ORIGINAL ARTICLE Annals of Nuclear Medicine Vol. 12. No. 5, 275-280, 1998 
Potential for qualitative diagnosis of tumors and tumorous lesions in the liver with Tc-99m-GSA SPECT -Correlation with pathological evaluation and MRI findings-
Kazuhiro SArro,*,** Kiyoshi KoIZUML** Kimihiko ABE,** Yasuyuki GoTo* and Tomoyuki SEKI*** 
*Department of Radiology, Musashino Red Cross Hospital **Department of Radiology and ***Department of Internal Medicine, Tokyo Medical College 
To evaluate the effect of technetium-99m-labeled DTPA-galactosyl human serum albumin (Tc-99m-GSA) SPECT imaging for qualitative diagnosis of hepatic lesions. The subjects were 29 patients with pathologically confirmed hepatic lesions (21 malignant and 8 benign lesions). SPECT data were obtained at about 30 minutes after injecting 185 MBq (5 mCi) of Tc-99m-GSA. The GSA SPECT findings were compared with those of pathological evaluation and T2-weighted MR images (T2WI). Of 29 lesions, 17 showed decreased accumulation, and three exhibited increased accumulation. The other nine lesions were undetectable. The malignant lesions which showed increased accumulation were all well differentiated hepatocellular carcinomas (HCCs). One of the eight benign lesions exhibited increased accumulation. The three lesions which showed increased accumulation of GSA exhibited hypointensity on T2WI, whereas the malignant lesions which showed decreased accumulation of GSA exhibited hyperintensity on T2WI. The GSA SPECT findings correlate well with those of T2WI. GSA SPECT may be useful for qualitative diagnosis of focal liver lesions. If a lesion is suspected of being HCC, increased accumulation ma indicate well differentiated HCC. 
Key words:  liver, SPECT, MRI, hepatocellular carcinoma, Tc-99m DTPA-galactosyl human serum albumin
INTRODUCTION 
TECHNETIUM-99m-labeled DTPA-galactosyl human se-rum albumin (Tc-99m-GSA) is a liver scintigraphic agent which binds specifically to asialoglycoprotein receptors (ASGPR) on the membrane of hepatocytes. It is mainly used for evaluating hepatic function,1-4 and has the potential for use in the qualitative diagnosis of hepatic tumors on the basis of the presence or absence of normal hepatocytes in the lesions.5-8 Several papers have reported that hepatocellular carcinoma (HCC) and metastatic liver tumors lack ASGPR on the cell membranes9'ro and that Tc-99m-GSA SPECT consequently showed decreased accumulation in those tumors 5,7,8,11 but there have been no reports of the use of Tc-99m-GSA SPECT imaging for the qualitative diagnosis of hepatic tumors, whereas magnetic resonance imaging (MRI) is useful for qualitative diagnosis of liver lesions. We compared the findings of Tc-99m-GSA SPECT with those of pathological evaluation and MRI, and evaluated the potential of Tc-99m-GSA SPECT imaging for qualitative diagnosis of hepatic tumors . 
MATERIALS AND METHODS 
The subjects were 29 patients with hepatic lesions which were pathologically confirmed: 14 lesions by surgery, 14 by biopsy and one by autopsy. Surgery and biopsies were performed within 2 months of radionuclide studies and MRI. Of the 29 lesions, 21 were malignant lesions: 10 HCCs (8 well differentiated, 2 moderately differentiated), seven metastatic adenocarcinomas (all from colorectal carcinomas), two cholangiocellular carcinomas and two carcinoid tumors. The remaining eight lesions were benign and included five cases of chronic hepatitis or cirrhosis, one non-neoplastic lesion in alcoholic hepatitis, one angiomyolipoma and one submassive necrosis. SPECT images were obtained with a rotating single head gamma camera (ZLC-DIGITRAC Orbiter, Shimadzu-Siemens, Kyoto, Japan) interfaced with a computer (SCINTIPAC-24000, Shimadzu. Kyoto, Japan). Immediately after the injection of 1 85 MBq (5 mCi) of Tc-99m-GSA (Nihon Medi-Physics, Nishinomiya, Japan) into an antecubital vein, dynamic acquisition over the heart and liver was performed for 30 minutes with the patients in the supine position to obtain sequential planar anterior images and generate time-activity curves for the heart and liver. The SPECT data were obtained in a matrix size of 64 x 64, in 64 views taken with 36deg. erotation, with 10 seconds imaging for each view. MRI was performed with a 1.O or 1.5 T superconduc-tive Magnex (Shimadzu, Kyoto, Japan) in 24 cases. T1-weighted image (TIWI: TR 500 msec/TE 20 msec), T2-weighted image (T2WI: TR 2000 msec/TE 90 msec) and dynamic images (TR 200 msec/TE 9 msec, flip angle 60deg.) were obtained. Lesion size was determined mainly with surgical specimens. In some unoperated cases, however, it was determined by MRI or ultrasonography (US). The largest diameter seen in these images was regarded as the lesion size. The findings of Tc-99m-GSA SPECT were compared with those of pathological evaluation and MRI, especially T2WI. Statistical comparison of the detectable lesions and undetectable lesions was done by Student's t-test. 
RESULT 
Comparison with pathologicalfindings Table 1 shows the relationships between the Tc-99m-GSA SPECT findings and the pathological findings. Of the 29 lesions, 15 exhibited decreased accumulation and three showed increased accumulation. The other 11 le-sions were undetectable. The mean size of the detectable lesions was 44.3+-29.2 mm (mean +- standard deviation), and that of the undetectable lesions was 16.2+-9.0 mm. A significant difference was found between the two groups (p = 0.0047). Regarding the 10 HCCs, five lesions (2 moderately differentiated, 3 well differentiated) showed decreased accumulation, but two well differentiated HCCs exhibited increased accumulation (Fig. 1 ). The other three well differentiated HCCs were not detected. Of the seven metastatic lesions, four showed decreased accumulation (Fig. 2) and the other three were not detected. All of the cholangiocellular carcinomas and carcinoid tumors exhibited decreased accumulation. Of the 21 malignant lesions, 13 (62%) showed decreased accumulation. The malignant lesions which exhibited increased accumulation were all well differentiated HCCs. In the five cases of chronic hepatitis and cirrhosis, which were suspected of being HCC on US or CT, GSA SPECT was unable to detect any focal lesions. The final diagnoses of these lesions were confirmed to be nontumor lesions by biopsy. One non-neoplastic lesion caused by alcoholic hepatitis showed increased accumulation (Fig. 3). This lesion was confirmed to be liver fibrosis by biopsy and was pathologically not different from the surrounding liver parenchyma. Two other benign lesions, i.e., angiomyolipoma and submassive necrosis, exhibited decreased accumulation. Of the eight benign lesions, one showed increased accumulation, two exhibited decreased accumulation and five were undetected. The size of the undetectable benign lesions was also significantly smaller than that of the detectable benign lesions. 
Comparison with MRlfindings Tables 2 and 3 show the relationships between the Tc-99m-GSA SPECT findings and the MRI findings. Three lesions which showed increased accumulation of GSA exhibited hypointensity on T2WI and hyperintensity on TIWI (Fig. 1, Fig. 3). Regarding the well differentiated HCCs, the hypointense lesions seen with T2WI also exclusively exhibited increased accumulation of GSA (Table 4). CT during arterial portography (CTAP) was performed in one of the two well differentiated HCCs which showed increased accumulation of GSA, and this disclosed hyperperfusion of the lesion (Fig. 1). The malignant lesions which showed decreased accumulation of GSA exhibited hyperintensity on T2WI. Of the eight lesions undetectable by GSA SPECT but evaluable by MRl, four lesions (50%) exhibited isointensity on T2WI and hyperintensity on T1WI. Biopsy confirmed three of these four lesions to be non-tumorous. 
DISCUSSION 
TC-99m-GSA has been used mainly for evaluating liver function.2-4 There have been a few reports evaluating the use of this scintigraphic agent for qualitative diagnosis of hepatic focal lesions.12 Because this scintigraphic agent binds specifically to ASGPR on the membrane of hepatocytes, evaluating the presence of hepatocytes in hepatic lesions offers the potential for qualitative diagnosis of the lesions .5-8,11 In this study, detectable lesions were significantly larger than undetectable lesions. SPECT imaging has been demonstrated to have poorer spatial resolution than other imaging procedures, such as US, CT and MRI. An earlier report pointed out that the rate detection of lesions smaller than 20 mm by SPECT is not very good,13 and severe liver dysfunction makes the accumulation of scintigraphic agents in the hepatic parenchyma heteroge-neous and consequently makes the detection of lesions difficult. As another possibility, undetectable lesions might ac-cumulate GSA equally with the surrounding liver, resulting in their escaping detection. Concerning the relation-ship with MR findings, 50% of the lesions which were undetectable with GSA SPECT showed isointensity on T2WI and hyperintensity on T1WI. This MR finding indicates that the lesions are adenomatous hyperplasia or early HCC, including well differentiated HCC.14 GSA might accumulate these lesions equally with the sur-rounding liver parenchyma, probably because these lesions have ASPGR.6 Decreased accumulation suggests the absence of or a decrease in the number of normal hepatocytes 8 All these lesions exhibited hyperintensity on T2WI. This MR finding is not specific to tumors, and it indicates the lesions are pathologically different from the surrounding liver paren-chyma. 
Three lesions exhibited increased accumulation of GSA. There might be two causes for this phenomenon. Well differentiated HCC is thought to have normally functioning hepatocytes and increased cellularity of the lesions, which causes increased accumulation of GSA. In fact, well differentiated HCC has been reported to show in-creased accumulation of GSA.15 Hyodo et al. reported several well differentiated HCCs having ASGPR.6 In the present study, these lesions exhibited hypointensity on T2WI, which indicates adenomatous hyperplasia or well differentiated HCC,14,16,17 The pathological findings of these two lesions are similar: increased cellularity with or without structural atypia is recognized in both lesions 18 From this viewpoint, it is expected that adenomatous hyperplasia also exhibits increased accumulation of GSA, but because the rate of cellularity in these two entities is different,18 the accumulation also might be different. In this study, adenomatous hyperplasia was not detected. More experience is necessary to distinguish well differentiated HCC from adenomatous hyperplasia. Another benign lesion also exhibited hypointensity on T2WI. Its GSA SPECT findings support hypercellularity. But when compared with the surrounding liver tissue, no specific finding was observed pathologically. Because the diagnosis was made by biopsy, only a portion of the lesion was evaluated; therefore, a precise pathological evaluation may not have been made. Pseudolesions are some-times observed in the anterior portion of the medial segment and caudate lobe when using CTAP, and hyper-plastic lesions arise in these lesions 19,20 In earlier reports, these hyperplastic lesions showed hyperintensity on T1WI and hypointensity on T2WI. The lesion we examined showed similar intensity to those reported cases. During radiological diagnosis, a hyperplastic lesion was suspected. If a biopsy were able to show the character of the entire lesion, another reason for the increased accumulation might be able to be speculated. The second cause of the increased accumulation may be increased blood perfusion. CTAP was performed in only one case of well differentiated HCC, and it confirmed increased portal blood flow. Previous reports discussed the accumulation seen in GSA as mainly being influenced by ASGPR, but there was no discussion of the blood flow.1,3 If increased focal blood flow were maintained, increased accumulation might also occur. In conclusion, GSA SPECT may be useful for achieving qualitative diagnosis of focal liver lesions as well as liver function. The findings of GSA SPECT correlate well with those of T2WI. Increased accumulation predicts hypercellularity and/or hyperperfusion of the lesions. If a lesion is suspected of being HCC, increased GSA accumulation may predict a well differentiated HCC. 
REFERENCES 
1. Kudo M, Todo A, Ikekubo K, Hino M. Receptor index viahepatic asialoglycoprotein receptor imaging: Correlation with chronic hepatocellular damage. Am J Gastroenterol 87: 865-870, 1992. 
2. Ha-Kawa SK, Tanaka Y. A quantitative model of techne-tium-99m-DTPA-galactosyl-HSA for the assessment of hepatic blood flow and hepatic binding receptor. J Nucl Med 32: 2233-2240, 1991 . 
3. Kudo M, Todo A, Ikekubo K, Hino M. Yonekura Y, Yamamoto K, et al. Functional hepatic imaging with receptor-binding radiopharmaceutical: Clinical potential as a measure of functioning hepatocyte mass. Gastroenterol Jpn 26: 734-741, 1991. 
4. Koizumi K, Uchiyama G. Arai T, AinodaT, Yoda Y. A new liver functional study using Tc-99m DTPA-galactosyl human serum albumin: Evaluation of the validity of several functional parameters. Ann Nucl Med 6: 83-87. 1992. 
5. Aburano T, Shuke N, Yokoyama K, Tonami N, Hisada K, Tanei M, et al. Discordant hepatic uptake of Tc-99m NGA and Tc-99m PMT in a patient with hepatoma. Clin Nucl Med 17: 793-796, 1992. 
6. Hyodo I, Mizuno M, Yamada G, Tsuji T. Distribution of asialoglycoprotein receptor in human hepatocellular carci-noma. Liver 13: 80-85, 1993. 
7. Stadalnik RC, Vera DR, Woodle ES, Trudeau WL, Porter BA, Ward RE, et al. Technetium-99m NGA functional hepatic imaging: Preliminary clinical experience. J Nucl Med 26: 1233-1242, 1985. 
8. Virgolini I, Muller C. Klepetko W, Angelberger P, Bergmann H, O'Grady J, et al. Decreased hepatic function in patients with hepatoma or liver metastasis monitored by a hepatocyte specific galactosylated radioligand. Br J Cancer 61 : 937-941, 1990. 
9. SawamuraT, Nakada H, Hazama H. Shiozaki Y, Sameshima Y. Tashiro Y. Hyperasialoglycoproteinemia in patients with chronic liver diseases and/or liver cell carcinoma. Asialoglycoprotein receptor in cirrhosis and liver cell car-cinoma. Gastroenterol 87: 1217-1221, 1984. 
10. Naitoh Y. Hyperasialoglycoproteinemia in patients with chronic liver disease. Acta Hepatol Jpn 28: 1179-1187, 1987. (in Japanese) 
11. Kubota Y, Kojima M, Hazama H, Kawa S, Nakazawa M, Nishiyama Y, et al. A new liver function test using the asialoglycoprotein-receptor system on the liver cell mem-brane: I. Evaluation of liver imaging using the Tc-99m-neoglycoprotein. K4KU IGAKU (Jpn J Nucl Med) 23: 899-905, 1986. (in Japanese) 
12. Kurtaran A, Muller C, Novacek G, Kaserer K, Mentes M, Raderer M, et al. Distinction between hepatic focal nodular hyperplasia and malignant liver lesions using technetium-99m-galactosyl-neoglycoalbumin. J Nucl Med 38: 1912-1915, 1997. 
13. Kudo M, Hirasa M, Takakuwa H. Ibuki Y, Fujimi K, Miyamura M, et al. Small hepatocellular carcinomas in chronic liver disease: Detection with SPECT. Radiology 159: 697-703. 1986. 
14. Muramatsu Y, Nawano S, Takayasu K, Moriyama N, Yamada T, Yamasaki S, et al. Early hepatocellular carci-noma: MR imaging. Radiology 181: 209-213, 1991 . 
15. Kudo M. Ikekubo K, Todo A, Mimura J. Okabe Y, Kashida H, et al. Clinical utility of receptor imaging in the assessment of liver function. Jpn J Gastroenterol 89: 1349-1359, 1992, (in Japanese) 
16. Mitchell DG, Stark DD. BenigD hepatocellular masses. In: Mitchell DG, Stark DD, eds. Hepatobiliary MRI: A Text-Atlas at Mid and High Field, St. Louis: Mosby Year Book, pp. 111-124, 1992. 
17. Matsui O, Kadoya M, Kameyama T. Yoshikawa J. Arai K. Gabata T, et al. Adenomatous hyperplastic nodules in the cirrhotic liver: Differentiation from hepatocellular carci-noma with MR imaging. Radiology 173: 123-126, 1989. 
18. Takayama T, Makuuchi M, Hirohashi S, Sakamoto M, Okazaki N. Takayasu K, et al. Malignant transformation of adenomatous hyperplasia to hepatocellular carcinoma. Lancet 336: 1150-1153, 1990. 
19. Paulson EK. Baker ME, Spritzer CE. Leder RA. Gulliver DJ. Meyers WC. Focal fatty infiltration: A cause of nontumorous defects in the left hepatic lobe during CT arterial portography. J Comput Assist Tomogr 17: 590-595. 1993. 
20. Matsui O, Takahashi S, Kadoya M, Yoshikawa J, Gabata T, Takashima T, et al. Pseudolesion in segment 4 of the liver at CT during arterial portography: Correlation with aberrant gastric venous drainage. Radiology 193: 31-35, 1994.