Author Affiliations: Department of Radiology, Beijing Jishuitan Hospital, the Fourth Clinical Medical College of Peking University, Beijing 100035, China (Bai RJ); Department of Radiology, Second Affiliated Hospital, Harbin Medical University, Harbin 150086, China (Li JP, Jiang HJ, Liu XD, Ling ZS and Feng GL); Department of Radiology, the First Hospital of Harbin, Harbin 150010, China (Ren SH); and Department of Pathology, Harbin Medical University, Harbin 150086, China (Huang Q)

Corresponding Author: Hui-Jie Jiang, MD, PhD, Department of Radiology, Second Affiliated Hospital, Harbin Medical University, Harbin 150086, China (Tel: +86-451-86605576; Email: jhjemail@163.com)

 

© 2014, Hepatobiliary Pancreat Dis Int. All rights reserved.

doi: 10.1016/S1499-3872(14)60298-8

Published online October 27, 2014.

 

Contributors: BRJ and LJP contributed equally to the article. JHJ proposed the study. BRJ, LJP and JHJ performed research and wrote the first draft. BRJ and LJP collected and analyzed the data. All authors contributed to the design and interpretation of the study and to further drafts. JHJ is the guarantor.

Funding: This study was supported by grants from the National Nature Science Foundation of China (81471736) and Heilongjiang Province Foundation for Returness (LC2013C38).

Ethical approval: This study was approved by the institutional review board of Harbin Medical University.

Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

BACKGROUND: The peripheral morphologic characteristics of hepatocellular carcinoma (HCC) reflect tumor growth patterns. Computed tomography (CT) perfusion is a new method to analyze hemodynamic changes in tissues. We assessed the relationship between CT perfusion and histopathologic findings in the periphery of HCC lesions.

 

METHODS: Non-contrast CT, enhanced dual-phase CT, and CT perfusion were performed on 77 subjects (47 patients and 30 controls). Based on the imaging findings of enhanced dual-phase CT, the tumor edges were classified into three types: type I (sharp); type II (blurry); and type III (mixed). The CT perfusion parameters included hepatic blood flow, hepatic arterial fraction, hepatic arterial perfusion, and hepatic portal perfusion. The tissue sections from resected specimens were subjected to routine hematoxylin and eosin staining and immunohistochemical staining for CD34. The correlations between microvessel density (MVD) and the CT perfusion parameters were analyzed using Pearson's product-moment correlation coefficient. Changes in the perfusion parameters in tumor edges of different tumor types were evaluated.

 

RESULTS: Type I (sharp): the pathologic findings showed fibrous connective tissue capsules in the tumor edges, and an MVD ≤30/mm². Type II (blurry): the histology showed that the edges were clear with no capsules and an MVD >30/mm². Type III (mixed): the pathology was similar to that of types I and II, and an MVD >30/mm². Hepatic blood flow, hepatic arterial fraction, hepatic arterial perfusion, and hepatic portal perfusion were significantly increased in the tumor edges of HCC patients compared to those of the controls (P<0.05). The correlation between CT perfusion parameters and MVD was higher in blurry tumor edges of type II than in those of types I or III.

 

CONCLUSION: CT perfusion imaging of tumor edges may be helpful in revealing histopathological features, and indirectly reflect angiogenic changes of HCCs.

 

(Hepatobiliary Pancreat Dis Int 2014;13:612-617)

 

KEY WORDS: hepatocellular carcinoma; tumor edge; computed tomography; X-ray computer; histopathology; perfusion imaging

The morphologic characteristics of hepatocellular carcinoma (HCC) can not only predict the biological behavior of the tumor, but also determine the effect of treatment. New insight into angiogenesis, especially at the edges between tumors and normal tissues, can provide information that can be used in selecting treatments.^{[1, 2]} Studies on HCC showed that HCC edges are the most active area of angiogenesis.^[3-6] Angiogenesis plays an important role in tumor growth and development and in tumor invasion and metastasis.^[7-9] Thus, it is important to understand tumor angiogenesis in tumor edges and to improve the radiologic assessment. Currently, the evaluation of the morphologic characteristics of tumor edges mainly relies on conventional ultrasonography, MRI and CT scanning, but the evaluation of angiogenesis of tumor edges remains difficult before morphologic changes occur.^{[10, 11]} In recent years, a functional computed tomography (CT) perfusion technique using multi-parametric imaging maps has been widely applied clinically because this technique can provide not only information about the morphologic properties of tumor edges, but also the hemodynamic changes caused by the tumor at tumor edges.^[12-14] The current study was to determine the relationship between several CT perfusion parameters and the corresponding histologic features of HCC lesion edges and to assess the clinical value of functional imaging.

 

 

Written informed consent was obtained from all the participants before the study. All participants were prospectively enrolled from June 2011 to July 2013. The control group included 30 subjects (8 women and 22 men; mean age 54.5 years, range 30-78) who were free of liver disease and underwent CT perfusion examinations of the abdomen for unrelated causes. The HCC group included 47 patients (33 men and 14 women; mean age 56.0 years, range 30-79) who needed surgical resections. Both control and HCC subjects were also evaluated by physical examination, laboratory screening and Doppler ultrasonography of the liver, and a thorough medical history was taken from each subject. The diagnosis of HCC was made according to the European Association for the Study of the Liver (EASL) criteria.^[15]

 

The 77 subjects went through CT scans using a low-dose (120 kVp, 60 mAs) spiral CT (Lightspeed 64-slice VCT; GE Healthcare, Milwaukee, WI, USA) pre- and post-contrast. The non-contrast scan first localized the central plane of the tumor, and the perfusion scan was performed on this slice. The CT parameters for perfusion imaging were as follows: 120 kV, 60 mAs; cine scan mode; volume coverage up to 40 mm³; 50 seconds of continuous scanning time commencing 5 seconds after the injection of contrast material; 1 second per 360° revolution; 5-mm slice thickness reconstruction; and a 512×512 pixel matrix size. The rate of injection of iohexol was 3.5-4.0 mL/s in the perfusion studies; the dose was 1.0 mL/kg of body weight. The injection began 5 seconds after the beginning of image acquisition. Images were reconstructed continuously over 50 seconds, generating a total of 396 slices. An enhanced dual-phase CT scan was performed 10 minutes after the CT perfusion scanning was completed. The contrast agent was intravenously injected according to the dosage of 1.0 mL/kg body weight; 3.5-4.0 mL/s flow rate; 5-mm layer thickness image reconstruction; pitch, 0.984:1; 120 kV; 280 mA; and matrix 512×512. The arterial phase (AP) was 15-20 seconds after injection. The portal venous phase (PVP) was acquired at 45-60 seconds.

 

The perfusion images were transmitted into a GE AW 4.3 workstation using a Perfusion 3 software package. The deconvolution algorithm allowed the calculation of CT perfusion parameters for each tissue region of interest (ROI). The 20-30 mm² ROIs were placed on the aorta, portal vein, and liver parenchyma. In each of the control subjects, 2 or 3 ROIs were placed randomly in the liver parenchyma. In patients with HCC, 2 or 3 ROIs were placed within the tumor edges identified with the enhanced dual-phase CT. Hepatic blood flow (HBF) was expressed as mL/min/100 mg body weight, and the hepatic arterial fraction (HAF) referred to the percentage of the total blood input from the arterial blood supply. Hepatic artery perfusion (HAP; mL/min/100 mg body weight) and hepatic portal perfusion (HPP; mL/min/100 mg body weight) were calculated based on the values of HBF and HAF. The perfusion parameters were measured three times at each time point for each ROI, and the mean of the three measurements was used in the analysis.

 

To evaluate the characteristics of dual-phase CT enhancement in the tumor edges, the tumor attenuation was classified as hyper-, hypo-, or iso-density compared with the surrounding liver parenchyma. According to the above CT characteristics of enhancement, we divided the tumor edges into the following three types: (1) sharp edges (type I) with the tumor boundary clearly displayed on enhanced dual-phase CT; (2) blurry edges (type II) with the boundary between the tumor and normal liver tissue blurry on enhanced dual-phase CT; and (3) mixed edges (type III), part of the tumor boundary was sharp, and another part of the tumor edge had a blurry boundary, as seen on enhanced dual-phase CT.

 

Three pieces of tissues were drawn on the edges of types I and II tumors; the tissues were drawn at different edge positions according to different edge morphologies for type III tumor. The tumor rim and tumor tissues were fixed in 10% neutral buffered formalin and embedded in paraffin for immunohistochemical evaluation, and hematoxylin and eosin and CD34 staining. Microvessel density (MVD) was determined using anti-CD34 staining. The MVD assessment was performed using a two-step protocol (Beijing Zhongshan Golden Bridge Biotechnology Co., Beijing, China) according to the method reported by Weidner et al.^[16] The CD34-stained sections were screened under low power (×40), and 3 areas with the most intense neovascularization (hot spots) were selected. Microvessel counts of these areas were performed under high power (×200). Brown-stained endothelial cells or cell clusters that were clearly separated from adjacent microvessels, tumor cells, and other connective tissue elements were counted as a single microvessel, irrespective of the presence of a vessel lumen. The mean microvessel count/mm² of the five richest vascular areas was taken as the MVD. The neovascularization was abundant in the tumor edges if MVD >30/mm².^{[17, 18]}

 

All analyses were performed using SPSS (version 11.5; SPSS Inc., Chicago, IL, USA). The correlations between MVD and the CT perfusion parameters were analyzed using Pearson's product-moment correlation coefficient. Changes in the different perfusion parameters in the tumor edges of different types were evaluated by analysis of variance (ANOVA). A P value less than 0.05 was considered statistically significant.

 

 

Type I (sharp edges): Of 16 patients, 5 showed hypo-density around the mass during AP and hyper-density during PVP. Six patients showed hyper-density during AP and hypo-density during PVP. Five patients exhibited hyper-density during AP and PVP. In the corresponding CT perfusion maps, tumor contours also appeared clear, and the dividing lines were clearly marked along the surrounding liver tissues (Fig. 1A-D). The histology showed that the tumors had fibrous connective capsules. All patients had an MVD ≤30/mm² (Table 1).

 

Type II (blurry edges): There were 20 patients. The tumor edges in 13 patients showed mixed-density during AP and PVP. Seven patients showed hyper-density during AP and mixed-density during PVP. In the corresponding CT perfusion maps, the tumor edges appeared blurry (Fig. 1E-H). The pathology showed no capsule formation, and there were no obvious boundaries between cancer and surrounding liver tissues. All patients had an MVD >30/mm² (Table 1).

 

Type III (mixed edges): There were 11 patients. Seven of them showed hyper-density during AP and hypo-density during PVP. Four patients showed hyper-density during AP but mixed-density during PVP. The tumor edges had a staggered distribution with the surrounding liver parenchyma in the corresponding CT perfusion maps (Fig. 1I-L). The histology showed that partial tumor edges were clear or coated, partial tumor edges were blurry and had no obvious boundary with normal liver tissues, and this type had performances of the above two types. All patients had an MVD >30/mm² (Table 1).

 

Perfusion parameters HBF, HAF, HAP and HPP in the tumor edges were higher than those in controls (P<0.05). The differences in HBF, HAF, HAP and HPP values for different tumor types were statistically significant (P<0.05). As the tumor edges varied from the sharp edges (type I) to blurry edges (type II), a significant increase in HBF, HAF and HAP, and a decrease in HPP in the tumor edges were observed (Fig. 2).

 

Positive linear correlations existed between the values of HBF, HAF, HAP and MVD (P<0.05). As expected, HPP demonstrated a negative correlation with MVD measurements (P<0.05)(Table 2).

 

 

Tumors need nutrients to survive. For HCC, angiogenesis is important for lesion development, growth, and metastasis.^[19] Many studies^[3-6] have shown that angiogenesis is concentrated in the edges of HCC lesions. Monitoring hemodynamic changes in the edges of HCC lesions may provide useful information to better evaluate the treatment response. MVD is used to evaluate the degree of tumor angiogenesis, but cannot be used to monitor the therapeutic effect because of its high dependence on immunohistochemical staining of pathologic specimens.^[20-22] CT perfusion imaging can be performed in vivo to monitor HCC neovascularization and other biological behaviors.^[23-25]

 

In the current study, for the type I tumors, the edges were clear, capsules were noted on histopathologic evaluation, and the MVD was the lowest (20.16±3.68/mm²). On enhanced dual-phase CT, tumor edges were clear, with 11 cases showing hyper-density during the AP and 10 showing hyper-density shadows during the PVP, indicating that the tumor edges obtained blood from the hepatic artery and portal vein simultaneously. Thus, in type I tumors, enhanced dual-phase CT reflects not only the growth pattern of HCC, but also the characteristics of the blood supply. The type II tumors lacked pseudocapsule and had blurry boundaries. On enhanced dual-phase CT, 13 patients had mixed-density during AP and PVP, indicating that liver cancer cannot be clearly shown on enhanced dual-phase CT. The correlation between CT perfusion parameters and MVD was significantly higher in type II tumors than in types I and III tumors, indicating that blurry tumor edges received the most perfusion and tumor angiogenesis was the most activated. This result showed that CT perfusion reflected neovascularization more accurately than the enhanced dual-phase CT did. Type III tumors showed the similar results as types I and II tumors. The tumor edges had a staggered distribution with the surrounding liver parenchyma on enhanced dual-phase CT. Partial tumor edges were clear or coated, blurry, and the boundaries were not obviously separated from normal liver tissues. There was high correlation between CT perfusion parameters and MVD. Thus, it is necessary to use both the enhanced dual-phase CT and CT perfusion to evaluate angiogenesis of HCC.

 

HBF, HAF, HAP and HPP were significantly increased in types I-III tumor edges compared with those in the control group; the edge areas for HCC patients exhibited high perfusion, and the correlation between CT perfusion parameters and MVD was higher in tumor edges than in controls. Therefore, it is essential to choose these edge regions to assess tumor angiogenesis and to improve the radiological assessment of many tumors. In our study, differences in HBF, HAF, HAP and HPP values in the tumor edges were statistically significant among the different tumor types (P<0.05). As the tumor edges changed from sharp (type I) to blurry (type II), HBF, HAF and HAP increased significantly. There was a decrease in HPP in blurry tumor edges (type II) compared to that in the sharp and mixed edges (types I and III; P<0.05). The correlation between CT perfusion parameters and MVD was higher in blurry tumor edges (type II) than in types I and III, indicating that the blurry tumor edges were richer in blood flow perfusion and tumor angiogenesis was more active. Thus, CT perfusion imaging of tumor edges assesses tumor characteristics more specifically, indirectly reflects pathologic changes, and finds a new method for the preoperative evaluation of HCC. Notably, these findings are also comparable to those in our previous study.^[24]

 

In summary, our data indicated that the CT perfusion maps showed various perfusion status and tumor angiogenesis was different in different tumor types. These findings underscore the importance of CT perfusion as a non-invasive tool to quantify angiogenesis changes at HCC tumor edges. Therefore, the assessment of the characteristic enhancement patterns and perfusion parameters can help doctors to better diagnose the tumor types and to make the best decision in therapeutic strategies.

 

 

1 Lu XY, Xi T, Lau WY, Dong H, Xian ZH, Yu H, et al. Pathobiological features of small hepatocellular carcinoma: correlation between tumor size and biological behavior. J Cancer Res Clin Oncol 2011;137:567-575. PMID: 20508947

2 Grizzi F, Franceschini B, Fiamengo B, Russo C, Dioguardi N. Vascular architecture: is it a helpful histopathological biomarker for hepatocellular carcinoma? J Zhejiang Univ Sci B 2007;8:217-220. PMID: 17444594

3 Naito H, Takara K, Wakabayashi T, Kawahara H, Kidoya H, Takakura N. Changes in blood vessel maturation in the fibrous cap of the tumor rim. Cancer Sci 2012;103:433-438. PMID: 22098620

4 Kan Z, Phongkitkarun S, Kobayashi S, Tang Y, Ellis LM, Lee TY, et al. Functional CT for quantifying tumor perfusion in antiangiogenic therapy in a rat model. Radiology 2005;237:151-158. PMID: 16183931

5 Matsubayashi R, Matsuo Y, Edakuni G, Satoh T, Tokunaga O, Kudo S. Breast masses with peripheral rim enhancement on dynamic contrast-enhanced MR images: correlation of MR findings with histologic features and expression of growth factors. Radiology 2000;217:841-848. PMID: 11110952

6 Ogawa M, Yamamoto H, Nagano H, Miyake Y, Sugita Y, Hata T, et al. Hepatic expression of ANG2 RNA in metastatic colorectal cancer. Hepatology 2004;39:528-539. PMID: 14768007

7 Hennedige T, Venkatesh SK. Imaging of hepatocellular carcinoma: diagnosis, staging and treatment monitoring. Cancer Imaging 2013;12:530-547. PMID: 23400006

8 Bishayee A, Darvesh AS. Angiogenesis in hepatocellular carcinoma: a potential target for chemoprevention and therapy. Curr Cancer Drug Targets 2012;12:1095-1118. PMID: 22873221

9 Matsui O, Kobayashi S, Sanada J, Kouda W, Ryu Y, Kozaka K, et al. Hepatocelluar nodules in liver cirrhosis: hemodynamic evaluation (angiography-assisted CT) with special reference to multi-step hepatocarcinogenesis. Abdom Imaging 2011;36: 264-272. PMID: 21267562

10 Seeff LB, Hoofnagle JH. Epidemiology of hepatocellular carcinoma in areas of low hepatitis B and hepatitis C endemicity. Oncogene 2006;25:3771-3777. PMID: 16799618

11 Waldman SA, Fortina P, Surrey S, Hyslop T, Kricka LJ, Graves DJ. Opportunities for near-infrared thermal ablation of colorectal metastases by guanylyl cyclase C-targeted gold nanoshells. Future Oncol 2006;2:705-716. PMID: 17155897

12 Kanda T, Yoshikawa T, Ohno Y, Fujisawa Y, Kanata N, Yamaguchi M, et al. Perfusion measurement of the whole upper abdomen of patients with and without liver diseases: initial experience with 320-detector row CT. Eur J Radiol 2012;81:2470-2475. PMID: 22055684

13 Ma GL, Bai RJ, Jiang HJ, Hao XJ, Dong XP, Li DQ, et al. Early changes of hepatic hemodynamics measured by functional CT perfusion in a rabbit model of liver tumor. Hepatobiliary Pancreat Dis Int 2012;11:407-411. PMID: 22893468

14 Lee JM, Yoon JH, Kim KW. Diagnosis of hepatocellular carcinoma: newer radiological tools. Semin Oncol 2012;39:399-409. PMID: 22846858

15 European Association For The Study Of The Liver; European Organisation For Research And Treatment Of Cancer. EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol 2012;56:908-943. PMID: 22424438

16 Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis--correlation in invasive breast carcinoma. N Engl J Med 1991;324:1-8. PMID: 1701519

17 Karahan OI, Yikilmaz A, Isin S, Orhan S. Characterization of hepatocellular carcinomas with triphasic CT and correlation with histopathologic findings. Acta Radiol 2003;44:566-571.

18 Kim HC, Kim TK, Sung KB, Yoon HK, Kim PN, Ha HK, et al. Preoperative evaluation of hepatocellular carcinoma: combined use of CT with arterial portography and hepatic arteriography. AJR Am J Roentgenol 2003;180:1593-1599.

19 Fukumura D, Jain RK. Imaging angiogenesis and the microenvironment. APMIS 2008;116:695-715. PMID: 18834413

20 Messerini L, Novelli L, Comin CE. Microvessel density and clinicopathological characteristics in hepatitis C virus and hepatitis B virus related hepatocellular carcinoma. J Clin Pathol 2004;57:867-871. PMID: 5280410

21 Zhao ZC, Zheng SS, Wan YL, Jia CK, Xie HY. The molecular mechanism underlying angiogenesis in hepatocellular carcinoma: the imbalance activation of signaling pathways. Hepatobiliary Pancreat Dis Int 2003;2:529-536. PMID: 14627514

22 Ippolito D, Sironi S, Pozzi M, Antolini L, Ratti L, Alberzoni C, et al. Hepatocellular carcinoma in cirrhotic liver disease: functional computed tomography with perfusion imaging in the assessment of tumor vascularization. Acad Radiol 2008;15:919-927. PMID: 18572129

23 Sahani DV, Holalkere NS, Mueller PR, Zhu AX. Advanced hepatocellular carcinoma: CT perfusion of liver and tumor tissue--initial experience. Radiology 2007;243:736-743. PMID: 17517931

24 Li JP, Zhao DL, Jiang HJ, Huang YH, Li DQ, Wan Y, et al. Assessment of tumor vascularization with functional computed tomography perfusion imaging in patients with cirrhotic liver disease. Hepatobiliary Pancreat Dis Int 2011;10:43-49. PMID: 21269934

25 Choi BI. Advances of imaging for hepatocellular carcinoma. Oncology 2010;78:46-52. PMID: 20616584