Natural Course of Hypovascular Nodules Detected on Gadoxetic Acid-enhanced MR Imaging: Presence of Fat is a Risk Factor for Hypervascularization

Transcription

1 Magn Reson Med Sci, Vol. 12, No. 4, pp , Japanese Society for Magnetic Resonance in Medicine doi: /mrms MAJOR PAPER Natural Course of Hypovascular Nodules Detected on Gadoxetic Acid-enhanced MR Imaging: Presence of Fat is a Risk Factor for Hypervascularization Dai JOISHI 1,3, Akihisa UENO 1,2, Akihiro TANIMOTO 1 *, Shigeo OKUDA 1, Yohei MASUGI 2, Katsura EMOTO 2,KiyoshiOKUMA 1, Michiie SAKAMOTO 2, Yutaka IMAI 3, and Sachio KURIBAYASHI 1 Departments of 1 Diagnostic Radiology and 2 Pathology, Keio University School of Medicine 35 Shinanomachi, Shinjuku, Tokyo , Japan 3 Department of Diagnostic Radiology, Tokai University School of Medicine (Received December 26, 2012; Accepted July 17, 2013; published online October 29, 2013) Purpose: Hypovascular that exhibit hypointensity in hepatocyte-phase images of gadoxetic acid-enhanced magnetic resonance (MR) imaging are frequently encountered in clinical practice. We investigated risk factors for the development of these into hypervascular hepatocellular carcinoma (HCC). Methods: We retrospectively reviewed our institutional database and identiˆed 302 patients who underwent gadoxetic acid-enhanced MR imaging for suspected or conˆrmed HCC from February 1, 2008 to January 30, We excluded patients who were examined for metastasis of other malignancies or for other hepatic tumors, such as focal nodular hyperplasia. We identiˆed hypovascular that were hypointense in hepatocytephase images, recorded their characteristics, and calculated the cumulative hypervascularization rate for that were followed up. Results: Of the 302 patients, 82 had hypovascular (178 ; mean size, 9.3 mm). Sixty were followed up for over 6 months, and eight progressed to hypervascular HCC. Hypervascularization occurred more frequently in with fat than those without (Pº0.01). The cumulative hypervascularization rate was 5.1z over a year. Conclusion: The presence of intralesional fat was found to be a risk factor for hypervascularization of hypovascular that exhibited hypointensity in the hepatocyte-phase images of gadoxetic acid-enhanced MR imaging. Keywords: gadoxetic acid, hepatocellular carcinoma, liver, magnetic resonance imaging Introduction Hepatocellular carcinoma (HCC) is the sixth most common cancer worldwide and the third leading cause of cancer-related deaths. 1 3 The prognosis forpatientswiththisdiseaseremainsdismalbecause of its aggressive nature and frequent association with underlying cirrhosis. To improve the possibility of therapeutic intervention and patient survival, early detection of HCC is vital. 4 6 However, despite ešorts to improve the diagnostic accuracy of noninvasive imaging methods, the sensitivity *Corresponding author, Phone: , Fax: , t-mri@tt.rim.or.jp and speciˆcity of dynamic multidetector computed tomography (CT) and magnetic resonance (MR) imaging for detecting HCCs smaller than 2 cm remain unsatisfactory. 7,8 The limitations of these modalities are mainly because of the di culty in detecting hypovascular and isovascular lesions that occur during the early stages of multistep hepatocarcinogenesis. 9 Gadoxetic acid is a recently developed MR contrast agent that is speciˆcally taken up by hepatocytes and provides a better lesion-liver contrast, which is not achievable with extracellular gadolinium-based agents Gadoxetic acid-enhanced MR imaging is known to have a higher sensitivity than dynamic CT for detecting HCC, especially le- 281

2 282 D. Joishi et al. sions smaller than 2 cm in diameter Therefore, this technique may more readily detect early-stage lesions than CT arterial portography, 16 which was believed to have the highest sensitivity for this purpose. 9 The increased use of gadoxetic acid for routine examination may lead to the more frequent detection of hypovascular that are hypointense on hepatocyte-phase (15 to 20 min after administration of gadoxetic acid) images of gadoxetic acidenhanced MR imaging. However, the management of these is a contentious issue. Several recent studies have reported the detection of hypovascular by gadoxetic acid-enhanced MR imaging, though the occurrence of hypervascular HCC varied. For this reason, we believe that results from dišerent groups of patients are important to identify a treatment or surveillance strategy for these small hypovascular. The purpose of the present study was to investigate the risk factors for hypovascular developing into hypervascular HCC. Methods Patients This study was conducted in accordance with the principles of the Declaration of Helsinki and approved by our institutional review board. The need for written informed consent was waived because of the retrospective design of the study. We retrospectively reviewed our institutional database and identiˆed 302 patients who underwent gadoxetic acid-enhanced MR imaging for suspected or conˆrmed HCC from February 1, 2008 to January 30, We excluded patients examined for metastasis of other malignancies or for other hepatic tumors, such as focal nodular hyperplasia. For all cases, we documented patient age, sex, etiology of chronic liver disease, Child-Pugh classiˆcation, tumor markers (alpha-fetoprotein [AFP] and protein induced by vitamin K absence or antagonist II [PIVKA II]), presence of other hypervascular HCCs, and history of interventional treatment. MR imaging protocol MR imaging in this study was performed with a 1.5- (Signa ES version 11.1, Signa HD TS version 12, GE Healthcare, Waukesha, WI, USA) or 3-tesla (Discovery MR750, GE Healthcare) scanner using either an 8- (1.5T) or 32-channel (3T) body-array surface coil. First, a T 1 -weighted fast spoiled gradient echo (T 1 W-FSPGR) sequence was obtained using in- and opposed-phase imaging. Next, noncontrast images with fat-suppressed T 1 -weighted gradient echo imaging with a 3D acquisition sequence (liver acquisition with volume acceleration [LAVA]) were obtained with a parallel imaging technique (array spatial sensitivity encoding technique [ASSET]; factor=2); scanning time was 16 to 20 s. For dynamic scans, the contrast agent (0.025 mmol/kg; 0.1 ml/kg) was administered over 4 seconds followed by a 30-mL saline ush using a power injector (Sonic Shot GX, Nemoto- Kyorindo, Tokyo, Japan). Fluoroscopic triggering was used to determine the timing of the arterialphase image. After administration of mmol/ kg gadoxetic acid, arterial-, portal venous- (60 s after injection), and late-phase (120 to 180 s after injection) images were sequentially obtained. Between the late phase and ˆnal hepatocyte phase (15 to 20 min), T 2 -weighted single-shot fast spin-echo sequence, fat-suppressed T 2 -weighted fast spin-echo sequence, and dišusion-weighted images (DWIs) were obtained. CT protocol CT was performed with 64-multidetector-row scanners LightSpeed VCT (GE Healthcare), Discovery 750 HD (GE Healthcare), or Aquilion 64 (Toshiba Medical Systems, Otawara, Japan). After acquisition of precontrast CT images, 2 ml/kg of a nonionic contrast medium was administered intravenously for 30 s using a power injector (Dual Shot Type-D, NemotoKyorindo). Arterial-phase images were obtained after a scan delay of 40 s, portal venous-phase images after 80 s, and equilibrium-phase images after 180 s. The slice thickness of precontrast images was 5.0 mm, and that of postcontrast images was 1.25 mm. Nodules We included that demonstrated (1) round or oval shape and hypointense appearance on hepatocyte-phase images in both the axial and coronal planes; (2) no hypervascularity on arterialphase images obtained using dynamic CT or gadoxetic acid-enhanced MR imaging; and (3) no distinct high signal intensity on T 2 -weighted images (T 2 WIs) that suggested hemangioma and cyst and that were (4) detected before interventional treatment. In the case of follow-up assessment, we included those monitored longer than 6 months. Assessment of nodule hypervascularization We assessed hypervascularization using gadoxetic acid-enhanced MR imaging or dynamic CT. Hypervascular HCC was deˆned when a nodule showed hyperintensity in arterial-phase images and Magnetic Resonance in Medical Sciences

3 Hypovascular Nodules on EOB-MRI hypointensity in hepatocyte-phase images of gadoxetic acid-enhanced MR imaging or in portal venous- or equilibrium-phase images of dynamic CT. A lesion that was isointense in arterial-phase images and hypointense in precontrast images was also deemed hypervascular HCC. Image interpretation Together, 2 radiologists (D.J. and A.U., with 8 and 11 years' experience in hepatobiliary imaging) evaluated the gadoxetic acid-enhanced MR imaging and dynamic CT using a picture archiving and communication systems (PACS). Any disagreement between the 2 readers was solved in conference with a third radiologist (A.T., 30 years' experience). Nodule size was evaluated on axial-plane hepatocyte-phase images. The major and minor axes were measured on the monitor. Measurements were repeated 3 times, and the median size was recorded. Nodules were evaluated for the presence of fat and signal changes on T 1 -weighted images (T 1 WIs), T 2 WIs, and DWIs. The presence of fat was considered positive if the signal intensity of the was decreased on opposed-phase T 1 WIs, which suggests the existence of an intracellular lipid component. In T 2 WIs and DWIs, the signal intensity of the nodule was visually classiˆed as high if it exceeded that of the liver parenchyma. On T 1 WIs, the nodule intensity was categorized as low if it was lower than that of the background liver on the precontrast LAVA images. For patients who underwent gadoxetic acid-enhanced MR imaging 6 months or more after initial assessment, the follow-up interval and follow-up period (total follow-up time) were ascertained. 283 Detected In 82 of the 302 patients, 178 hypovascular were detected on gadoxetic acid-enhanced MR imaging. The mean number of per patient was 2.2±1.9, ranging from one to 11. Eighty-one in 48 patients were examined by gadoxetic acid-enhanced MR imaging only once, and 37 in 12 patients were examined at least twice with a follow-up time of less than 6 months. The remaining 32 patients with 60 underwent gadoxetic acid-enhanced MR imaging twice or moreandwerefollowedupafter6months(fig.1). Ten patients had at least one nodule that was followed up and one that was examined only once. Followed-up The median follow-up interval for the 60 followed up longer than 6 months was 188 days (range, 76 to 823 days), and the median follow-up period was 492 days (range, 199 to 1002 days). The mean initial size of the was 7.8±3.1 mm (range, 4 to 17 mm); 47 were smaller than 10 mm (47/60, 78z) and 58 were smaller than 15 mm (58/60, 97z). The mean number of per patient was 1.9±1.3 (range, one to 5). No reduction in lesion size was found during the follow-up period. Thirty-one increased in size eight progressed to hypervascular HCC (hypervascularized ), and 23 remained hypovascular (grown ). Twenty-nine remained stable in size (stable ) during the follow-up Statistical analysis We performed statistical analysis using IBM SPSS statistics version (IBM Co., Armonk, NY, USA). We used Mann-Whitney U test to compare 2 continuous values, Kruskal-Wallis one-way analysis of variance to compare 3 continuous values, and Fisher's exact test to compare proportions. Multivariate analysis of nodule imaging ˆndings was performed using logistic regression analysis. Kaplan-Meier time-to-event curves were used to estimate the cumulative rate of hypervascularization and were compared using the log-rank test. A 2-sided P valueº0.05 was considered signiˆcant in all analyses. Results Fig. 1. Flowchartusedinthisstudy Vol. 12 No. 4, 2013

4 284 D. Joishi et al. period. The median time for the to become hypervascularized was 463 days (range, 196 to 853 days). No signiˆcant dišerences in age, sex, etiology of liver disease, Child-Pugh classiˆcation, tumor markers, presence of other hypervascular HCC, past history of intervention, follow-up interval, or follow-up period were found among patients with only stable, grown, or a combination of grown and stable or with hypervascularized. The presence of fat was most frequently seen in the hypervascularized (Fig. 2), as determined by univariate and multivariate analyses (Pº0.01 for both). No statistically signiˆcant dišerences in initial nodule size, nodule signal intensity, follow-up interval, or follow-up period were observed among the stable, grown, or hypervascularized (Table 1). Cumulative hypervascularization rate The cumulative hypervascularization rates for the followed-up were 5.1z±2.9z at one year, 13.1z±5.2z at 2 years, and 37.6z±15.9z at 3 years. Here, we considered 1002 days as 3 years in calculating the 3-year cumulative rate. Comparative analysis of with and without fat showed that hypervascularization occurred earlier in those with fat (P=0.012, Fig. 3). The curves of larger than 10 mm and those smaller than or equal to 10 mm did not dišer signiˆcantly. Discussion Gadoxetic acid-enhanced MR imaging can enable dynamic contrast and liver-speciˆc imaging. 22 A higher lesion-to-liver contrast in the hepatocyte phase may improve the sensitivity of the detection of hepatocellular lesions, 14 16,23 especially early HCC. 24 In our study, we observed hypovascular that were hypointense in the hepatocytephase images in 27.2z (82/302) of the patients who underwent gadoxetic acid-enhanced MR imaging for suspected or conˆrmed HCC. Recent studies have shown that OATP8 is the most probable uptake transporter 25,26 and that its expression gradually decreases during multistep Fig. 2. Hypovascular nodule with minimal fat, hypervascularized during the follow-up period. Images for a 72-year-old woman show an initial hypovascular hypointense nodule with minimal fat and hypervascularization on follow-up after 2 years and 4 months. (a e) Gadoxetic acid-enhanced magnetic resonance (MR) imaging at initial assessment. Minimal signal reduction in the nodule (dotted circle) on an opposed-phase T 1 -weighted image (b) comparedwithaninphase T 1 -weighted image (a) indicates the presence of fat. Signal changes are not seen on the precontrast LAVA (liver acquisition with volume acceleration) image (c) or arterial-phase image (d). The hepatocyte-phase image shows hypointense measuring 6 4 mm in the Spiegel lobe (e, arrow).(f h) Followup gadoxetic acid-enhanced MR imaging 2 years and 4 months after the initial assessment. The lesion is isointense with the surrounding liver on the precontrast LAVA image (f) and shows hypervascularity on the arterial-phase image (g, arrow). The nodule is enlarged and measures 10 9 mm on the hepatocytephase image (h, arrow). Magnetic Resonance in Medical Sciences

5 Hypovascular Nodules on EOB-MRI 285 Table 1. Comparison of stable, grown, and hypervascularized per patient and nodule P value Hypervascularized Grown Stable Per nodule P value hypervascularized grown c only stable Per patient Univariate Multivariate Number of patients Number of ± ± ± Initial size (mean±sd) [mm] Presence of fat +/- 2/27 4/19 5/ T1WI low +/- 6/23 2/21 1/ N.A. T2WI high +/- 1/28 3/20 1/ N.A. DWI high +/- 0/29 2/21 0/ N.A. Age (mean±sd) [y] 66.7± ± ± Sex M/F 7/4 10/4 6/ Etiology a A/B/C 0/3/8 3/2/9 0/2/ Child-Pugh A/B, C 10/1 12/2 6/ AFP (ng/ml) b À20/Ã20 5/5 9/5 4/ PIVKA II (mau/ml) b Æ40/º40 8/2 11/3 6/ POHH +/- 4/7 7/7 4/ Past history of intervention +/- 1/10 5/9 2/ ( ) 145(75 823) 231(85 823) N.A. Follow-up interval median (range) (days) 199( ) 151(76 693) 232(85 823) Follow-up interval median (range) (days) 543( ) 464( ) 463( ) N.A. Total follow-up time median (range) (days) 538( ) 465( ) 463( ) Total follow-up time median (range) (days) a A, alcoholic; B, hepatitis B virus; C, hepatitis C virus. b Thirty-one cases in which alpha-fetoprotein (AFP) and protein induced by vitamin K absence or antagonist II (PIVKA II) were assayed. c grown but without hypervascularized. POHH, presence of other hypervascular hepatocellular carcimomas (HCCs). SD, standard deviation. T1WI/T2WI/DWI=T1-/T2-/diŠusion-weighted image. N.A., not assessed Presence of fat seemed to be positive if showed signal reduction on opposed-phase T1W fast spoiled gradient echo (FSPGR) images compared to in-phase images. Fig. 3. Cumulative hypervascularization rate for hypovascular that appeared hypointense on the hepatocyte-phase images of gadoxetic acid-enhanced magnetic resonance (MR) imaging, depending on the presence of fat. hepatocarcinogenesis. 27 However, the question remains whether all hypovascular detected by gadoxetic acid-enhanced MR imaging are premalignant lesions, such as high grade dysplastic nodule (HGDN) and early HCC. Because nearly half the followed-up were stable in size, it cannot be assumed that all detected were premalignant lesions. To determine the timing of treatment for HCCs, it is important to ascertain lesion hypervascularity. 28 Several recent studies have reported the hypervascularization of hypovascular detected on gadoxetic acid-enhanced MR imaging Table 2 summarizes those study results and ours. One risk factor for hypervascularization is nodule size. The results of these studies showed that larger than 10 mm were at a higher risk for hypervascularization. We did not ˆnd initial size of to be a predictor for hypervascularization, possibly because 78z of the followed-up were smaller than 10 mm and the number of the was relatively small. Another risk factor for nodule hypervascularization is the presence of fat, a ˆnding of our study as well as Motosugi's. 17 Because fatty changes in are a characteristic histopathological feature of HGDN and early HCC 29,30 and a gradual decrease in OATP8 expression has been shown during multistep hepatocarcinogenesis, 27 it is reasonable to consider most hypovascular with fat detected by gadoxetic acid-enhanced MR imaging to be HGDN or early HCC. In Fig. 3, 3 without fat became hyper- Vol. 12 No. 4, 2013

6 286 D. Joishi et al. Table 2. Summary of the results of 6 studies Number of Initial size of the nodule mean±sd (range) (mm) Number of hypervascularized One-year cumulative rate of hypervascularization Risk factor of hypervascularization Present study Motosugi et al. (17) Kumada et al. (18) Akai et al. (19) Takayama et al. (20) Takechi et al. (21) ±3.1 (4 to 17) 8(11.7z) 5.1z±2.9z Presence of fat ±3.6 (3 to 26) 16(11.9z) 15.6z±4.2z À10 mm Presence of fat 49 Hypervascularized 20 13(26.5z) 43.5z À15 mm (4 to 40) Absent 14 (8 to 40) (5 23) 17(13.1z) 3.2z À15 mm (not signiˆcant) (5 30) 31(30.1z) 18.4z±6.8z À9mm 112 Hypervascularized 10.9±3.5 Absent 7.1±2.4 27(24.1z) 30.4z(À10 mm) 5.2z(À10 mm) À10 mm vascular before with fat. Their initial sizes ranged between 5 to 8 mm, and all had synchronous or previous hypervascular HCC. Hence, we suspected them as intrahepatic metastatic in which the pattern of progression dišered from that of the typical early stage lesion of multistep hepatocarcinogenesis. Our study has several limitations. First, we obtained pathologic conˆrmation in only a few cases. However, radiological ˆndings obtained using gadoxetic acid-enhanced MR imaging are probably adequately informative for routine clinical use. Second, the follow-up interval was not ˆxed because of the retrospective study design. Third, because there were many small lesions, some measurement errors associated with nodular size may have occurred; however, we carefully repeated the measurements and magniˆed the images on the PACS monitor to minimize errors. Lastly, we did not include hyperintense in the hepatocyte phase. In conclusion, the presence of intralesional fat was found to be a risk factor for hypervascularization of hypovascular that exhibited hypointensity in the hepatocyte-phase images of gadoxetic acid-enhanced MR imaging. Acknowledgements We wish to thank Dr. Takayuki Abe for his kind advice regarding statistical analysis. References 1. Venook AP, PapandreouC, FuruseJ, deguevara LL. The incidence and epidemiology of hepatocellular carcinoma: a global and regional perspective. Oncologist 2010; 15 Suppl 4: Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, CA Cancer J Clin 2005; 55: Bosch FX, Ribes J, Diaz M, Cláeries R. Primary liver cancer: worldwide incidence and trends. Gastroenterology 2004; 127 (5 Suppl 1):S5-S Yuen MF, Cheng CC, Lauder IJ, Lam SK, Ooi CG, Lai CL. Early detection of hepatocellular carcinoma increases the chance of treatment: Hong Kong experience. Hepatology 2000; 31: Chan AC, Poon RT, Ng KK, Lo CM, Fan ST, Wong J. Changing paradigm in the management of hepatocellular carcinoma improves the survival beneˆt of early detection by screening. Ann Surg 2008; 247: Sato T, Tateishi R, Yoshida H, et al. Ultrasound surveillance for early detection of hepatocellular carcinoma among patients with chronic hepatitis C. Hepatol Int 2009; 3: Krinsky GA, Lee VS, Theise ND, et al. Transplantation for hepatocellular carcinoma and cirrhosis: sensitivity of magnetic resonance imaging. Liver Transpl 2002; 8: de Láedinghen V, Laharie D, Lecesne R, et al. Detection of in liver cirrhosis: spiral computed tomography or magnetic resonance imaging? A prospective study of 88 in 34 patients. Eur J Gastroenterol Hepatol 2002; 14: Kudo M. Multistep human hepatocarcinogenesis: correlation of imaging with pathology. J Gastroenterol 2009; 44 Suppl 19: Huppertz A, Haraida S, Kraus A, et al. Enhancement of focal liver lesions at gadoxetic acid-enhanced MR imaging: correlation with histopatho- Magnetic Resonance in Medical Sciences

7 Hypovascular Nodules on EOB-MRI 287 logic ˆndings and spiral CT initial observations. Radiology 2005; 234: Bluemke DA, Sahani D, Amendola M, et al. E cacy and safety of MR imaging with liverspeciˆc contrast agent: U.S. multicenter phase III study. Radiology 2005; 237: Vogl TJ, K äummel S, Hammerstingl R, et al. Liver tumors: comparison of MR imaging with Gd-EOB- DTPA and Gd-DTPA. Radiology 1996; 200: Di Martino M, Marin D, Guerrisi A, et al. Intraindividual comparison of gadoxetate disodium-enhanced MR imaging and 64-section multidetector CT in the detection of hepatocellular carcinoma in patients with cirrhosis. Radiology 2010; 256: Ichikawa T, Saito K, Yoshioka N, et al. Detection and characterization of focal liver lesions: a Japanese phase III, multicenter comparison between gadoxetic acid disodium-enhanced magnetic resonance imaging and contrast-enhanced computed tomography predominantly in patients with hepatocellular carcinoma and chronic liver disease. Invest Radiol 2010; 45: Kim YK, Kim CS, Han YM, et al. Detection of hepatocellular carcinoma: gadoxetic acid-enhanced 3-dimensional magnetic resonance imaging versus multi-detector row computed tomography. J Comput Assist Tomogr 2009; 33: Kogita S, Imai Y, Okada M, et al. Gd-EOB- DTPA-enhanced magnetic resonance images of hepatocellular carcinoma: correlation with histological grading and portal blood ow. Eur Radiol 2010; 20: Motosugi U, Ichikawa T, Sano K, et al. Outcome of hypovascular hepatic revealing no gadoxetic acid uptake in patients with chronic liver disease. J Magn Reson Imaging 2011; 34: Kumada T, Toyoda H, Tada T, et al. Evolution of hypointense hepatocellular observed only in the hepatobiliary phase of gadoxetate disodiumenhanced MRI. AJR Am J Roentgenol 2011; 197: Akai H, Matsuda I, Kiryu S, et al. Fate of hypointense lesions on Gd-EOB-DTPA-enhanced magnetic resonance imaging. Eur J Radiol 2012; 81: Takayama Y, Nishie A, Nakayama T, et al. Hypovascular hepatic nodule showing hypointensity in the hepatobiliary phase of gadoxetic acidenhanced MRI in patients with chronic liver disease: prediction of malignant transformation. Eur J Radiol 2012; 81: Takechi M, Tsuda T, Yoshioka S, et al. Risk of hypervascularization in small hypovascular hepatic showing hypointense in the hepatobiliary phase of gadoxetic acid-enhanced MRI in patients with chronic liver disease. Jap J Radiol 2012; 30: Tanimoto A, Lee JM, Murakami T, Huppertz A, Kudo M, Grazioli L. Consensus report of the 2nd International Forum for Liver MRI. Eur Radiol 2009; 19 Suppl 5:S975-S Ahn SS, Kim MJ, Lim JS, Hong HS, Chung YE, Choi JY. Added value of gadoxetic acid-enhanced hepatobiliary phase MR imaging in the diagnosis of hepatocellular carcinoma. Radiology 2010; 255: Sano K, Ichikawa T, Motosugi U, et al. Imaging study of early hepatocellular carcinoma: usefulness of gadoxetic acid-enhanced MR imaging. Radiology 2011; 261: Narita M, Hatano E, Arizono S, et al. Expression of OATP1B3 determines uptake of Gd-EOB- DTPA in hepatocellular carcinoma. J Gastroenterol 2009; 44: Kitao A, Zen Y, Matsui O, et al. Hepatocellular carcinoma: signal intensity at gadoxetic acid-enhanced MR imaging correlation with molecular transporters and histopathologic features. Radiology 2010; 256: Kitao A, Matsui O, Yoneda N, et al. The uptake transporter OATP8 expression decreases during multistep hepatocarcinogenesis: correlation with gadoxetic acid enhanced MR imaging. Eur Radiol 2011; 21: Kudo M. The 2008 Okuda lecture: management of hepatocellular carcinoma: from surveillance to molecular targeted therapy. J Gastroenterol Hepatol 2010; 25: Eguchi A, Nakashima O, Okudaira S, Sugihara S, Kojiro M. Adenomatous hyperplasia in the vicinity of small hepatocellular carcinoma. Hepatology 1992; 15: Kutami R, Nakashima Y, Nakashima O, Shiota K, Kojiro M. Pathomorphologic study on the mechanism of fatty change in small hepatocellular carcinoma of humans. J Hepatol 2000; 33: Vol. 12 No. 4, 2013