Diagnosis: Multiple hepatocellular adenomas (HCAs), one of the telangectatic focal nodular hyperplasia (FNH)-type and the others of steatotic type.
Referral Sources: AMR seminar n. 52-case 5 (slides labelled 88903-99, and MB-316), contributed by Michele Bisceglia, MD, San Giovanni Rotondo, Italy.
In March 1999 a 17-year–old Italian girl presented at Casa Sollievo della Sofferenza Hospital, San Giovanni Rotondo, Italy, with an 11 cm twisted pedunculated liver mass, which was surgically resected as an emergency procedure (the first lesion) (Figs. 1A, C). There was no history of oral contraception, Fanconi anemia, glycogen storage disease, familial adenomatous polyposis, or diabetes mellitus. At surgery, a second 3 cm mass was noted on the dome of the liver at the junction of the right and left lobes, between segments VII and VIII (the second lesion). This lesion was not resected.
PATHOLOGIC FINDINGS OF THE FIRST LESION
A central scar was grossly apparent on the cut surface along with some hemorrhagic foci and a small blood filled cavity (Figs. 1B, D). Histologically, the mass was diagnosed as FNH based on the central scar (Fig. 2), abnormal vessels, and the patchy ductular proliferation at the interface between the fibrous septa of the central scar and the tumor cells. Peliotic or peliosis-like areas, focal hemorrhages, and infarction were considered to be secondary to the torsion (Figs. 3A–D). Both HCA and well differentiated hepatocellular carcinoma were also considered in the differential diagnosis. Reticulin stain showed regularly organized trabeculae 1 to 2-cells thick (Figs. 4A, B). The presence of bile ductules was subsequently confirmed by positive immunostaining for CK7 (Figs. 5A–D), which was performed years later (in 2006 while planning the second surgery—see below).
Three months postoperatively, a magnetic resonance imaging (MRI) with intravenous contrast showed the second lesion to be nodular with lobulated margins and dishomogeneous—contrast enhancement, which was hypointense in the late venous phase. Follow-up ultrasound revealed that the tumor was growing slowly. The patient was asymptomatic, but the levels of serum alkaline phosphatase and γ-glutamyl–transferase were slightly and consistently elevated.
In September 2006, 7 years after the first surgery, MRI with intravenous contrast showed that the second lesion had enlarged to 7 cm in the T1-weighted images and was persistently hypointense in gradient echo images in the late phase, suggestive of a diffuse lipid component (Figs. 6A, B). Surgical excision was advised because of the progressive enlargement and the risk of bleeding.
While planning the second surgical intervention, slides of the first lesion were sent in consultation to 7 specialized International Liver Centers. The diverse opinions rendered included classic FNH, HCA of telangiectatic type (previously termed telangiectatic-FNH), and classic HCA. Two of the specialists who diagnosed HCA of classic type also suspected a well-differentiated hepatocellular carcinoma. These latter 2 consultants also performed immunostaining for CD34 (both), which showed no sinusoidal staining and Glypican-3 (1 of the 2), which was also negative. All wondered what the second lesion would turn out to be.
The second lesion was resected at the Department of Surgery, Ospedale Civile Maggiore, Verona, Italy, in December 2006. The excised mass was then 11 cm in size, with no visible central scar and, on sectioning was poorly delimited and showed foci of clear, ballooned hepatocytes, steatosis around thin-walled venules, and smaller eosinophilic cells disposed along arterial branches. Biliary rosettes, abortive bile ducts, and ductules were seen (Figs. 7A–D), well highlighted by immunostaining with antibody to CK7 (Figs. 8A–D). Multiple, minute, hyperplastic-like nodular foci of clear/steatotic hepatocytes were also seen in the adjacent host liver (Fig. 9).
HCA-not otherwise specified was the diagnosis favoured by the Verona pathologist (A.T.) who provided several representative blocks for the referral centers.
Slides from this second lesion were sent to 5 of the previous centers. The 3 diagnoses returned were: classic HCA; HCA with CK7+ biliary ductules; and liver cell adenomatous hyperplasia. The 3 consultants who answered all had regarded the first lesion as HCA. No answer was received from 2 of the centers. On the basis of the above mentioned several smaller nodules present at the periphery of the current main mass and given the first large mass previously seen in 1999, the 2 consultants who diagnosed this second lesion as a HCA favoured the whole process to be liver adenomatosis. One of these latter consultants stressed that the satellite peripheral nodules in this second liver lesion resembled nodular regenerative hyperplasia, and also suggested that some published cases of liver cell adenomatosis actually may have been examples of nodular regenerative hyperplasia (according to some synonymous with adenomatous hyperplasia). Subsequently 2 other liver pathologists also reviewed sections of both lesions. One diagnosed the first mass as HCA of telangiectatic-FNH type and the second as HCA(s) steatotic-type (variant-3 and variant-1 of the new classification of HCA, respectively: see below). The other diagnosed HCA with ductal differentiation on both specimens (“synonymous with HCA of telangiectatic-FNH type”—see below) and interpreted the “central scar” of the first tumor as the result of so-called (local) congestive hepatopathy.
In June 2007, 6 months after the second surgical intervention, the patient underwent MRI with contrast medium. No tumors or other lesions were seen, apart from changes owing to the previous surgery. As then she has been followed up for every 6 months by ultrasonography and no more lesions have been seen so far. She is active and well as of February 2009.
FNH and HCA are benign hepatocellular proliferations that may be multiple and occasionally pedunculated. Classic FNH is a firm, well-delimited but not encapsulated nodular hepatocellular lesion composed of 2-cell thick plates, characterized by a central scar containing abnormal arteries with medial hypertrophy and intimal hyperplasia, associated with ductular proliferation and often lymphoid infiltration.1 However, it is now widely acknowledged that key diagnostic features may be either “totally lacking, partly present, or inconspicuous” in some cases or variants.2–4 Imaging studies frequently distinguish FNH from other hepatocellular nodules,5,6 but atypical morphologic features often occur.7–9
HCA is composed of benign hepatocytes arranged in plates up to 3-cell thick with thin-walled supplying arteries scattered throughout the tumor and no acinar architecture. There are no portal tracts or central veins, there is no connection with the biliary system, whereas bile ducts and ductules are typically absent.1,10 Most HCAs are solitary, but multiple lesions are present in 10% to 20% of patients. HCA mainly affects women, and almost all examples occur in a setting of hepatocellular stimulation by medications (oral contraceptives, anabolic steroids), or abnormal carbohydrate metabolism (familial diabetes mellitus, glycogen storage disease type 1a, and galactosemia).10 so that “if none of these conditions is present, the diagnosis is doubtful.”10 Occasional HCAs arising outside the hepatocellular stimulation context are termed “spontaneous.” Rare cases of malignant transformation in solitary adenomas have also been observed,11,12 and it is recommended to record the size as the risk of the development of malignancy increases with lesional size.1
Liver adenomatosis was described as a separate clinical entity around 2 decades ago, and was arbitrarily defined as the presence of more than 10 HCAs in an otherwise normal liver in patients with no history of steroid intake or metabolic disease.13,14 Liver adenomatosis is rare with a total of only about 70 published cases in all.15 There is an almost equal male to female distribution, and while some lesions may undergo hemorrhage or necrosis, and the serum alkaline phosphatase and γ-glutamyl–transferase are frequently raised,13–15 many patients are asymptomatic. There are 3 histologic types; steatotic, peliotic, and mixed. All 3 can also be distinguished by organ imaging.15 Although liver adenomatosis is not considered to be precancerous justifying preventive liver transplantation,14 one case of malignant degeneration in an adenoma is on record,16 and the risk of malignancy—as above stated for solitary HCAs—is size-related, regardless of the number of lesions.1 Most cases of liver adenomatosis can be managed and followed up conservatively, but large tumors should be removed because of the risk of hemorrhage.14 “Liver transplantation is indicated in highly symptomatic massive forms with serious complications or in liver disease impairing socio-professional life in young patients.”14
Indeed a morphologic spectrum with different histologic forms have been described for FNH and HCA. Occasionally even the association of FNH and HCA has been reported.2
In 1999, on the basis of the standard definition of FNH as a lesion composed of hyperplastic hepatic parenchyma subdivided into nodules by fibrous septa, often forming stellate scars and always containing small bile ducts and abnormally thick-walled vessels,2,17 Nguyen et al,2 retrospectively studied 305 surgically resected specimens of FNH and divided them into 2 groups: the classical type (∼80% or 245 cases) and the non-classical type (∼20% or 60 cases). In the category of FNH-classical type of this study only a proportion (153 lesions) showed 1 to 3 gross central scars (∼63%), whereas others (ie, 92 lesions) did not (∼37%). All the cases with gross central scar histologically showed architectural nodular distortion, malformed arterial vessels, and bile ductular reaction; 43 of the 92 cases without a gross central scar also showed the same microscopic features, with evidence of fibrous septa, architectural nodularity, and ductular reaction. The remaining 49 cases without any gross central scar, usually small nodules, “also had the classical characteristics, but on a subtle scale,” with only a vague nodular aspect, arteries that were only moderately abnormal, and discrete bile ductular proliferation only visible at high magnification. The category of FNH-nonclassical type was defined as lacking nodular architecture or malformed vessels, but always presenting some degree of interlobular bile ductular proliferation, definitely considering this latter histologic feature as the hallmark of the lesion.2 FNH of the nonclassical type was further subdivided into telangiectatic FNH (T-FNH, ∼15%, ie, 47 cases of the total 305 from the study), mixed hyperplastic and adenomatous FNH (1% to 2%), and FNH with cytologic atypia (2% to 3%).2 This distinguished French group stressed the detailed description of variants, hoping the categorization might prove useful in difficult diagnostic cases.
However, molecular studies published in 2004 by Paradis et al18 showed that T-FNH is closer to HCA than to FNH, by demonstrating monoclonality, and suggested that the name be changed to telangiectatic HCA. Their findings were soon corroborated by others19,20 so the former T-FNH is now regarded as a form of HCA.
In 2007, after a meeting held in Bordeaux in 2005 (“Bordeaux” update), Bioulac-Sage et al1 delineated 4 variants of HCA. Given the classic (archetypal-monoclonal) type as architecturally composed of liver cell plates up to 3-cell thick, with little cytologic atypia allowed, and supplied by thin-walled arteries without any other portal elements or bile ductular reaction, the 4 variants were categorized both molecularly (on the basis of mutations of HNF1α and β-catenin genes) and histologically as the following:
(1) Variant-1 (HNF1α mutation)—histologic pattern the same as in archetypal type with marked steatosis/clear cells;
(2) Variant-2 (β-catenin mutation)—histologic pattern the same as in archetypal type with more pronounced cytologic atypia, and less steatosis than in 1 (also called HCA/hepatocellular carcinoma, “atypical” HCA, borderline lesion);
(3) Variant-3 (no known mutation)—histologically characterized by telangiectasia and a variable inflammatory infiltrate—may contain CK7+ ductules (previously “telangiectatic focal nodular hyperplasia”);
(4) Variant-4 (no known mutation)—with no specific morphologic trait (and without inflammatory infiltrate).1
HCA variant-3 corresponds to T-FNH, which some people also called “progressive focal nodular hyperplasia,”21 and may contain CK7+ bile ductules (“adenoma with ductal differentiation”),1 as well documented also in our case. The term T-FNH was introduced for the first time in 1989 in the publication of a series of 13 cases of syndromic FNH22 in contrast to the usual “solid” type of FNH (which notoriously is pale on cut surface, “paler” than the surrounding liver).
T-FNH is a monoclonal lesion, prone to bleeding, similar to adenomas not carrying the HNF1α mutation, which on the basis of morphologic and molecular data needs to be considered a separate entity.19 T-FNH also differs from typical FNH at imaging, as seen in a study comparing imaging features [ultrasonography, computed tomography, magnetic resonance (MR)] with histopathologic results from resected specimens.23 Differential imaging features in T-FNH are: lack of a central scar, lesion heterogeneity, hyperintensity on T1-weighted MR images, and persistent contrast enhancement on delayed contrast-enhanced computed tomography or T1-weighted MR images.23
The Bordeaux update also assessed the morphologic spectrum of FNH: the classic (“archetypal”) type, characterized by circumscription, central scar(s) with abnormal arteries, lymphocytic inflammation, and bile ductular reaction, in the absence of interlobular bile ducts; variants are recognized as forms with “lack of, inconspicuous or only some of the key diagnostic features listed in the archetypal type,” and forms with “unusual features (steatosis),” acknowledging the difficulty of the differential diagnosis of the FNH variants with HCA and sometimes even the impossibility (small lesions <1 cm). Conclusively, FNH is now represented only by the archetypal type (as above defined by Bordeaux update), corresponding to the classical type of others' classification,19 also termed as FNH “solid” type.9,10
FINAL DIAGNOSIS AND CONCLUSIONS
The final histologic diagnosis in this case was: multiple spontaneous HCA, with 2 major lesions (the first lesion was of the T-FNH type and the second lesion was the HCA steatotic-variant) and several minor nodules seen at the periphery of the second major lesion, also of steatotic type. Although some authors require at least 10 lesions for a diagnosis of the liver adenomatosis, others consider the current definition as “difficult to apply in practice,”1 and Ishak et al24 say that more than 3 will suffice. The diagnosis of multiple adenomas (liver adenomatosis) in this case was clinically supported by the raised alkaline phosphatase and γ-glutamyl–transferase. Furthermore, our case was not related to hormone ingestion or metabolic diseases, as the current definition of liver adenomatosis requires. Malignancy was excluded because of the absence of atypia, the intact reticulin framework, the regular disposition of trabeculae 1 to 2-cells thick, the negative staining with Glypican-3,25–31 the minimal sinusoidal staining with CD34, and the MIB-1 labeling of very occasional nuclei.
CLUB'S OPINIONS AND COMMENTS
* Spontaneous adenomatosis of liver. Thanks for the extensive discussion. I suspect that the classification of these lesions will only get more complex with time.
* I am not so familiar with these lesions (nodular hyperplasia and adenomas of the liver). This case is very illustrative of how difficult the diagnosis of these entities can be.
* This exercise shows to what extent people can give differing interpretations to the same slides. I think that a lot of the differential diagnosis is semantic, conjectural, and/or hypothetical, and in a given circumstance no one suggested solution can be conclusively proven over any other one, and that many established entities are either biologically the same or closely similar. One can ask whether each and every case has one and only one correct resolution. But try explaining that to the patient if he gets wind of the fact that one consultant did indeed call it malignant!! Whatever this condition is, it apparently is capable of recurring but it doesn't look malignant. The main fear would be massive hemorrhage in the event of a third recurrence.
* Agree with the diagnosis. I have seen at least 2 other cases of multiple hepatic adenomas (hepatic adenomatosis) unrelated to steroid therapy.
* An exhaustive write-up. Much thanks. Glypican-3 is negative in adenomas and positive in carcinoma (Cancer Cytopath 2007;111:316).
* I would have no problems calling this adenoma. Excellent review of the problem.
* Thank you for this excellent presentation and review. I was not aware of the reference of the Bordeaux update and the molecular studies of telangiectatic focal nodular hyperplasia.
* Thanks for a great update on benign hepatocellular proliferations. Our hospital system is now stressing profiliation among its 4 nodes. Sadly, I don't get to see any liver pathology any more.
* Thanks for going over a difficult and controversial subject. I do not have much experience with this subject, but it seems that there are cases in which it may be difficult to draw a line not only between hepatocellular carcinoma and adenomas, but focal nodular hyperplasia and hepatocellular carcinoma/adenoma. This case reflects the issue.
* Hepatocellular adenomas. Lacks the proliferating bile ductules of true focal nodular hyperplasia.
* A very interesting case. Thank you very much for detailed description. My first impression of the first slide was fibrolamellar hepatocellular carcinoma.
* Interesting case. Results of IHC studies for CD34 and glypican might be informative in this case. It greatly facilitates the accuracy of distinguishing between malignant hepatic lesions and benign mimicker.
* I agree that there is no carcinoma. Using good old-fashioned criteria, I would have called the 1999 lesion as you did, focal nodular hyperplasia and the 2006 nodules as “hyperplastic nodules” perhaps regenerative or some other unknown etiology. The new classification seems a bit complex, but I don't regard myself as an expert on this field. Perhaps I want too much, like clonality studies, to be convinced about multiple adenomas/adenomatosis.
* I was thinking well-differentiated hepatocellular carcinoma but it is not in my field of expertise. However, after reading your text, I believe adenoma is probably a better diagnosis.
* The first biopsy looks like focal nodular hyperplasia but the other two are impossible!
* I have very little to add to your scholarly discussion. I think that the nodules are definitely not hepatocellular carcinomas, and I eventually convinced myself that they look better for hepatocellular adenomas than for focal nodular hyperplasia.
* Thanks for this typically in-depth review of the issues relating to hepatic adenomas and focal nodular hyperplasia. I can't argue with your final interpretation.
* Multiple hepatic adenomas versus focal nodular hyperplasia.
* Very difficult case! Thank you for the learned discussion. Don't know that any of our current understanding of these lesions makes any sense to me and I get the feeling that much of what we think we know about these lesions is arbitrary and that the arguments could go both ways.
* I was completely unaware of the molecular studies in adenoma. Thank you for the wonderful discussion.
The authors are indebted to all the consultants who reviewed the material and shared their opinions and insight with them.
1. Bioulac-Sage P, Balabaud C, Bedossa P, et al. Pathological diagnosis of liver cell adenoma and focal nodular hyperplasia: Bordeaux update. J Hepatol. 2007;46:521–527.
2. Nguyen BN, Fléjou JF, Terris B, et al. Focal nodular hyperplasia of the liver: a comprehensive pathologic study of 305 lesions and recognition of new histologic forms. Am J Surg Pathol. 1999;23:1441–1454.
3. Bioulac-Sage P, Balabaud C, Wanless IR. Diagnosis of focal nodular hyperplasia: not so easy. Am J Surg Pathol. 2001;25:1322–1325.
4. Lepreux S, Laurent C, Balabaud C, et al. Focal nodular hyperplasia lacking some key histopathological features making the diagnosis difficult. Virchows Arch. 2002;440:445–446.
5. Mortelé KJ, Praet M, Van Vlierberghe H, et al. Focal nodular hyperplasia of the liver: detection and characterization with plain and dynamic-enhanced MRI. Abdom Imaging. 2002;27:700–707.
6. Hussain SM, Terkivatan T, Zondervan PE, et al. Focal nodular hyperplasia: findings at state-of-the-art MR imaging, US, CT, and pathologic analysis. Radiographics. 2004;24:3–17; discussion 18–19.
7. Vilgrain V, Fléjou JF, Arrivé L, et al. Focal nodular hyperplasia of the liver: MR imaging and pathologic correlation in 37 patients. Radiology. 1992;184:699–703.
8. Grazioli L, Morana G, Federle MP, et al. Focal nodular hyperplasia: morphologic and functional information from MR imaging with gadobenate dimeglumine. Radiology. 2001;221:731–739.
9. Ba-Ssalamah A, Schima W, Schmook MT, et al. Atypical focal nodular hyperplasia of the liver: imaging features of nonspecific and liver-specific MR contrast agents. AJR Am J Roentgenol. 2002;179:1447–1456.
10. International Working Party. Terminology of nodular hepatocellular lesions. Hepatology. 1995;22:984–993.
11. Ferrell LD. Hepatocellular carcinoma arising in a focus of multilobular adenoma. A case report. Am J Surg Pathol. 1993;17:525–529.
12. Foster JH, Berman MM. The malignant transformation of liver cell adenomas. Arch Surg. 1994;129:712–727.
13. Flejou JF, Barge J, Menu Y, et al. Liver adenomatosis. An entity distinct from liver adenoma? Gastroenterology. 1985;89:1132–1138.
14. Chiche L, Dao T, Salamé E, et al. Liver adenomatosis: reappraisal, diagnosis, and surgical management: eight new cases and review of the literature. Ann Surg. 2000;231:74–81.
15. Lewin M, Handra-Luca A, Arrivé L, et al. Liver adenomatosis: classification of MR imaging features and comparison with pathologic findings. Radiology. 2006;241:433–440.
16. Leese T, Farges O, Bismuth H. Liver cell adenomas. A 12-year surgical experience from a specialist hepato-biliary unit. Ann Surg. 1988;208:558–564.
17. Gibson JB. Histological typing of tumours of the liver, biliary tract and pancreas. In: Sobin LH, ed. International Histological Classification of Tumors. Vol 20. Geneva: World Health Organization; 1978:29.
18. Paradis V, Benzekri A, Dargère D, et al. Telangiectatic focal nodular hyperplasia: a variant of hepatocellular adenoma. Gastroenterology. 2004;126:1323–1329.
19. Bioulac-Sage P, Rebouissou S, Sa Cunha A, et al. Clinical, morphologic, and molecular features defining so-called telangiectatic focal nodular hyperplasias of the liver. Gastroenterology. 2005;128:1211–1218.
20. Zucman-Rossi J, Jeannot E, Nhieu JT, et al. Genotype-phenotype correlation in hepatocellular adenoma: new classification and relationship with HCC. Hepatology. 2006;43:515–524.
21. Sadowski DC, Lee SS, Wanless IR, et al. Progressive type of focal nodular hyperplasia characterized by multiple tumors and recurrence. Hepatology. 1995;21:970–975.
22. Wanless IR, Albrecht S, Bilbao J, et al. Multiple focal nodular hyperplasia of the liver associated with vascular malformations of various organs and neoplasia of the brain: a new syndrome. Mod Pathol. 1989;2:456–462.
23. Attal P, Vilgrain V, Brancatelli G, et al. Telangiectatic focal nodular hyperplasia: US, CT, and MR imaging findings with histopathologic correlation in 13 cases. Radiology. 2003;228:465–472.
24. Ishak KG, Goodman ZD, Stocker JT. Tumors of the Liver and Intrahepatic Bile Ducts. Fascicle 31. Atlas of Tumor Pathology: 3rd Series. Washington DC: AFIP; 2001:21–22.
25. Yamauchi N, Watanabe A, Hishinuma M, et al. The glypican 3 oncofetal protein is a promising diagnostic marker for hepatocellular carcinoma. Mod Pathol. 2005;18:1591–1598.
26. Libbrecht L, Severi T, Cassiman D, et al. Glypican-3 expression distinguishes small hepatocellular carcinomas from cirrhosis, dysplastic nodules, and focal nodular hyperplasia-like nodules. Am J Surg Pathol. 2006;30:1405–1411.
27. Wang HL, Anatelli F, Zhai QJ, et al. Glypican-3 as a useful diagnostic marker that distinguishes hepatocellular carcinoma from benign hepatocellular mass lesions. Arch Pathol Lab Med. 2008;132:1723–1728.
28. Kandil D, Leiman G, Allegretta M, et al. Glypican-3 immunocytochemistry in liver fine-needle aspirates: a novel stain to assist in the differentiation of benign and malignant liver lesions. Cancer. 2007;111:316–322.
29. Coston WM, Loera S, Lau SK, et al. Distinction of hepatocellular carcinoma from benign hepatic mimickers using Glypican-3 and CD34 immunohistochemistry. Am J Surg Pathol. 2008;32:433–444.
30. Ligato S, Mandich D, Cartun RW. Utility of glypican-3 in differentiating hepatocellular carcinoma from other primary and metastatic lesions in FNA of the liver: an immunocytochemical study. Mod Pathol. 2008;21:626–631.
31. Shafizadeh N, Ferrell LD, Kakar S. Utility and limitations of glypican-3 expression for the diagnosis of hepatocellular carcinoma at both ends of the differentiation spectrum. Mod Pathol. 2008;21:1011–1018.
© 2009 Lippincott Williams & Wilkins, Inc.