associated marker is Ki-67. Ki-67 is a nuclear antigen present only in proliferating cells. A detailed cell cycle analysis showed that the Ki-67 antigen is expressed in cells during the G1, S, and G1-M phases but not during the G0 phase of the cell cycle [9, 10]. The diagnostic and prognostic value of Ki-67 immunostaining of human tumors has been well documented [11]. In HCC, Ki-67 expression was found to correlate closely with tumor growth rate [12] and was an independent prognostic indicator of patient disease-free and overall survival rates [13].
In this study, we hypothesized that arterial flow might, in general, decrease in the late stage of HCC development (moderately differentiated HCC to poorly differentiated HCC), whereas it increases in the early stage (well-differentiated HCC to moderately differentiated HCC), as previously described [2-4]. The purpose of this study was to evaluate the hemodynamic patterns of HCCs on CT hepatic arteriography and CTAP during the development of HCC, with a particular emphasis on its late stage, in correlation with the number of unpaired arteries and the expression of Ki-67 (proliferative activity index).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tumor size was not taken into consideration for inclusion criteria. The mean size (± SD) of the well-differentiated HCCs, moderately differentiated HCCs, and poorly differentiated HCCs was 1.8 ± 0.6, 2.6 ± 1.5, and 4.2 ± 2.9 cm, respectively. The mean time interval between imaging (i.e., CT hepatic arteriography and CTAP) and surgical resection was 14.2 ± 4.2 days. Our institutional review board does not require its approval or informed consent for the retrospective evaluation of patients' records and images.
CT Protocols
CT hepatic arteriography and CTAP were performed using a 4-MDCT unit
(Aquilion, Toshiba) immediately after unenhanced scanning of the liver. The
parameters for scanning were as follows: collimation, 3 mm; reconstruction, 5
mm; pitch, 5.5: amperage, 300 mAs; kilovoltage, 120 kVp; and table speed, 16.5
mm per rotation. For CTAP, data acquisition was started 30 seconds after the
initiation of a transcatheter arterial injection of 100 mL of nonionic
contrast material (iopamidol, 150 mg I/mL [Iopamiron 150, Bayer HealthCare])
at 2.5 mL/s using an automated power injector. Before the contrast material
was injected, a transarterial infusion of prostaglandin E1 (10 mg)
as a vasodilator was performed.
Data acquisition of CT hepatic arteriography began 15 (first phase) and 30 (second phase) seconds after the initiation of a transcatheter hepatic arterial injection of 20-50 mL of nonionic contrast material at a rate of 1-2.5 mL/s using an automated power injector. The appropriate injection rate for CT hepatic arteriography was determined to be the maximum injection rate, which basically depended on the vessel caliber, that would not cause backward flow of contrast material on hepatic arteriography. The duration of the arterial injection was 20 seconds.
Pathologic Evaluation
For all 60 nodules, two experienced pathologists with 2 and 7 years of
experience with liver pathology examined the resected specimens with no
knowledge of the preoperative CT findings in a consensus fashion. Unpaired
arteries in neoangiogenesis were not associated with the portal tract, fibrous
tissue, septa, or other units (e.g., bile duct). The unpaired arteries were
assessed according to established methods of vascular counting as follows
[15]: the three
neovascularized areas (hot spots) were originally identified at low
magnification (x40) and were circled in pen, then the absolute number of
unpaired arteries was counted in a single field at the highest magnification
(x200) within every hot spot area.
Immunohistochemistry
Ki-67 was detected immunohistochemically by the streptavidin-biotin
peroxidase method [16] using
anti-Ki-67 (MIB-1, DakoCytomation). To measure the Ki-67 labeling index, more
than 1,000 cells were counted in randomly selected high-power fields, and the
number of Ki-67-positive nuclei was recorded in each field. The labeling index
was defined as the ratio of Ki-67-labeled cells to the total number of cells
counted and was expressed as percentages.
Assessment
Images of the largest axial sections of the tumor were evaluated by two
radiologists with 11 and 19 years of experience with CT of the liver in a
consensus fashion. All interpretations were independently made from the axial
CT images on a monitor (Coronis 3MP, Barco), and the observers were completely
blinded to the pathologic and clinical findings for each case. On the monitor,
images obtained with routine window settings (window level, 100 H; window
width, 500 H) were viewed. All lesions were classified in one of the following
groups on the basis of the enhancement patterns observed on CT hepatic
arteriography and CTAP: group 1, nodules that showed preserved portal
perfusion on CTAP and were hypo- or isoattenuating on CT hepatic arteriography
(Fig. 1A,
1B); group 2, portal perfusion
defect on CTAP and nodules were very hyperattenuating compared with
noncancerous liver on CT hepatic arteriography (Fig.
2A,
2B); group 3, portal perfusion
defect on CTAP and nodules were hyperattenuating compared with noncancerous
liver on CT hepatic arteriography (Fig.
3A,
3B); and group 4, portal
perfusion defect on CTAP and nodules were hypo- or isoattenuating compared
with noncancerous liver on CT hepatic arteriography (Fig.
4A,
4B,
4C).
|
|
|
|
|
|
|
|
|
Statistical Analysis
Continuous variables were expressed as means ± SD and tested using
the Student's t test. The level of significance was set at a
p value of less than 0.05 for all tests. Cramer's V with chi-square
was used for radiologic pattern and tumor differentiation. The value of the
level of association was interpreted as follows: 0.00, no relationship;
0.00-0.10, not generally useful; 0.10-0.20, weak; > 0.20-0.25, moderate;
> 0.25-0.30, moderately strong; > 0.30-0.35, strong; > 0.35-0.40,
very strong; > 0.40-0.45, worrisomely strong; > 0.45-9.99, redundant;
and 1.00, perfect relationship. Spearman's rank correlation coefficient test
was used to determine the correlation between the number of unpaired arteries
and the Ki-67 labeling index. A correlation coefficient value of up to 0.20
showed virtually no correlation; 0.21-0.40, weak correlation; 0.41-0.60,
moderate correlation; 0.61-0.80, high correlation; and 0.81 or greater, very
high correlation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
The relationship between the radiologic findings and the Ki-67 labeling index of all HCCs is shown in Figure 5. The Ki-67 labeling indexes of group 2 (9.8 ± 8.3), group 3 (16.4 ± 16.4), and group 4 (35.1 ± 19.6) were higher than that of group 1 (7.1 ± 5.5) (p = 0.0002, p = 0.0005, and p < 0.0001, respectively). The Ki-67 labeling index of group 4 was higher than those of group 2 and group 3 (p = 0.0006 and 0.0244, respectively). The Ki-67 labeling index of group 3 tended to be higher than that of group 2, but the difference was not significant (p = 0.167).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As shown in Table 1, well-differentiated HCCs mainly belonged to group 1, moderately differentiated HCCs mainly belonged to group 2 followed by group 3, and poorly differentiated HCCs mainly belonged to group 3 followed by group 4. The number of unpaired arteries increased as the histologic grade progressed from well-differentiated HCC to moderately differentiated HCC, but it decreased as the tumor grade progressed from moderately differentiated HCC to poorly differentiated HCC. These results suggest that the vascularity of poorly differentiated HCC tends to decrease in comparison with moderately differentiated HCC both radiologically and pathologically.
In the current study, regarding hypervascular advanced HCC, the Ki-67 labeling index of group 2 (10.6 ± 9.0) was higher than that of group 3 (6.6 ± 6.7) in moderately differentiated HCC, but the Ki-67 labeling index of group 2 (9.1 ± 7.8) was lower than that of group 3 (26.1 ± 17.1) in poorly differentiated HCC (the difference was not significant). Furthermore, the Ki-67 labeling index of group 4 in poorly differentiated HCC was the highest of all the groups. The number of unpaired arteries of well-differentiated HCC and moderately differentiated HCC and of poorly differentiated HCC showed a strong correlation and a moderate inverse correlation with Ki-67 labeling index, respectively. These results also suggest that arterial blood supply increases in the early stage of advancement but that, conversely, it decreases in the late stage of advancement.
According to this concept, both groups 3 and 4 may theoretically consist of two different subgroups: one with moderate tumor aggressiveness exhibiting an intermediate Ki-67 labeling index and the other with high tumor aggressiveness exhibiting a high Ki-67 labeling index. A schematic presentation of this concept regarding the sequential changes of intratumoral arterial blood supply is summarized in Figure 6. The numbers of HCCs belonging to groups 3 and 4 and showing intermediate Ki-67 labeling index (here, it was defined as < 10)—that is, those considered to be on the left side of the peak (group 3' and group 4')—were 10 (eight moderately differentiated HCCs and two poorly differentiated HCCs) and one (moderately differentiated HCC), respectively. Two cases of well-differentiated HCC also belonged to group 3'. The numbers of HCCs belonging to group 3 and group 4 and showing high Ki-67 labeling index (3 10)—that is, those considered to be on the right side of the peak (group 3 and group 4)—were 11 (one moderately differentiated HCC and 10 poorly differentiated HCCs) and five (all were poorly differentiated HCCs), respectively. Thus, the HCCs with the group 3' or group 4' hemodynamic pattern and intermediate Ki-67 labeling index were well-differentiated HCC (n = 2) or moderately differentiated HCC (n = 9) (total = 11) rather than poorly differentiated HCC (n =2) and those with the group 3 or group 4 hemodynamic pattern and high Ki-67 labeling index were poorly differentiated HCC (n = 15) rather than moderately differentiated HCC (n =1) (p = 0.0001, Fisher's exact test).
|
0.0001, Spearman's rank correlation); however, when the sequence was changed
to groups 1, 4', 3', 2, 3, and 4, the correlation became stronger (rho =
0.872, p < 0.0001) (Fig.
7). This finding may support our hypothetical concept shown in
Figure 6. Although strict
differentiation between these two subgroups of group 3 and group
4—namely, group 3' versus group 3 or group 4' versus group 4, in other
words, moderately differentiated HCC versus poorly differentiated
HCC—solely by the presented radiologic method (CT hepatic arteriography
and CTAP) may be difficult, we believe these imaging techniques can provide
important information in an approximate estimation of the proliferative
activity of the tumor (Fig.
5). A new radiologic approach, such as diffusion-weighted imaging
or calculated apparent diffusion coefficient values on MRI
[17], may be helpful in this
strict differentiation of group 3' versus group 3 or of group 4' versus group
4.
|
From the standpoint of implications for patients, Jonas et al. [6] reported that vascular invasion and histopathologic grade of differentiation were identified as the only prognostic parameters in liver transplant recipients. Thus, in planning liver transplantation, evaluation of arterial blood flow can contribute to predict a worse prognosis in patients with hypovascular advanced HCC. Furthermore, transarterial infusion chemotherapy may be not effective for hypovascular advanced HCCs compared with hypervascular HCCs. We should choose alternative treatments as soon as possible because such tumors will show rapid progression due to the high proliferative activity of the tumor cells.
In malignant neoplasms, angiogenesis is also required for tumor growth, progression, and metastasis [18-20]. The supply of oxygen and nutrition from the tumor vessels is believed to be essential for the growth of solid tumors [21]. During multistep hepatocarcinogenesis, the arterial blood flow increases from hypovascular well-differentiated HCC to hypervascular classical HCC [2-4]. During this process, the proliferative activity of the tumor cells also increases. In contrast, the Ki-67 labeling index showed a moderate inverse correlation with arterial blood flow in poorly differentiated HCC. Mineura et al. [21] previously reported that high-grade gliomas tended to show lower oxygen consumption than lower-grade gliomas using 15O PET, and they speculated that high-grade gliomas may have anaerobic metabolism in contrast to the aerobic metabolism of low-grade tumors [21]. Considering our results, a similar mechanism may apply to HCC. Yasuda et al. [22] showed that glycolysis induced by hypoxia inducible factor-1 (HIF-1) is the predominant energy source under the hypoxic environment in metastatic liver cancers, and some clinicians have pretreated HCC by transarterial embolization. It may be presumed that in some poorly differentiated HCCs, tumor cells acquire a metabolic profile of increased glycolysis through the HIF-1-regulated pathway, thus enabling them to proliferate more rapidly under hypoxia. In other words, aerobic metabolizing as the energy source in hypervascular HCC may switch to glycolysis in some poorly differentiated HCCs that show less arterial blood flow.
The present study has some limitations. First, there is a possibility that the number of unpaired arteries does not directly reflect the attenuation level of HCC on CT hepatic arteriography. The degree of contrast enhancement of the nodular HCC in the arterial phase tended to correlate with the number of unpaired arteries [8]; however, in our recent study [7], the number of unpaired arteries showed weak correlation with the attenuation level on CT hepatic arteriography and it has been suggested that in addition to arterial blood flow, microvascular permeability is important in the contrast enhancement of a lesion [8, 23]. Second, we did not use a quantitative measure of the tumor on CT hepatic arteriography. However, individual differences in the degree of liver fibrosis can greatly affect the attenuation of the liver tumor and liver parenchyma. This difficulty in evaluating these hepatocellular lesions on the basis of blood supply may add limitations to this study [2].
In conclusion, we showed that arterial blood flow decreased in the late stage of HCC development both radiologically and pathologically. Although exact differentiation between poorly differentiated HCC and moderately differentiated HCC based on CTAP and CT hepatic arteriography is still limited, hypovascular advanced HCC showed higher proliferative activity than hypervascular advanced HCC.
Acknowledgments
We thank Yoshihiko Maehara, Professor and Chair, Department of Surgery and
Science, Kyushu University, for providing clinical data and Tsuneyoshi
Masazumi, Professor and Chair, Department of Anatomic Pathology, Kyushu
University, for providing pathologic data. We also thank Tsuyoshi Tajima,
Akihiro Nishie, Masakazu Hirakawa, Kousei Ishigami, and Daisuke Kakihiara,
Department of Clinical Radiology, Kyushu University, for editing this
manuscript.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
10)—that is, those considered to be on right side of peak (group 3 and
group 4)—mainly consisted of poorly differentiated HCCs. (Strictly
speaking, it seems to be incompatible to place poorly differentiated HCC
showing group 3' or group 4' into a portion labeled as well-differentiated HCC
or moderately differentiated HCC. However, such tumor is rare.) Moderately
differentiated HCC on line c shows same degree of arterial blood flow as
poorly differentiated HCC on line d, but proliferative activity of HCC on line
d is higher than that of HCC on line c. (It is difficult to distinguish d from
c solely by means of CT hepatic arteriography and CT during
arterioportography.) Poorly differentiated HCC on line e shows hypovascular
and very high proliferative activity. It is possible to point out poorly
differentiated HCC on line e, and thus we can predict proliferative activity
of this type of tumor.