| Abstract|| |
The differential diagnosis of hepatic mass lesions is broad and arriving at the right diagnosis can be challenging, especially on needle biopsies. The differential diagnosis of liver tumors in children is different from adults and is beyond the scope of this review. In adults, the approach varies depending on the age, gender, and presence of background liver disease. The lesions can be divided broadly into primary and metastatic (secondary), and the primary lesions can be further divided into those of hepatocellular origin and nonhepatocellular origin. The first category consists of benign and malignant lesions arising from hepatocytes, while the second category includes biliary, mesenchymal, hematopoietic, and vascular tumors. Discussion of nonepithelial neoplasms is beyond the scope of this review. The hepatocytic lesions comprise dysplastic nodules, focal nodular hyperplasia, hepatic adenoma, and hepatocellular carcinoma, and the differential diagnosis can be challenging requiring clinicopathological correlation and application of immunohistochemical (IHC) markers. Liver is a common site for metastasis, sometimes presenting with an unknown primary site, and proper workup is the key to arriving at the correct diagnosis. The correct diagnosis in this setting requires a systematic approach with attention to histologic features, imaging findings, clinical presentation, and judicious use of IHC markers. The list of antibodies that can be used for this purpose keeps on growing continually. It is important for pathologists to be up to date with the sensitivity and specificity of these markers and their diagnostic role and clinical implications. The purpose of this review is to outline the differential diagnosis of hepatic masses in adults and discuss an algorithmic approach to make a right diagnosis.
Keywords: Adenoma, carcinoma, cholangiocarcinoma, hepatocellular
|How to cite this article:|
Vyas M, Jain D. A practical diagnostic approach to hepatic masses. Indian J Pathol Microbiol 2018;61:2-17
| Introduction|| |
A variety of nonneoplastic as well as neoplastic lesions can present as hepatic masses and the differential diagnosis is very broad. It includes a variety of primary hepatic lesions and metastases. The differential diagnosis for liver mass lesions is different in children and will not be discussed here. The diagnosis requires correlation with clinical history, age, gender, history of medication including hormone use, imaging features, and the presence of any background liver disorder. The nature of the background liver disorder provides an important clue about the likely possibilities, and the presence or absence of cirrhosis is of utmost importance in formulating the differential diagnosis. The diagnosis of hepatic mass lesions can not only be challenging on biopsies but sometimes also on resections.
While the differential diagnosis has to be formulated on a case-by-case basis, an algorithmic approach is generally very useful. Simplistically, hepatic masses can be broadly classified as hepatocellular and nonhepatocellular in origin. This distinction is possible in many cases, purely based on morphology, while in some cases, it requires immunohistochemical (IHC) markers. In some poorly/undifferentiated cases, it may not be possible to definitively identify the cell lineage/differentiation despite extensive workup. Another practical way of approaching the differential diagnosis is to classify the lesions that arise in cirrhotic livers and those that arise in noncirrhotic livers [Table 1]. In cirrhotic livers, the main consideration is to rule out hepatocellular carcinoma (HCC) and the differential diagnosis includes dysplastic nodule (DN) and macroregenerative nodule (MRN). In noncirrhotic livers, especially in young women, hepatic adenoma (HA) and focal nodular hyperplasia (FNH) remain the top differential diagnoses, while in elderly patients, metastases are most common. In addition, a variety of vascular and mesenchymal lesions and nonneoplastic processes such as liver abscesses, hydatic cysts, and heterotopias can also present as hepatic masses and need to be considered in appropriate clinical setting. A brief discussion of key histologic features of various hepatic mass lesions and an algorithmic approach to their diagnosis is outlined in the subsequent section.
| Masses of Hepatocytic Origin|| |
As discussed above, the lesions can be broadly classified as those arising in cirrhotic and noncirrhotic livers as the differential diagnosis is sufficiently different.
Masses in cirrhotic livers
Role of radiology in the diagnosis of hepatic nodular lesions
Nodules in a cirrhotic liver represent a spectrum of lesions. Screening and surveillance of patients with cirrhosis for HCC has led to early detection of such suspicious nodules on imaging. Since many nodules can be diagnosed as HCC based on imaging findings, biopsy confirmation is not necessary in every case. Lesions <1 cm can be followed up by imaging for up to 2 years. Lesions 1–2 cm in size require two concordant radiological tests for a noninvasive diagnosis of HCC, otherwise biopsy is indicated., Most nodules >2 cm in cirrhotic liver are HCC and can be reasonably diagnosed on contrast-enhanced imaging (computed tomography or magnetic resonance imaging) by the presence of typical hypervascularity in arterial phase and washout in portal venous phase. Biopsy is required only when the imaging is not typical of HCC.,,,
Role of pathology in the diagnosis of hepatic nodular lesions
These atypical nodules on imaging represent a histologic spectrum ranging from regenerative nodules to DNs to HCC. While HA and FNH have also been described in cirrhotic livers, due to rarity of these lesions in cirrhosis, the diagnosis should be largely restricted to the examination of resection specimens. Early diagnosis of HCC is desirable due to the high mortality associated with advanced disease. The International Consensus Group for Hepatocellular Neoplasia classified nodular lesions in cirrhotic livers into MRN, low-grade DN (LGDN), and high-grade DN (HGDN). They also defined small HCC as all tumors <2 cm in size. The diagnostic criteria for MRN, DN, and early HCC are qualitative, and hence very subjective, especially difficult to apply on liver biopsies.
MRNs represent cirrhotic nodules that are >5 mm and devoid of any cytological or architectural atypia., Any nodule demonstrating some cytological or architectural atypia that is insufficient for a diagnosis of HCC is considered to be a DN. The cytologic atypia can be in the form of small cell change or large cell change. Studies suggest that small cell change is a form of dysplasia with higher malignant potential, while large cell change is a reactive phenomenon.
LGDNs are often distinct from the surrounding liver because of fibrous tissue around the nodule. They show mild increase in cell density and lack cytologic atypia (small cell change), although they may show large cell change. Unpaired arteries are sometimes present in small numbers., The cell plates display normal thickness and the lesion contains portal tracts and central veins. LGDNs are difficult to separate from MRNs on needle biopsies and the morphological criteria have proved unreliable and not reproducible.,
HGDNs show cytologic atypia in the form of small cell change and increased cell density, but the features are insufficient to diagnose HCC. More features of angiogenesis are seen in HGDN as compared to LGDN. Partial sinusoidal capillarization along with few nontriadal arteries can be seen. Pseudoacini are absent or rare. Secondary nodules arising in HGDN (“nodule in nodule”) most likely represent HCC arising in HGDN., The distinction between HGDN and early HCC can be difficult in some cases, and the presence of stromal invasion into portal tracts, fibrous septa, or adjacent parenchyma helps make the diagnosis of HCC, but these features may be absent in biopsies. The presence of pseudoacini formation or thickened hepatic cell plates may point to the diagnosis of HCC over HGDN in such cases., The distinction between MRN, LGDN, and HGDN assumes significance as the risk of malignant transformation is higher with HGDN as compared with MRN and LGDN. Recent genetic and molecular data suggest that HGDN is a precancerous lesion of HCC, and that LGDN, HGDN, “nodule-in-nodule” early HCC, and HCC is the likely sequence of carcinogenesis in HCC.
As mentioned earlier, HCCs >2 cm can be diagnosed solely on radiology if classic features are present and do not require biopsy confirmation. This 2-cm cutoff is also used to classify HCC as small HCC (<2 cm) and conventional HCC (>2 cm). Small HCCs have been further classified into early HCC and progressed HCC, based on morphological features. Early HCC has a vague nodular appearance with incomplete or absent fibrous capsule, well-differentiated morphology, and often steatosis. Progressed HCCs, on the other hand, are clearly malignant appearing with distinct nodular architecture, destructive or pushing growth, and frequent vascular invasion. These lesions are usually nonsteatotic in view of being completely neoarterialized. Progressed HCCs can arise de novo, from preexisting HGDN or from an early HCC (giving the appearance of “nodule in nodule”). Early HCC has a longer time to recurrence and a higher 5-year survival rate than progressed HCC and hence their separation into distinct categories. Despite this difference, most centers including ours do not use this subclassification of HCC in their routine clinical practice and clinical decisions are made simply based on size, stage, differentiation, and other clinical parameters.
Conventional HCCs (>2 cm) can be well, moderate, or poorly differentiated [Figure 1]. In well- differentiated and moderately differentiated HCCs, the tumor cells are clearly hepatocytic in nature and demonstrate obvious cytologic atypia, nuclear pleomorphism, thick hepatic cords (>3 cell plates' thick), pseudoacini, and loss of reticulin fibers [Figure 1]. Cells may have eosinophilic, steatotic, or sometimes clear cytoplasm. A variety of intrahepatocytic inclusions may be seen that include Mallory hyaline, bile, D-PAS (Periodic Acid Schiff with diastase)-positive eosinophilic globules, and pale bodies. HCC may show several cytoarchitectural patterns as follows: trabecular, macrotrabecular, acinar/pseudoglandular, solid, clear cell, scirrhous, or steatohepatitic type. Well-to-moderately differentiated HCCs in a background of cirrhosis usually do not require any IHC stains to confirm their hepatocytic differentiation; however, occasionally, one can use HepPar-1 antigen [Figure 2]c or Arginase in case of slightest doubt. However, in very well-differentiated lesions, IHC markers that support malignancy are often required to differentiate them from benign nodules. Reticulin stain also helps by delineating the thick trabeculae and loss of reticulin network [Figure 2]a.
|Figure 1: Examples of morphologic subtypes of hepatocellular carcinoma. (a) Moderately differentiated hepatocellular carcinoma – a trabecular pattern. Bile formation is noted (arrow) and cells very closely resemble hepatocytes. The nuclear atypia is mild and hepatic cords are thickened. (b) Hepatocellular carcinoma with macrovesicular steatosis with ballooned hepatocytes and lobular inflammation (steatohepatitic type). (c) Hepatocellular carcinoma showing extensive acinar differentiation. (d) Poorly differentiated hepatocellular carcinoma where the tumor cells have hardly any resemblance to normal hepatocytes|
Click here to view
|Figure 2: Key stains in the diagnosis of hepatocellular carcinoma. (a). Reticulin showing thickening of hepatic cell plates and focal loss of reticulin fibers. (b) CD34-diffuse capillarization of sinusoids in the tumor (left) (inset – intense cytoplasmic staining of sinusoidal endothelial cells). (c) Hep Par1-intense granular cytoplasmic staining in moderately differentiated hepatocellular carcinoma (inset – high power). (d) Glypican 3 diffuse cytoplasmic positivity in a well-differentiated hepatocellular carcinoma (inset – high power). (e) heat shock protein 70-strong cytoplasmic and nuclear positivity in moderately differentiated hepatocellular carcinoma (inset– high power). (f) Polyclonal CEA-canalicular positivity in the part of the tumor (left) and membranous positivity (right) (inset – canalicular staining)|
Click here to view
Poorly differentiated HCCs may show a trabecular or solid growth pattern with marked cellular pleomorphism. The malignant nature of these lesions is immediately obvious based on histology; however, IHC is often needed to confirm their hepatocytic nature [Table 2]. The markers that are useful in this context are Arginase, HepPar1, and polyclonal CEA/CD10 [canalicular staining pattern, [Figure 2]f, although HepPar-1 and pCEA have lower sensitivity in PD-HCC. Although GPC3 can be positive in other tumors, when combined with Arginase, their sensitivity nears >80%, and in this setting, this combination is very helpful.
|Table 2: Markers aiding diagnosis of hepatocytic lesions[15,19,21,92,108]|
Click here to view
Immunohistochemistry in differentiation of high-grade dysplastic nodule versus early hepatocellular carcinoma
The distinction between high-grade dysplastic nodule (HDN) and early HCC can be tricky in some cases, especially on biopsies. Even though stromal invasion is the most reliable feature of small HCC, it is not always evident, especially in a biopsy specimen. Loss or fragmentation of reticulin framework or pseudoacini formation is a useful feature when identified. However, this feature is more often seen in progressed HCC than early HCC. Absence of ductular reaction in HCC can be highlighted with a CK7 and/or CK19 stain which helps differentiate it from MRNs and DNs. A panel of three immunostains, namely, glypican 3 (GPC3), glutamine synthetase (GS), and heat shock protein 70 (HSP70), has been recommended. GS is an enzyme involved in the nitrogen metabolism pathway in the liver and is also a target of β-catenin. Observing the pattern of expression of GS is important as it may be normally positive in the hepatic cell plates around the central veins. For diagnosis of HCC, GS should be diffusely positive in at least 50% of the tumor cells. HSP70 is an antiapoptotic protein which is abundantly upregulated in HCC but may be focally positive in HDNs and also in normal periseptal hepatocytes and stellate cells. For a diagnosis of HCC, HSP70 should show strong nuclear positivity [Figure 2]e. GPC3 is a membrane-anchored heparin sulfate proteoglycan normally expressed in fetal tissue and placenta but not in normal adult liver. The positivity for GPC3 (glypican 3) increases from DN to early or well-differentiated HCC to moderate-to-poorly differentiated HCC. GPC 3 is expressed in HCC in a diffuse cytoplasmic, membranous, or Golgi pattern [Figure 2]d. Staining with at least two of these three markers is considered supportive of a diagnosis of HCC; however, the sensitivity of this combination is only up to 50%. Diffuse endothelial staining for CD34 [Figure 2]b, signifying capillarization of the sinusoids, is indicative of HCC and, in our experience, is more helpful than HSP70 stain, although there may be some overlap in the staining pattern of early HCC and HDN. Some studies have shown that the diagnostic accuracy of the three marker panels (GPC3, GS, and HSP70) can be increased by addition of a fourth stain, clathrin heavy chain, especially in cases of small HCCs. However, the data on utility of these markers in practice still remain preliminary and scant. The practical issue remains that in cases that are very well differentiated and the diagnosis is difficult on HE stains, the special stains, including GPC3, tend to be equivocal or unhelpful. In our experience, GPC3 is positive in <30% of cases that have diagnostic difficulties and HSP70 is seldom useful. In this setting, reticulin and CD34 are the most useful special stains in our hands [Figure 2]a and [Figure 2]b. A canalicular staining pattern when stained with polyclonal CEA [Figure 2]f is particularly seen in HCC. Similar staining pattern is also observed when stained with CD10. Despite all the available tools, there may be instances when definitive diagnosis is not possible on biopsy samples. In such cases, correlation with clinical and radiological information may be helpful; however, sometimes, a definitive diagnosis may have to be reserved for the resection specimen.
Lesions arising in noncirrhotic livers
In the noncirrhotic livers, the possibility of other neoplasms besides HCC that can involve the liver becomes significant and includes various benign hepatocytic and biliary proliferations, vascular and mesenchymal tumors, and metastases.
HCC, although most commonly occurs in cirrhotic livers, can also occur in noncirrhotic livers. Studies suggest that the incidence of HCC in noncirrhotic livers has been increasing, largely attributed to the increasing incidence of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis and decreasing incidence of viral hepatitis. These tumors are usually well to moderately differentiated and are larger in size at presentation as compared with their counterparts arising in cirrhotic livers., However, from a diagnostic standpoint, the approach is no different from those seen in cirrhotic livers. The differential diagnosis for the well-differentiated lesions in this setting includes HA and FNH, but does not include DN or MRN. In noncirrhotic livers, especially in young individuals, a possibility of fibrolamellar variant of HCC also needs to be kept in mind.
Fibrolamellar hepatocellular carcinoma
FLC is a distinct variant of HCC which typically arises in a younger age group without cirrhosis, although sometimes they can arise in older patients, or rarely in combination with conventional HCCs. Grossly, these tumors are larger than conventional HCCs and appear tan-white with often a central scar with calcifications. The tumor cells are polygonal, large with abundant eosinophilic granular cytoplasm, show conspicuous eosinophilic macronucleoli, and are invested within abundant collagenous stroma that has a lamellar appearance. This triad is virtually diagnostic of FLC. Slightly pale eosinophilic, fibrinogen containing cytoplasmic inclusions (pale bodies) are frequently present, but are not specific.,,, These tumors are positive for HepPar-1, polyclonal CEA (canalicular pattern), and arginase. Granular cytoplasmic staining with CD68 is sensitive but not specific and the cells frequently express CK7 and CK19, although there is no evidence of true cholangiolar differentiation., Recently, a novel somatic fusion gene DNAJB1-PRKACA has been identified in these tumors and its detection is considered the gold standard for making the diagnosis. However, testing for this fusion gene is currently available only in limited centers. It has been suggested that when this fusion gene cannot be demonstrated, CD68 and CK7 IHC stains are very helpful in making the diagnosis with suggestive histology, especially on biopsies. In the hands of experienced pathologists who are familiar with this neoplasm, the diagnosis can be made on HE stains alone in most cases, provided the histology is classic.
HA is a rare, benign neoplasm that is most often seen in women in the reproductive age group and is strongly associated with oral contraceptive use. Other predisposing causes include use of anabolic steroids, antiepileptic drugs, vascular (e.g., Budd–Chiari syndrome, portal vein shunts, and agenesis,) and biliary disorders (primary sclerosing cholangitis), and metabolic and genetic disorders (familial adenomatous polyposis, Peutz–Jeghers syndrome, Klinefelter syndrome, glycogen storage disease, obesity, and diabetes MODY3).,,, During its evolution, HA may remain stable and increase in size or regress. Transformation into HCC has been reported, mostly in tumors related to androgen or anabolic steroids' exposure and in β-catenin-activated HA; however, this still remains controversial as some of these reported tumors likely represent well-differentiated HCC from the very beginning., HA can be single or multiple; the term “adenomatosis” is appropriate when >10 tumors are present.
Grossly, HAs are well-defined lesions that stand out from the background noncirrhotic liver parenchyma. They are usually unencapsulated, homogeneous, tan-white to brown lesions, and sometimes with areas of necrosis or hemorrhage. Some tend to be yellowish due to steatosis on gross examination while others appear pale red to dark red. Histologically, the neoplasm is sometimes demarcated from adjacent normal liver by a pseudocapsule or compressed liver, and is composed of relatively monomorphic hepatocytes, lacks true portal tracts, and shows thin-walled unpaired arteries [Figure 3]. Similar to HCC, bile, steatosis, lipofuscin, Dubin–Johnson syndrome-like coarsely granular brown pigment, Mallory hyaline, peliosis, and extramedullary hematopoiesis may be seen.
|Figure 3: (a) Low magnification of hepatic adenoma (right) showing similar morphology but absence of portal tracts as compared to the normal (left) (b) Higher magnification of the hepatic adenoma showing very bland hepatocytes in trabecular arrangement and absence of portal tracts. A single unpaired artery devoid of fibrous sheath (naked artery) is surrounded by hepatocytes (c) Inflammatory hepatic adenoma showing marked telangiectasia and portal tract-like structures with only arteries and chronic inflammation. Bile ductular proliferation may be seen. (d) Hepatic adenoma with macrovesicular steatosis is often seen in hepatocyte nuclear factor 1α-mutated adenomas|
Click here to view
Subclassification of hepatic adenoma
In 2006, the French collaborative group proposed the classification of HAs, based on molecular and genomic data that have significantly changed our clinical practice. The HAs are now subclassified based on their histologic features, IHC markers, and molecular phenotype. A detailed discussion of HA subclassification is beyond the scope of this discussion and readers are referred to excellent reviews on this topic., The IHC stains used in this regard include liver fatty binding protein (LFABP), serum amyloid A-associated protein (SAA), C-reactive protein (CRP), β catenin, and GS. Though besides identifying β-catenin-activated HA, which carries a high risk of malignant transformation, in most cases, the subtyping of HA may be of more academic interest.,
Hepatocyte nuclear factor 1α-inactivated hepatic adenoma hepatocyte nuclear factor 1α (HNF1α) is a tumor suppressor gene and its inactivation promotes lipogenesis by increasing fatty acid synthesis and downregulation of LFABP. LFABP is a small protein involved in the intracellular transport of long-chain fatty acids in the liver and its downregulation promotes intralesional steatosis, which is often a histologic feature of H-HA.,,, While most HNF-1-inactivated HAs show steatosis, it is not an universal finding. Most H-HAs result from biallelic inactivating somatic mutations and constitute approximately 35%–40% of all HAs.,,,, Less than 10% of patients have germline mutation in one of the genes and tend to present at a younger age. They frequently have maturity-onset diabetes of the young type 3 (MODY3) and a family history of liver adenomatosis.,,,,,,
Histologically, the hallmark of this subtype of HA is moderate to marked steatosis (hence the synonym “steatotic adenomas”) with periarterial sparing [Figure 3]d., Normal liver parenchyma stains faintly with LFBAP, while loss of LFABP staining in the lesion is characteristic of these lesions [Figure 4]a. However, loss of LFABP can also be seen in HCC and the marker is not useful in that regard. GS expression is seen in normal pericentral distribution.
|Figure 4: Subtyping of hepatocellular adenoma. (a) Loss of liver fatty acid binding protein in hepatocyte nuclear factor 1α-mutated hepatic adenoma (b) Strong granular cytoplasmic expression of C-reactive protein in inflammatory hepatic adenoma. Serum amyloid A expression is similar. (c) Strong aberrant nuclear staining of β-catenin in β-catenin-mutated hepatic adenoma. (d) β-catenin-mutated hepatic adenomas show diffuse positivity for glutamine synthetase (GS), a surrogate marker for Wnt signaling activation. (e) Example of hepatic adenoma that is negative for glutamine synthetase with patchy peripheral staining, while the adjacent liver shows normal perivenular staining pattern (f) Focal nodular hyperplasia showing geographic (map like) staining pattern for glutamine synthetase|
Click here to view
β catenin-activated hepatic adenoma
β-catenin is a component of the Wnt/B-catenin signaling pathway which has an important role in hepatocellular development and physiology. In normal cells, β-catenin is rapidly degraded. Mutations in either β-catenin gene or the gene coding for proteins which degrade β-catenin (axins, glycogen S kinase 3, and adenomatous polyposis coli) lead to its constitutive activation and nuclear accumulation. B-HA constituted about 10%–15% of all HAs in the French cohort; however, their reported prevalence in the USA and in our own experience is much lower. In contrast with other subtypes, B-HA occurs more commonly in older men and has a higher risk of malignant transformation., Histologically, in addition to the features of typical HA, these lesions may show pseudoacini and mild cytologic atypia.
On IHC, B-HA shows a distinct pattern of heterogeneous cytoplasmic and aberrant nuclear staining for β-catenin [Figure 4]c, although this finding is almost always very focal. β-catenin mutation also leads to activation of GLUL gene that results in constitutional activation of GS seen as diffuse strong GS staining in the lesion and serves as a surrogate marker for this subgroup [Figure 4]d.,, GS expression is considered to be a more reliable marker of activation of Wnt/B-catenin signaling than nuclear β-catenin staining as the staining can be limited to only few hepatocytes and some cases are negative for β-catenin mutation, suggesting other mechanisms for activation of Wnt signaling. A variety of mutations in β-catenin have been identified that lead to variable degree of β-catenin activation and in turn variable staining for GS. Mutations in exon 3 appear to be the most common with deletions and amino acid substitutions within the ß-TRCP binding site (D32-S37) that are associated with strongest activation of β-catenin and diffuse and homogeneous staining for GS. T41 mutations are associated with moderate β-catenin activation and S45, K335, and N387 mutations are associated with weaker β-catenin activation. Weak β-catenin activation is associated with heterogeneous and weaker staining for GS. It is important to note that 20%–30% of HCCs also show β-catenin mutations leading to aberrant nuclear expression, suggesting evolution along this pathway, and this stain cannot be used for differentiating HA from HCC.,, Pattern of GS staining also helps in differentiation from FNH as FNH shows a characteristic “map-like” positivity around the fibrotic bands [Figure 4]f. We feel that many cases of β-catenin-activated HAs represent well-differentiated HCCs. Their occurrence in older patients, often men and with cytological and architectural atypia (pseudoacini), supports this notion.
Inflammatory hepatic adenoma
This category represents the previously so-called “telangiectatic HA” and accounts for 40%–60% of all HAs., The underlying mechanism is a mutation in the IL6 transducer gene which encodes the glycoprotein-130 (gp-130) component of IL6 receptor. The resulting activation of IL6 leads to acute-phase inflammatory response through the STAT3 signaling pathway. Some cases of tumors with activating GNAS mutations, with or without associated McCune–Albright syndrome, have also been shown to possess the inflammatory phenotype. Clinically, these patients often have a high body mass index and a history of alcohol consumption. These patients may clinically present with features of systemic inflammatory response such as anemia, fever, leukocytosis, and elevated serum CRP or SAA., Morphologically, these tumors show patchy inflammatory infiltrates, composed of lymphocytes, histiocytes, rare plasma cells, and neutrophils, often grouped around thick-walled arteries and/or portal tract-like areas [Figure 3]c. These areas of thickened vasculature have been termed as “pseudoportal tracts” as the portal vein and bile duct are lacking. There is variable ductular reaction around the vasculature, similar to FNH, but a central scar typical of FNH is lacking. Due to this morphologic similarity, these lesions were once considered a variant of FNH.,, Areas of sinusoidal dilatation and peliosis are often seen justifying the use of the term “telangiectatic “ to describe these lesions. True to their nature, these lesions strongly express acute-phase inflammatory proteins such as SAA and CRP [Figure 4]b compared to weaker expression in the adjacent liver parenchyma. CRP has higher sensitivity (~100%); however, diffuse expression can also be seen in up to 15% of cases of FNH and periseptal staining in all cases. The sensitivity of SAA is around 90%. LFABP expression is normal. Approximately 10% of IHAs harbor β-catenin mutations and will show aberrant nuclear β-catenin expression. CK7 immunostain is useful to highlight the ductular reaction surrounding the vasculature. The CD34 stain highlights the vasculature and can stain the sinusoidal endothelium, sometimes extensively, mimicking a HCC.
Hepatic adenoma, unclassified
This category comprises about 5%–10% of all the HAs. By IHC, they show cytoplasmic LFABP, pericentral GS, and no nuclear β-catenin, SAA, or CRP. Approximately 4% of cases in this group have recently been shown to be associated with the activation of sonic Hedgehog pathway, resulting from the fusion of INHBE and GLI1 genes. Henriet et al. have described strong expression of arginosuccinate syntase-1 (ASS 1) in a subset of unclassified HAs, suggesting another distinct subtype. It appears that expression of ASS-1 correlates with an increased risk of bleeding. The role of arginine in the synthesis of proteins that are precursors of nitrous oxide has been suggested as the basis of increased hemorrhagic tendency associated with these tumors.
Immunohistochemistry in the diagnosis of hepatic adenomas
The role of IHC in the diagnosis of HA is twofold.
- Confirmation of diagnosis of HA and its distinction from HCC and FNH
- Subclassification of HA.
Diagnosis of hepatic adenoma and differentiation from hepatocellular carcinoma
Distinction of HAs from well-differentiated HCCs can be difficult, particularly on needle biopsies. Presence of hepatocyte plate thickening (>3 cells), pseudoacini, small cell change, or loss of reticulin framework is indicative of HCC. However, one needs to be aware that HAs with steatosis can show focal loss of reticulin, which can be misleading, especially on small biopsies. IHC markers such as GPC3, GS, HSP70, and CD34 can be helpful in this regard. Though a positive GPC3 is useful in making the diagnosis of HCC, a negative result cannot rule it out. Studies suggest that while GPC is positive in >90% progressed HCCs, the positivity dips down to <50% in well-differentiated HCCs. Similarly, arterialization of sinusoids highlighted by CD34 is supportive of HCC, but not entirely specific as it can also be seen in some HAs and FNHs, especially in periportal and periseptal areas., Hence, the diagnosis cannot be based on a single immunostain, and a panel of markers in addition to clinical features is required to make a definitive diagnosis. Finally, it is worth acknowledging that some HAs cannot be reliably differentiated from well-differentiated HCCs, especially on small biopsies, and are best reported in clinical practice as “well-differentiated hepatocytic neoplasm/lesion” with a comment explaining the differentials. Some have suggested the designation “hepatocytic neoplasm of uncertain malignant potential (HUMP)” for such lesions. Clinical scenario and treatment implications need to be carefully considered before rendering opinion in such cases. The diagnosis of HA in women over 50 years and men of any age should be made with extreme caution, unless a predisposing cause is known (glycogen storage disease or estrogen/androgen use)., For differentiation of HA from FNH, see the subsequent section on FNH.
Role of immunohistochemistry in the subclassification of hepatic adenoma
The immunostains that help in subclassification include LFABP, CRP, SAA, β-catenin, and GS. It is worth noting that the stains useful in subclassification of HA should be used solely for this purpose and that they are not helpful in differentiating HA from HCC or other benign lesions (except GS in FNH). Their pattern of staining and role in each subtype are discussed above [Table 3].
|Table 3: Markers for subclassification of hepatocellular adenomas[19,50]|
Click here to view
Focal nodular hyperplasia
FNH is a benign lesion arising as a result of a regenerative response to vascular abnormality. Though these lesions commonly arise in noncirrhotic livers, FNH-like lesions have been rarely described in cirrhotic livers as well., It is more common in young women, with an estimated prevalence of 0.9% which is ten times greater than HA.,, Portal tract injury and arterialization of sinusoids result in disorganized growth of hepatocytes around them. Increased incidence of FNH can be seen in hereditary syndromes associated with vascular abnormalities, such as hemorrhagic telangiectasia (Rendu–Osler–Weber disease) or congenital absence of portal vein.,,, The true nature of these lesions has been a matter of debate; however, it has been shown that these lesions are polyclonal by studies of X-chromosome inactivation. The characteristic features include a central scar, containing thick-walled vessels, with radiating fibrous bands with nodular regeneration of hepatic parenchyma, giving a pseudocirrhotic appearance. A marked ductular reaction and inflammatory infiltrate are seen at the interface between the fibrous bands and hepatocytic nodules. Differentiating FNH from HA or well-differentiated HCC can be challenging in some cases, especially on biopsies which may not demonstrate a central scar, but only few fibrous bands. IHC can be useful in such cases. A “map-like,” geographic pattern of GS staining is more characteristic of FNH. Rarely, the geographic pattern can become extensive mimicking diffuse staining that can be misleading and some cases may show a pseudo-geographic pattern. The ductular reaction adjacent to the fibrous bands can be highlighted with CK7 and CK19, a feature usually absent in HA (except inflammatory subtype) and HCC.,, The LFABP, CD34, SAA, and CRP stains are not useful in differentiating FNH from either adenomas or HCCs.
Lesions of nonhepatocellular origin
Adenocarcinoma of the bile duct, also commonly referred to as cholangiocarcinoma (CC), is the second most common primary epithelial malignancy of the liver after HCC, accounting for about 15% of all primary malignant hepatic neoplasms. Known risk factors include chronic biliary inflammation, the most common being primary sclerosing cholangitis in the Western World and liver parasitic infestation (Opisthorchis viverrini, Clonorchis sinensis, and Schistosoma japonica) in endemic areas.,, Other causes of chronic biliary inflammation such as hepatolithiasis, choledochal cyst, and Caroli's disease and toxic agents such as Thorotrast have been implicated. Cirrhosis and chronic hepatitis C have been associated with ten-fold and four-fold increased risk of CC, respectively. However, more than 80% of CC cases have no associated predisposing factors.,
The most common histologic pattern of CC is tubular or glandular in a desmoplastic stroma [Figure 5]. The glandular component can also show cribriform, nesting, cord-like, or papillary patterns. Other variants recognized in the WHO classification include mucinous, squamous, clear cell, papillary, oncocytic, and spindle/sarcomatoid. While solid or undifferentiated variants are rare, which can cause a diagnostic confusion with HCC, the major differential diagnosis remains metastatic adenocarcinoma.
|Figure 5: Cholangiocarcinoma. (a) Classic cholangiocarcinoma involving the liver. Note the fibrotic background and moderately differentiated glandular profiles. (b) Cholangiocarcinoma showing a more hepatoid appearance at the periphery of the tumor with glandular differentiation mimicking a trabecular pattern, and tumor cells look more like hepatocytes. This can be a potential pitfall on small biopsies. Other areas in the tumor and immunohistochemical showed features of a typical cholangiocarcinoma|
Click here to view
Differentiating cholangiocarcinoma from hepatocellular carcinoma
Negativity for the hepatocytic markers (HepPar1 antigen, Arg1 and GPC3, AFP, CD10, and pCEA) helps distinguish CC from HCC.,, Another marker, bile salt export pump (BSEP), an ATP-bound transport protein expressed in the canalicular aspect of hepatocyte, has also been found to be a very sensitive and specific marker for hepatocellular differentiation.
Differentiating cholangiocarcinoma from metastatic carcinoma
Presence of a desmoplastic stroma may be a clue for the diagnosis of CC, but is rather nonspecific. Immunophenotype of CC is not distinct either. In fact, distinction of intrahepatic CC from metastatic adenocarcinoma originating in stomach, gallbladder, pancreas, or extrahepatic bile ducts may not be possible based on histology and IHC. CC is usually positive for CK7, CK19, CK8/18, MOC31, and MUC 1, 2, and 3. They may also express CK20 and CDX2 in a subset of cases. Villin is expressed in adenocarcinomas of the colon, stomach, and pancreatobiliary tract, and hence has low specificity. It can also be expressed in the brush border of approximately 50% of CC, particularly in intrahepatic CC., A variety of other markers are available in practice that can help narrow down the possible primary sites in conjunction with the morphology. A list of all these markers, including markers for hepatocytic differentiation and their specificity, is shown in [Table 2] and [Table 4]. Often the markers are used in combination or as a predecided panel; however, one needs to keep this in mind that none of the markers is 100% sensitive or specific. It is imperative that the pathologist is aware of the potential pitfalls (false positivity and false negativity) associated with each marker and should not make the diagnosis on the basis of a marker alone.
|Table 4: Lineage-specific markers helpful in the diagnosis of metastatic tumors to the liver|
Click here to view
Combined hepatocellular/cholangiocarcinoma (mixed hepatocellular carcinoma-cholangiocarcinoma)
These are the tumors that contain unequivocal, intimately mixed elements of both HCC and CC. This diagnosis excludes the collision tumor with HCC and CC, and hence intimate admixture is a strict requirement. The subclassification and nomenclature of these tumors is a source of controversy and confusion; however, each of the components need to be recognized on the basis of morphology and supported by IHC. The HCC and CC components can show a variable degree of differentiation. The CC component is typically positive for CK7, CK19, and MOC31, while the HCC component shows variable expression of any of the hepatocytic differentiation markers discussed above.,, Aberrant expression of hepatocytic and biliary markers in morphologically HCC and CC components tends to occur not infrequently and can confuse the unwary. Making correct diagnosis and distinction from HCC is clinically important as these tumors are more likely to involve lymph nodes compared to HCCs and have an intermediate prognosis between HCC and CC. Another subtype of such mixed tumors is the tumor that shows a predominance of areas with combined or transitional features between HCC and CC areas. Such tumors are called HCC-CC with stem cell features. The cells in these transitional areas resemble stem cells both morphologically (small size, scant cytoplasm, oval shape, and absence of nuclear atypia) and immunohistochemically (positivity for NCAM, KIT, and CD133, in addition to CK19 and MOC31). The arrangement of these cells can vary and gives rise to various morphologic subtypes such as typical (peripheral arrangement of stem cell-like cells), intermediate (solid nests, strands, or trabeculae), or cholangiolar (tubular or anastomosing cord-like pattern).
Primary biliary mixed adenocarcinoma and neuroendocrine carcinoma
Mixed adenocarcinoma and neuroendocrine carcinomas (MANECs) are malignant tumors composed of adenocarcinoma/squamous cell carcinoma components along with a neuroendocrine carcinoma, each component comprising at least 30% of the tumor. MANECs are rare in general and involvement of the biliary tract is even rarer. On histology, they may show areas of intestinal- or pancreatobiliary-type adenocarcinoma intermingled with high-grade neuroendocrine tumor. Immunostains can clearly demarcate the adenocarcinoma component with cytokeratin 7, 19, CEA, and CA19.9, and the neuroendocrine component can be ascertained with at least one out of the three major IHC markers (synaptophysin, chromogranin A, and CD56). MANECs are aggressive tumors and the treatment is driven by the most aggressive histologic component.
Other benign lesions of biliary lineage
Bile duct adenoma and von Meyenburg complex/bile duct hamartoma
Both these lesions are most often detected as incidental subcapsular lesions on laparotomy. Bile duct adenomas (BDAs) are usually 0.5–1.5 cm in size and tend to be larger than von Meyenburg Complex (VMC) but smaller than most CCs. They are most commonly located subcapsularly and grossly seen as tan-white to yellow nodules. These lesions are well circumscribed and composed of irregular and somewhat angulated ductular structures in variably collagenized stroma. BDA lacks infiltrative growth or an expansile growth pattern, and long-standing lesions may be markedly sclerotic. In some cases, the biliary epithelium may show endocrine, oncocytic, or clear cell features.,, BDA has generally been considered a benign biliary neoplasm occupying an intermediate position between VMC and CC, although one study proposed it to be a benign proliferation of peribiliary glands and thus considered these lesions “peribiliary gland hamartomas.” In this study, the glandular structures of BDA, unlike VMC, were found to be immunoreactive to “peri-biliary gland-specific” antigens (D10 and 1F6), which do not react with normal bile duct epithelium.
VMCs are thought to represent remnants of biliary epithelium of the ductal plate that fail to involute and in some cases are associated with ductal plate malformation and/or polycystic liver diseases. They are asymptomatic and are most often incidentally identified as tiny nodules or cysts during imaging studies, abdominal surgery, or at autopsy. These are usually smaller than biliary adenomas (<0.5 cm) and comprise irregular and branching ductular profiles, some of which are dilated/cystic with inspissated bile, in a sclerotic stroma. Some of the larger lesions can be more cellular with more compact glands with scanty stroma mimicking BDA, and in some cases, the distinction is arbitrary. Preserved portal structures are identified in some areas of the lesion, along with few aggregates of benign lymphocytes. Nuclear pleomorphism is minimal and mitoses are absent.
BDAs can cause potential diagnostic confusion with VMC at one end and CC at the other end of the spectrum  and the distinction can be difficult on biopsies. IHC markers are not helpful in this setting. Both VMC and BDA have a uniform bland ductular proliferation with no significant atypia, mitosis, or infiltrative growth. The lining epithelial cells in VMC tend to be more cuboidal or flattened and the lesion appears less cellular than a BDA due to fewer ducts with more abundant stroma. Compared to BDAs, the CCs are larger, show more nuclear pleomorphism, and identification of mitotic figures and presence of infiltrative edges are the key differentiating features.
Intrahepatic mucinous biliary neoplasm hepatobiliary cystadenoma
Although intrahepatic biliary mucinous neoplasms (IHBN) are considered the most common cystic neoplasms of the liver, they are still rare, seen almost exclusively in women and very similar to their pancreatic counterparts. These lesions are usually multilocular and more common in the right lobe of liver., The cyst is typically lined by biliary or gastric foveolar-type epithelium, with focal intestinal or Paneth cell metaplasia. The characteristic feature is presence of ovarian-type stroma underlying the lining epithelium, which is similar to its pancreatic counterpart. The noninvasive lesions are classified into low or high grade based on the degree of cytologic atypia and architectural complexity of the lining epithelium. These lesions are considered premalignant with 25% rate of transformation to cystadenocarcinoma.,, Any invasive component implies a diagnosis of carcinoma arising in a IHBN. These lesions are large and cystic and rarely encountered in biopsies. In resections, when the entire cyst lining is replaced by malignant epithelium, cystic CC and metastatic adenocarcinoma with cystic degeneration need to be considered and excluded.,, Presence of ovarian-type stroma is often the key diagnostic feature that helps in the differentiation and sometimes may require IHC markers (estrogen [ER] and progesterone receptors [PRs], inhibin) for its identification in challenging cases.
Tumors that are poorly differentiated and of unclear lineage
Sometimes, the tumor cells may be very poorly differentiated or undifferentiated, without offering any histologic clue to the lineage differentiation or site of origin. Choosing the right IHC panel in such cases is the key in arriving at the right diagnosis or differential diagnosis.
In the diagnosis of metastatic tumors, a twofold approach is again applicable. First, excluding primary hepatic neoplasm (HCC or CC) and, second, determining the primary site in case of a metastasis. The hepatocytic lineage can be excluded from the differential diagnosis based on clinical history, tumor morphology, and immunostains. Negative hepatocyte markers such as Arginase-1, Hepar-1, and GPC-3, and positive MOC31 are useful to rule out a diagnosis of HCC. MOC31 is a reliable marker in making this distinction as MOC31 is usually negative or only weakly expressed in HCC. Differentiating CC from metastasis can be a little more challenging. CCs are usually positive for CK7, CK19, and MOC31, but all of these markers are nonspecific. In fact, both CK7 and/or CK19 can be expressed in the subsets of HCC in addition to metastatic carcinomas from other body sites. Clinical and radiological information is useful to make a diagnosis.
Detailed discussion about approach to metastatic lesions in the liver is beyond the scope of this review and only a brief practical outline is presented here. Broadly speaking, metastatic lesions in the liver can be split into four broad categories based on morphological features as follows:
- Adenocarcinomas showing clear glandular or acinar pattern mimicking CC
- Lesions with epithelioid/polygonal/clear cells mimicking HCC
- Metastatic small cell/neuroendocrine tumors
- Poorly differentiated neoplasms with no specific pattern.
Adenocarcinomas showing clear glandular or acinar pattern mimicking cholangiocarcinoma
Although CC versus metastatic adenocarcinoma is the main consideration, in difficult cases with atypical staining patterns, it is worth keeping in mind that HCC can sometimes show extensive pseudo-acinar formation and be mistaken for adenocarcinoma. Negative hepatocytic markers are useful to rule it out.
It is virtually impossible to distinguish CC from metastatic adenocarcinoma, especially from the upper gastrointestinal, extrahepatic biliary tract, pancreas, and gall bladder. Those arising from the lower gastrointestinal tract can be distinguished from CC by positivity for CK20 and CDX2, but this is also not very specific [Figure 6]d.
|Figure 6: Metastatic lesions in liver. (a) Lung adenocarcinoma showing positivity with TTF-1 (inset), (b) Mammary origin carcinoma exhibiting positivity for GATA-3 (inset), (c) Metastatic carcinoid confirmed by diffuse positivity for chromogranin A (inset), (d) Metastatic adenocarcinoma from colon demonstrating CDX2 positivity (inset). (e) Renal clear cell carcinoma showing polygonal cells with clear cytoplasm. The tumor was positive for carbonic anhydrase and vimentin (inset) (f) Metastatic poorly differentiated neuroendocrine carcinoma of the small cell type. Inset shows diffuse positive for chromogranin|
Click here to view
A variety of IHC markers for mucin are available and have virtually replaced histochemical mucin staining in practice. MUC 1 is positive in the apical membranous aspect of a wide variety of adenocarcinomas that include breast, ovary, endometrium, lung, endocervix, bladder, kidney, majority of pancreaticobiliary carcinomas and CCs, numerous esophageal and gastric adenocarcinomas, and very rarely colonic carcinomas. Tumors consistently negative include adrenocortical carcinoma and HCC. MUC2 expression is suggestive of intestinal differentiation and is expressed largely in gastrointestinal tract carcinomas, including colon, stomach, and esophagus, with a cytoplasmic staining pattern; however, any other carcinoma with intestinal differentiation from other sites (e.g., ovary, lung, and urinary bladder) can express MUC2. MUC5AC expression is suggestive of gastric differentiation and its expression is seen in tumors of gastrointestinal, pancreaticobiliary, and endocervical origin. The mucin markers should not be interpreted in isolation and always be interpreted in conjunction with other IHC markers and morphology.
Commonly used IHC markers for recognizing metastatic lung adenocarcinoma include TTF1 [Figure 6]a and napsin A, which show similar sensitivity (75% each). Lung mucinous carcinomas are usually negative for napsin A and TTF1 (70%), but express CK 7 (94%), MUC2 (56%), and CK20 (48%). CDX2 is generally negative and can be helpful in differentiating these from a colonic primary.
Thyroid carcinomas can metastasize to the liver, though presentation as an unknown primary is rare. Thyroid tumors are generally CK7+/CK20 − and TTF1 and thyroglobulin positive.
To make a diagnosis of metastatic breast carcinoma, a panel that includes GCDFP15, mammaglobin, and GATA3, in addition to hormone receptors ER and PR, is most useful. GCDFP15 has a high specificity (95%) but a lower sensitivity (50%–74%). In contrast, mammaglobin, overexpressed in 48%–84% of breast carcinoma, is more sensitive, but less specific than GCDFP15. GATA3 is frequently expressed in breast and urothelial carcinomas with strong nuclear staining [Figure 6]b and is a very good marker for recognizing breast metastasis with appropriate morphology; however its expression, although moderate to weak, can be seen with cervical, lung, and anal squamous cell carcinomas.,
Female genital tract
PAX8 is one of the most useful markers to distinguish tumors of female genital tract origin from the rest. For diagnosis of an ovarian primary, a panel consisting of PAX8, CA125, and WT1, in addition to ER and PR, is useful. Though WT1 may be negative in endometrioid carcinoma, CA125 is still useful. ER and PR when expressed may also be useful; however, these are not specific by any means. Aberrant p53 expression is indicative of a serous carcinoma. Ovarian mucinous tumors of intestinal phenotype tend to express colonic markers such as CK20 and CDX2, along with mCEA. SMAD4/DPC4 loss, when seen, is indicative of a pancreaticobiliary origin.
A combination of prostate-specific antigen, prostate-specific acid phosphatase, and alpha-methylacyl-CoA racemase is used to make a diagnosis of metastatic prostate carcinoma.
Lesions with epithelioid/polygonal/clear cells mimicking hepatocellular carcinoma
HCC needs to be ruled out when the constituting cells are undifferentiated and polygonal, especially when arranged in a trabecular or nested pattern. This is especially true when the background liver is cirrhotic and serum AFP is elevated (seen in 60%–80% of cases). Morphological clues that the tumor may be of hepatocytic origin are presence of canalicular bile, Mallory hyaline-like inclusions, and steatosis. The major differential diagnosis for such tumors include the following:
Renal cell carcinoma
HCC with clear cell morphology can mimic RCC. Carbonic anhydrase, PAX2, and PAX8 are the best immunostains for distinguishing RCC from HCC [Figure 6]e. In addition to positivity for these markers, RCCs are negative for hepatocytic markers, except CD10. Within the category of renal cell carcinomas, there is a slight variation in the immunoprofile of different morphologic subtypes of renal cell carcinomas [Table 4] and it is important to be aware of these variations while interpreting the results.
Germ cell tumors
PLAP, OCT4, HCG, and CD30 are commonly used markers for germ cell tumors; HCG is positive in all nonseminomatous germ cell tumors and is negative in seminoma. Placental-like alkaline phosphatase (PLAP) is also very sensitive, but not very specific for germ cell tumors. Octamer-binding transcription factor 4 (OCT4) is expressed in 80%–100% of embryonal carcinomas and all classical seminomas (nuclear staining); it is not expressed in yolk sac tumors and choriocarcinomas. CD117 and SALL4 are expressed in seminomas and not in nonseminomatous germ cell tumors. CD30 is strongly expressed in embryonal carcinomas. GPC-3 is expressed in yolk sac tumors and is important to bear this in mind when HCC is in the differential diagnosis.,
Adrenal cortical carcinoma
Inhibin and Melan-A (A103) are good markers in this regard with good sensitivity (60%–100%). In addition, they stain for keratin AE1/3 and are negative for hepatocytic markers, except a subset which may express GPC-3.
The most useful melanocytic markers in practice include Melan A, HMB45, S100, and SOX10. Other markers include MITF (microphthalmia-induced transcription factor) and tyrosinase.
Squamous cell carcinoma
Positive markers confirming SCC are CK5/6, p63, and p40.
Epithelioid AML can mimic HCC, especially when vascular and lipomatous components are scant. Smooth muscle actin and positive melanocytic markers such as HMB45 and Melan A are useful in making this diagnosis.
This unique variant that very closely mimics HCC most commonly arises from the stomach, but can also arise at other sites in the gastrointestinal and urogenital tracts. Hepar-1 is uniformly positive in these tumors with variable staining with other hepatocytic markers. The differentiation from a true HCC is based on location of the tumor, presence of any associated adenocarcinoma component, and lack of a liver lesion. The diagnosis becomes almost impossible when they metastasize to the liver.
Bladder transitional cell carcinoma
GATA3, p63, and uroplakin are most useful in recognizing metastatic bladder transitional cell carcinomas.
Lesions with neuroendocrine differentiation
Neuroendocrine tumors can be distinguished from HCC by the expression of endocrine markers such as chromogranin A [Figure 6]c and [Figure 6]f and synaptophysin and negative hepatocellular markers. Most NETs in addition will have positive staining for CK19 and MOC31. Utility of CD56 is limited in this setting due to expression seen in HCC. It is important to note that Hepar-1 may be also be expressed in a small subset of NETs, so use of >1 hepatocytic markers is recommended. Use of transcription factors (CDX2, TTF1, NESP55, and PDX1) has been shown to be useful in predicting the possibly primary site in metastatic endocrine tumors.,
Poorly differentiated neoplasms with no specific pattern
This is a category that often requires a much wider IHC panel and not infrequently the site of origin remains unknown despite extensive workup. A careful review of clinical workup and presentation is required to form a clinically relevant differential diagnosis in this setting. Depending on the degree of differentiation, one may have to use markers to identify hematolymphoid, mesenchymal, epithelial, or germ cell nature of the tumor. In some cases, it may only be possible to suggest that the lesion is a carcinoma, which sometimes may be all that is necessary. While some centers have a standard broad panel of IHC markers that besides keratins (CK5/6, CK7, and CK20), endocrine markers (chromogranin and synaptophysin), and melanoma markers (SOX10 and HMB45) also includes markers for breast and genitourinary tract (GATA3), lung (TTF1 and napsin), gastrointestinal tract (CDX2), and gynecologic tract (PAX8 and WT1). The number of markers that can be added to such a list is endless and is only limited by the amount of tissue present in the biopsy. There are no studies that suggest that use of a very broad panel has any advantage over more carefully selected panel based on clinical judgment; hence, in our practice, we follow the latter approach.
| Summary and Conclusion|| |
This was a whirlwind review of hepatic masses in adults, and a practical diagnostic approach is outlined in [Figure 7]. Though this algorithm is an oversimplification and in practice the diagnosis of hepatic lesions is more challenging, a simple approach can be useful in identification and diagnosis of many, if not most cases. Furthermore, the spectrum of entities discussed in this review is still not all inclusive and other tumor types such as soft-tissue neoplasms, hematolymphoid tumors, heterotopias, hamartomas, and germ cell tumors can involve the liver and must be kept in mind. Clinical features such as patient age, gender, history of medication or chemical exposure, and background liver disease are very important in formulating the appropriate differential diagnosis and diagnosis. Careful histologic observations and directed, judicious use of IHC is the key to making the right diagnosis of hepatic masses. IHC is a very useful adjunct in making the diagnosis; however, it should never be interpreted in isolation.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Mitchell K, Jain D. Diagnostic approach to needle biopsies of hepatic mass lesions. Diagn Histopathol 2008;14:598-608.
Sherman M, Colombo M. Hepatocellular carcinoma screening and diagnosis. Semin Liver Dis 2014;34:389-97.
Tang A, Valasek MA, Sirlin CB. Update on the liver imaging reporting and data system: What the pathologist needs to know. Adv Anat Pathol 2015;22:314-22.
Schraml C, Kaufmann S, Rempp H, Syha R, Ketelsen D, Notohamiprodjo M, et al.
Imaging of HCC-current state of the art. Diagnostics (Basel) 2015;5:513-45.
Bruix J, Sherman M, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma: An update. Hepatology 2011;53:1020-2.
Forner A, Vilana R, Ayuso C, Bianchi L, Solé M, Ayuso JR, et al.
Diagnosis of hepatic nodules 20 mm or smaller in cirrhosis: Prospective validation of the noninvasive diagnostic criteria for hepatocellular carcinoma. Hepatology 2008;47:97-104.
Bruix J, Sherman M, Practice Guidelines Committee, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma. Hepatology 2005;42:1208-36.
de Lope CR, Tremosini S, Forner A, Reig M, Bruix J. Management of HCC. J Hepatol 2012;56 Suppl 1:S75-87.
Sherman M. Hepatocellular carcinoma: Epidemiology, surveillance, and diagnosis. Semin Liver Dis 2010;30:3-16.
Walther Z, Jain D. Molecular pathology of hepatic neoplasms: Classification and clinical significance. Patholog Res Int 2011;2011:403929.
International Consensus Group for Hepatocellular Neoplasia. The International Consensus Group for Hepatocellular Neoplasia. Pathologic diagnosis of early hepatocellular carcinoma: A report of the international consensus group for hepatocellular Neoplasia. Hepatology 2009;49:658-64.
Park YN. Update on precursor and early lesions of hepatocellular carcinomas. Arch Pathol Lab Med 2011;135:704-15.
Niu ZS, Niu XJ, Wang WH, Zhao J. Latest developments in precancerous lesions of hepatocellular carcinoma. World J Gastroenterol 2016;22:3305-14.
Park YN, Yang CP, Fernandez GJ, Cubukcu O, Thung SN, Theise ND, et al.
Neoangiogenesis and sinusoidal “capillarization” in dysplastic nodules of the liver. Am J Surg Pathol 1998;22:656-62.
Marginean EC, Gown AM, Jain D. Diagnostic approach to hepatic mass lesions and role of immunohistochemistry. Surg Pathol Clin 2013;6:333-65.
Roncalli M, Terracciano L, Di Tommaso L, David E, Colombo M; Gruppo Italiano Patologi Apparato Digerente (GIPAD), et al.
Liver precancerous lesions and hepatocellular carcinoma: The histology report. Dig Liver Dis 2011;43 Suppl 4:S361-72.
Kojiro M, Nakashima O. Histopathologic evaluation of hepatocellular carcinoma with special reference to small early stage tumors. Semin Liver Dis 1999;19:287-96.
Shafizadeh N, Kakar S. Hepatocellular carcinoma: Histologic subtypes. Surg Pathol Clin 2013;6:367-84.
Choi WT, Ramachandran R, Kakar S. Immunohistochemical approach for the diagnosis of a liver mass on small biopsy specimens. Hum Pathol 2017;63:1-3.
Park YN, Kojiro M, Di Tommaso L, Dhillon AP, Kondo F, Nakano M, et al.
Ductular reaction is helpful in defining early stromal invasion, small hepatocellular carcinomas, and dysplastic nodules. Cancer 2007;109:915-23.
Di Tommaso L, Roncalli M. Tissue biomarkers in hepatocellular tumors: Which, when, and how. Front Med (Lausanne) 2017;4:10.
Di Tommaso L, Destro A, Fabbris V, Spagnuolo G, Laura Fracanzani A, Fargion S, et al.
Diagnostic accuracy of clathrin heavy chain staining in a marker panel for the diagnosis of small hepatocellular carcinoma. Hepatology 2011;53:1549-57.
Perumpail RB, Wong RJ, Ahmed A, Harrison SA. Hepatocellular carcinoma in the setting of non-cirrhotic nonalcoholic fatty liver disease and the metabolic syndrome: US experience. Dig Dis Sci 2015;60:3142-8.
van Meer S, van Erpecum KJ, Sprengers D, Klümpen HJ, Jansen PL, Ijzermans JN, et al.
Hepatocellular carcinoma in noncirrhotic livers is associated with steatosis rather than steatohepatitis: Potential implications for pathogenesis. Eur J Gastroenterol Hepatol 2016;28:955-62.
Leung C, Yeoh SW, Patrick D, Ket S, Marion K, Gow P, et al.
Characteristics of hepatocellular carcinoma in cirrhotic and non-cirrhotic non-alcoholic fatty liver disease. World J Gastroenterol 2015;21:1189-96.
Torbenson M. Review of the clinicopathologic features of fibrolamellar carcinoma. Adv Anat Pathol 2007;14:217-23.
Klein WM, Molmenti EP, Colombani PM, Grover DS, Schwarz KB, Boitnott J, et al.
Primary liver carcinoma arising in people younger than 30 years. Am J Clin Pathol 2005;124:512-8.
Ross HM, Daniel HD, Vivekanandan P, Kannangai R, Yeh MM, Wu TT, et al.
Fibrolamellar carcinomas are positive for CD68. Mod Pathol 2011;24:390-5.
Graham RP, Jin L, Knutson DL, Kloft-Nelson SM, Greipp PT, Waldburger N, et al.
DNAJB1-PRKACA is specific for fibrolamellar carcinoma. Mod Pathol 2015;28:822-9.
Reznik Y, Dao T, Coutant R, Chiche L, Jeannot E, Clauin S, et al.
Hepatocyte nuclear factor-1 alpha gene inactivation: Cosegregation between liver adenomatosis and diabetes phenotypes in two maturity-onset diabetes of the young (MODY) 3 families. J Clin Endocrinol Metab 2004;89:1476-80.
Bioulac-Sage P, Balabaud C, Zucman-Rossi J. Subtype classification of hepatocellular adenoma. Dig Surg 2010;27:39-45.
Bioulac-Sage P, Laumonier H, Laurent C, Zucman-Rossi J, Balabaud C. Hepatocellular adenoma: What is new in 2008. Hepatol Int 2008;2:316-21.
Dhingra S, Fiel MI. Update on the new classification of hepatic adenomas: Clinical, molecular, and pathologic characteristics. Arch Pathol Lab Med 2014;138:1090-7.
Rebouissou S, Bioulac-Sage P, Zucman-Rossi J. Molecular pathogenesis of focal nodular hyperplasia and hepatocellular adenoma. J Hepatol 2008;48:163-70.
Foster JH, Berman MM. The malignant transformation of liver cell adenomas. Arch Surg 1994;129:712-7.
Bioulac-Sage P, Cubel G, Balabaud C. Pathological diagnosis of hepatocellular adenoma in clinical practice. Diagn Histopathol 2011;17:521-9.
Paradis V. Hepatocellular adenomas: WHO classification and immunohistochemical workup. Surg Pathol Clin 2013;6:311-31.
Bioulac-Sage P, Laumonier H, Couchy G, Le Bail B, Sa Cunha A, Rullier A, et al.
Hepatocellular adenoma management and phenotypic classification: The Bordeaux experience. Hepatology 2009;50:481-9.
Bioulac-Sage P, Rebouissou S, Thomas C, Blanc JF, Saric J, Sa Cunha A, et al.
Hepatocellular adenoma subtype classification using molecular markers and immunohistochemistry. Hepatology 2007;46:740-8.
Arnason T, Fleming KE, Wanless IR. Peritumoral hyperplasia of the liver: A response to portal vein invasion by hypervascular neoplasms. Histopathology 2013;62:458-64.
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.
Bellanné-Chantelot C, Carette C, Riveline JP, Valéro R, Gautier JF, Larger E, et al.
The type and the position of HNF1A mutation modulate age at diagnosis of diabetes in patients with maturity-onset diabetes of the young (MODY)-3. Diabetes 2008;57:503-8.
Bioulac-Sage P, Balabaud C, Bedossa P, Scoazec JY, Chiche L, Dhillon AP, et al.
Pathological diagnosis of liver cell adenoma and focal nodular hyperplasia: Bordeaux update. J Hepatol 2007;46:521-7.
Bacq Y, Jacquemin E, Balabaud C, Jeannot E, Scotto B, Branchereau S, et al.
Familial liver adenomatosis associated with hepatocyte nuclear factor 1alpha inactivation. Gastroenterology 2003;125:1470-5.
Cho SJ, Ferrell LD, Gill RM. Expression of liver fatty acid binding protein in hepatocellular carcinoma. Hum Pathol 2016;50:135-9.
Rebouissou S, Franconi A, Calderaro J, Letouzé E, Imbeaud S, Pilati C, et al.
Genotype-phenotype correlation of CTNNB1 mutations reveals different ß-catenin activity associated with liver tumor progression. Hepatology 2016;64:2047-61.
Bioulac-Sage P, Rebouissou S, Sa Cunha A, Jeannot E, Lepreux S, Blanc JF, et al.
Clinical, morphologic, and molecular features defining so-called telangiectatic focal nodular hyperplasias of the liver. Gastroenterology 2005;128:1211-8.
Paradis V, Benzekri A, Dargère D, Bièche I, Laurendeau I, Vilgrain V, et al.
Telangiectatic focal nodular hyperplasia: A variant of hepatocellular adenoma. Gastroenterology 2004;126:1323-9.
Nault JC, Couchy G, Balabaud C, Morcrette G, Caruso S, Blanc JF, et al.
Molecular classification of hepatocellular adenoma associates With risk factors, bleeding, and malignant transformation. Gastroenterology 2017;152:880-94.e6.
Henriet E, Abou Hammoud A, Dupuy JW, Dartigues B, Ezzoukry Z, Dugot-Senant N, et al.
Arginosuccinate synthase 1 (ASS1): A marker of unclassified hepatocellular adenoma and high bleeding risk. Hepatology 2017;66:2016-28.
Fujiwara M, Kwok S, Yano H, Pai RK. Arginase-1 is a more sensitive marker of hepatic differentiation than hepPar-1 and glypican-3 in fine-needle aspiration biopsies. Cancer Cytopathol 2012;120:230-7.
Kakar S, Gown AM, Goodman ZD, Ferrell LD. Best practices in diagnostic immunohistochemistry: Hepatocellular carcinoma versus metastatic neoplasms. Arch Pathol Lab Med 2007;131:1648-54.
Kong CS, Appenzeller M, Ferrell LD. Utility of CD34 reactivity in evaluating focal nodular hepatocellular lesions sampled by fine needle aspiration biopsy. Acta Cytol 2000;44:218-22.
Balabaud C, Bioulac-Sage P, Ferrell L, Kakar S, Paradis V, Quaglia A, et al.
Well-differentiated hepatocellular neoplasm of uncertain malignant potential. Hum Pathol 2015;46:634-5.
An SL, Wang LM, Rong WQ, Wu F, Sun W, Yu WB, et al.
Hepatocellular adenoma with malignant transformation in male patients with non-cirrhotic livers. Chin J Cancer 2015;34:217-24.
Roncalli M, Sciarra A, Tommaso LD. Benign hepatocellular nodules of healthy liver: Focal nodular hyperplasia and hepatocellular adenoma. Clin Mol Hepatol 2016;22:199-211.
Libbrecht L, Cassiman D, Verslype C, Maleux G, Van Hees D, Pirenne J, et al.
Clinicopathological features of focal nodular hyperplasia-like nodules in 130 cirrhotic explant livers. Am J Gastroenterol 2006;101:2341-6.
Quaglia A, Tibballs J, Grasso A, Prasad N, Nozza P, Davies SE, et al.
Focal nodular hyperplasia-like areas in cirrhosis. Histopathology 2003;42:14-21.
Bioulac-Sage P, Balabaud C, Zucman-Rossi J. Focal nodular hyperplasia, hepatocellular adenomas: Past, present, future. Gastroenterol Clin Biol 2010;34:355-8.
Altavilla G, Guariso G. Focal nodular hyperplasia of the liver associated with portal vein agenesis: A morphological and immunohistochemical study of one case and review of the literature. Adv Clin Path 1999;3:139-45.
Buscarini E, Danesino C, Plauchu H, de Fazio C, Olivieri C, Brambilla G, et al.
High prevalence of hepatic focal nodular hyperplasia in subjects with hereditary hemorrhagic telangiectasia. Ultrasound Med Biol 2004;30:1089-97.
De Gaetano AM, Gui B, Macis G, Manfredi R, Di Stasi C. Congenital absence of the portal vein associated with focal nodular hyperplasia in the liver in an adult woman: Imaging and review of the literature. Abdom Imaging 2004;29:455-9.
Paradis V, Laurent A, Flejou JF, Vidaud M, Bedossa P. Evidence for the polyclonal nature of focal nodular hyperplasia of the liver by the study of X-chromosome inactivation. Hepatology 1997;26:891-5.
Bioulac-Sage P, Laumonier H, Rullier A, Cubel G, Laurent C, Zucman-Rossi J, et al.
Over-expression of glutamine synthetase in focal nodular hyperplasia: A novel easy diagnostic tool in surgical pathology. Liver Int 2009;29:459-65.
Joseph NM, Ferrell LD, Jain D, Torbenson MS, Wu TT, Yeh MM, et al.
Diagnostic utility and limitations of glutamine synthetase and serum amyloid-associated protein immunohistochemistry in the distinction of focal nodular hyperplasia and inflammatory hepatocellular adenoma. Mod Pathol 2014;27:62-72.
Ahmad I, Iyer A, Marginean CE, Yeh MM, Ferrell L, Qin L, et al.
Diagnostic use of cytokeratins, CD34, and neuronal cell adhesion molecule staining in focal nodular hyperplasia and hepatic adenoma. Hum Pathol 2009;40:726-34.
Iyer A, Robert ME, Bifulco CB, Salem RR, Jain D. Different cytokeratin and neuronal cell adhesion molecule staining patterns in focal nodular hyperplasia and hepatic adenoma and their significance. Hum Pathol 2008;39:1370-7.
Nakanuma Y, Sripa B, Vatanasapt B, Leong A, Ponchon T, Ishak K. Intrahepatic cholangiocarcinoma. In: Hamilton SR, Aaltonen LA, editors. WHO Classification of Tumors, Pathology and Genetics Tumors of the Digestive System. Lyon: IARC Press; 2000. p. 173-80.
Aljiffry M, Walsh MJ, Molinari M. Advances in diagnosis, treatment and palliation of cholangiocarcinoma: 1990-2009. World J Gastroenterol 2009;15:4240-62.
Bosch FX, Ribes J, Díaz M, Cléries R. Primary liver cancer: Worldwide incidence and trends. Gastroenterology 2004;127:S5-16.
Poomphakwaen K, Promthet S, Kamsa-Ard S, Vatanasapt P, Chaveepojnkamjorn W, Klaewkla J, et al.
Risk factors for cholangiocarcinoma in Khon Kaen, Thailand: A nested case-control study. Asian Pac J Cancer Prev 2009;10:251-8.
Gatto M, Bragazzi MC, Semeraro R, Napoli C, Gentile R, Torrice A, et al.
Cholangiocarcinoma: Update and future perspectives. Dig Liver Dis 2010;42:253-60.
Ryu HS, Lee K, Shin E, Kim SH, Jing J, Jung HY, et al.
Comparative analysis of immunohistochemical markers for differential diagnosis of hepatocellular carcinoma and cholangiocarcinoma. Tumori 2012;98:478-84.
Suriawinata AA, Thung SN. Malignant liver tumors. Clin Liver Dis 2002;6:527-54, ix.
Chu PG, Arber DA, Weiss LM. Expression of T/NK-cell and plasma cell antigens in nonhematopoietic epithelioid neoplasms. An immunohistochemical study of 447 cases. Am J Clin Pathol 2003;120:64-70.
Werling RW, Yaziji H, Bacchi CE, Gown AM. CDX2, a highly sensitive and specific marker of adenocarcinomas of intestinal origin: An immunohistochemical survey of 476 primary and metastatic carcinomas. Am J Surg Pathol 2003;27:303-10.
Karabork A, Kaygusuz G, Ekinci C. The best immunohistochemical panel for differentiating hepatocellular carcinoma from metastatic adenocarcinoma. Pathol Res Pract 2010;206:572-7.
Zhang F, Chen XP, Zhang W, Dong HH, Xiang S, Zhang WG, et al.
Combined hepatocellular cholangiocarcinoma originating from hepatic progenitor cells: Immunohistochemical and double-fluorescence immunostaining evidence. Histopathology 2008;52:224-32.
Acosta AM, Wiley EL. Primary biliary mixed adenoneuroendocrine carcinoma (MANEC): A Short review. Arch Pathol Lab Med 2016;140:1157-62.
Brunt EM. Benign tumors of the liver. Clin Liver Dis 2001;5:1-15, v.
Allaire GS, Rabin L, Ishak KG, Sesterhenn IA. Bile duct adenoma. A study of 152 cases. Am J Surg Pathol 1988;12:708-15.
Albores-Saavedra J, Hoang MP, Murakata LA, Sinkre P, Yaziji H. Atypical bile duct adenoma, clear cell type: A previously undescribed tumor of the liver. Am J Surg Pathol 2001;25:956-60.
Arena V, Arena E, Stigliano E, Capelli A. Bile duct adenoma with oncocytic features. Histopathology 2006;49:318-20.
Bhathal PS, Hughes NR, Goodman ZD. The so-called bile duct adenoma is a peribiliary gland hamartoma. Am J Surg Pathol 1996;20:858-64.
Mortele KJ, Ros PR. Benign liver neoplasms. Clin Liver Dis 2002;6:119-45.
Blonski W, Reddy KR. Evaluation of nonmalignant liver masses. Curr Gastroenterol Rep 2006;8:38-45.
Devaney K, Goodman ZD, Ishak KG. Hepatobiliary cystadenoma and cystadenocarcinoma. A light microscopic and immunohistochemical study of 70 patients. Am J Surg Pathol 1994;18:1078-91.
Yu FC, Chen JH, Yang KC, Wu CC, Chou YY. Hepatobiliary cystadenoma: A report of two cases. J Gastrointestin Liver Dis 2008;17:203-6.
Hiatt JR, Furnas H, Tompkins RK. Serous adenocarcinoma of the ovary presenting as a large liver cyst. West J Med 1986;144:609-10.
Lee JH, Kim KS, Chung CW, Park YN, Kim BR. Hepatic resection of metastatic tumor from serous cystadenocarcinoma of the ovary. J Korean Med Sci 2002;17:415-8.
Tang Y, Yamashita Y, Ogata I, Namimoto T, Abe Y, Urata J, et al.
Metastatic liver tumor from cystic ovarian carcinomas: CT and MRI appearance. Radiat Med 1999;17:265-70.
Chan ES, Yeh MM. The use of immunohistochemistry in liver tumors. Clin Liver Dis 2010;14:687-703.
Lau SK, Weiss LM, Chu PG. Differential expression of MUC1, MUC2, and MUC5AC in carcinomas of various sites: An immunohistochemical study. Am J Clin Pathol 2004;122:61-9.
Ordóñez NG. Napsin A expression in lung and kidney neoplasia: A review and update. Adv Anat Pathol 2012;19:66-73.
Inamura K, Satoh Y, Okumura S, Nakagawa K, Tsuchiya E, Fukayama M, et al.
Pulmonary adenocarcinomas with enteric differentiation: Histologic and immunohistochemical characteristics compared with metastatic colorectal cancers and usual pulmonary adenocarcinomas. Am J Surg Pathol 2005;29:660-5.
Saad RS, Cho P, Silverman JF, Liu Y. Usefulness of Cdx2 in separating mucinous bronchioloalveolar adenocarcinoma of the lung from metastatic mucinous colorectal adenocarcinoma. Am J Clin Pathol 2004;122:421-7.
Fischer S, Asa SL. Application of immunohistochemistry to thyroid neoplasms. Arch Pathol Lab Med 2008;132:359-72.
Mehra R, Varambally S, Ding L, Shen R, Sabel MS, Ghosh D, et al.
Identification of GATA3 as a breast cancer prognostic marker by global gene expression meta-analysis. Cancer Res 2005;65:11259-64.
Yang M, Nonaka D. A study of immunohistochemical differential expression in pulmonary and mammary carcinomas. Mod Pathol 2010;23:654-61.
Mittal K, Soslow R, McCluggage WG. Application of immunohistochemistry to gynecologic pathology. Arch Pathol Lab Med 2008;132:402-23.
Buza N, Hui P. Immunohistochemistry in gynecologic pathology: An example-based practical update. Arch Pathol Lab Med 2017;141:1052-71.
Epstein JI, Egevad L, Humphrey PA, Montironi R; Members of the ISUP Immunohistochemistry in Diagnostic Urologic Pathology Group. Best practices recommendations in the application of immunohistochemistry in the prostate: Report from the international society of urologic pathology consensus conference. Am J Surg Pathol 2014;38:e6-e19.
Alshenawy HA. Immunohistochemical panel for differentiating renal cell carcinoma with clear and papillary features. Pathol Oncol Res 2015;21:893-9.
Ulbright TM, Tickoo SK, Berney DM, Srigley JR, Members of the ISUP Immunohistochemistry in Diagnostic Urologic Pathology Group. Best practices recommendations in the application of immunohistochemistry in testicular tumors: Report from the International Society of Urological Pathology Consensus Conference. Am J Surg Pathol 2014;38:e50-9.
Amin MB, Trpkov K, Lopez-Beltran A, Grignon D, Members of the ISUP Immunohistochemistry in Diagnostic Urologic Pathology Group. Best practices recommendations in the application of immunohistochemistry in the bladder lesions: Report from the international society of urologic pathology consensus conference. Am J Surg Pathol 2014;38:e20-34.
Chan ES, Alexander J, Swanson PE, Jain D, Yeh MM. PDX-1, CDX-2, TTF-1, and CK7: A reliable immunohistochemical panel for pancreatic neuroendocrine neoplasms. Am J Surg Pathol 2012;36:737-43.
Srivastava A, Hornick JL. Immunohistochemical staining for CDX-2, PDX-1, NESP-55, and TTF-1 can help distinguish gastrointestinal carcinoid tumors from pancreatic endocrine and pulmonary carcinoid tumors. Am J Surg Pathol 2009;33:626-32.
Wang F, Jing X, Wang T, Li G, Li T, Zhang Q, et al.
Differential diagnostic value of GPC3-CD34 combined staining in small liver nodules with diameter less than 3 cm. Am J Clin Pathol 2012;137:937-45.
Department of Pathology, Yale University School of Medicine, 310 Cedar Street, New Haven, CT-06510
Source of Support: None, Conflict of Interest: None
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]
[Table 1], [Table 2], [Table 3], [Table 4]