Case Report Hepatic adrenal rest tumor: Diagnostic pitfall and proposed algorithms to prevent misdiagnosis as lipid-rich hepatocellular carcinoma
Tomoko Sugiyama,1 Takuma Tajiri,1 Shinichiro Hiraiwa,1 Chie Inomoto,2 Hiroshi Kajiwara,2 Seiichiro Kojima,3 Kouske Tobita4 and Naoya Nakamura2
1Department of Diagnostic Pathology, 3Department of Internal Medicine, 4Department of Surgery,
Tokai University Hachioji Hospital, Tokyo, and 2Department of Pathology, Tokai University School of Medicine, Kanagawa, Japan
We present a case of adrenal rest tumor of the liver in which differential diagnosis from lipid rich-hepatocellular carcinoma (HCC) was challenging. The patient was a 50-year-old woman in whom a 3-cm tumorous mass was discovered in segment 7 of the liver during computed tomography evaluation of a uterine leiomyoma. The preop- erative diagnosis was HCC, and subsegmental liver resec- tion was performed. The tumor appeared as a well- demarcated golden yellow nodule consisting of clear or partially eosinophilic cells arranged in a trabecular pattern. The initial impression of this lesion was that of clear cell type or lipid-rich type HCC because it stained positive for Hep Par1, but negative for arginase-1 and positive for CD56 which is one of the neuroendocrine markers. The lesion also stained positive for SF-1 and 3-HSD, both of which are markers of adrenocortical tissue. The final diag- nosis was hepatic adrenal rest tumor. Hepatic adrenal rest tumor should be considered in the differential diagnosis of segment 7 tumor. A diagnostic algorithm that includes immunohistochemical staining for CD56 and arginase-1 is to rule out the possibility of lipid-rich HCC.
Key words: adrenal rest tumor of the liver, arginase-1, CD56, hepatocellular carcinoma
Adrenal rest tumor of the liver is quite rare. It is found most often in relatively young adults, and it occurs with equal frequency between sexes.1,2 Hepatic adrenal rest tumors are generally benign, but malignant tumors are some- times seen.3,4 Macroscopically, hepatic adrenal rest tumors
Correspondence: Tomoko Sugiyama, MD, Department of Diagnostic Pathology, Tokai University Hachioji Hospital, 1838 Ishikawa-machi, Hachioji, Tokyo 192-0032, Japan. Email: sugitomo@tokai-u.jp
Received 29 September 2014. Accepted for publication 1 Novem- ber 2014.
appears as well-circumscribed, yellowish, nodular lesions and are which usually located beneath the capsule of the right hepatic lobe (segment 7). Histologically, they are composed of cords of round to polygonal, pale, lipid-rich cells separated by vascular channels or bands of collagen.
Adrenal rest is thought to arise from ectopic adrenal tissue.5,6 It is usually a minute lesion of less than 7 mm in diameter and is generally detected incidentally at surgery or autopsy, and usually found in the celiac axis, testis or broad ligament, the walls of which are related to the embryogenic development of the adrenal gland. The adrenal cortex is derived from the coelomic mesoderm of the urogenital ridge at approximately 5 weeks’ gestational age and then sepa- rates from it during week 8. Ectopic adrenal tissue occurs when a fragment of the primitive adrenal gland breaks off during development. The fragment may come to rest in any visceral organ, but it most often locates in the kidney, liver, or gonad. Ectopic adrenal tissue may contain both cortex and medulla, if the breaking event occurs after migration of neural crest tissue into the cortex; it may contain only cortex if the breaking occurs before the medulla and cortex are joined. An adrenal rest consists solely of adrenocortical cells.
We report a rare case of a hepatic adrenal rest tumor that mimicked a lipid-rich hepatocellular carcinoma (HCC) histo- logically. We also strongly propose a diagnostic algorithm that incorporates immunohistochemistry to prevent misdiag- nosis of a hepatic adrenal rest tumor as a hepatic segment 7 tumor.
CLINICAL SUMMARY
A 50-year old woman was admitted to our hospital after a liver mass had been incidentally discovered upon computed
a
b
c
d
tomography (CT) performed for preoperative evaluation of a uterine leiomyoma. There were no marked abnormalities upon routine blood chemistry tests, and test results for anti-hepatitis B surface (HBs) antigen and hepatits C virus (HCV) antibody were negative. Levels of tumor markers alpha-fetoprotein (AFP) and protein-induced by vitamin K absence or antagonist-II(PIVKA-II) were within normal range. The patient was not an abuser of alcohol. Ultrasonography was performed and revealed a 3-cm heterogenous mass in the posterior- superior subsegment of the right hepatic lobe (segment 7). CT revealed that it appeared to be a well-demarcated low-density fat-containing tumor (Fig. 1a, b). Normal right adrenal gland was observed in the correct place. The radiologic features were suggestive of angiomyolipoma (AML), but the possibility of HCC was not ruled out. Thus, liver biopsy was performed. Histologically, the biopsy tissue consisted of hepatoid cells with clear or foamy cytoplasm and round nuclei and arranged in a trabecular pattern (Fig. 1c). Immunohistochemistry was performed, and the cell cytoplasm stained positive for Hep- Par1 (Fig. 1d). Together, the findings were consistent with HCC with fatty deposition; thus, subsegmental resection of the right hepatic lobe was undertaken.
PATHOLOGICAL FINDINGS
Macroscopically, a well-demarcated golden yellow tumor, measuring 20 x 30 mm, was located in the liver, which was noncirrhotic (Fig. 2a). The tumor showed exophytic growth, but surrounded by the liver capsule. No hemorrhage or
necrosis was seen. Microscopically, the tumor appeared to be surrounded by a thin fibrous capsule. (Fig. 2b). The tumor was composed mainly of clear and partially eosinophilic cells (Fig. 2c), arranged in a trabecular or alveolar fashion with sinusoid-like vasculature. The cytoplasm of the clear cells had a foamy appearance (Fig. 2d), and it did not stain with periodic acid-Schiff (PAS), indicating lipid-rich vacuoles. No bile plugs were seen. The nuclei were small without atypia, and mitotic figures were few. Immunohistochemically, the cytoplasm of the tumor cells was partially and weakly positive for Hep Par1 (Fig. 3a).
Although we initially thought it to be an HCC with marked fatty change, we noted that the tumor cell cytoplasm was strongly foamy and bubbly, suggesting the possibility of an adrenal rest tumor of the liver. Thus, we re-examined the surgical specimen and looked for adrenal cortical and liver specific markers as well as neuroendocrine markers because it is difficult to distinguish HCC from adrenocortical cells on the basis of histologic analysis alone. We initially performed immunohistochemical staining for liver marker Arginase-1 (Fig. 3b) as well as chromogranin-A, synaptophysin, and CD56 (Fig. 3c), which are neuroendo- crine markers, because adrenocortical cells derive from the endocrine system and stain positive for neuroendocrine markers. We also stained for adrenal markers steroidogenic factor-1 (SF-1) and 3ß-hydroxysteroid dehydrogenase (3ß- HSD) (Fig. 3d). The tumor cells were positive for vimentin, Hep Par1, synaptophysin, CD 56, MART-1, SF-1, and 3ß-HSD, and negative for keratin AE1/3, arginase-1, HMB45 and chromogranin A.
a
b
c
d
a Hep-Par 1
b Arginase-1
c CD56
d 3ß-HSD
Figure 3 Immunohistochemistry find- ings. (a) The cytoplasm of tumor cells was positive for Hep Par1. (b) The cytoplasm of tumor cells was negative for arginase-1, but that of hepatocytes was positive for arginase-1. (c) The cytoplasmic mem- brane of tumor cells was positive for CD56, whereas that of hepatocytes was negative. (d) The cytoplasm of tumor cells was also positive for 3ß-hydroxysteroid dehydrogenase (3ß-HSD).
Differential diagnosis and final diagnosis
In our differential diagnosis we considered the following:
Clear cell type HCC
We evaluated the possibility of a lipid-rich or clear cell-type HCC resembling adrenal rest tumor. However, @ 2015 Japanese Society of Pathology and Wiley Publishing Asia Pty Ltd
immunohistochemically, HCC is positive for arginase-1, which is a specific marker for hepatocytes.
Angiomyolipoma
We also considered the possibility of angiomyolipoma. Immunohistochemically, angiomyolipoma is positive for HMB 45 and MART-1,7 whereas adrenal rest tumor is negative for
HMB 45. It is important to bear in mind that adrenal rest tumor also is positive for MART-1.
Metastatic clear cell carcinoma of the kidney
Clear cell carcinoma of the kidney is PAS-positive; PAS stains glycogen-rich granules in cytoplasm. In our patient, however, no mass was depicted radiologically in the kidney, and there had been no clinical history of renal (clear-) cell carcinoma.
Histologically and immunohistochemically, the tumor cells had the appearance of adrenal cortical cells and were shown to be steroid-producing on the basis of SF-1 and 3ß-HSD. In addition, liver specific marker arginase-1 was absent. Although is it is difficult to discriminate between adrenal rest tumor and non-tumorous adrenal tissue such as ectopic adrenal tissue, the lesion was only composed of adrenal cortical cells (without medulla cells), and that was character- ized by an expanding growth surrounded by a fibrous capsule. Furthermore, nests of clear and eosinophilic tumor cells were randomly intermingled, without the typical zonal arrangement of adrenal gland cortical cells. So, we think this case is a benign neoplasm (adenoma). Thus, the final histo- pathologic diagnosis in this case was adrenal rest tumor of the liver.
The patient’s post-operative course was uneventful. Twenty months have passed since the surgery, and there has been no recurrence.
DISCUSSION
Histologic features of the tumor in this case mimicked HCC with fatty change, and cells were arranged in a trabecular pattern with sinusoid-like vasculature and lipid-rich cyto- plasm.8,9 It is impossible to distinguish adrenal rest tumor from HCC on the basis of histologic analysis alone. However, immunohistochemistry showed the tumor cells to be weakly positive for Hep Par1, which can be considered a specific marker for HCC.
Hep Par1 is indeed a sensitive marker for hepatocytes, 10 but other tumors, such as adenocarcinoma of the lung, gall- bladder, pancreas, and stomach, and adrenal gland carci- noma are also positive for Hep Par1.11 Most adrenal gland
tumors (including adenoma and hyperplasia) we have exam- ined have been positive for Hep Par1, especially eosinophilic cells of the zona reticularis. These histologic and immuno- histochemical similarities can pose a diagnostic pitfall when adrenal rest tumor of the liver is encountered.
Recently, a more specific marker, arginase-1, has been reported for HCC and hepatocytes. Arginase-1 is a binuclear manganese metalloenzyme that catalyzes the hydrolysis of arginine to ornithine and urea.12 Non-hepatocellular tumors including neuroendocrine tumor, melanoma, gastric adeno- carcinoma, and adrenocortical carcinoma, all of which can mimic HCC, have been found to be less reactive to anti- arginase-1 than to anti-Hep Par1.13 Specific adrenocortical cell markers are SF-1, which is a steroidogenic-tissue- specific-transcription-factor, detected in steroidogenic tissues such as adrenal gland, testis, ovary, and placenta,14 and steroidogenic enzymes 3ß-HSD, which is a protein antigen related to adrenocortical steroidogenesis.15 The tumor in our case was positive for these markers, but tests for these adrenal markers are not usually available in most pathologic laboratories. Other sensitive markers of adrenal cortical cells are MART-1,16 synaptophysin, inhibin and CD56,17 all of which were present in the tumor cells in our case. Although the tumor cells in our case were selectively positive (lipid-rich cytoplasm) for synaptophysin, MART-1 and inhibin, and were strongly positive for CD56 (in cytoplasmic membrane). In particular, CD56 is an excellent marker for distinguishing adrenocortical cells from hepatocytes.
We searched in the PubMed data base and found only four case reports of hepatic adrenal rest tumor (Table 1). All four tumors were located in the posterior segment (segment 7).14,18-20 Mean age at the time of diagnosis was 60.8 years. Mean tumor diameter was 24.4 mm. None of the four tumors was malignant, but malignancy of such tumors at other sites including the retroperitoneum has been reported. Tumor fea- tures in our case matched those of the four reported cases.
Adrenal rest tumor should be included in the differential diagnosis in of a segment 7 liver tumor. Macroscopic obser- vation of golden yellow tissue resembling adrenal cortex is also helpful. Because of the histologic similarity between HCC, angiomyolipoma and adrenal rest tumor and it is impossible to differentiate only by histology. Immunohisto- chemical analysis must be added to the diagnostic protocol for a final determination of adrenal rest tumor. For differential
| Author | Year | Gender | Chief complaints | Clinical diagnosis | Location | Tumor size | Follow up |
|---|---|---|---|---|---|---|---|
| Arai K et al.14 | 62 | M | HCV follow up no symptoms | AML susp | post segment | 25 mm | uneventful |
| Tajima T et al.18 | 55 | F | no symptoms | AML susp | post segment | 25 mm | uneventful |
| Baba Y et al.19 | 67 | F | medical check symptoms (-) | HCC or AML susp | post segment | 17 mm | uneventful |
| Yong MS20 | 62 | M | (-) | HCC susp | post segment | 30 mm | uneventful |
Diagnostic algorithm for a segment 7 liver tumor
Liver tumor is golden yellow (segment 7 )
☒
CD56 (-) Arginase-1 (+) HMB 45 (-)
CD56 (+) Arginase-1 (-) HMB 45 (-)
CD56 (-) Arginase-1 (-) HMB 45 (+)
Hepatocellular carcinoma
SF-1(+) 3BHSD(+)
Angiomyolipoma
Adrenal rest tumor
diagnosis, here we strongly propose a diagnostic algorithm that includes CD56, arginase-1, and HMB 45 (MART-1) immunostaining (Fig. 4). This immunostaining pattern will prove to be very useful and practical to ordinary pathologists in preventing misdiagnosis of a hepatic adrenal rest tumor as a lipid-rich HCC. There have been no documented case reports covering the diagnostic pitfall that arises when an adrenal rest tumor is encountered and none introducing an algorithm to prevent the misdiagnosis of such tumors.
In conclusion, we describe a rare case of adrenal rest tumor in segment 7 of the liver as well as the diagnostic hurdles that must be overcome. We must keep in mind that Hep Par1 is not a reliable marker for differentiating adrenal rest tumor from HCC. We need to use arginase-1 as a specific marker to rule out HCC and CD56 to properly identify the adrenal rest tumor. In a case of low density fat-containing segment 7 liver nodule, we must consider hepatic adrenal rest tumor by one of the clinicopathologic differential diagnosis.
DISCLOSURE
There is no possible conflict of interest related to this paper.
REFERENCES
1 Ishak KG, Goodman ZD, Stocker JT. Atlas of Tumor Pathology: Tumors of the Liver and Intrahepatic Bile Ducts, Third Series,
Fascicle 31. Washington, DC: Armed Forces Institute of Pathol- ogy, 1999; 152-4.
2 Hyams VJ, Thomas DF, Sarkisian SS. Adrenal rest tumor of the liver. Am J Surg 1960; 99: 960-3.
3 Goren E, Engelburg IS, Eidelman A. Adrenal rest carcinoma in hilum of kidney. Urology 1991; 38: 187-90.
4 Yokoyama H, Adachi T, Tsubouchi K, Tanaka M, Sasano H. Non-functioning adrenocortical carcinoma arising in an adrenal rest: Immunohistochemical study of an adult patient. Tohoku J Exp Med 2013; 229: 267-70.
5 Moore KL, Persaud TVN. The Developing Human: Clinically Oriented Embryology, 7th edn. Philadelphia: Saunders, 2003; 288-306.
6 Ye H, Yoon GS, Epstein JI. Intrarenal ectopic adrenal tissue and renal-adrenal fusion: A report of nine cases. Mod Pathol 2009; 22: 175-81.
7 Tsui WM, Colombari R, Portmann BC et al. Hepatic angiomyolipoma: A clinicopathologic study of 30 cases and delineation of unusual morphologic variants. Am J Surg Pathol 1999; 23: 34-48.
8 Honjo K, Koike J, Odanaka M et al. Large hepatocellular carci- noma with diffuse fatty deposition, a case report. Shindan Byouri 2004; 21: 325-8 (in Japanese).
9 Tanaka S, Kitamara T, Imaoka S, Sasaki Y, Taniguchi H, Ishiguro S. Hepatocellular carcinoma: Sonographic and histo- logic correlation. AJR Am J Roentgenol 1983; 140: 701-7.
10 Wennerberg AE, Nalesnic MA, Coleman WB. Hepatocyte par- affin 1: A monoclonal antibody that reacts with hepatocytes and can be used for differential diagnosis of hepatic tumors. Am J Pathol 1993; 143: 1050-4.
11 Lugli A, Tornillo L, Mirlacher M, Bundi M, Sauter G, Terracciano LM. Hepatocyte paraffin 1 expression in human normal and neoplastic tissues: Tissue microarray analysis on 3,940 tissue samples. Am J Clin Pathol 2004; 122: 721-7.
12 Yan BC, Gong C, Song J et al. Arginase-1: A new immunohis- tochemical marker of hepatocytes and hepatocellular neo- plasms. Am J Surg Pathol 2010; 34: 1147-54.
13 Alessandro L, Luigi T, Martina M, Marcel B, Guido S, Luigi MT. Hepatocytte paraffin 1 expression in human normal and neo- plastic tissues. Tissue microarray analysis on 3940 tissue samples. Am J Clin Pathol 2004; 122: 721-27.
14 Arai K, Muro H, Suzuki M, Oba N, Ito K, Sasano H. Adrenal rest tumor of the liver: A case report with immunohistochemical investigation of steroidogenesis. Pathol Int 2000; 50: 244-8.
15 Sasano H, Mason JI, Sasano N, Nagura H. Immunolocalization of 3ß-hydroxysteroid dehydrogenase in human adrenal cortex and in its disorders. Endocr Pathol 1990; 1: 94-101.
16 Makhlouf HR, Ishak KG, Shekar R, Sesterhenn IA, Young DY, Fanburg-Smith J. Melanoma markers in angiomyolipoma of the liver and kidney: A comparative study. Arch Pathol Lab Med 2002; 126: 49-55.
17 Sangoi AR, McKenney JK. A Tissue microarray-based comparative analysis of novel and traditional immunohisto- chemical markers in the distinction between adrenal cortical lesions and pheochromocytoma. Am J Surg Pathol 2010; 34: 423-32.
18 Tajima T, Funakoshi A, Ikeda Y et al. Nonfunctioning adrenal rest tumor of the liver: Radiologic appearance. J Comput Assist Tomogr 2001; 25: 98-101.
19 Baba Y, Beppu T, Imai K et al. A case of adrenal rest tumor of the liver: Radiological imaging and immunohistochemical study of steroidogenic enzymes. Hepatol Res 2008; 38: 1154-8.
20 Yong MS. Hepatic adrenal rest tumor mimicking hepatocellular carcinoma. Korean J Hepatol 2010; 16: 338-41.
Copyright of Pathology International is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.