Va

Wa

b

a

ARRA

KHVPD

1

l

DsV

((khM

R

h0

European Journal of Radiology 85 (2016) 2036–2041

Contents lists available at ScienceDirect

European Journal of Radiology

j ourna l h o mepage: www.elsev ier .com/ locate /e j rad

EGFR-2 expression in HCC, dysplastic and regenerative liver nodules,nd correlation with pre-biopsy Dynamic Contrast Enhanced CT

.M. Thaissa,∗, S. Kaufmanna, C. Klotha, K. Nikolaoua, H. Bösmüllerb,1, M. Horgera,1

Eberhard Karls University, Department of Radiology, Diagnostic and Interventional Radiology, Hoppe-Seyler-Str. 3, D-72076 Tuebingen, Germany2

Eberhard Karls University, Department of Pathology, Liebermeisterstraße 8, D-72076 Tuebingen, Germany

r t i c l e i n f o

rticle history:eceived 8 June 2016eceived in revised form 6 September 2016ccepted 13 September 2016

eywords:CCEGFR-2erfusion imagingCE-CT

a b s t r a c t

Purpose: To evaluate whether VEGFR-2-expression in hepatocellular carcinoma (HCC), dysplastic (DLN)and regenerative liver nodules (RLN) correlates with pre-histology, in vivo Dynamic Contrast Enhanced-Computed Tomography (DCE-CT) data as VEGFR-2-expression affects prognosis and therapeutic options.Materials and methods: 34 patients (63.6 ± 8.9 years, 7 females) underwent liver biopsy or surgerydue to suspected HCC or dysplastic nodules after DCE-CT between 2009 and 2015 with no previouschemo- or interventional therapy. Immunohistochemistry staining for VEGFR-2 was performed usingImmunoreactive-Remmele-Stegner-Score (IRS) for quantification. A 128-row CT-scanner was used forDCE-CT with assessment of perfusion parameters blood flow (BF), blood volume (BV), arterial liverperfusion (ALP), portal venous perfusion (PVP), and hepatic perfusion index (HPI).Results: Histology confirmed HCC (n = 10), DLN (n = 7) and RLN (n = 34). Mean IRS for VEGFR-2 in HCCswas 9.1 ± 3.0, 7.3 ± 1.6 for DLN and 5.2 ± 2.8 for RLN (p = 0.0004 for HCC vs. RLN). Perfusion values variedsignificantly between all three groups for BF and HPI (p < 0.001 and p < 0.0001) and for BV in HCC vs.RLN (p < 0.0001) and DLN vs. RLN (p = 0.0019). Strong correlations between VEGFR-2-IRS and perfusionparameters were observed for BF in HCC (r = 0.88, p < 0.01) and HPI in HCC and DLN (r = 0.85, p < 0.04;r = 0.9, p < 0.01).

Conclusion: Immunostaining revealed different VEGFR-2-expression levels in HCC, dysplastic and regen-erative liver nodules. Perfusion markers blood flow, blood volume and hepatic perfusion index correlatedwell with VEGFR-2-immunostaining. This non-invasive discrimination between regenerative and dys-plastic/HCC nodules might open new perspectives for diagnosis, therapy planning, and anti-VEGFR therapy monitoring.

. Introduction

Hepatocellular carcinoma (HCC) is the most common primaryiver cancer [1] and diagnosis at very early stages with potential

Abbreviations: ALP, arterial liver perfusion; BF, blood flow; BV, blood volume;LN, Dysplastic Liver Nodule; HCC, Hepatocellular Carcinoma; HPI, Hepatic Perfu-

ion Index; PVP, portal venous perfusion; RLN, Regenerative Liver Nodule; VEGFR-2,ascular Endothelial Growth Factor Receptor 2; VPCT, Volume Perfusion CT.∗ Corresponding author.

E-mail addresses: wolfgang.thaiss@med.uni-tuebingen.deW.M. Thaiss), sascha.kaufmann@med.uni-tuebingen.deS. Kaufmann), christopher.kloth@med.uni-tuebingen.de (C. Kloth),onstantin.nikolaou@med.uni-tuebingen.de (K. Nikolaou),ans.boesmueller@med.uni-tuebingen.de (H. Bösmüller),arius.Horger@med.uni-tuebingen.de (M. Horger).1 These authors share senior authorship.2 https://www.medizin.uni-tuebingen.de/en/Referral+Guide/Hospitals/adiology.html

ttp://dx.doi.org/10.1016/j.ejrad.2016.09.012720-048X/© 2016 Elsevier Ireland Ltd. All rights reserved.

© 2016 Elsevier Ireland Ltd. All rights reserved.

curative therapy is still a challenge. Although recent advances havebeen made concerning diagnostic accuracy and detection of smalllesions, the need for earlier identification of dysplastic nodules andearly HCC is pressing with regard to rising incidences [1].

Imaging modalities such as MRI and CT with the optionof dynamic contrast enhanced (DCE) examinations have gainedmomentum, providing quantifiable perfusion parameters andthereby enabling the detection of pre-malignant or malignantlesions below <1 cm, which are currently considered non-classifiable [2,3]. DCE-CT uses repetitive CT scans of the liver tocalculate perfusion parameters from imaging data acquired dur-ing the passage of contrast agent [4–7]. Such perfusion parametershave proven beneficial with regard to therapy monitoring [8]and detection of early recurrence, e.g. after trans-arterial chemoembolization (TACE) [9].

The enhanced vascularization of HCC has its histopathologiccorrelate − among others − in the enhanced expression of vascu-lar endothelial growth factor receptors (VEGFR), especially type 2

nal of

[ls

igli

fredliq

Voes

otRdao

2

2

6lD2twt

ot(

2

Hessmto(tflp

2

hrrd

W.M. Thaiss et al. / European Jour

10]. Several investigations have demonstrated upregulated mRNAevels, as well as expression of VEGFR-2 in HCC and surroundinginuses [11,12].

Consequently, therapeutic efforts include the use of agentsnterfering with VEGF-receptors to inhibit tumor vasculaturerowth. Especially Sorafenib has been increasingly used over theast years [13,14]. Therefore, the expression levels of VEGFR directlymplicate the effectiveness of anti-angiogenetic treatment [12].

Besides HCC, other nodular liver lesions may be challengingor diagnosis. Among them, dysplastic nodules (DLN) are clinicallyelevant due to their potential for malignant transformation. How-ver, atypical cytologic features and structural changes make themifficult to differentiate from early HCC. In contrast, regenerative

iver nodules (RLN) are a form of non-neoplastic lesions that arisen a cirrhotic liver and may progress with time to DLN and subse-uently to HCC [15].

Conclusively, profound knowledge is available for the role ofEGFR-2 in HCC and its carcinogenesis as well as for the utilizationf DCE-CT in detecting arterialized liver nodules. To our knowl-dge, however, no link between VEGFR-2 immunohistochemistrytaining and DCE-CT has been described.

In this study, we investigated whether in vivo perfusion valuesbtained from DCE-CT imaging correlate with the histological dis-ribution of VEGFR-2 expression in samples from HCC, DLN, andLN obtained from patients without previous therapy. in order toetermine whether prior non-invasive DCE-CT can complement thenalysis of VEGFR-2 expression in the assessment of therapeuticptions for pre-malignant and malignant hepatic lesions.

. Materials and methods

.1. Study population

Retrospective database search identified 34 patients (mean age3.6 years, range 45.7 − 85.6 years, 7 females) that had undergone

iver biopsy or surgery due to suspected HCC or DLN after previousCE-CT had been performed at our institution between 2009 and015. Perfusion studies were approved by the local Ethics Commit-ee for HCC with histological confirmation. All patients providedritten informed consent including information about the radia-

ion exposure.Patients with previous treatment with anti-VEGF therapy, TACE,

r any other sort of interventional therapy were excluded. Meanime between DCE-CT and histology sample isolation was 52.5 daysrange 6–209 days).

.1.1. CT protocolA 128-row CT scanner (Somatom Definition AS+, Siemens

ealthcare, Forchheim, Germany) was used. A low-dose non-nhanced CT (60 mAs, 100 kVp, 128 × 0.6 mm collimation, 5 mmlice thickness) was used for DCE-CT planning. Adaptive spiralcanning technique with 80 kVp, 100–120 mAs, 64 × 0.6 mm colli-ation, scan range of 6.9 cm coverage and a scan time of 40 s at a

ime resolution of 1.5 s per spiral data set was used with a delayf 7 s after contrast agent injection. A dual-head pump injectorMedtron, Saarbruecken, Germany) was used for the administra-ion of 50 mL Ultravist 370 (Bayer, Leverkusen, Germany) with aow rate of 5 mL/s. Examinations resulted in a mean dose-lengthroduct of 489.6 mGy x cm.

.1.2. DCE-CT analysisSyngo Volume Perfusion CT Body (Siemens Healthcare, Forch-

eim, Germany) was used for subsequent motion correction, noiseeduction, and threshold-based exclusion of bone, fat and air [16]. Aegion of interest (ROI) was drawn at the maximal lesion outline asescribed previously [17,18] and the ROI-area (cm2) was recorded.

Radiology 85 (2016) 2036–2041 2037

Perfusion parameters for blood flow (BF; mL/100 mL/min),blood volume (BV; mL/100 mL), arterial liver perfusion (ALP;mL/100 mL/min), portal venous perfusion (PVP; mL/100 mL/min)and hepatic perfusion index (HPI; %) were determined for HCC, DLNand RLN.

Calculation of ALP and PVP considering the dual blood supply ofthe liver by hepatic artery and portal vein was performed by usingthe time of peak splenic enhancement as a separation point of arte-rial and portal-venous phase with ROIs in portal vein and spleen.The arterial time density curve for ALP was calculated by dividingthe maximum arterial slope by the maximum aortic enhancement.Portal-venous time density curve for PVP was calculated by divid-ing the maximum portal-venous slope by the maximum portal-veinenhancement. HPI represents the quotient of ALP divided by thesum of ALP and PVP.

2.1.3. HistologyFor the semiquantitative evaluation of VEGFR-2 staining, we

used the Remmele/Stegner immunoreactive score (IRS score) [19].Based on the staining of the lesion tissue, the intensity of cytoplas-mic expression was graded from 0 to 3, with 0 as negative staining,grade 1 as weak, grade 2 as medium and grade 3 as strong inten-sity. The score was obtained by multiplying this grade with a factordetermined by the percentage of positive tumor cells (0–10%/1;10–50%/2; 50–80%/3; 80–100%/4). Combined scores 8, 9 and 12were considered strongly positive, score 4 and 6 moderately posi-tive, and score 1, 2, 3 weakly positive.

The following primary antibody and dilutions were used forimmunohistochemistry: VEGFR-2 (1:50, monoclonal rabbit, 55B11,Cell signaling, Danvers, USA). The tissue sections were pre-treatedwith EDTA-buffer solution (pH 8.6) at 95 ◦C for 64 min. Immunos-taining was performed on an automated immunostainer accordingto the manufacturer’s instructions (Ventana Medical Systems Inc.,Tucson, AZ, USA).

2.1.4. Statistical analysesStatistics were calculated with Prism (GraphPad software, La

Jolla, CA, USA). All data are reported as mean ± standard deviation(SD). One-way ANOVA with Tukey correction was used for groupcomparison. Pearson’s r was used for correlation analysis and ROC.For all tests, p values smaller than 0.05 were considered significant.

3. Results

3.1. Patient characteristics

All patients showed histological and morphological signs of liverfibrosis with an etiology of hepatitis C (n = 11), hepatitis B (n = 4,one patient with B and C), alcohol abuse (n = 12), non-alcoholicsteatohepatitis (n = 1), primary biliary cirrhosis (n = 1) and crypto-genic cirrhosis (n = 6). Child Pugh Score was A = 13, B = 13, C = 4, notassessed in 4 patients.

3.2. Immunostaining for VEGFR-2

Histology confirmed HCC (n = 10), dysplastic liver nodules (n = 7)and regenerative liver nodules (n = 34). Mean IRS for VEGFR-2 inHCC was 9.1 (SD 3.0), 7.3 (SD 1.6) for DLN and 5.2 (SD 2.8) for RLN(p = 0.0004 for HCC vs. RLN; p = 0.06 for HCC vs. DLN, not significant).An example is given in Fig. 1, groups are compared in Fig. 2.

3.3. DCE-CT parameters

The average ROI area derived from DCE-CT measurements inthe HCC group was 3.1 ± 2.3 cm2 (range 0.8–6.4 cm2), DLN group

2038 W.M. Thaiss et al. / European Journal of Radiology 85 (2016) 2036–2041

Fig. 1. 78-year old patient with HCC. DEC-CT maps for blood blow (A), blood volume (B), arterial (C) and portal venous perfusion (D) are shown. VEGFR-2 immunostainingfrom the nodule with increased expression in HCC (arrow) compared to minimal staining in an adjacent regenerative nodule (arrowhead). 50× magnification.

Table 1Dynamic Contrast Enhanced-CT perfusion values of HCC (n = 10), dysplastic (n = 7) and regenerative liver nodules (n = 34).

Mean SD Minimum Maximum

HCC BF 73.34 22.05 44.84 115.23 mL/100 mL/minBV 15.25 3.55 9.52 20.18 mL/100 mLALP 49.37 15.92 22.75 70.40 mL/100 mL/minPVP 1.367 1.81 0 4.56 mL/100 mL/minHPI 97.75 3.05 92.16 100 %

DLN BF 46.25 3.35 42.06 49.85 mL/100 mL/minBV 13.76 0.98 12.52 14.83 mL/100 mLALP 40.90 9.29 27.31 48.22 mL/100 mL/minPVP 26.68 23.52 11.90 61.76 mL/100 mL/minHPI 74.39 18.44 46.75 84.54 %

RLN BF 27.67 9.48 8.10 57.20 mL/100 mL/minBV 5.741 4.35 1.00 22.10 mL/100 mLALP 13.21 5.84 2.70 26.30 mL/100 mL/minPVP 67.66 34.15 14.8 193.20 mL/100 mL/minHPI 17.84 9.25 6.20 54.60 %

SD = standard deviation, BF = blood blow, BV = blood volume, ALP = arterial liver perfusioncarcinoma, DLN = dysplastic liver nodule, RLN = regenerative liver nodule.

Fs

2(

a

ig. 2. Immunoreactive Remmele-Stegner Score (IRS) for VEGFR-2 expressionhows significant differences between HCC and regenerative liver nodules.

.3 ± 1.1 cm2 (range 1.4–3.7 cm2) and RLN group 2.9 ± 1.8 cm2

range 1.2–6.4 cm2).Perfusion parameters for HCC revealed an aver-

ge BF of 73.3 ± 22.1 mL/100 mL/min, average BV of

, PVP = portal venous perfusion, HPI = hepatic perfusion index. HCC = hepatocellular

15.3 ± 3.6 mL/100 mL and ALP with 49.4 ± 15.9 mL/100 mL/min,PVP 1.4 ± 1.8 mL/100 mL/min and HPI of 97.8 ± 3.0%.

For DLN BF was 46.3 ± 3.4 mL/100 mL/min, BV13.8 ± 1.0 mL/100 mL, ALP with 40.9 ± 9.3 mL/100 mL/min, PVP26.7 ± 23.5 mL/100 mL/min and HPI of 74.4 ± 18.4%.

RLN demonstrated BF values of 27.7 ± 9.5 mL/100 mL/min, aver-age BV of 5.7 ± 4.4 mL/100 mL, ALP 13.2 ± 5.8 mL/100 mL/min, PVP67.7 ± 34.2 mL/100 mL/min and HPI of 17.8 ± 9.3%. Values are sum-marized in Table 1.

Perfusion values varied significantly between all three groups(Table 2, Fig. 3) except the group separation between HCC and DLNin BV, ALP and HPI.

3.4. Correlations between IRS and DCE-CT perfusion parameters

As VEGFR-2 expression should be directly associated for perfu-sion properties of the tissue of interest, we performed correlation

analysis between DEC-CT perfusion values and immunostainingresults.

Strong correlations between BF and VEGFR-2-IRS values in HCCwere observed (r = 0.88, p < 0.01). In DLN, this correlation did not

W.M. Thaiss et al. / European Journal of Radiology 85 (2016) 2036–2041 2039

Fig. 3. Dynamic contrast enhanced CT perfusion parameters. BF = blood blow, BV = bloodperfusion index. HCC = hepatocellular carcinoma, DLN = dysplastic liver nodule, RLN = rege

Table 2ANOVA results for perfusion parameter group analysis.

Blood Flow (BF)F 44.48P value <0.0001

Tukey’s multiple comparisons test Mean Diff. 95% CIBF HCC vs. BF DLN 27.1 9.024 to 45.17BF HCC vs. BF RLN 45.67 33.70 to 57.63BF DLN vs. BF RLN 18.57 3.332 to 33.81

Blood volume (BV)F 19.68P value <0.0001

Tukey’s multiple comparisons test Mean Diff. 95% CIBV HCC vs. BV DLN 1.494 −4.711 to 7.699 n.s.BV HCC vs. BV RLN 9.510 5.342 to 13.68BV DLN vs. BV RLN 8.016 2.737 to 13.30

Hepatic Perfusion Index (HPI)F 208P value <0.0001

Tukey’s multiple comparisons test Mean Diff. 95% CIHPI HCC vs. HPI DLN 23.36 8.040 to 38.67HPI HCC vs. HPI RLN 79.91 69.38 to 90.44HPI DLN vs. HPI RLN 56.56 43.99 to 69.12

Arterial liver Perfusion (ALP)F 67.1P value <0.0001

Tukey’s multiple comparisons test Mean Diff. 95% CIALP HCC vs. ALP DLN 8.476 −4.185 to 21.14 n.s.ALP HCC vs. ALP RLN 36.16 27.78 to 44.55ALP DLN vs. ALP RLN 27.69 17.01 to 38.36

Portal venous Perfusion (PVP)F 13.21P value <0.0001

Tukey’s multiple comparisons test Mean Diff. 95% CIPVP HCC vs. PVP DLN −25.31 −74.37 to 23.74 n.s.PVP HCC vs. PVP RLN −66.3 −100.0 to −32.57PVP DLN vs. PVP RLN −40.98 −81.22 to −0.7492

HCC = hepatocellular carcinoma, DLN = dysplastic liver nodule, RLN = regenerativeliver nodule, n.s. = not significant

volume, ALP = arterial liver perfusion, PVP = portal venous perfusion, HPI = hepaticnerative liver nodule.

reach statistical significance (r = 0.84, p = 0.16). HPI in HCC and DLNcorrelated with IRS (r = 0.85, p < 0.04 for HCC and r = 0.9, p < 0.01for DLN). Inverse correlations existed between PVP and IRS for HCCand DLN (r = −0.79, p = 0.05 for HCC and r = −0.99, p = 0.006 for DLN).RLN IRS values and perfusion values showed no correlation.

3.5. ROC-Analysis

From the results above, we hypothesized that the quantitativeparameters from DCE-CT might predict the presence of a DLN orhigher grade malignancy (HCC) in separation form RLN when takinghistology as gold standard. Statistical analysis revealed a sensi-tivity/specificity for HPI >46.75% of 90.0%/96.9% (likelihood ratio29.7, area under the curve 0.99) and for BF >41.2 mL/100 mL/minof 100%/97.1% (likelihood ratio 34.0, area under the curve 0.98).

4. Discussion

Vascular endothelial growth factor (VEGF) is known as one ofthe major players involved in tumor angiogenesis. The interactionwith its receptors VEGFR-1 and in particular VEGFR-2 induces pro-liferation, migration and survival of endothelial cells [20].

In the present study, we examined perfusion parameters inHCC and its precursors (dysplastic nodules) including regenera-tive liver nodules by DCE-CT and correlated the results with thoseof VEGFR-2-expression determined by immunochemistry using animmunoreactive score (IRS).

Our results confirm significantly increased VEGFR-2-expressionin HCC and to a lesser degree in dysplastic nodules compared toregenerative liver nodules. At the same time, the magnitude of mostof the measured perfusion parameters (BF, BV, ALP and HPI) provedto be higher in HCC and DLN in comparison to those registered inRLN.

This data is in line with previous reports showing higher VEGFR-

2-expression levels in tumor cells and tumorous blood vesselswhen compared to normal tissue [21]. Nakamura et al. reporteddefuse VEGFR-2 expression in dysplastic hepatocytes, includingcapillarized areas and more pronounced in endothelial cells of

2 nal of

Hp

acAaqcpiwa

topaddibad

tcispuusr

ct[tm[ob

tmhb

icpbD2fs

C

flne

[

[

[

[

[

[

[

[

[

[

[

[

[

[

040 W.M. Thaiss et al. / European Jour

CC, whereas the expression was faint in the surrounding liverarenchyma [22].

Unfortunately, VEGFR-2-measurements in tissues are invasivend therefore a surrogate non-invasive parameter that wouldlosely resemble results of immunochemistry would be desirable.s such, DCE-CT as an established imaging technique with highccuracy of its quantification tools could substitute for VEGFR-2uantification. Starting from this hypothesis, our data confirmedorrelation between VEGFR-2 expression, blood flow, and hepaticerfusion index as markers for the degree of tumor arterialization

n HCC as well as HPI in DLN, but not in RLN. Inverse correlationas found for portal-venous perfusion and VEGFR-2-IRS in HCC

nd DLN.These results are in line with the concept of vasculariza-

ion in hepatocarcinogenesis, which claims that the blood supplyf focal liver lesions is steadily increasing from predominantlyortal-venous to arterial blood supply due to tumor-relatedngiogenesis accompanying the dedifferentiation from low-gradeysplastic nodule to high-grade dysplastic nodule, early HCC, well-ifferentiated HCC, and moderately differentiated HCC [23]. This

nformation has practical significance, as it helps to discriminateetween HCC, DLN and RLN, which is mandatory for patient man-gement. As DLN are considered premalignant lesions of HCC, theirifferentiation from RLN has also clinical relevance [22].

Moreover, VEGFR-2 is a key target in modern anti-cancerreatment [24]. Non-invasive assessment of VEGFR-2-expressionould thus be a potential tool for patient pre-selection beforenitiation of these therapeutic agents and for achieving a moreensitive response monitoring. Our data support the theory thaterfusion parameters such as BF and HPI might indeed beseful to pre-evaluate liver nodules in risk stratification, as val-es >41.2 mL/100 mL/min for BF and HPI >46.75% showed highensitivity and specificity for the presence of DLN or HCC in sepa-ation from RLN.

In a more advanced scenario, more specific lesion detection andharacterization in a cirrhotic liver using VEGFR-2-labeled con-rast agents (e.g. via ultrasound) could benefit from this knowledge25,26]. Finally, the level of tissue VEGFR-2 expression in HCC seemso have prognostic implications, as VEGFR-2 has been related to

ore aggressive tumor behavior and the risk of distant metastases27]. Serum VEGFR-2 has already been successfully used as a markerf response in HCC-patients undergoing transarterial chemoem-olisation (TACE) [28].

There are certain limitations of our study, in particular due tohe retrospective design and relatively small sample size, which is

ainly due to the small number of DEC-CTs directly followed byistology without intermediate therapy or alternative initial MRIefore therapy.

In conclusion, we found different expression levels for VEGFR-2n HCC, dysplastic liver nodules and regenerative liver nodules thatorrelate well with the non-invasively assessed tumor perfusionarameters using DCE-CT. Both parameters enable discriminationetween HCC and its precursors. The strong correlations betweenCE-CT perfusion parameters and the expression levels of VEGFR-

in immunohistochemistry are expected to be of practical useor implementation and monitoring of VEGFR-2-targeted therapytrategies.

onflict of interest

We wish to confirm that none of the authors has a known con-ict of interest associated with this publication and there has beeno significant financial support for this work that could have influ-nced this outcome.

[

Radiology 85 (2016) 2036–2041

Institutional review board approval was obtained for this retro-spective data analysis.

References

[1] R. Siegel, J. Ma, Z. Zou, A. Jemal, Cancer statistics, 2014, CA. Cancer J. Clin. 64(1) (2014) 9–29.

[2] A.H. Davarpanah, J.C. Weinreb, The role of imaging in hepatocellularcarcinoma: the present and future, J. Clin. Gastroenterol. 47 (2013) S7–S10(Suppl.).

[3] D. Ippolito, S. Sironi, M. Pozzi, L. Antolini, L. Ratti, C. Alberzoni, E.B. Leone, F.Meloni, M.G. Valsecchi, F. Fazio, Hepatocellular carcinoma in cirrhotic liverdisease: functional computed tomography with perfusion imaging in theassessment of tumor vascularization, Acad. Radiol. 15 (7) (2008) 919–927.

[4] K.A. Miles, C. Charnsangavej, F.T. Lee, E.K. Fishman, K. Horton, T.Y. Lee,Application of CT in the investigation of angiogenesis in oncology, Acad.Radiol. 7 (10) (2000) 840–850.

[5] K.A. Miles, Perfusion CT for the assessment of tumour vascularity: whichprotocol? Br. J. Radiol. 76 (2003) S36–42 (Spec No 1).

[6] A.R. Kambadakone, D.V. Sahani, Body perfusion CT: technique, clinicalapplications, and advances, Radiol. Clin. North Am. 47 (1) (2009) 161–178.

[7] G. Petralia, L. Preda, G. D’Andrea, S. Viotti, L. Bonello, R. De Filippi, M. Bellomi,CT perfusion in solid-body tumours. Part I: Technical issues, La Radiol. Med.115 (6) (2010) 843–857.

[8] D. Ippolito, D. Fior, P.A. Bonaffini, C. Capraro, D. Leni, R. Corso, S. Sironi,Quantitative evaluation of CT-perfusion map as indicator of tumor responseto transarterial chemoembolization and radiofrequency ablation in HCCpatients, Eur. J. Radiol. 83 (9) (2014) 1665–1671.

[9] J.A. Lee, W.K. Jeong, Y. Kim, S.Y. Song, J. Kim, J.N. Heo, C.K. Park, Dual-energyCT to detect recurrent HCC after TACE: initial experience of color-codediodine CT imaging, Eur. J. Radiol. 82 (4) (2013) 569–576.

10] D. Sia, C. Alsinet, P. Newell, A. Villanueva, VEGF signaling in cancer treatment,Curr. Pharm. Des. 20 (17) (2014) 2834–2842.

11] J.H. Kwon, N. Lee, J.Y. Park, Y.S. Yu, J.P. Kim, J.H. Shin, D.S. Kim, J.W. Joh, D.S.Kim, K.Y. Choi, K.J. Kang, G. Kim, Y.H. Moon, H.J. Wang, Actionable geneexpression-based patient stratification for molecular targeted therapy inhepatocellular carcinoma, PLoS One 8 (6) (2013) e64260.

12] S. Peng, Y. Wang, H. Peng, D. Chen, S. Shen, B. Peng, M. Chen, R. Lencioni, M.Kuang, Autocrine vascular endothelial growth factor signaling promotes cellproliferation and modulates sorafenib treatment efficacy in hepatocellularcarcinoma, Hepatology (Baltimore Md.) 60 (4) (2014) 1264–1277.

13] J. Bruix, M. Reig, M. Sherman, Evidence-based diagnosis, staging, andtreatment of patients with hepatocellular carcinoma, Gastroenterology 150(4) (2016) 835–853.

14] J. Chen, R. Jin, J. Zhao, J. Liu, H. Ying, H. Yan, S. Zhou, Y. Liang, D. Huang, X.Liang, H. Yu, H. Lin, X. Cai, Potential molecular, cellular andmicroenvironmental mechanism of sorafenib resistance in hepatocellularcarcinoma, Cancer Lett. 367 (1) (2015) 1–11.

15] R.F. Hanna, D.A. Aguirre, N. Kased, S.C. Emery, M.R. Peterson, C.B. Sirlin,Cirrhosis-associated hepatocellular nodules: correlation of histopathologicand MR imaging features, Radiographics 28 (3) (2008) 747–769.

16] A. Chandler, W. Wei, E.F. Anderson, D.H. Herron, Z. Ye, C.S. Ng, Validation ofmotion correction techniques for liver CT perfusion studies, Br. J. Radiol. 85(1016) (2012) e514–e522.

17] S. Kaufmann, T. Horger, A. Oelker, C. Kloth, K. Nikolaou, M. Schulze, M. Horger,Characterization of hepatocellular carcinoma (HCC) lesions using a novelCT-based volume perfusion (VPCT) technique, Eur. J. Radiol. 84 (6) (2015)1029–1035.

18] D. Spira, L. Grunwald, W. Vogel, A. Sauter, S.M. Spira, C.D. Claussen, M. Horger,Midtreatment evaluation of lymphoma response to chemotherapy by volumeperfusion computed tomography, J. Comput. Assist. Tomogr. 38 (1) (2014)123–130.

19] W. Remmele, H.E. Stegner, [Recommendation for uniform definition of animmunoreactive score (IRS) for immunohistochemical estrogen receptordetection (ER-ICA) in breast cancer tissue], Der Pathologe 8 (3) (1987)138–140.

20] T. Takahashi, H. Ueno, M. Shibuya, VEGF activates protein kinaseC-dependent, but Ras-independent Raf-MEK-MAP kinase pathway for DNAsynthesis in primary endothelial cells, Oncogene 18 (13) (1999) 2221–2230.

21] J. Huang, X. Zhang, Q. Tang, F. Zhang, Y. Li, Z. Feng, J. Zhu, Prognosticsignificance and potential therapeutic target of VEGFR-2 in hepatocellularcarcinoma, J. Clin. Pathol. 64 (4) (2011) 343–348.

22] K. Nakamura, Y. Zen, Y. Sato, K. Kozaka, O. Matsui, K. Harada, Y. Nakanuma,Vascular endothelial growth factor, its receptor Flk-1, and hypoxia induciblefactor-1alpha are involved in malignant transformation in dysplastic nodulesof the liver, Hum. Pathol. 38 (10) (2007) 1532–1546.

23] T. Inoue, M. Kudo, M. Komuta, S. Hayaishi, T. Ueda, M. Takita, S. Kitai, K.Hatanaka, N. Yada, S. Hagiwara, H. Chung, T. Sakurai, K. Ueshima, M.Sakamoto, O. Maenishi, T. Hyodo, M. Okada, S. Kumano, T. Murakami,

Assessment of Gd-EOB-DTPA-enhanced MRI for HCC and dysplastic nodulesand comparison of detection sensitivity versus MDCT, J. Gastroenterol. 47 (9)(2012) 1036–1047.

24] A. Kiselyov, K.V. Balakin, S.E. Tkachenko, VEGF/VEGFR signalling as a target forinhibiting angiogenesis, Expert Opin. Investig. Drugs 16 (1) (2007) 83–107.

nal of

[

[

[

[28] Y.B. Zheng, Q.W. Meng, W. Zhao, B. Liu, J.W. Huang, X. He, Y. Li, B.S. Hu, L.G. Lu,Prognostic value of serum vascular endothelial growth factor receptor 2response in patients with hepatocellular carcinoma undergoing transarterial

W.M. Thaiss et al. / European Jour

25] C. Grouls, M. Hatting, A. Rix, S. Pochon, W. Lederle, I. Tardy, C.K. Kuhl, C.Trautwein, F. Kiessling, M. Palmowski, Liver dysplasia: US molecular imagingwith targeted contrast agent enables early assessment, Radiology 267 (2)(2013) 487–495.

26] K. Sugimoto, F. Moriyasu, Y. Negishi, N. Hamano, H. Oshiro, N.G. Rognin, T.Yoshida, N. Kamiyama, Y. Aramaki, Y. Imai, Quantification in molecular

ultrasound imaging: a comparative study in mice between healthy liver and ahuman hepatocellular carcinoma xenograft, J. Ultrasound Med. 31 (12) (2012)1909–1916.

27] M. Baron Toaldo, V. Salvatore, S. Marinelli, C. Palama, M. Milazzo, L. Croci, L.Venerandi, M. Cipone, L. Bolondi, F. Piscaglia, Use of VEGFR-2 targeted

Radiology 85 (2016) 2036–2041 2041

ultrasound contrast agent for the early evaluation of response to sorafenib ina mouse model of hepatocellular carcinoma, Mol. Imaging Biol. MIB 17 (1)(2015) 29–37.

chemoembolization, Med. Oncol. (Northwood, London, England) 31 (3) (2014)843.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具