Section 6 -

Small/Early Hepatocellular Carcinoma and Malignancy-Associated Hepatocellular Lesions Neil Theise, M.D. Romil Saxena, M.D.

Case 6A 54 year old man with hepatitis C cirrhosis. Ill defined, 1.0 cm, distinctive nodule identified in explanted liver. Diagnosis: High grade dysplastic nodule (n.b. foci of small cell change, pseudogland formation) containing early hepatocellular carcinoma (stromal invasion). Case 6B: 43 year old woman with hepatitis C cirrhosis. Well defined, distinctive nodule measuring 1.9 cm in explanted liver. Diagnosis: Low grade dysplastic nodule. Dysplastic Nodules And Hepatocarcinogenesis In North America, dysplastic nodules (DNs), premalignant hepatocellular lesions usually found in cirrhosis, have largely been a topic only for specialists in hepatopathology, having little relevance for even the hepatologist, let alone the general surgical pathologist's routine practice. Hepatitis C will change all that, however. In Japan, where hepatitis C and other causes of chronic liver disease keep hepatocellular carcinoma (HCC) the most common form of malignancy in adults, radiographic screening for early malignant or pre-malignant lesions has long been a high priority. As a result, it was in Japan that pathologists first drew attention to these lesions. Now, however, the hepatitis C epidemic is causing hepatocellular carcinoma to become increasingly common in North America. Early diagnosis thus becomes relevant and radiologists in liver referral centers are beginning to screen patients with advanced scarring and, as a result, small hepatocellular nodules are being biopsied. As this practice will certainly extend to the wider, general medical community, under the pressure of an epidemic, it is time for the general pathologist to become familiar with these curious, literally distinctive (see "Definitions" below) hepatocellular nodules. Definitions and Observations DNs are defined grossly as large hepatic nodules that are distinct from the surrounding liver parenchyma in terms of size, color, texture, or the degree to which they bulge from the cut surface of the liver. Confirmation that a nodule is in fact a DN comes with histologic examination and the identification of intact portal structures distributed through the lesion. The number of these portal structures may be mildly or greatly reduced compared to a similar area of non-diseased hepatic parenchyma. Since these lesions are most often found in the setting of cirrhosis they seem to correspond to what Edmondson called "adenomatous hyperplasia" (AH), though he viewed these lesions as having "limited growth potential." AH has been most widely used, though not exclusively, by researchers from Japan. A different term, "macroregeneraqtive nodule" (MRN), coined by Furuya et al for the first Japanese autopsy study of the nodules, became the one most widely accepted in the earliest publications from North America and Europe. In 1997 a consensus document co-authored by investigators from Japan, North America, and Europe rejected these terms as being imprecise if not actually misleading, suggesting the term "dysplastic nodule", with further classification as low or high grade, to replace them. Though researchers outside of Japan have generally accepted this new terminology, investigators in Japan have inconsistently adopted it, usually continuing to use AH as the nomenclature of choice. In his review, dysplastic nodule will be the preferred term, however, recognizing at the same time that it, too, is problematic as it misleadingly suggests that the lesions comprise nodules of dysplastic appearing hepatocytes. More on this issue, below. DNs have been found in a wide variety of chronic liver diseases, including processes which are hepatitic (hepatitis B, C, and autoimmune hepatitis), cholangitic (primary biliary cirrhosis, primary sclerosing cholangitis), metabolic (alpha-1-antitrypsin deficiency, primary hemochromatosis), and toxic (alcoholic liver injury). Typically, livers with DNs contain a small number of these nodules, rarely more than ten, although there are exceptions which will be dicussed further on. DNs may be sub-classified as low grade or high grade according to features listed in Table 1. Low grade lesions may be devoid of atypia or display features of large cell change. They may be iron or copper retentive or they may be diffusely steatotic. They should not contain nodule-in-nodule type lesions or small cell change, let alone features closely associated with HCC; these changes would be indicative of the lesion being high grade. It is usually not possible at this time to radiographically distinguish small HCCs from DNs or high grade DNs from low grade DNs with complete confidence, nor is it usually possible to reliably make such distinctions on the basis of gross morphology. Histologic examination, either by biopsy or examination of a resected specimen, is required for accurate classification. Histologic Features of Low Grade DNs Low grade DNs are very well defined nodules, being surrounded by a condensed rim of fibrous tissue similar to that of surrounding cirrhotic nodules. The nodules are thus not truly encapsulated. Portal tracts, present in virtually all DNS, are most often uniformly distributed in low grade DNs and may even be distributed in a virtually normal fashion with regularly intervening terminal hepatic venules. In some nodules, portal structures may be caught up in fibrous septa which partially subdivide the nodule. The hepatocytes of low grade DNs tend to be of comparable size to hepatocytes outside the lesion. The hepatocytes may display changes characteristic of the underlying liver disease affecting the surrounding liver, such as fatty change, Mallory bodies, or increased iron or copper deposition. These changes will be distributed in the DN as they are in surrounding cirrhotic nodules. Occasionally, a DN in a non-siderotic liver may contain increased iron or a DN in an otherwise siderotic liver will be iron free; it would still be classified as low grade. Rarely, a similar increase in copper, in an otherwise copper-free liver, is also seen. These changes in DNs, when not confined to a subnodule within the lesion, i.e. copper or iron retention and fat accumulation, may represent a marker of the clonality in the DN hepatocytes. We will return to this subject below. Studies of large cell change in cirrhosis and statistical evaluation of the association of this feature in DNs and HCCindicate the liklihood that large cell change is usually a reactive, not a premalignant change (though it also may serve as a marker for livers at increased risk for HCC). Therefore, it has been recommended that DNs containing large cell change without other atypical features be classified as low grade lesions. Scirrhous changes, stellate fibrosis of DN portal tracts or diffuse pericellular fibrosis in regions of DN parenchyma, have previously been thought to be features of high grade DNs; however, careful analysis of scirrhous change highlights that it does not segregate with other features of high grade lesions, appearing in otherwise low grade or high grade lesions equally. Thus we would recommend that scirrhous change be excluded from the diagnostic criteria for high grade DNs. Histologic Features of High Grade DNs High grade DNs are defined by the presence of small cell change and/or architectural atypia. They are usually well circumscribed and surrounded by a condensed rim of fibrous tissue, like low grade lesions, though some may merge focally with adjacent liver parenchyma. The atypical features in high grade DNs may take a variety of forms and may be diffuse throughout the nodule or focal. Diffuse changes most often fall into the category of cellular atypia. The definition of cellular atypia in this setting should be limited to small cell change: small, crowded hepatocytes with basophilic cytoplasm and an increased nuclear:cytoplasmic ratio. Small cell change appears to be consistently related to the development of HCC in a variety of studies and should remain a criterion for "atypical." This cellular feature is reported more frequently in studies from Japan than those from other countries; this discrepancy remains unexplained. Pseudoacinar structures resembling those seen in well differentiated HCC are a form of architectural atypia in high grade DNs and may be either focal or diffuse. Focal atypia may merge with the surrounding DN parenchyma, but it more often occurs instead as a "nodule-in-nodule" lesion. Such subnodules often appear to compress the adjacent DN parenchyma and studies of proliferative rates of the cells making up these lesions indicate that they are proliferating more rapidly than the surrounding tissue. These subnodules may display small cell change, but may also show changes which are not classically "atypical" including fatty change, clear cell change, clusters of hepatocytes with Mallory's hyaline, increased iron uptake within the DN, iron resistance in an otherwise siderotic nodule, and accumulation of copper binding protein. Some expansile subnodules do not display any distinctive cytological features though, architecturally, they may display a pseudoacinar growth pattern. We have argued that all subnodules, with or without distinctive cellular changes, are appropriately defined as architectural atypia on the basis of the expansile growth and should warrant classification of the entire DN as a high grade lesion. HCC may be identified in high grade DNs. These microfoci of HCC may display any of the features seen in larger HCCs, though they are usually well-differentiated . Typical growth patterns include pseudoacinus formation, thickened trabeculae or a scirrhous growth. Common cytological features include intracytoplasmic Mallory's hyaline, fatty change, clear cell change, iron resistance, and multinucleation. Multiple foci of HCC may also be found in a single DN and the histologic features of these foci are often different from each other. The DN parenchyma surrounding a microfocus of HCC will usually contain portal tracts and may consist of normal appearing hepatocytes, suggesting a background of a low grade MRN, or may show atypia, indicating a high grade background. Either way, by convention, DNs containing foci of HCC are classified as high grade. The Premalignant Nature Of DNs The association of DNs with HCC is demonstrated in two ways. First, DNs are sometimes found in livers which also contain grossly apparent HCC elsewhere. Second, as mentioned above, DNs sometimes contain one or multiple microscopic foci of HCC. Both of these relationships have been found to achieve statistical significance. Beyond this statistical correlation, the atypical features often found in DNs include many which have previously been thought to be premalignant. Clustering of hepatocytes containing Mallory's hyaline and foci of iron resistance in siderotic nodules have been independently described as premalignant changes. Small LCD, considered premalignant on the basis of morphometric analysis, is also seen and, in some series, is very common. Other features which are commonly identified in mature HCC are found in high grade DNs. For example, immunohistochemical studies of DN sinusoids reveals increasing degrees of "capillarization" (i.e. loss of endothelial fenestration, deposition of basement membrane and expression of antigens such as factor VIII and CD34) with the development of atypia and HCC. Studies of ploidy indicate increased frequency of aneuploidy in high grade DNs. Immunostaining for AFP is usually negative, but may be positive in relatively rare nodules with atypical foci and microfoci of HCC. Most importantly, longitudinal studies now demonstrate that, when followed over time by serial biopies or biopsy and then radiographic changes indicative of expansile growth, there is in fact progression from at least high grade DN to HCC. In these longitudinal studies, it is interesting to note that low grade DNs are not as obviously premalignant or as indicative of neighboring malignant transformation as are high grade DNs. Speculations on the Early Stages of Human Hepatocarcinogenesis How DNs actually form has not yet been firmly established. One early view suggested that an ordinary regenerative nodule in cirrhosis becomes more rapidly proliferative, therefore becoming larger. In turn, with the increased proliferation it also becomes at greater risk for the carcinogenic "hits", thereby giving rise to atypia and carcinoma. While simple, this hypothesis does not take into consideration three known facts about DNS. First, they can be found in livers in advance of cirrhosis and therefore do not always arise from a pre-existent regenerative nodule. Second, the presence of many intact portal tracts in most DNs, which have not yet been demonstrated to fully reconstitute after scarring and injury, suggests that they must be pre-existent to the formation of the DN, making it unlikely that a small cirrhotic nodule with few if any portal tracts could enlarge to a nodule with many portal tracts. Third, some DNs have been demonstrated to be clonal lesions, not hyperplastic phenomenon. To account for these features, we have suggested an alternative process of DN development (Figure 1). This alternate hypothesis has been able to predict some previously unexplored features of DNs and can be broadened to explain other types of borderline or early carcinoma features. It is currently the most widely accepted model for this mode of early hepatocarcinogenesis. The steps of this hypothesis are as follows: 1) A clonal expansion of hepatocytes follows on the earliest carcinogenic events in response to any diffuse injury of the liver which leads to increased hepatocyte turn over; 2) These early hits lead to a clonal expansion of hepatocytes which spreads around adjacent portal structures rather than displacing them; 3) As the rest of the liver becomes scarred, progressing to later stages of disease and eventually cirrhosis, the island of clonal hepatocytes, if resistant to the scarring affecting the rest of the liver, would remain intact -- an island of relatively preserved hepatic parenchyma made up of neoplastic, clonal hepatocytes; 4) With establishment of cirrhosis in the adjacent liver, the clonal expansion takes on the appearance of a large cirrhotic nodule; 5) Having already undergone the earliest transforming events of hepatocarcinogenesis, the clonal, hepatocyte expansion remains at increased risk for later developments and, thus, the lesion becomes the likeliest site of full malignant transformation. Figure 1: Pathways of Hepatocarcinogenesis Clonal, neoplastic expansion preceeds or coincides with develop-ment of cirrhosis; thus, cirrhosis is not a "pre-malignant condition" as has commonly been taught. Rather, they are parallel, related processes arising as a result of chronic liver disease. The morpho-logic appearance of the pre-malignant, clonal expansion depends on the relative stellate cell activation within the neoplasia and the sur-rounding reactive hepatic parenchyma. Distinguishing High Grade DNs from Early HCC This more generalized concept of early hepatocarcinogenesis can help to reconcile lesions, most often described by Japanese investigators, which do not match the classic descriptions of DNs (Figure 1). Some of these lesions consist of carcinoma arising simultaneously within nearby, normal-appearing cirrhotic nodules, thus, in the absence of a recognizable DN. More notably, lesions described as "early hepatocellular carcinoma", which often have indistinct borders or are identified in the absence of fully established cirrhosis, may simply reflect differing stages of scarring within the neoplastic expansions as opposed to the surrounding liver. The indistinct nature of the lesion suggests to the non-specialist that they are a different neoplastic pathway, while they are, in fact, the result of similar process. The trickiest diagnostic dilemma, of course, is where to draw the diagnostic line between "dysplasia" and "carcinoma." An informal comparison of diagnoses for a range of hepatocellular neoplastic lesions amongst Japanese and non-Japanese pathologists, shows a lack of consistency within and between those two groups when it comes to deciding "cancer" vs. "not yet cancer." But there is great consensus amongst all the reviewing pathologists regarding more established carcinomas and low grade lesions. Not surprisingly, then, the "borderline" category is the current area of controversy and leads to the greatest confusion when trying to compare studies from different investigators. Additional uncertainty in nodule characterization lies in the redundancy of so many adaptive or pathological changes in the liver, with each of many morphologic changes arising from very different causes. For example, fatty change in dysplastic nodules or early HCC may arise from clonal changes in lipid metabolism or sensitivity to alcohol toxicity, or to changes in the ratio of arterial to portal blood flow as the latter is lost during the neoplastic progression. Mallory bodies may also represent a clonal marker of neoplastic alterations in gene expression, but could arise from chronic cholestasis in a nodule with incomplete biliary drainage, or an altered response to alcohol or lipid metabolism. Hepatocytes within portal tract or septal stroma may represent either invasion, or entrapment of non-neoplastic hepatocytes by scarring, or hepatocyte regeneration from hepatic stem/progenitor cells (the "hepatocellular buds" described by Wanless in regression of cirrhosis). A perhaps very useful technique for distinguishing these non-neoplastic or pre-malignant lesions from early HCC with stromal invasion follows from our own study demonstrating that intraseptal hepatocytes in cirrhosis do largely represent such stem/progenitor-associated regeneration. We have previously shown that such hepatocytes are likely to arise from an hepatic stem/progenitor cell activation in the form of a ductular reaction. These reactions are readily stainable with antibodies which detect biliary-type cytokeratins (e.g. 7, 19). If such staining is applied to a nodule and hepatocytes within portal/septal stroma at the margins of the lesion or within it are devoid of such a reaction, then it serves to identify true stromal invasion and, thus, early hepatocellular carcinoma. Table 1. Features found in distinctive nodules in cirrhotic livers LGDN HGDN eHCC Diffuse hemosiderosis X rare rare Diffuse (or zonal) fatty change X Angiogenesis ("unpaired arteries") X XX XXX Scirrhous change X X X Large cell change X X X Small cell change X X Pseudoacinar growth X XX Iron resistence in otherwise siderotic nodule X X Mallory body clustering (with/without steatosis, PMNs) X X Nodule-in-nodule expansile growth (with steatosis or other changes above) X X Stromal invasion X LGDN: low grade dysplastic nodule; HGDN: high grade dysplastic nodule; eHCC: early hepatocellular carcinoma. X's: when multiple implies increased over the same lesion in lower grade nodules. Suggestions for further reading This very limited list contains studies or reviews that highlight particular histopathologic, immunohistochemica, or clinical features of low and high grade dysplastic nodules which should be of help to the general pathologist in understanding the fine points that will make diagnosis a less daunting proposition. They are listed in alphabetical order of first author, not by perceived importance. Deugnier Y, Charalambous P, le Quilleuc D, et al. Preneoplastic significance of hepatic iron-free foci in genetic hemochromatosis: A study of 185 patients. Hepatology 18: 1363-1369, 1993. Falkowski O, An HJ, Ianus IA, et al. Regeneration of hepatocyte 'buds' in cirrhosis from intrabiliary stem cells. J Hepatol. 2003; 39: 357-64. Hytiroglou P, Theise ND, Schwartz M, et al. Macroregenerative nodules in a series of adult cirrhotic liver explants: Issues of classification and nomenclature. Hepatology 21: 703-708, 1995. Kojiro M. Pathology of early HCC-progression from early to advanced. Hepato-Gastroenterology, 45;1203-1205, 1998. Krinsky G, Lee VS, Theise N. Focal lesions in the cirrhotic liver-high resolution ex-vivo MR imaging with pathologic correlation Imaging. J Comput Assist Tomogr 24: 189-196, 2000. Park YN, Chae KJ, Kim YB, et al. Apoptosis and proliferation in hepatocarcinogenesis related to cirrhosis. Cancer. 2001; 92: 2733-8. Park YN, Yang CP, Fernandez GJ, et al. Neoangiogenesis and sinusoidal "capillarization" in dysplastic nodules of the liver. Am J Surg Pathol. 22(6): 656-62, 1998. Sakamoto M, Hirohashi S, Shimosato Y. Early stages of multistep hepatocarcino-genesis: Adenomatous hyperplasia and early hepatocellular carcinoma. Hum Pathol 22: 172-178, 1991. Sakamoto M, Hirohashi S. Natural history and prognosis of adenomatous hyperplasia and early hepatocellular carcinoma: multi-institutional analysis of 53 nodules followed up for more than 6 months and 141 patients with single early hepatocellular carcinoma treated by surgical resection or percutaneous ethanol injection. Jpn J Clin Oncol. 28(10):604-8, 1998. Seki S, Sakaguchi H, Kitada T, et al. Outcomes of dysplastic nodules in human cirrhotic liver: a clinicopathological study. Clin Cancer Res 6: 3469-73, 2000. Takayama T, Makuuchi M, Hirohashi S, et al. Malignant transformation of adenomatous hyperplasia to hepatocellular carcinoma. Lancet 336(8724): 1150-3, 1990. Theise N, Fiel I, Hytiroglou P, et al. Macroregenerative nodules in cirrhosis are not associated with elevated serum or stainable tissue alpha-fetoprotein. Liver 15: 30-34, 1994. Theise ND, Park YN, Kojiro M. Dysplastic nodules and hepatocarcinogenesis. Clin Liver Dis. 2002 May;6(2):497-512. Review. Wanless I, Callea R, Craig JR, et al. Terminology of nodular lesions of the liver: Recommendations of the World Congress of Gastroenterology Working Group. Hepatology 22: 983-93. 1995.