- Email: [email protected]
Steatotic hepatocellular adenomas with different phenotypic subtypes: A case report
Clinics and Research in Hepatology and Gastroenterology (2015) 39, xxx.e17—xxx.e22
Available online at
ScienceDirect www.sciencedirect.com
CASE REPORT
Steatotic hepatocellular adenomas with different phenotypic subtypes: A case report N. Alberti a,∗, C. Castain b, A. Crombe a, N. Frulio a a
Department of Radiology, hôpital St-André, CHU de Bordeaux, 1, rue Jean-Burguet, 33000 Bordeaux, France b Department of Pathology, hôpital Pellegrin, CHU de Bordeaux, place Amélie-Raba-Léon, 33000 Bordeaux, France Available online 26 September 2014
Summary Hepatocellular adenomas (HCAs) are heterogeneous group of benign tumors; three pathomolecular subtypes have been identified so far: hepatocyte nuclear factor 1 ␣—inactivated HCA (H-HCA) (35—40%), inflammatory HCA (I-HCA) (> 50%), -catenin activated HCA (10%). Ten percent of I-HCA are also -catenin activated. We report a rare case of three histologically confirmed steatotic HCAs of three different phenotypes: I-HCA, -catenin activated I-HCA, and H-HCA in a 36-year-old woman. This observation outlines that in the same patient, HCA may be of different subtypes. The predisposition to develop different HCA hypothetically caused by a ‘‘benign tumorigenic field effect’’ may result, even rarely, in different genotypes/phenotypes such as H-HCA and I-HCA. Moreover, we illustrate the very high specificity of MR for subtyping HCA. © 2014 Elsevier Masson SAS. All rights reserved.
Introduction Hepatocellular adenoma (HCA) is a rare type of benign monoclonal liver neoplasm, occurring mostly (85%) in young women using oral contraceptives. The estimated prevalence is three to four per 100,000 in Europe and North America
∗ Corresponding author. Tel.: +33 5 56 79 58 00; fax: +33 5 56 79 58 99. E-mail address: [email protected] (N. Alberti).
http://dx.doi.org/10.1016/j.clinre.2014.08.007 2210-7401/© 2014 Elsevier Masson SAS. All rights reserved.
[1—3]. Three pathomolecular subtypes have been identified so far [1]: hepatocyte nuclear factor 1 ␣—inactivated (HHCA) (35—40%), inflammatory HCA (I-HCA) (>50%), -catenin activated HCA (10%). In addition, 10% of I-HCAs are -catenin activated. HCAs of any subtype can be solitary or multiple. Multiple HCAs are usually of the same subtype. Steatosis may be observed, to a varying degree, in all the subtypes of HCAs. In this paper, we report an unusual association of three ‘‘steatotic HCAs’’ that belonged to three different phenotypes (-catenin activated I-HCA, I-HCA and H-HCA) in a 36-year-old woman.
xxx.e18
N. Alberti et al.
Case history A 36-year-old woman was admitted to undergo evaluation of 2 hyper echogenic liver nodules detected during abdominal ultrasonography realised for abdominal pain in mid-2013. She was taking oral contraception for 18 years until the last 4 months (tubal ligation). Clinical examination was normal, excepted BMI value of 29. Laboratory analysis revealed elevated C-reactive protein (CRP) (50 UI/L; normal, 0—5), and abnormal liver function test with elevated alanine aminotransferase (ALT) and aspartate aminotransferase (AST) (120 and 320 U/L respectively). Serum alpha-fetoprotein was normal and there was no serological evidence of active hepatitis B or C. MRI showed the two liver nodules in segment II/III (76 mm; lesion 1) and in segments I, IV and VIII (53 mm; lesion 2) and also a third one in segment VI (8 mm; lesion 3). On T2-weighted images (Fig. 1), lesion 1 was heterogeneous, showing a marked hyperintensity (defined as at least as intense as the spleen signal) both in the periphery (‘‘atoll-like’’) and in the center (nodular aspect, asterisk). Lesion 2 was homogeneous and markedly hyperintense. Lesion 3 was also homogenous and slightly hyperintense. On chemical shift sequences (Fig. 2) signal dropout was found in opposed-phase in the 3 lesions, assessing the presence of fatty deposition: focally for lesions 1 (percentage of estimated volume: less than 5%) and 2 (percentage of estimated volume: more than 90%), diffuse and homogeneous for lesion 3. After injection, in the arterial phase (Fig. 3), lesion 1 showed a diffuse enhancement, more marked in the
Figure 1 Axial T2-weighted image SSFSE (Single Shot Fat Spin Echo). Two nodules localized in segment II/III (76 mm; lesion 1) and in segments I, IV and VIII (53 mm; lesion 2). Lesion 1 showed an ‘‘atoll sign’’, characterized by a hyperintensy band on T2weighted images in the periphery of the lesion, and also an hyperintense center (‘‘nodule-in-nodule’’ appearance). Lesion 2 shows marked hyper intensity of the entire nodule.
central nodular area (asterisk); lesion 2 presented peripheral intense enhancement, whereas no enhancement was observed for lesion 3. On delayed-phase (5mn), no wash out was found in any lesion; lesion 1 showed a diffuse persistent
Figure 2 Chemical shift sequences. Signal dropout in opposed-phase was found in the 3 lesions, assessing the presence of fatty deposition: focally for lesions 1 (less than 5% of the nodule) and 2 (more than 90%) (white arrows and asterisk respectively); diffuse and homogeneous for lesion 3, localized in segment VI (8 mm) (black arrows).
Steatotic hepatocellular adenomas with different phenotypic subtypes: A case report
Figure 3 Axial T1-weighted image after chelates of gadolinium contrast injection, with fat suppression (arterial phase). Lesion 1 showed a diffuse enhancement, more marked in the central nodular area (asterisk); lesion 2 presented peripheral intense enhancement.
enhancement, and the peripheral rim of lesion 2 was also hyperintense. The patient underwent a left hepatectomy and a segment VI tumorectomy two months later. Lesion 1 (Fig. 4) was resected and corresponded to a well-differentiated
xxx.e19
hepatocellular proliferation, intermingled with inflammatory infiltrates, and marked sinusoidal dilatation with congestion; the whole was characteristic of I-HCA, confirmed by positive expression of CRP in tumoral hepatocytes. In addition, in the center of the tumor, there were some cholestatic changes, hepatocyte plates were slightly thicker, the reticulin network was focally irregular and decreased but there was no obvious cytological atypia. Moreover, hepatocytes exhibited strong glutamine synthetase (GS) staining, associated with aberrant nuclear and cytoplasmic expression of -catenin whereas the rest of I-HCA does not. The final diagnosis was I-HCA associated with an activation of the Wnt/-catenin pathway restricted to an area of the tumor which exhibited some features of atypical HCA. Lesion 2 (Fig. 5) was biopsied. Major part of biopsy showed severe steatosis and the other part exhibited classical characteristics of I-HCA. Liver fatty acid binding protein (LFABP) staining was conserved and the whole biopsy was positive for CRP. Therefore, this lesion corresponded to an I-HCA with important steatotic changes. Lesion 3 (Fig. 6) was resected and corresponded to a benign hepatocytic nodule with marked steatosis, no inflammatory infiltrates, with complete lack of LFABP, contrasting with normal LFABP expression in surrounding non-tumoral liver, leading to the diagnosis of H-HCA. In addition on the resected specimen of the left hepatic lobe, one micro H-HCA (few mm) and two micro I-HCAs (few mm) were found. The non-tumoral liver showed discrete lesions of steatohepatitis.
Figure 4 Lesion 1 (resected specimen): I-HCA with a -catenin activated focus: a: gross section (fixed specimen): nodule with congestive alterations and a central green area; b—d: microscopy (square of a): area with cholestatic changes limited by arrows in the center of I-HCA: b: HES (inset: canalicular cholestasis, arrow head); c: positive CRP expression in the whole tumor; d: strong and diffuse GS expression only in the cholestatic area with aberrant nuclear staining of beta-catenin (inset). CRP: C-reactive protein; GS: glutamine synthetase.
xxx.e20
N. Alberti et al.
Figure 5 Lesion 2 (biopsy): steatotic I-HCA: a: HES: major steatosis (*) on half of the fragments; b: expression of LFABP is conserved; c: CRP is diffusely expressed; CRP: C-reactive protein; LFABP: liver fatty acid binding protein.
The postoperative course of the patient was uneventful. The patient was quickly asymptomatic. One month later, liver function tests were within normal ranges. Because of its location (deep central part of the liver) and the vascular proximity (inferior vena cava and sus-hepatic veins), the lesion 2 could not be resected, and we decided to follow-up our patient by MRI. At initial follow-up (6 months), lesion 2
Figure 6 Lesion 3 (resected specimen): H-HCA: a: diffuse steatosis (HES); b: lack of LFABP immunostaining in the tumor, constrasting with normal expression in the surrounding parenchyma; LFABP: liver fatty acid binding protein.
was stable and no other lesion appeared in the remaining liver.
Discussion In HCA, multiple nodules are frequent, and are generally of the same subtype. Here, we report one patient with the coexistence of three different subtypes of fat-containing HCAs, namely H-HCA, I-HCA and -catenin activated I-HCA. Specific MRI features are closely associated with H-HCA or I-HCA, the two most frequently identified subtypes of HCA. Thus, H-HCA can be recognized with confidence if a homogeneous fat distribution is observed, particularly on chemical shift sequences, whereas the characteristic MRI pattern of I-HCA included a markedly hyperintense signal on T2 W images, strong arterial enhancement and persistent enhancement in the delayed-phase, related to sinusoidal dilatation on pathology. Steatosis may also be present in I-HCA, but is usually not as extensive as in the H-HCA. When present, steatosis is usually distributed irregularly within the I-HCA, and when there are multiple tumours, the amount of steatosis varies among the tumours in an individual patient [2,3]. -catenin activated HCA appears as homogeneous or heterogeneous hypervascular mass with persistent or non-persistent enhancement during the delayed-phase images. Signal intensity on T2 and on T1 pre-contrast sequences is variable but mainly heterogeneously hyper-and hypointense, respectively. Malignant transformation simulates hepatocarcinoma on imaging and does not show peculiar findings [4—7].
Steatotic hepatocellular adenomas with different phenotypic subtypes: A case report MRI made the diagnosis of HCA for the 3 nodules. Lesion 3 showed a typical pattern of H-HCA (homogeneous and diffuse signal dropout in opposed-phase), related to diffuse fat repartition on pathology. The diagnosis of IHCA was likely for nodule 1 due to the presence of the « atoll sign » characterized by a hyperintensy band on T2weighted images in the periphery of the lesion [8]. Yet, the center of the lesion showed a nodular aspect (marked hyperintensity on T2-weighted images associated with a hypervascular appearance on T1 post-contrast sequences). This aspect was not quite typical and we thought that it was an indication for a possible malignant transformation or borderline lesion (‘‘nodule-in-nodule’’ appearance). On pathology, this part of the tumor corresponded to a cholestatic area with aberrant nuclear staining of -catenin. Diagnosis of I-HCA for nodule 2 was both a diagnosis by default after elimination of H-HCA (heterogeneous distribution of steatosis) and a reasonable one (hyperarterialisation on T1 post-contrast sequences and association with another I-HCA). The ‘‘atoll sign’’ was not observed in lesion 2, but this sign is very inconstant in I-HCA (only seen in 43% of cases) [8]. The natural history of HCA is not completely understood and treatment strategies are not completely defined. In patients with adenomas measuring < 5.0 cm, most groups recommend the suspension of contraceptives and observation based on reports showing tumour regression or even disappearance [9,10]. However, in HCA sized > 5.0 cm in diameter, surgical treatment is recommended because of the risk for malignant transformation and the risk for rupture, observed in 21—29% of cases [9,11,12]. In the case of our patient, the lesion 2 could not be resected because of its location in a difficult region to resect (central liver and due to vascular proximity). Follow-up by MRI was decided. Moreover, the patient was asymptomatic after the surgery, and we thought that the suspension of contraceptives could lead to a tumor regression (tubal ligation 4 months ago). Lesion 2 was stable at initial follow-up (6 months). Liver function tests are seldom abnormal in HCAs but mild elevations in gamma-glutamyl transferase and alkaline phosphatase have already been observed [13,14]. In our case, the non-tumoral liver showed discrete lesions of steatohepatitis, and we think that it could explain the increased rates of ALT and AST. This observation: • illustrates the utility of immunohistochemistry to confirm doubtful HCA subtypes; • confirms the very high specificity of MR for subtyping HCA (especially for the two major subtypes, H-HCA and I-HCA); • outlines that in the same patient with several nodules, HCA may be of different subtypes [15]. The predisposition to develop different HCA such as H-HCA and I-HCA is hypothetically caused by a ‘‘benign tumorigenic field effect’’. This predisposition could be the result of a conjunction between genetic and genotoxic factors leading to proliferation of benign hepatocytes [1]. According to some investigators, the association of I-HCA and -catenin activated I-HCA could indicate that -catenin mutation could occur as a second step [16], as in our case.
xxx.e21
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
Acknowledgements We thank H. Trillaud, P. Bioulac-Sage, C. Balabaud, J. Saric, F. Esperabe-Vigneau and A. Reffet. Funding information: no funding was received for this report. The work presented here was carried out in collaboration between all authors. NF and CC defined the research theme. CC, NA, AC and NF designed methods and analyzed the data, interpreted the results and wrote the paper. CC reviewed all histological data. All authors have contributed to, seen and approved the manuscript.
References [1] Nault J-C, Bioulac-Sage P, Zucman-Rossi J. Hepatocellular benign tumors-from molecular classification to personalized clinical care. Gastroenterology 2013;144:888—902. [2] Sempoux C, Chang C, Gouw A, Chiche L, Zucman-Rossi J, Balabaud C, et al. Benign hepatocellular nodules: what have we learned using the pathomolecular classification. Clin Res Hepatol Gastroenterol 2013;37(4):322—7, http://dx.doi.org/10.1016/j.clinre.2013.04.008 [Epub 2013 Jul 19]. [3] Bioulac-Sage P, Cubel G, Balabaud C, Zucman-Rossi J. Revisiting the pathology of resected benign hepatocellular nodules using new immunohistochemical markers. Semin Liver Dis 2011;31:91—103. [4] Ronot M, Paradis V, Duran R, Kerbaol A, Vullierme M-P, Belghiti J, et al. MR findings of steatotic focal nodular hyperplasia and comparison with other fatty tumours. Eur Radiol 2013;23:914—23. [5] Ronot M, Bahrami S, Calderaro J, Valla D-C, Bedossa P, Belghiti J, et al. Hepatocellular adenomas: accuracy of magnetic resonance imaging and liver biopsy in subtype classification. Hepatology 2011;53:1182—91. [6] Laumonier H, Bioulac-Sage P, Laurent C, Zucman-Rossi J, Balabaud C, Trillaud H. Hepatocellular adenomas: magnetic resonance imaging features as a function of molecular pathological classification. Hepatology 2008;48:808—18. [7] Grazioli L, Olivetti L, Mazza G, Bondioni MP. MR imaging of hepatocellular adenomas and differential diagnosis dilemma. Int J Hepatol 2013:374170. [8] Van Aalten SM, Thomeer MGJ, Terkivatan T, Dwarkasing RS, Verheij J, de Man RA, et al. Hepatocellular adenomas: correlation of MR imaging findings with pathologic subtype classification. Radiology 2011;261:172—81. [9] Dokmak S, Paradis V, Vilgrain V, Sauvanet A, Farges O, Valla D, et al. A single-center surgical experience of 122 patients with single and multiple hepatocellular adenomas. Gastroenterology 2009;137:1698—705. [10] Aseni P, Sansalone CV, Sammartino C, Benedetto FD, Carrafiello G, Giacomoni A, et al. Rapid disappearance of hepatic adenoma after contraceptive withdrawal. J Clin Gastroenterol 2001;33:234—6. [11] Cho SW, Marsh JW, Steel J, Holloway SE, Heckman JT, Ochoa ER, et al. Surgical management of hepatocellular adenoma: take it or leave it? Ann Surg Oncol 2008;15:2795—803.
xxx.e22 [12] Deneve JL, Pawlik TM, Cunningham S, Clary B, Reddy S, Scoggins CR, et al. Liver cell adenoma: a multicenter analysis of risk factors for rupture and malignancy. Ann Surg Oncol 2009;16:640—8. [13] Weimann A, Ringe B, Klempnauer J, Lamesch P, Gratz KF, Prokop M, et al. Benign liver tumors: differential diagnosis and indications for surgery. World J Surg 1997;21:983—90 [Discussion 990—991]. [14] Flejou JF, Barge J, Menu Y, Degott C, Bismuth H, Potet F, et al. Liver adenomatosis. An entity distinct from liver adenoma? Gastroenterology 1985;89:1132—8.
N. Alberti et al. [15] Sakamoto A, Hayashi H, Sakamoto I, Isomoto I, Eguchi S, Takatsuki M, et al. Multiple hepatocellular adenomas in a patient with glycogen storage disease type I: various enhancement patterns in MRI with Gd-EOB-DTPA. Abdom Imaging 2012;37:239—43. [16] Castain C, Sempoux C, Brunt EM, Causse O, Heitzmann A, Hernandez-Prera JC, et al. Coexistence of inflammatory hepatocellular adenomas with HNF1␣-inactivated adenomas: is there an association? Histopathology 2014;64: 890—5.
Available online at
ScienceDirect www.sciencedirect.com
CASE REPORT
Steatotic hepatocellular adenomas with different phenotypic subtypes: A case report N. Alberti a,∗, C. Castain b, A. Crombe a, N. Frulio a a
Department of Radiology, hôpital St-André, CHU de Bordeaux, 1, rue Jean-Burguet, 33000 Bordeaux, France b Department of Pathology, hôpital Pellegrin, CHU de Bordeaux, place Amélie-Raba-Léon, 33000 Bordeaux, France Available online 26 September 2014
Summary Hepatocellular adenomas (HCAs) are heterogeneous group of benign tumors; three pathomolecular subtypes have been identified so far: hepatocyte nuclear factor 1 ␣—inactivated HCA (H-HCA) (35—40%), inflammatory HCA (I-HCA) (> 50%), -catenin activated HCA (10%). Ten percent of I-HCA are also -catenin activated. We report a rare case of three histologically confirmed steatotic HCAs of three different phenotypes: I-HCA, -catenin activated I-HCA, and H-HCA in a 36-year-old woman. This observation outlines that in the same patient, HCA may be of different subtypes. The predisposition to develop different HCA hypothetically caused by a ‘‘benign tumorigenic field effect’’ may result, even rarely, in different genotypes/phenotypes such as H-HCA and I-HCA. Moreover, we illustrate the very high specificity of MR for subtyping HCA. © 2014 Elsevier Masson SAS. All rights reserved.
Introduction Hepatocellular adenoma (HCA) is a rare type of benign monoclonal liver neoplasm, occurring mostly (85%) in young women using oral contraceptives. The estimated prevalence is three to four per 100,000 in Europe and North America
∗ Corresponding author. Tel.: +33 5 56 79 58 00; fax: +33 5 56 79 58 99. E-mail address: [email protected] (N. Alberti).
http://dx.doi.org/10.1016/j.clinre.2014.08.007 2210-7401/© 2014 Elsevier Masson SAS. All rights reserved.
[1—3]. Three pathomolecular subtypes have been identified so far [1]: hepatocyte nuclear factor 1 ␣—inactivated (HHCA) (35—40%), inflammatory HCA (I-HCA) (>50%), -catenin activated HCA (10%). In addition, 10% of I-HCAs are -catenin activated. HCAs of any subtype can be solitary or multiple. Multiple HCAs are usually of the same subtype. Steatosis may be observed, to a varying degree, in all the subtypes of HCAs. In this paper, we report an unusual association of three ‘‘steatotic HCAs’’ that belonged to three different phenotypes (-catenin activated I-HCA, I-HCA and H-HCA) in a 36-year-old woman.
xxx.e18
N. Alberti et al.
Case history A 36-year-old woman was admitted to undergo evaluation of 2 hyper echogenic liver nodules detected during abdominal ultrasonography realised for abdominal pain in mid-2013. She was taking oral contraception for 18 years until the last 4 months (tubal ligation). Clinical examination was normal, excepted BMI value of 29. Laboratory analysis revealed elevated C-reactive protein (CRP) (50 UI/L; normal, 0—5), and abnormal liver function test with elevated alanine aminotransferase (ALT) and aspartate aminotransferase (AST) (120 and 320 U/L respectively). Serum alpha-fetoprotein was normal and there was no serological evidence of active hepatitis B or C. MRI showed the two liver nodules in segment II/III (76 mm; lesion 1) and in segments I, IV and VIII (53 mm; lesion 2) and also a third one in segment VI (8 mm; lesion 3). On T2-weighted images (Fig. 1), lesion 1 was heterogeneous, showing a marked hyperintensity (defined as at least as intense as the spleen signal) both in the periphery (‘‘atoll-like’’) and in the center (nodular aspect, asterisk). Lesion 2 was homogeneous and markedly hyperintense. Lesion 3 was also homogenous and slightly hyperintense. On chemical shift sequences (Fig. 2) signal dropout was found in opposed-phase in the 3 lesions, assessing the presence of fatty deposition: focally for lesions 1 (percentage of estimated volume: less than 5%) and 2 (percentage of estimated volume: more than 90%), diffuse and homogeneous for lesion 3. After injection, in the arterial phase (Fig. 3), lesion 1 showed a diffuse enhancement, more marked in the
Figure 1 Axial T2-weighted image SSFSE (Single Shot Fat Spin Echo). Two nodules localized in segment II/III (76 mm; lesion 1) and in segments I, IV and VIII (53 mm; lesion 2). Lesion 1 showed an ‘‘atoll sign’’, characterized by a hyperintensy band on T2weighted images in the periphery of the lesion, and also an hyperintense center (‘‘nodule-in-nodule’’ appearance). Lesion 2 shows marked hyper intensity of the entire nodule.
central nodular area (asterisk); lesion 2 presented peripheral intense enhancement, whereas no enhancement was observed for lesion 3. On delayed-phase (5mn), no wash out was found in any lesion; lesion 1 showed a diffuse persistent
Figure 2 Chemical shift sequences. Signal dropout in opposed-phase was found in the 3 lesions, assessing the presence of fatty deposition: focally for lesions 1 (less than 5% of the nodule) and 2 (more than 90%) (white arrows and asterisk respectively); diffuse and homogeneous for lesion 3, localized in segment VI (8 mm) (black arrows).
Steatotic hepatocellular adenomas with different phenotypic subtypes: A case report
Figure 3 Axial T1-weighted image after chelates of gadolinium contrast injection, with fat suppression (arterial phase). Lesion 1 showed a diffuse enhancement, more marked in the central nodular area (asterisk); lesion 2 presented peripheral intense enhancement.
enhancement, and the peripheral rim of lesion 2 was also hyperintense. The patient underwent a left hepatectomy and a segment VI tumorectomy two months later. Lesion 1 (Fig. 4) was resected and corresponded to a well-differentiated
xxx.e19
hepatocellular proliferation, intermingled with inflammatory infiltrates, and marked sinusoidal dilatation with congestion; the whole was characteristic of I-HCA, confirmed by positive expression of CRP in tumoral hepatocytes. In addition, in the center of the tumor, there were some cholestatic changes, hepatocyte plates were slightly thicker, the reticulin network was focally irregular and decreased but there was no obvious cytological atypia. Moreover, hepatocytes exhibited strong glutamine synthetase (GS) staining, associated with aberrant nuclear and cytoplasmic expression of -catenin whereas the rest of I-HCA does not. The final diagnosis was I-HCA associated with an activation of the Wnt/-catenin pathway restricted to an area of the tumor which exhibited some features of atypical HCA. Lesion 2 (Fig. 5) was biopsied. Major part of biopsy showed severe steatosis and the other part exhibited classical characteristics of I-HCA. Liver fatty acid binding protein (LFABP) staining was conserved and the whole biopsy was positive for CRP. Therefore, this lesion corresponded to an I-HCA with important steatotic changes. Lesion 3 (Fig. 6) was resected and corresponded to a benign hepatocytic nodule with marked steatosis, no inflammatory infiltrates, with complete lack of LFABP, contrasting with normal LFABP expression in surrounding non-tumoral liver, leading to the diagnosis of H-HCA. In addition on the resected specimen of the left hepatic lobe, one micro H-HCA (few mm) and two micro I-HCAs (few mm) were found. The non-tumoral liver showed discrete lesions of steatohepatitis.
Figure 4 Lesion 1 (resected specimen): I-HCA with a -catenin activated focus: a: gross section (fixed specimen): nodule with congestive alterations and a central green area; b—d: microscopy (square of a): area with cholestatic changes limited by arrows in the center of I-HCA: b: HES (inset: canalicular cholestasis, arrow head); c: positive CRP expression in the whole tumor; d: strong and diffuse GS expression only in the cholestatic area with aberrant nuclear staining of beta-catenin (inset). CRP: C-reactive protein; GS: glutamine synthetase.
xxx.e20
N. Alberti et al.
Figure 5 Lesion 2 (biopsy): steatotic I-HCA: a: HES: major steatosis (*) on half of the fragments; b: expression of LFABP is conserved; c: CRP is diffusely expressed; CRP: C-reactive protein; LFABP: liver fatty acid binding protein.
The postoperative course of the patient was uneventful. The patient was quickly asymptomatic. One month later, liver function tests were within normal ranges. Because of its location (deep central part of the liver) and the vascular proximity (inferior vena cava and sus-hepatic veins), the lesion 2 could not be resected, and we decided to follow-up our patient by MRI. At initial follow-up (6 months), lesion 2
Figure 6 Lesion 3 (resected specimen): H-HCA: a: diffuse steatosis (HES); b: lack of LFABP immunostaining in the tumor, constrasting with normal expression in the surrounding parenchyma; LFABP: liver fatty acid binding protein.
was stable and no other lesion appeared in the remaining liver.
Discussion In HCA, multiple nodules are frequent, and are generally of the same subtype. Here, we report one patient with the coexistence of three different subtypes of fat-containing HCAs, namely H-HCA, I-HCA and -catenin activated I-HCA. Specific MRI features are closely associated with H-HCA or I-HCA, the two most frequently identified subtypes of HCA. Thus, H-HCA can be recognized with confidence if a homogeneous fat distribution is observed, particularly on chemical shift sequences, whereas the characteristic MRI pattern of I-HCA included a markedly hyperintense signal on T2 W images, strong arterial enhancement and persistent enhancement in the delayed-phase, related to sinusoidal dilatation on pathology. Steatosis may also be present in I-HCA, but is usually not as extensive as in the H-HCA. When present, steatosis is usually distributed irregularly within the I-HCA, and when there are multiple tumours, the amount of steatosis varies among the tumours in an individual patient [2,3]. -catenin activated HCA appears as homogeneous or heterogeneous hypervascular mass with persistent or non-persistent enhancement during the delayed-phase images. Signal intensity on T2 and on T1 pre-contrast sequences is variable but mainly heterogeneously hyper-and hypointense, respectively. Malignant transformation simulates hepatocarcinoma on imaging and does not show peculiar findings [4—7].
Steatotic hepatocellular adenomas with different phenotypic subtypes: A case report MRI made the diagnosis of HCA for the 3 nodules. Lesion 3 showed a typical pattern of H-HCA (homogeneous and diffuse signal dropout in opposed-phase), related to diffuse fat repartition on pathology. The diagnosis of IHCA was likely for nodule 1 due to the presence of the « atoll sign » characterized by a hyperintensy band on T2weighted images in the periphery of the lesion [8]. Yet, the center of the lesion showed a nodular aspect (marked hyperintensity on T2-weighted images associated with a hypervascular appearance on T1 post-contrast sequences). This aspect was not quite typical and we thought that it was an indication for a possible malignant transformation or borderline lesion (‘‘nodule-in-nodule’’ appearance). On pathology, this part of the tumor corresponded to a cholestatic area with aberrant nuclear staining of -catenin. Diagnosis of I-HCA for nodule 2 was both a diagnosis by default after elimination of H-HCA (heterogeneous distribution of steatosis) and a reasonable one (hyperarterialisation on T1 post-contrast sequences and association with another I-HCA). The ‘‘atoll sign’’ was not observed in lesion 2, but this sign is very inconstant in I-HCA (only seen in 43% of cases) [8]. The natural history of HCA is not completely understood and treatment strategies are not completely defined. In patients with adenomas measuring < 5.0 cm, most groups recommend the suspension of contraceptives and observation based on reports showing tumour regression or even disappearance [9,10]. However, in HCA sized > 5.0 cm in diameter, surgical treatment is recommended because of the risk for malignant transformation and the risk for rupture, observed in 21—29% of cases [9,11,12]. In the case of our patient, the lesion 2 could not be resected because of its location in a difficult region to resect (central liver and due to vascular proximity). Follow-up by MRI was decided. Moreover, the patient was asymptomatic after the surgery, and we thought that the suspension of contraceptives could lead to a tumor regression (tubal ligation 4 months ago). Lesion 2 was stable at initial follow-up (6 months). Liver function tests are seldom abnormal in HCAs but mild elevations in gamma-glutamyl transferase and alkaline phosphatase have already been observed [13,14]. In our case, the non-tumoral liver showed discrete lesions of steatohepatitis, and we think that it could explain the increased rates of ALT and AST. This observation: • illustrates the utility of immunohistochemistry to confirm doubtful HCA subtypes; • confirms the very high specificity of MR for subtyping HCA (especially for the two major subtypes, H-HCA and I-HCA); • outlines that in the same patient with several nodules, HCA may be of different subtypes [15]. The predisposition to develop different HCA such as H-HCA and I-HCA is hypothetically caused by a ‘‘benign tumorigenic field effect’’. This predisposition could be the result of a conjunction between genetic and genotoxic factors leading to proliferation of benign hepatocytes [1]. According to some investigators, the association of I-HCA and -catenin activated I-HCA could indicate that -catenin mutation could occur as a second step [16], as in our case.
xxx.e21
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
Acknowledgements We thank H. Trillaud, P. Bioulac-Sage, C. Balabaud, J. Saric, F. Esperabe-Vigneau and A. Reffet. Funding information: no funding was received for this report. The work presented here was carried out in collaboration between all authors. NF and CC defined the research theme. CC, NA, AC and NF designed methods and analyzed the data, interpreted the results and wrote the paper. CC reviewed all histological data. All authors have contributed to, seen and approved the manuscript.
References [1] Nault J-C, Bioulac-Sage P, Zucman-Rossi J. Hepatocellular benign tumors-from molecular classification to personalized clinical care. Gastroenterology 2013;144:888—902. [2] Sempoux C, Chang C, Gouw A, Chiche L, Zucman-Rossi J, Balabaud C, et al. Benign hepatocellular nodules: what have we learned using the pathomolecular classification. Clin Res Hepatol Gastroenterol 2013;37(4):322—7, http://dx.doi.org/10.1016/j.clinre.2013.04.008 [Epub 2013 Jul 19]. [3] Bioulac-Sage P, Cubel G, Balabaud C, Zucman-Rossi J. Revisiting the pathology of resected benign hepatocellular nodules using new immunohistochemical markers. Semin Liver Dis 2011;31:91—103. [4] Ronot M, Paradis V, Duran R, Kerbaol A, Vullierme M-P, Belghiti J, et al. MR findings of steatotic focal nodular hyperplasia and comparison with other fatty tumours. Eur Radiol 2013;23:914—23. [5] Ronot M, Bahrami S, Calderaro J, Valla D-C, Bedossa P, Belghiti J, et al. Hepatocellular adenomas: accuracy of magnetic resonance imaging and liver biopsy in subtype classification. Hepatology 2011;53:1182—91. [6] Laumonier H, Bioulac-Sage P, Laurent C, Zucman-Rossi J, Balabaud C, Trillaud H. Hepatocellular adenomas: magnetic resonance imaging features as a function of molecular pathological classification. Hepatology 2008;48:808—18. [7] Grazioli L, Olivetti L, Mazza G, Bondioni MP. MR imaging of hepatocellular adenomas and differential diagnosis dilemma. Int J Hepatol 2013:374170. [8] Van Aalten SM, Thomeer MGJ, Terkivatan T, Dwarkasing RS, Verheij J, de Man RA, et al. Hepatocellular adenomas: correlation of MR imaging findings with pathologic subtype classification. Radiology 2011;261:172—81. [9] Dokmak S, Paradis V, Vilgrain V, Sauvanet A, Farges O, Valla D, et al. A single-center surgical experience of 122 patients with single and multiple hepatocellular adenomas. Gastroenterology 2009;137:1698—705. [10] Aseni P, Sansalone CV, Sammartino C, Benedetto FD, Carrafiello G, Giacomoni A, et al. Rapid disappearance of hepatic adenoma after contraceptive withdrawal. J Clin Gastroenterol 2001;33:234—6. [11] Cho SW, Marsh JW, Steel J, Holloway SE, Heckman JT, Ochoa ER, et al. Surgical management of hepatocellular adenoma: take it or leave it? Ann Surg Oncol 2008;15:2795—803.
xxx.e22 [12] Deneve JL, Pawlik TM, Cunningham S, Clary B, Reddy S, Scoggins CR, et al. Liver cell adenoma: a multicenter analysis of risk factors for rupture and malignancy. Ann Surg Oncol 2009;16:640—8. [13] Weimann A, Ringe B, Klempnauer J, Lamesch P, Gratz KF, Prokop M, et al. Benign liver tumors: differential diagnosis and indications for surgery. World J Surg 1997;21:983—90 [Discussion 990—991]. [14] Flejou JF, Barge J, Menu Y, Degott C, Bismuth H, Potet F, et al. Liver adenomatosis. An entity distinct from liver adenoma? Gastroenterology 1985;89:1132—8.
N. Alberti et al. [15] Sakamoto A, Hayashi H, Sakamoto I, Isomoto I, Eguchi S, Takatsuki M, et al. Multiple hepatocellular adenomas in a patient with glycogen storage disease type I: various enhancement patterns in MRI with Gd-EOB-DTPA. Abdom Imaging 2012;37:239—43. [16] Castain C, Sempoux C, Brunt EM, Causse O, Heitzmann A, Hernandez-Prera JC, et al. Coexistence of inflammatory hepatocellular adenomas with HNF1␣-inactivated adenomas: is there an association? Histopathology 2014;64: 890—5.