Accepted Manuscript
Immunohistochemical Approach for the Diagnosis of a Liver Mass on Small Biopsy Specimens
Won-Tak Choi, Rageshree Ramachandran, Sanjay Kakar
| PII: | S0046-8177(17)30014-X |
| DOI: | doi: 10.1016/j.humpath.2016.12.025 |
| Reference: | YHUPA 4107 |
| To appear in: | Human Pathology |
| Received date: | 21 September 2016 |
| Revised date: | 18 December 2016 |
| Accepted date: | 28 December 2016 |
Human PATHOLOGY
W: B. Saunders
Please cite this article as: Choi Won-Tak, Ramachandran Rageshree, Kakar Sanjay, Immunohistochemical Approach for the Diagnosis of a Liver Mass on Small Biopsy Spec- imens, Human Pathology (2017), doi: 10.1016/j.humpath.2016.12.025
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Immunohistochemical Approach for the Diagnosis of a Liver Mass on Small Biopsy Specimens
Won-Tak Choi, MD, PhDt,*, Rageshree Ramachandran, MD, PhDt, and Sanjay Kakar, MD+
*Department of Pathology,
University of California at San Francisco, San Francisco, CA 94143
Running Title: Immunohistochemical Approach for the Diagnosis of a Liver Mass
Conflicts of Interest and Source of Funding: None Declared
* Corresponding author:
Won-Tak Choi, MD, PhD
University of California at San Francisco, Department of Pathology, 505 Parnassus Avenue, M552, Box 0102, San Francisco, CA 94143 Tel: 415-353-9533; Fax: 415-353-1200
E-mail address: Won-Tak.Choi@ucsf.edu
ACCEPTED MANUSCRIPT
ABSTRACT
Well-differentiated hepatocellular carcinoma (HCC) shares overlapping histological features with benign hepatocellular lesions, including hepatocellular adenoma (HCA) and focal nodular hyperplasia (FNH) in non-cirrhotic liver, and with high-grade dysplastic nodule (HGDN) in cirrhotic liver. Several metastatic tumors, such as neuroendocrine tumor, renal cell carcinoma, adrenocortical carcinoma, melanoma, and epithelioid angiomyolipoma, can be indistinguishable from HCC on histologic grounds. Since this distinction has important therapeutic implications, judicious use of immunohistochemical markers plays an important role in establishing an accurate diagnosis, especially when limited material of tumor is available on cell block or a small core biopsy. This review describes commonly used immunohistochemical markers used a in the diagnosis of HCC, highlighting advantages and disadvantages of each marker, and suggests appropriate immunohistochemical panels for specific clinicopathologic situations.
KEYWORDS - Hepatocellular carcinoma; hepatocellular adenoma; focal nodular hyperplasia; well-differentiated hepatocellular lesion; immunohistochemistry
ACCEPTED MANUSCRIPT
1. INTRODUCTION
The distinction of hepatocellular carcinoma (HCC) from intrahepatic cholangiocarcinoma or metastatic tumor can be challenging on cell block or core biopsy specimens. This review summarizes the challenges in the use of immunohistochemistry for the diagnosis of HCC and suggests specific panels depending on the clinical setting. The review is divided into five parts:
(A) Advantages and pitfalls of commonly used antibodies.
(B) Selection of antibody panel for diagnosis of HCC.
(C) HCC versus other polygonal cell tumors.
(D) Immunohistochemistry in histologic variants of HCC and other related settings.
(E) HCC versus benign hepatocellular lesions.
2. ADVANTAGES AND PITFALLS OF COMMONLY USED ANTIBODIES
2.1. Commonly Used Markers for Hepatocellular Carcinoma
Hepatocyte Paraffin-1 (Hep Par-1): Hep Par-1 is a monoclonal antibody developed using formalin-fixed tissue from failed liver allografts [1]. It recognizes hepatocyte specific antigen, which is now known to be carbamoyl phosphate synthetase I, a urea cycle enzyme [2]. Hep Par-1 is a useful marker for diagnosis of HCC with sensitivity and specificity exceeding 80% [3-8]. It shows a diffuse cytoplasmic granular staining pattern, which is easy to interpret. Most adenocarcinomas (including pancreaticobiliary, colorectal, breast, urothelial, prostatic) are negative or show focal
ACCEPTED MANUSCRIPT
weak positivity [1]. Similarly, most tumors that morphologically mimic HCC (referred to as “polygonal cell tumors” in this review), such as neuroendocrine tumor (NET), renal cell carcinoma (RCC), adrenocortical carcinoma (ACC), melanoma, and epithelioid angiomyolipoma (AML), are negative or only focally positive [1]. However, Hep Par-1 cannot distinguish neoplastic hepatocellular lesions from benign liver parenchyma [1]. Hep Par-1 has low sensitivity for poorly-differentiated HCC (50-60%) and scirrhous HCC (20-30%) [8, 9]. While most adenocarcinomas show negative staining for Hep Par- 1, strong positivity for Hep Par-1 can be seen in a subset of adenocarcinomas (including gastric, esophageal, pulmonary), and less commonly in intrahepatic cholangiocarcinoma and pancreatic adenocarcinomas. Hepatoid carcinomas are usually positive for Hep Par-1 [1,6,7,9,10].
Arginase-1 (Arg-1): Arg-1 is a binuclear manganese metalloenzyme that hydrolyzes arginine to ornithine and urea as a component of the urea cycle, and it is found only in liver [8]. Arg-1 is the most sensitive immunohistochemical marker (> 90%) for HCC (Figures 1A-C), including poorly-differentiated HCC and scirrhous HCC [8, 9]. It is easy to interpret on small biopsies, as it displays diffuse cytoplasmic staining with variable nuclear reactivity [8] (Figure 1C). Arg-1 has also proved to be the most specific marker (> 90%) for HCC, as most other tumors are negative [11]. However, weak or focal positivity has been observed in rare cases of pancreatic, colorectal, breast, prostatic, and biliary adenocarcinomas. Majority of the hepatoid carcinomas are positive for Arg-1 [10, 12, 13].
Glypican-3 (GPC-3): GPC-3 is a membrane-anchored heparan sulfate proteoglycan normally expressed in fetal liver but not in adult liver [14, 15]. Since GPC-
ACCEPTED MANUSCRIPT
3 is an oncofetal antigen, it can be expressed in a wide variety of tumors, including primary melanoma, nonseminomatous germ cell tumors (such as yolk sac tumor and choriocarcinoma), adenocarcinoma (including stomach and intrahepatic cholangiocarcinoma), and squamous carcinomas (lung, larynx, cervix), and in around 5% of intrahepatic cholangiocarcinoma [9, 16]. Although GPC-3 is not a specific hepatocellular marker, its combined use with Arg-1 can be very helpful for the diagnosis of poorly-differentiated HCC and scirrhous HCC due to it high sensitivity (both > 80%) in these settings [8, 9, 17]. GPC-3 is expressed in HCC in a diffuse cytoplasmic, membranous, or Golgi pattern, but it is not expressed in normal liver or benign hepatocellular lesions such as hepatocellular adenoma (HCA) and focal nodular hyperplasia (FNH) [16-21]. The sensitivity for well-differentiated HCC is relatively low at 50-60% [8, 17]. GPC-3 can occasionally stain cirrhotic nodules and regions of active inflammation in non-neoplastic liver [17, 22], which can be mistaken for HCC on small biopsies.
Polyclonal Carcinoembryonic Antigen (pCEA): CEA is a glycoprotein found in the glycocalyx of fetal epithelial cells and in small amounts in normal adult cells [8]. Most adenocarcinomas show diffuse cytoplasmic, membranous, and/or luminal expression. The polyclonal antibody cross reacts with biliary glycoprotein and yields a distinct canalicular staining pattern which is thought to be specific for HCC (Figure 1D) [5, 8, 23, 24]. Like Hep Par-1, the sensitivity of pCEA is high for well- and moderately- differentiated HCC (> 80%) [8]. However, the staining can be difficult to interpret in some cases and can be confused with the luminal or membranous staining seen in adenocarcinoma. Diffuse cytoplasmic staining in addition to the canalicular pattern can
ACCEPTED MANUSCRIPT
be observed in up to half of HCCs, further making the interpretation difficult [1]. Like Hep Par-1, the sensitivity of pCEA is low in poorly-differentiated HCC (54%) and scirrhous HCC (37%) [8, 9]. Monoclonal antibodies to CEA (mCEA) do not react against HCC, and their sensitivity for adenocarcinoma is low (60%).
Other Markers: Alpha-fetoprotein (AFP) is an oncofetal protein produced by liver and yolk sac visceral endoderm [1]. Although it is a marker of hepatocellular differentiation, germ cell tumors (such as hepatoid yolk sac tumor) can express this protein [1]. AFP has a low sensitivity of 30-50% for HCC, and its staining tends to be patchy with high background staining [4, 7, 23, 25]. As such, we do not recommend an AFP stain in diagnosing HCC. CD10 and villin show a distinct canalicular staining pattern (Figure 1D) [5, 23]. A similar pattern is seen with bile salt export protein (BSEP), an ATP-binding cassette (ABC) transporter expressed exclusively on the hepatocyte canalicular membrane (Figure 1D) [26]. Due to low sensitivity and availability of better alternatives, the use of these antibodies is not indicated in most clinical settings [8, 10, 23, 25, 27].
Heat shock protein (HSP)-70, an anti-apoptotic protein, is overexpressed in up to 80% of early HCCs in resection specimens, and in around 50% of biopsy specimens [28]. On the other hand, positive staining is seen in 5-10% of high-grade dysplastic nodules (HGDN) [29]. The combined use of HSP-70, glutamine synthetase (GS), and GPC-3 has been advocated for distinction of well-differentiated HCC from HCA and HGDN [30, 31]. Although HSP-70 can be useful in a small minority of cases, it has low overall sensitivity and its routine use is not necessary. GS is discussed in detail in the HCA section.
ACCEPTED MANUSCRIPT
Since albumin is synthesized only in the liver, in situ hybridization (ISH) to detect albumin mRNA is specific for hepatocellular differentiation [32]. Recently, a branched chain albumin ISH assay has been described which may have higher sensitivity for HCC, and is also positive in intrahepatic cholangiocarcinoma [33, 34]. Albumin ISH has high sensitivity (> 90%) for HCC and is negative in most metastatic adenocarcinomas. However, this assay has not been widely used and has limited availability.
Normal liver sinusoids do not stain for CD34, except for patchy staining in the periportal region. Arterialization of the blood supply in benign lesions (HCA and FNH) as well as HCC leads to sinusoidal positivity with CD34, a phenomenon referred to as “capillarization” of sinusoids [35, 36]. This phenomenon can be very useful in distinguishing lesional hepatocellular tissue from the non-lesional liver parenchyma in limited biopsy specimens. However, CD34 has limited utility in distinguishing HCC from benign hepatocellular lesions or non-hepatocellular tumors.
2.2. Commonly Used Markers for Adenocarcinoma
Cytokeratins (CK7, CK19, CK20): The majority of HCCs are negative for CK7 and CK19 [37], but one or both markers can be positive in 10-20% of cases, especially in poorly-differentiated HCC. CK19 expression is associated with a poor outcome [38]. By contrast, the majority of cholangiocarcinomas express CK7 (97%) and CK19 (77%) [37]. Similarly, up to 15% of HCCs can aberrantly express CK20, which can represent a potential diagnostic pitfall in a liver biopsy, particularly when imaging studies are not typical of HCC [39, 40].
ACCEPTED MANUSCRIPT
MOC-31: This antibody recognizes epithelial cell adhesion molecule (EpCAM) and shows diffuse membranous staining in more than 90% of cholangiocarcinomas and metastatic adenocarcinomas (including gastric, pancreatic, colorectal, lung, breast, and ovarian primaries), as well as majority of urothelial carcinomas, NETs, and RCCs [5, 41- 44]. Although the majority of HCCs are negative, MOC-31 positivity has been observed in 10-20% of HCCs [5, 41, 42].
Other Epithelial Markers: Ber-EP4 (epithelial-specific membrane antigen) and B72.3 (tumor-associated glycoprotein-72) are less useful due to positive staining in a subset of HCCs (up to 20-30%, respectively) [5, 23, 24].
3. SELECTION OF ANTIBODY PANEL FOR DIAGNOSIS
If hepatocellular differentiation is obvious based on histologic features (e.g., bile production), immunohistochemistry is not necessary for diagnosis, especially in the setting of cirrhosis. In most other situations especially in non-cirrhotic liver, a panel of 4 antibodies is recommended, including two hepatocellular markers (preferably Arg-1 and GPC-3) and two markers that are more commonly seen in adenocarcinoma (such as CK19 and MOC-31). If limited tissue is available, a two-stain approach using Arg-1 and CK19 can be very useful for initial evaluation. The high sensitivity and specificity of Arg-1 makes it the marker of first choice for demonstrating hepatocellular differentiation [8]. Arg-1 is known to maintain high sensitivity (81-84%) in cytology specimens [12, 45]. Based on the results of these two stains, cases fall into one of the following four
ACCEPTED MANUSCRIPT
categories, and further workup will be dictated in a case-dependent manner based on the overall clinical setting (Table 1).
Group 1: Arg-1 positive, CK19 negative: In most cases, this pattern establishes the diagnosis of HCC (Table 1). However, if morphologic features are not typical, staining patterns are weak or focal, or the clinical and imaging data are discordant, additional hepatocellular markers (Hep Par-1 and GPC-3) may be necessary to confirm the diagnosis.
Group 2: Arg-1 negative, CK19 positive: With this profile, a non-hepatocellular tumor is more likely, and the possibilities include cholangiocarcinoma, metastatic adenocarcinoma, and other polygonal cell tumors (Table 1). Additional immunohistochemistry should be chosen based on morphology and clinical setting, including (1) CK20 and/or CDX2 (caudal-related homeobox transcription factor 2) for colon and rectum; (2) CK7, TTF (thyroid transcription factor)-1, and/or napsin A for lung; (3) PSA (prostate-specific antigen), p501s (prostein), PAP (prostatic acid phosphatase), and/or NKX3.1 for prostate; (4) ER (estrogen receptor), mammaglobin, GATA (GATA-binding protein)-3, and/or GCDFP (gross cystic disease fluid protein) for breast; (5) CK7, ER, and/or WT (Wilms tumor)-1 for ovary; (6) CK7, TTF-1, and/or thyroglobulin for thyroid; (7) DPC (deleted in pancreatic carcinoma)-4 loss for pancreas; (8) PAX (paired box gene)-2, PAX-8, and/or RCC for RCC; and (9) synaptophysin and/or chromogranin for NET. An upfront wide panel using these markers is strongly discouraged in biopsy specimens as it often exhausts the tissue and does not permit use of additional more pertinent markers. Approximately 10-20% of HCCs can be positive for CK19 [38], but only a small fraction of CK19-positive HCCs will be Arg-1 negative. If
ACCEPTED MANUSCRIPT
HCC is considered likely based on morphology or clinical setting, additional hepatocellular markers (Hep Par-1 and GPC-3) can be considered.
Group 3: Arg-1 positive, CK19 positive: In most instances, this phenotype represents CK19-positive HCC. Liver metastasis from a hepatoid carcinoma in the stomach or pancreas can be virtually indistinguishable from HCC morphologically and immunophenotypically [46, 47]. Rarely, metastatic adenocarcinoma and intrahepatic cholangiocarcinoma may show aberrant Arg-1 staining, but the staining is weak or focal [10, 12, 13]. Additional hepatocellular markers (Hep Par-1 and GPC-3), cytokeratins (CK7, CK19, CK20), and/or site-specific markers (depending on history) are necessary for these cases.
Group 4: Arg-1 negative, CK19 negative: The differential diagnosis is wide in this category and can be divided into two groups based on pancytokeratin staining: (1) pancytokeratin-positive: Arg-1-negative HCC, CK19-negative adenocarcinoma, NET, RCC, and other carcinomas like squamous cell carcinoma and urothelial carcinoma; (2) pancytokeratin-negative: ACC, melanoma, AML, epithelioid gastrointestinal stromal tumor (GIST), or sarcomas with epithelioid morphology (Table 1). A broader work-up with additional hepatocellular markers (Hep Par-1 and GPC-3) and site-specific markers can be obtained for the pancytokeratin-positive group.
4. HCC VERSUS OTHER POLYGONAL CELL TUMORS
ACCEPTED MANUSCRIPT
The term ‘polygonal cell tumors’ refers to tumors that morphologically resemble HCC and includes epithelial neoplasms like RCC and NET as well as non-epithelial tumors like melanoma, ACC, AML, and sarcomas with epithelioid morphology.
HCC versus RCC: HCC can have clear cells and mimic clear cell RCC (CC- RCC) on morphology. This distinction is easy by immunohistochemistry as clear cell HCC expresses hepatocellular markers similar to classical HCC [7, 48]. The transcription factors, PAX-2 and PAX-8, are positive in 70-80% of CC-RCC, but not in HCC [49, 50] (Table 1). The combined use of Arg-1 and PAX-2/PAX-8 can distinguish clear cell HCC from CC-RCC in the majority of cases. RCC antigen is less useful due to its low sensitivity (20%) for metastatic CC-RCC, and positive results in non-renal tumors, including breast carcinoma, parathyroid adenoma, and embryonal carcinoma [51]. CD10 shows a distinct canalicular pattern in HCC [5, 23] which may be confused with the membrane staining in RCC-CC, making interpretation difficult.
HCC versus NET: The trabecular growth pattern and moderate to abundant eosinophilic cytoplasm in NET can mimic HCC. Primary hepatic NETs are extremely rare [52]. In addition to synaptophysin and chromogranin, most NETs are positive for CK19 and MOC-31 (Table 1), while hepatocellular markers like Arg-1 are negative [1]. Aberrant staining with Hep Par-1 can be seen in NET. Some HCCs can show focal staining with neuroendocrine markers [53], usually synaptophysin, but most cases are negative or weakly positive. CD56 is not helpful as a neuroendocrine marker, as it can be positive in HCC [1].
HCC versus Melanoma: Metastatic epithelioid melanoma may lack melanin pigment and can closely mimic HCC. GPC-3 can be positive in a small subset of
ACCEPTED MANUSCRIPT
melanomas [54]. Using the two-stain approach, melanoma belongs to the Arg-1- negative/CK19-negative group, and is pancytokeratin negative (Table 1). Melanocytic markers, such as SOX-10 (SRY-related HMG-box 10), S-100, HMB (human melanoma black)-45, and/or Melan-A, can establish the diagnosis.
HCC versus ACC: Typically, ACCs are negative or focally positive for pancytokeratin (Table 1). Most ACCs express inhibin (75-86%) and Melan-A (50-96%) [55, 56], while hepatocellular markers (including Arg-1 and Hep Par-1) are negative. Positive results with GPC-3 can be seen in 14% of ACCs [16], and rare cases of HCC, especially high-grade tumors, can be inhibin-positive [55].
HCC versus AML: Epithelioid AML can be monotypic and may lack lipomatous and angiomatous components, making it difficult to distinguish from HCC. Negative results with pancytokeratin, hepatocellular markers, CK19, and S100, coupled with co- expression of smooth muscle actin and melanocytic markers (HMB-45 and Melan-A) establish the diagnosis (Table 1) [57].
HCC versus Hepatoid Carcinoma: Hepatoid carcinoma is a rare extrahepatic carcinoma that closely mimics HCC both morphologically and immunohistochemically [46, 47]. It is often indistinguishable from HCC on histologic grounds; focal bile pigment and occasional cytoplasmic hyaline globules can be seen in the tumor [46]. Hepatoid carcinoma most commonly occurs in the stomach but can be seen in many other organs, including esophagus, pancreas, gallbladder, colon, lung, urinary bladder, uterus, and ovary [46, 58-62]. The HCC-like component can be accompanied by areas of conventional adenocarcinoma. Hep Par-1 is positive in nearly all of the cases, while variable reactivity has been reported with Arg-1 (63%), GPC-3 (50%), pCEA (38%), and
ACCEPTED MANUSCRIPT
albumin ISH [34, 46]. SALL4 (Sal-like protein 4) is often positive in gastric hepatoid carcinomas, but can be positive in HCC as well [63]. There is no specific immunohistochemical marker that can reliably distinguish HCC from hepatoid carcinoma, and correlation with clinical and radiologic findings is necessary to establish the diagnosis.
5. IMMUNOHISTOCHEMISTRY IN HISTOLOGIC VARIANTS OF HCC AND OTHER RELATED SETTINGS
Scirrhous HCC: Scirrhous HCC is a rare variant of HCC characterized by a prominent stromal component, which comprises more than 50% of the tumor (Figure 2A) [9, 64-66]. Hep Par-1 and pCEA have low sensitivities in scirrhous HCC (26% and 37%, respectively) [9]. Scirrhous HCC is positive for CK7, CK19, and MOC-31 in more than half of the cases (Figure 2B). The abundant stroma and aberrant immunophenotype can lead to an erroneous diagnosis of metastatic adenocarcinoma or intrahepatic cholangiocarcinoma [9]. It is important to include Arg-1 and GPC-3 in the panel in such situations, as they have high sensitivity (> 80%) for scirrhous HCC, and their combined use will identify nearly all cases (Figure 2C) [9].
Fibrolamellar carcinoma (FLM): This variant is characterized by a triad of histologic features: lamellar stromal fibrosis, large cells with abundant eosinophilic granular cytoplasm, and prominent eosinophilic nucleoli [66]. FLM is usually positive for hepatocellular makers like Arg-1, Hep Par-1, and pCEA [66], while GPC-3 is positive in two-thirds of cases. CK7 is positive in virtually all cases [66, 67]. CD68, a histiocytic
ACCEPTED MANUSCRIPT
marker, is positive in nearly all cases of FLM, but the staining can be patchy and is also seen in 25% of classical HCCs [66, 68]. Cholangiocarcinoma is negative for CD68 [66, 68]. Since positive staining for CK7 and CD68 is seen in nearly all FLM cases, the diagnosis should be rendered with caution if a suspected FLM case is negative for these markers. Recently, a specific gene fusion has been described in FLM that results from an ~400 kilobase deletion on chromosome 19, creating an in-frame fusion of the promoter and first exon of DNAJB1, a heat shock protein, and the trailing nine exons of PRKACA, the catalytic domain of protein kinase A [69]. The DNAJB1-PRKACA fusion transcript and the rearrangements of the PRKACA locus can be identified by reverse transcription polymerase chain reaction (RT-PCR) and fluorescence in situ hybridization (FISH), respectively [70]. This abnormality is seen in more than 80% of FLM cases, and has not been described in classical HCC, hepatoblastoma, or intrahepatic cholangiocarcinoma [70]. The FISH assay can be obtained for cases in which the features are suggestive but not diagnostic of FLM.
Combined HCC-Cholangiocarcinoma (HCC-CC): As per Word Health Organization (WHO) 2010 criteria, morphologic features of both HCC and CC components that are confirmed by immunohistochemistry are essential for the diagnosis of HCC-CC. A combination of hepatocellular markers is recommended, which should include Arg-1. Diagnosis of the CC component should be based on the presence of discrete glands (with or without mucin), staining for one or more markers typically positive in adenocarcinomas (CK7, CK19, and/or MOC31), and negative or focal staining with hepatocellular markers like Arg-1. Positive staining for CK7, CK19, or MOC31, even if strong and diffuse, is not sufficient for diagnosis of the CC component in the
ACCEPTED MANUSCRIPT
absence of supportive morphologic features. It is important to use strict criteria for this diagnosis due to therapeutic implications. The presence of the CC component is likely to lead to lymph node dissection at surgery, and the use of a gemcitabine-based chemotherapy regimen [71]. For patients with tumors arising in cirrhotic liver, many centers consider cholangiocarcinoma to be a contraindication for transplantation due to the high risk of recurrence [71-73].
Early HCC: The differentiation of early HCC from HGDN can be difficult on small biopsy samples. Early HCC is defined by stromal invasion, which is often not evident on biopsies. The combined use of HSP-70, GPC-3, and GS stains has been advocated for the distinction of early HCC from HGDN [30]. HSP-70 is a highly conserved protein that is overexpressed in HCC, resulting in strong nuclear staining in tumor cells. Weak patchy staining should be considered as negative. Normal bile ducts stain strongly with HSP-70 and can serve as a useful internal control. Diffuse staining with GS (moderate to strong staining in more than 50% of tumor cells) shows strong correlation with activation of ß-catenin [74, 75]. Positive staining with two of these three markers supports HCC with high specificity (Figure 3), although absence of staining does not exclude HCC as the sensitivity using the criterion of two positive markers is around 50%. Up to 22% of HGDN cases may show positive staining for one marker, but two or more markers are not positive [30]. In our experience, HSP-70 and GPC-3 are useful in a small minority of cases (10% or less). Most cases that are positive for HSP-70 and/or GPC-3 can be diagnosed as HCC based on morphology and reticulin stain. Diffuse staining for CD34 can also provide supportive evidence for HCC, but there is a significant overlap between early HCC and HGDN. CK7 may be helpful if the interface
ACCEPTED MANUSCRIPT
of tumor with adjacent non-tumor liver is present. Most cirrhotic nodules and HGDN show CK7-positive ductular reaction at the interface with non-lesional liver, while this finding is absent in HCC [76, 77].
6. HCC VERSUS BENIGN HEPATOCELLUAR LESIONS
Hepatocellular adenoma (HCA): The absence of cytologic atypia (small cell change, nuclear pleomorphism), lack of architectural abnormalities (thick cell plates, prominent pseudoacinar change), and an intact reticulin framework distinguish HCA from HCC [75, 78, 79]. Immunohistochemistry can be helpful in distinguishing HCA from HCC in small samples, and also helps in the classification of HCA into four categories based on the WHO 2010 classification.
(1) Hepatocyte nuclear factor 1-alpha (HNFla)-inactivated HCA: This tumor is characterized by mutations in the HNFla gene. Most of these tumors occur in young women and show prominent steatosis, although fat may be absent in some cases. Sinusoidal dilatation and/or cytologic atypia are typically absent, and association with HCC is low [79-81]. HNFla gene mutation leads to negative regulation of liver fatty acid binding protein (LFABP), and negative results with LFABP stain support the diagnosis of this variant (Table 2). Normal hepatocytes shows cytoplasmic staining with LFABP [81]. Since LFABP can be lost in HCC [82], it should not be used for distinction of HCA from HCC but only for subclassification once the diagnosis of HCA has been established.
ACCEPTED MANUSCRIPT
(2) Inflammatory (telangiectatic) HCA: This subtype is characterized by mutations that activate the interleukin (IL)-6 signaling pathway, most commonly in the IL6ST gene that encodes the signaling co-receptor gp130. Inflammatory HCA is characterized by sinusoidal dilatation, patchy inflammatory cell infiltrate, and variable steatosis (Figure 4A) [79, 81]. This subtype often shows overlapping features with FNH such as fibrous septa and ductular reaction, making this distinction challenging on histologic grounds alone [83]. Inflammatory HCA shows strong and diffuse positivity for acute phase reactants, serum amyloid A (SAA) and C-reactive protein (CRP) (Figure 4B; Table 2) [81, 83]. Although CRP has higher sensitivity (~100%) compared to SAA (~90%), it is less specific with periseptal staining seen in the majority of FNH, and diffuse staining in 15% of FNH [83]. Mutation in the CTNNB1 gene encoding ß-catenin is seen in 10% of inflammatory HCAs. These cases show diffuse GS staining in addition to SAA positivity and are considered to be high risk lesions similar to ß-catenin-activated HCA without inflammatory features [81, 84].
(3) ß-catenin-activated HCA: This subtype is characterized by activation of ß- catenin, which usually results from mutation in the CTNNB1 gene that encodes ß-catenin. This variant is more likely to show cytoarchitectural abnormalities, such as prominent pseudoacinar formation, small cell change, and cytologic atypia (Figure 5A) [79]. Association with concurrent or subsequent HCC has been noted in up to 70% of cases [81]. In the absence of ß-catenin activation, the protein is seen in a submembranous location by immunohistochemistry. ß-catenin activation leads to nuclear translocation of the protein, which can manifest as aberrant nuclear B-catenin staining [75, 81]. However, nuclear ß-catenin staining is not a sensitive marker for ß-catenin activation, and can be
ACCEPTED MANUSCRIPT
restricted to a few isolated neoplastic hepatocytes when present, limiting its utility in small biopsies. Diffuse GS staining (more than 50% of tumor cells showing moderate to strong cytoplasmic staining without the map-like pattern) is a more reliable marker to identify B-catenin-activated HCA (Figure 5B; Table 2) [74, 75].
(4) Unclassified: These tumors show features of HCA, but lack the defining characteristics of other subtypes.
Focal Nodular Hyperplasia (FNH): The hallmark features of FNH are nodular architecture, fibrous septa with or without central scar, ductular reaction, absence of interlobular bile ducts, and aberrant thick-walled blood vessels (Figure 6A) [79]. FNH shows a highly characteristic pattern of staining with GS manifested by interconnecting large groups of hepatocyte while sparing small islands of hepatocytes adjacent to fibrous septa (the so-called “map-like” pattern) (Figure 6B; Table 2) [17, 83]. In normal liver, GS staining is confined to 1-3 layers of hepatocytes around the central vein. Most HCAs also show pericentral staining that is often expanded compared to normal, and they may show patchy GS staining away from the veins, but without the typical map-like pattern. Diffuse GS staining typical of ß-catenin activation needs to be distinguished from the map-like pattern; this may be challenging in very limited samples like cell blocks. Patchy weak staining with SAA can be seen in a minority of FNHs [83], but map-like GS staining distinguishes FNH from inflammatory HCA in most cases.
Practical Considerations:
A. Selection of an antibody panel: The comprehensive use of all markers listed above for subclassification of HCA is not required in every case. If the histologic features clearly point towards a benign hepatocellular lesion and cytoarchitectural
ACCEPTED MANUSCRIPT
features of HCC are absent, the use of GS and SAA stains may suffice, and leads to the following scenarios:
(a) GS map-like, SAA negative: This supports FNH in the setting of appropriate histology, and further stains are not required. If GS is map-like and SAA is positive, the diagnosis is still likely to be FNH but special attention is warranted to confirm the map- like pattern. If CRP is added to the panel, it is important to note that periseptal staining is common in FNH.
(b) GS patchy (not map-like or diffuse), SAA positive: This supports inflammatory HCA in the setting of appropriate histology, and further stains are not required.
(c) GS patchy (not map-like or diffuse), SAA negative: This is likely a SAA- negative inflammatory HCA, HNF1a-inactivated HCA, or unclassified HCA. If definite subtyping of HCA is desirable, CRP for inflammatory HCA and LFABP for HNFla- inactivated HCA can be obtained.
(d) GS diffuse, SAA positive/negative: This supports a ß-catenin-activated neoplasm, and careful review of morphology and reticulin stain is essential to exclude a well-differentiated HCC. The combined use of HSP-70, GS, and GPC-3 has been advocated in distinguishing typical HCA from atypical hepatocellular neoplasms and well-differentiated HCC [31, 85]. Positive staining with both GS and HSP-70 has been described in nearly half of well-differentiated HCCs and 10% of atypical hepatocellular neoplasms [31]. GPC-3 is negative in most cases of atypical hepatocellular neoplasms. Hence, the use of HSP-70 and GPC-3 can be considered in atypical settings, but the yield is low and routine use of these markers is not necessary. For tumors that show obvious
ACCEPTED MANUSCRIPT
malignant features, GS and HSP-70 are not helpful for the diagnosis of HCC as these are often positive in non-hepatocellular malignant neoplasms including intrahepatic cholangiocarcinoma [86].
B. Interpretation of diffuse GS staining: The 50% cutoff for the definition of diffuse GS staining is based on the original French study [81]. Diffuse GS staining can be characterized as diffuse homogeneous or diffuse heterogeneous. The former is characterized by moderate to strong GS staining in more than 90% of tumor cells (Figure 5B). This pattern strongly correlates with ß-catenin activation, and is more likely to be observed with deletions and point mutations in the D32-S37 region of the CTNNB1 gene [87]. The diffuse heterogeneous pattern, which is characterized by moderate to strong GS staining in 50-90% of tumor cells, is more commonly observed with exon 3 point mutations such as T41 and S45 (Figure 5C) [83, 88, 89]. In 20-30% of cases, diffuse GS staining is present in the absence of ß-catenin mutation. These cases are currently regarded as high risk, and the diffuse GS staining may be due to mutations in other Wnt signaling pathway genes (such as APC, AXIN). In 10% of HCA cases, there are mutations in exons 7 and 8 of the CTNNB1 gene (K335 and N387). As opposed to the exon 3 mutations, these mutations leading to weak ß-catenin activation are currently not thought to be associated with high risk for HCC. Diffuse GS staining is typically not observed in these cases [90]. Although the use of GS has become a standard practice in clinical work-up, there are pitfalls in the use of this stain for risk assessment. For example, a minority of cases with exon 3 mutations such as S45 may not show a diffuse pattern and can be missed by immunohistochemistry. In some cases, diffuse versus non- diffuse pattern of GS staining is difficult to determine especially on biopsy specimens,
ACCEPTED MANUSCRIPT
and these cases can be regarded as indeterminate for -catenin activation (Table 3). In the future, further work-up of these cases for mutations involving ß-catenin or other Wnt signaling genes may be useful along with determination of other molecular changes associated with HCC such as TERT promoter mutation and cytogenetic changes.
C. Terminology for diagnosis: Since ß-catenin activation is a high-risk feature, it has been suggested that the tumors with ß-catenin activation (nuclear ß-catenin or diffuse GS staining) should not be classified as HCA, but as ‘atypical hepatocellular neoplasm’ or ‘hepatocellular neoplasm with uncertain malignant potential’ (HUMP) [88, 91, 92] to emphasize that these tumors have borderline characteristics between HCA and HCC. This designation can also be used for tumors that have borderline histologic features, but that do not show ß-catenin activation. If sufficient cytoarchitectural abnormalities and/or reticulin loss are present, the case should be classified as HCC. There are no widely accepted guidelines on when to use the ‘atypical’ designation; our recommendations are included in Table 4.
D. Staining with HCA subtyping antibodies in HCC: Loss of LFABP is not restricted to HCA and can be observed in HCC [82]. Similarly, SAA and CRP can be positive in HCC. These markers are useful for subclassification of HCA, but should not be used for distinction of HCA and HCC.
E. SAA and CRP staining in adjacent liver: Both SAA and CRP staining can be seen in non-lesional liver adjacent to a mass lesion. If the biopsy misses the mass and samples the non-lesional liver, SAA and/or CRP staining in this portion can be mistaken for inflammatory HCA. This can be especially true for limited biopsies or if sinusoidal dilatation, inflammation, and/or ductular reaction are present related to mass-effect and
ACCEPTED MANUSCRIPT
obscure the portal tracts. In these cases, the perivenular pattern of GS staining and lack of diffuse sinusoidal staining with CD34 can help to confirm the non-lesional liver.
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
REFERENCES
[1] 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.
[2] Butler SL, Dong H, Cardona D, Jia M, Zheng R, Zhu H, et al. The antigen for Hep Par 1 antibody is the urea cycle enzyme carbamoyl phosphate synthetase 1. Lab Invest 2008;88:78-88.
[3] Wennerberg AE, Nalesnik MA, Coleman WB. Hepatocyte paraffin 1: a monoclonal antibody that reacts with hepatocytes and can be used for differential diagnosis of hepatic tumors. Am J Pathol 1993;143:1050-4.
[4] Minervini MI, Demetris AJ, Lee RG, Carr BI, Madariaga J, Nalesnik MA. Utilization of hepatocyte-specific antibody in the immunocytochemical evaluation of liver tumors. Mod Pathol 1997;10:686-92.
[5] Morrison C, Marsh WJ, Frankel W. A comparison of CD10 to pCEA, MOC-31, and hepatocyte for the distinction of malignant tumors in the liver. Mod Pathol 2002;15:1279-87.
[6] Fan Z, van de Rijn M, Montgomery K, Rouse RV. Hep par 1 antibody stain for the differential diagnosis of hepatocellular carcinoma: 676 tumors tested using tissue microarrays and conventional tissue sections. Mod Pathol 2003;16:137-44.
[7] Kakar S, Muir T, Murphy LM, Lloyd RV, Burgart LJ. Immunoreactivity of Hep Par 1 in hepatic and extrahepatic tumors and its correlation with albumin in situ hybridization in hepatocellular carcinoma. Am J Clin Pathol 2003;119:361-6.
ACCEPTED MANUSCRIPT
[8] Nguyen T, Phillips D, Jain D, Torbenson M, Wu TT, Yeh MM, et al. Comparison of 5 Immunohistochemical Markers of Hepatocellular Differentiation for the Diagnosis of Hepatocellular Carcinoma. Arch Pathol Lab Med 2015;139:1028-34.
[9] Krings G, Ramachandran R, Jain D, Wu TT, Yeh MM, Torbenson M, et al. Immunohistochemical pitfalls and the importance of glypican 3 and arginase in the diagnosis of scirrhous hepatocellular carcinoma. Mod Pathol 2013;26:782-91.
[10] Fujikura K, Yamasaki T, Otani K, Kanzawa M, Fukumoto T, Ku Y, et al. BSEP and MDR3: Useful Immunohistochemical Markers to Discriminate Hepatocellular Carcinomas From Intrahepatic Cholangiocarcinomas and Hepatoid Carcinomas. Am J Surg Pathol 2016;40:689-96.
[11] Timek DT, Shi J, Liu H, Lin F. Arginase-1, HepPar-1, and Glypican-3 are the most effective panel of markers in distinguishing hepatocellular carcinoma from metastatic tumor on fine-needle aspiration specimens. Am J Clin Pathol 2012;138:203-10
[12]
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.
[13] Yan BC, Gong C, Song J, Krausz T, Tretiakova M, Hyjek E, et al. Arginase-1: a new immunohistochemical marker of hepatocytes and hepatocellular neoplasms. Am J Surg Pathol 2010;34:1147-54.
[14] Capurro M, Wanless IR, Sherman M, Deboer G, Shi W, Miyoshi E, et al. Glypican-3: a novel serum and histochemical marker for hepatocellular carcinoma. Gastroenterology 2003;125:89-97.
ACCEPTED MANUSCRIPT
[15] Nakatsura T, Yoshitake Y, Senju S, Monji M, Komori H, Motomura Y, et al. Glypican-3, overexpressed specifically in human hepatocellular carcinoma, is a novel tumor marker. Biochem Biophys Res Commun 2003;306:16-25.
[16] Baumhoer D, Tornillo L, Stadlmann S, Roncalli M, Diamantis EK, Terracciano LM. Glypican 3 expression in human nonneoplastic, preneoplastic, and neoplastic tissues: a tissue microarray analysis of 4,387 tissue samples. Am J Clin Pathol 2008; 129:899-906.
[17] Shafizadeh N, Ferrell LD, Kakar S. Utility and limitations of glypican-3 expression for the diagnosis of hepatocellular carcinoma at both ends of the differentiation spectrum. Mod Pathol 2008;21:1011-8.
[18] Libbrecht L, Severi T, Cassiman D, Vander Borght S, Pirenne J, Nevens F, et al. Glypican-3 expression distinguishes small hepatocellular carcinomas from cirrhosis, dysplastic nodules, and focal nodular hyperplasia-like nodules. Am J Surg Pathol 2006;30:1405-11.
[19] Shirakawa H, Kuronuma T, Nishimura Y, Hasebe T, Nakano M, Gotohda N, et al. Glypican-3 is a useful diagnostic marker for a component of hepatocellular carcinoma in human liver cancer. Int J Oncol 2009;34:649-56.
[20] Wang XY, Degos F, Dubois S, Tessiore S, Allegretta M, Guttmann RD, et al. Glypican-3 expression in hepatocellular tumors: diagnostic value for preneoplastic lesions and hepatocellular carcinomas. Hum Pathol 2006;37:1435- 41.
ACCEPTED MANUSCRIPT
[21] Yamauchi N, Watanabe A, Hishinuma M, Ohashi K, Midorikawa Y, Morishita Y, et al. The glypican 3 oncofetal protein is a promising diagnostic marker for hepatocellular carcinoma. Mod Pathol 2005;18:1591-8.
[22] Abdul-Al HM, Makhlouf HR, Wang G, Goodman ZD. Glypican-3 expression in benign liver tissue with active hepatitis C: implications for the diagnosis of hepatocellular carcinoma. Hum Pathol 2008;39:209-12.
[23] Lau SK, Prakash S, Geller SA, Alsabeh R. Comparative immunohistochemical profile of hepatocellular carcinoma, cholangiocarcinoma, and metastatic adenocarcinoma. Hum Pathol 2002;33:1175-81.
[24] Ma CK, Zarbo RJ, Frierson HFJ, Lee MW. Comparative immunohistochemical study of primary and metastatic carcinomas of the liver. Am J Clin Pathol 1993;99:551-7.
[25]
Chu PG, Ishizawa S, Wu E, Weiss LM. Hepatocyte antigen as a marker of hepatocellular carcinoma: an immunohistochemical comparison to carcinoembryonic antigen, CD10, and alpha-fetoprotein. Am J Surg Pathol 2002;26:978-88.
[26] Wang R, Sheps JA, Ling V. ABC transporters, bile acids, and inflammatory stress in liver cancer. Curr Pharm Biotechnol 2011;12:636-46.
[27] Lagana SM, Salomao M, Remotti HE, Knisely AS, Moreira RK. Bile salt export pump: a sensitive and specific immunohistochemical marker of hepatocellular carcinoma. Histopathology 2015;66:598-602.
ACCEPTED MANUSCRIPT
[28] Chuma M, Sakamoto M, Yamazaki K, Ohta T, Ohki M, Asaka M, et al. Expression profiling in multistage hepatocarcinogenesis: identification of HSP70 as a molecular marker of early hepatocellular carcinoma. Hepatology 2003;37:198-207.
[29] Di Tommaso L, Franchi G, Park YN, Fiamengo B, Destro A, Morenghi E, et al. Diagnostic value of HSP70, glypican 3, and glutamine synthetase in hepatocellular nodules in cirrhosis. Hepatology 2007;45:725-34.
[30] Di Tommaso L, Destro A, Seok JY, Balladore E, Terracciano L, Sangiovanni A, et al. The application of markers (HSP70 GPC3 and GS) in liver biopsies is useful for detection of hepatocellular carcinoma. J Hepatol 2009;50:746-54.
[31] Nguyen TB, Roncalli M, Di Tommaso L, Kakar S. Combined use of heat-shock protein 70 and glutamine synthetase is useful in the distinction of typical hepatocellular adenoma from atypical hepatocellular neoplasms and well- differentiated hepatocellular carcinoma. Mod Pathol 2016;29:283-92.
[32] Krishna M, Lloyd RV, Batts KP. Detection of albumin messenger RNA in hepatic and extrahepatic neoplasms. A marker of hepatocellular differentiation. Am J Surg Pathol 1997;2:147-52.
[33] Ferrone CR, Ting DT, Shahid M, Konstantinidis IT, Sabbatino F, Goyal L, et al. The Ability to Diagnose Intrahepatic Cholangiocarcinoma Definitively Using Novel Branched DNA-Enhanced Albumin RNA In Situ Hybridization Technology. Ann Surg Oncol 2016;23:290-6.
[34] Shahid M, Mubeen A, Tse J, Kakar S, Bateman AC, Borger D, et al. Branched chain in situ hybridization for albumin as a marker of hepatocellular
ACCEPTED MANUSCRIPT
differentiation: evaluation of manual and automated in situ hybridization platforms. Am J Surg Pathol 2015;39:25-34.
[35] 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.
[36] Kimura H, Nakajima T, Kagawa K, Deguchi T, Kakusui M, Katagishi T, et al. Angiogenesis in hepatocellular carcinoma as evaluated by CD34 immunohistochemistry. Liver 1998;18:14-9.
[37] Maeda T, Kajiyama K, Adachi E, Takenaka K, Sugimachi K, Tsuneyoshi M. The expression of cytokeratins 7, 19, and 20 in primary and metastatic carcinomas of the liver. Mod Pathol 1996;9:901-9.
[38] Durnez A, Verslype C, Nevens F, Fevery J, Aerts R, Pirenne J, et al. The clinicopathological and prognostic relevance of cytokeratin 7 and 19 expression in hepatocellular carcinoma. A possible progenitor cell origin. Histopathology 2006;49:138-51.
[39] Mourra N, Azizi L. CK20 positivity in hepatocellular carcinoma: a potential diagnostic pitfall in liver biopsy. Appl Immunohistochem Mol Morphol 2013;21:94-5.
[40] Karabork A, Kaygusuz G, Ekinci C. The best immunohistochemical panel for differentiating hepatocellular carcinoma from metastatic adenocarcinoma. Pathol Res Pract 2010;206:572-7.
ACCEPTED MANUSCRIPT
[41] Niemann TH, Hughes JH, De Young BR. MOC-31 aids in the differentiation of metastatic adenocarcinoma from hepatocellular carcinoma. Cancer 1999;87:295- 8.
[42] Proca DM, Niemann TH, Porcell AI, De Young BR. MOC31 immunoreactivity in primary and metastatic carcinoma of the liver. Report of findings and review of other utilized markers. Appl Immunohistochem Mol Morphol 2000;8:120-5.
[43] Ordóñez NG. The diagnostic utility of immunohistochemistry in distinguishing between mesothelioma and renal cell carcinoma: a comparative study. Hum Pathol 2004;35:697-710.
[44] Ordóñez NG. Value of the MOC-31 monoclonal antibody in differentiating epithelial pleural mesothelioma from lung adenocarcinoma. Hum Pathol 1998;29:166-9.
[45] Mcknight R, Nassar A, Cohen C, Siddiqui MT. Arginase-1: a novel immunohistochemical marker of hepatocellular differentiation in fine needle aspiration cytology. Cancer Cytopathol 2012;120:223-9.
[46] Chandan VS, Shah SS, Torbenson MS, Wu TT. Arginase-1 is frequently positive in hepatoid adenocarcinomas. Hum Pathol 2016;55:11-6.
[47] Terracciano LM, Glatz K, Mhawech P, Vasei M, Lehmann FS, Vecchione R, et al. Hepatoid adenocarcinoma with liver metastasis mimicking hepatocellular carcinoma: an immunohistochemical and molecular study of eight cases. Am J Surg Pathol 2003;27:1302-12.
ACCEPTED MANUSCRIPT
[48] Murakata LA, Ishak KG, Nzeako UC. Clear cell carcinoma of the liver: a comparative immunohistochemical study with renal clear cell carcinoma. Mod Pathol 2000;13:874-81.
[49] Mazal PR, Stichenwirth M, Koller A, Blach S, Haitel A, Susani M. Expression of aquaporins and PAX-2 compared to CD10 and cytokeratin 7 in renal neoplasms: a tissue microarray study. Mod Pathol 2005;18:535-40.
[50] Barr ML, Jilaveanu LB, Camp RL, Adeniran AJ, Kluger HM, Shuch B. PAX-8 expression in renal tumours and distant sites: a useful marker of primary and metastatic renal cell carcinoma? J Clin Pathol 2015;68:12-7.
[51] McGregor DK, Khurana KK, Cao C, Tsao CC, Ayala G, Krishnan B, et al. Diagnosing primary and metastatic renal cell carcinoma: the use of the monoclonal antibody ‘Renal Cell Carcinoma Marker’. Am J Surg Pathol 2001;25:1485-92.
[52]
Sano K, Kosuge T, Yamamoto J, Shimada K, Takayama T, Yamasaki S, et al. Primary hepatic carcinoid tumors confirmed with long-term follow-up after resection. Hepatogastroenterology 1999;46:2547-50.
[53] Wang JH, Dhillon AP, Sankey EA, Wightman AK, Lewin JF, Scheuer PJ. ‘Neuroendocrine’ differentiation in primary neoplasms of the liver. J Pathol 1991;163:61-7.
[54] Kandil D, Leiman G, Allegretta M, Evans M. Glypican-3 protein expression in primary and metastatic melanoma: a combined immunohistochemistry and immunocytochemistry study. Cancer 2009;117:271-8.
ACCEPTED MANUSCRIPT
[55] Renshaw AA, Granter SR. A comparison of A103 and inhibin reactivity in adrenal cortical tumors: distinction from hepatocellular carcinoma and renal tumors. Mod Pathol 1998;11:1160-4.
[56] Ghorab Z, Jorda M, Ganjei P, Nadji M. Melan A (A103) is expressed in adrenocortical neoplasms but not in renal cell and hepatocellular carcinomas. Appl Immunohistochem Mol Morphol 2003;11:330-3.
[57] Tsui WM, Colombari R, Portmann BC, Bonetti F, Thung SN, Ferrell LD, et al. Hepatic angiomyolipoma: a clinicopathologic study of 30 cases and delineation of unusual morphologic variants. Am J Surg Pathol 1999;23:34-48.
[58] Su JS, Chen YT, Wang RC, Wu CY, Lee SW, Lee TY. Clinicopathological characteristics in the differential diagnosis of hepatoid adenocarcinoma: a literature review. World J Gastroenterol 2013;19:321-7.
Ke La Sua Capa
[59] Haninger DM, Kloecker GH, Bousamra LM, Nowacki MR, Slone SP. Hepatoid adenocarcinoma of the lung: report of five cases and review of the literature. Mod Pathol 2014;27:535-42.
[60] Marchegiani G, Gareer H, Parisi A, Capelli P, Bassi C, Salvia R. Pancreatic hepatoid carcinoma: a review of the literature. Dig Surg 2013;30:425-33.
[61] Sekino Y, Mochizuki H, Kuniyasu H. A 49-year-old woman presenting with hepatoid adenocarcinoma of the urinary bladder: a case report. J Med Case Rep 2013;7:12.
[62] Wang L, Zhong Y, Sun L, Zhou H, Chen W, Zhang X. Clinical and pathological features of hepatoid carcinoma of the ovary. World J Surg Oncol 2013;11:29.
ACCEPTED MANUSCRIPT
[63] Osada M, Aishima S, Hirahashi M, Takizawa N, Takahashi S, Nakamura K, et al. Combination of hepatocellular markers is useful for prognostication in gastric hepatoid adenocarcinoma. Hum Pathol 2014;45:1243-50.
[64] Matsuura S, Aishima S, Taguchi K, Asayama Y, Terashi T, Honda H, et al. ‘Scirrhous’ type hepatocellular carcinomas: a special reference to expression of cytokeratin 7 and hepatocyte paraffin 1. Histopathology 2005;47:382-90.
[65] Kurogi M, Nakashima O, Miyaaki H, Fujimoto M, Kojiro M. Clinicopathological study of scirrhous hepatocellular carcinoma. J Gastroenterol Hepatol 2006;21:1470-7.
[66] Shafizadeh N, Kakar S. Hepatocellular Carcinoma: Histologic Subtypes. Surg Pathol Clin 2013;6:367-84.
[67] Górnicka B, Ziarkiewicz-Wróblewska B, Wróblewski T, Wilczynski GM, Koperski L, Krawczyk M, et al. Carcinoma, a fibrolamellar variant — immunohistochemical analysis of 4 cases. Hepatogastroenterology 2005;52:519- 23.
[68] 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.
[69] Honeyman JN, Simon EP, Robine N, Chiaroni-Clarke R, Darcy DG, Lim II, et al. Detection of a recurrent DNAJB1-PRKACA chimeric transcript in fibrolamellar hepatocellular carcinoma. Science 2014;343:1010-4.
ACCEPTED MANUSCRIPT
[70] 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.
[71] Kassahun WT, Hauss J. Management of combined hepatocellular and cholangiocarcinoma. Int J Clin Pract 2008;62:1271-8.
[72] Wu CH, Yong CC, Liew EH, Tsang LL, Lazo M, Hsu HW, et al. Combined Hepatocellular Carcinoma and Cholangiocarcinoma: Diagnosis and Prognosis After Resection or Transplantation. Transplant Proc 2016;48:1100-4.
[73] Groeschl RT, Turaga KK, Gamblin TC. Transplantation versus resection for patients with combined hepatocellular carcinoma-cholangiocarcinoma. J Surg Oncol 2013;107:608-12.
[74] 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.
[75] Shafizadeh N, Kakar S. Diagnosis of well-differentiated hepatocellular lesions: role of immunohistochemistry and other ancillary techniques. Adv Anat Pathol 2011;18:438-45.
[76] 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.
[77] Park YN, Yang CP, Fernandez GJ, Cubukcu O, Thung SN, Theise ND. Neoangiogenesis and sinusoidal “capillarization” in dysplastic nodules of the liver. Am J Surg Pathol 1998;22:656-62.
ACCEPTED MANUSCRIPT
[78] Ferrell LD, Crawford JM, Dhillon AP, Scheuer PJ, Nakanuma Y. Proposal for standardized criteria for the diagnosis of benign, borderline, and malignant hepatocellular lesions arising in chronic advanced liver disease. Am J Surg Pathol 1993;17:1113-23.
[79] 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.
[80] Zucman-Rossi J, Jeannot E, Nhieu JT, Scoazec JY, Guettier C, Rebouissou S, et al. Genotype-phenotype correlation in hepatocellular adenoma: new classification and relationship with HCC. Hepatology 2006;43:515-24.
[81] 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.
[82] Cho SJ, Ferrell LD, Gill RM. Expression of liver fatty acid binding protein in hepatocellular carcinoma. Hum Pathol 2016;50:135-9.
[83] 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.
[84]
Rebouissou S, Amessou M, Couchy G, Poussin K, Imbeaud S, Pilati C, et al. Frequent in-frame somatic deletions activate gp130 in inflammatory hepatocellular tumours. Nature 2009;457:200-4.
ACCEPTED MANUSCRIPT
[85] Lagana SM, Salomao M, Bao F, Moreira RK, Lefkowitch JH, Remotti HE. Utility of an immunohistochemical panel consisting of glypican-3, heat-shock protein-70, and glutamine synthetase in the distinction of low-grade hepatocellular carcinoma from hepatocellular adenoma. Appl Immunohistochem Mol Morphol 2013;21:170-6.
[86]
Lagana SM, Moreira RK, Remotti HE, Bao F. Glutamine synthetase, heat shock protein-70, and glypican-3 in intrahepatic cholangiocarcinoma and tumors metastatic to liver. Appl Immunohistochem Mol Morphol 2013;21:254-7.
[87] 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.
[88]
Evason KJ, Grenert JP, Ferrell LD, Kakar S. Atypical hepatocellular adenoma- like neoplasms with ß-catenin activation show cytogenetic alterations similar to well-differentiated hepatocellular carcinomas. Hum Pathol 2013;44:750-8.
[89] Hale G, Liu X, Hu J, Xu Z, Che L, Solomon D, et al. Correlation of exon 3 ß- catenin mutations with glutamine synthetase staining patterns in hepatocellular adenoma and hepatocellular carcinoma. Mod Pathol 2016;29:1370-80.
[90] Pilati C, Letouzé E, Nault JC, Imbeaud S, Boulai A, Calderaro J, et al. Genomic profiling of hepatocellular adenomas reveals recurrent FRK-activating mutations and the mechanisms of malignant transformation. Cancer Cell 2014;25:428-41.
ACCEPTED MANUSCRIPT
[91] Bedossa P, Burt AD, Brunt EM, Callea F, Clouston AD, Dienes HP, et al. Well- differentiated hepatocellular neoplasm of uncertain malignant potential: proposal for a new diagnostic category. . Hum Pathol 2014;45:658-60.
[92] Kakar S, Evason KJ, Ferrell LD. Well-differentiated hepatocellular neoplasm of uncertain malignant potential: proposal for a new diagnostic category — reply. Hum Pathol 2014;45:660-1.
ACCEPTED MANUSCientositalocal
ACCEPTED MANUSCRIPT
FIGURE LEGENDS
Figure 1. (A) HCC typically shows thickened cell plates (greater than 3 cells thick), small cell change, cytologic atypia, and mitotic figures (H&E, 200x). (B) Although Hep Par-1 is a highly sensitive and specific marker for HCC, it can be negative in a subset of HCC (200x). (C) Arg-1 is the most sensitive marker for HCC, and it shows diffuse cytoplasmic and variable nuclear staining (200x). (D) HCC shows a distinct canalicular staining pattern with pCEA, CD10, villin, or BSEP (200x).
Figure 2. (A) Scirrhous HCC shows dense stromal fibrosis (H&E, 200x). (B) CK7 is frequently positive in scirrhous HCC as are other adenocarcinoma markers, including CK19 and MOC-31 (200x). (C) GPC-3 (and/or Arg-1) is the preferred hepatocellular marker for scirrhous HCC (200x).
Figure 3. (A) Well-differentiated HCC shows pseudoacinar architecture and mild cytologic atypia, and it can be difficult to distinguish from HGDN (H&E, 200x). (B, C) Combined use of GS (B), HSP-70 (C), and GPC-3 can be useful, as positive staining with at least 2 markers is significantly associated with early HCC (200x).
Figure 4. (A) Inflammatory/telangiectatic HCA demonstrates sinusoidal
dilatation/congestion and patchy inflammatory cell infiltrate (H&E, 200x). (B) Strong and diffuse granular cytoplasmic staining for SAA is the most specific marker for inflammatory/telangiectatic HCA (200x).
ACCEPTED MANUSCRIPT
Figure 5. (A) ß-catenin-activated HCA shows more prominent pseudoacinar formation and cytologic atypia compared to other variants of HCA (H&E, 200x). (B) ß-catenin- activated HCA may show diffuse homogenous GS staining involving more than 90% of tumor cells (400x). (C) Diffuse heterogeneous staining involving 50-90% of tumor cells can also be seen in a subset of ß-catenin-activated HCAs (200x).
Figure 6. (A) FNH shows nodular architecture with thick fibrous septa and ductular reaction (H&E, 200x). (B) FNH demonstrates moderate to strong “map-like” GS staining pattern, whereas normal liver shows positive staining around central veins (400x).
ACCEPTED MANUSCRIPT
| Expression Pattern | Leading Diagnosis | Other Considerations |
|---|---|---|
| Arg-1+, CK19- | If Arg-1 positivity is strong and diffuse: HCC | Consider non-hepatocellular tumors if Arg-1 positivity is weak or focal |
| Arg-1-, CK19+ | Cholangiocarcinoma Metastatic adenocarcinoma Colorectal: CK20, CDX2 Lung: CK7, TTF-1, Napsin Prostate: PSA, p501s, PAP, NKX3.1 Breast: ER, Mammaglobin, GATA-3, GCDFP Ovary: CK7, ER, WT-1 Thyroid: CK7, TTF-1, Thyroglobulin Pancreas: DPC-4 loss Upper GI: CK7 Polygonal cell tumors: NET: Chromogranin, Synaptophysin RCC: PAX-2, PAX-8, RCC marker | Arg-1-negative HCC with aberrant CK19 expression. If HCC is still a consideration morphologically, additional markers like Hep Par-1 and GPC-3 can be obtained |
| Arg-1+, CK19+ | 0 HCC with CK19 expression | Non-hepatocellular tumor with aberrant Arg-1 expression (uncommon) |
| Arg-1-, CK19- | Pancytokeratin-positive: Arg-1-negative HCC CK19-negative adenocarcinoma Polygonal cell tumor: NET: Chromogranin, Synaptophysin RCC: PAX-2, PAX-8, RCC marker Other carcinomas: Urothelial, SCC | Pancytokeratin-negative: ACC: Inhibin, Melan-A Melanoma: SOX-10, S100, HMB- 45, Melan-A AML: SMA, HMB-45, Melan-A Epithelioid GIST: CD117/c-KIT, DOG-1 Other sarcomas with epithelioid morphology |
Abbreviations: ACC, adrenocortical carcinoma; AML, angiomyolipoma; Arg-1, arginase- 1; CK, cytokeratin; CDX2, caudal-related homeobox transcription factor 2; DOG-1, discovered on GIST-1; DPC-4, deleted in pancreatic carcinoma-4; ER, estrogen receptor; GATA-3, GATA-binding protein-3; GCDFP, gross cystic disease fluid protein; GIST,
ACCEPTED MANUSCRIPT
gastrointestinal stromal tumor; GPC-3, glypican-3; HCC, hepatocellular carcinoma; HMB-45, human melanoma black-45; NET, neuroendocrine tumor; PAP, prostatic acid phosphatase; PAX, paired box gene; PSA, prostate-specific antigen; p501s, prostein; RCC, renal cell carcinoma; SCC, squamous cell carcinoma; SMA, smooth muscle actin; SOX-10, SRY-related HMG-box 10; TTF-1, thyroid transcription factor-1; WT-1, Wilms tumor-1
NS’enER RIPT
Table 2. Immunophenotypic Characteristics of Benign Hepatocellular Lesions
| GS | SAA | CRP | LFABP | ß-catenin | |
|---|---|---|---|---|---|
| FNH | Map-like | Patchy + or - | Periseptal | + ☒ | Membranous |
| Inflammatory HCA | Patchy (< 50%) with perivascular accentuation | Diffuse or patchy | Diffuse or patchy | + ☒ | Membranous (nuclear if ß- catenin-activated) |
| HNF-1a- inactivated HCA | Patchy (< 50%) with perivascular accentuation | – | – | – | Membranous |
| ß-catenin- activated HCA | Diffuse | – | – | + ☒ | Nuclear (subset) |
| Unclassified HCA | Patchy (< 50%) with perivascular accentuation | – | – | + ☒ | Membranous |
Abbreviations: CRP, C-reactive protein; FNH, focal nodular hyperplasia; GS, glutamine synthetase; HCA, hepatocelluar adenoma; HNF, hepatocyte nuclear factor; LFABP, liver- fatty acid binding protein; SAA, serum amyloid A; +, positive staining; - , loss of staining
| Staining Pattern | Interpretation |
|---|---|
| Diffuse homogeneous | - Moderate to strong cytoplasmic staining in 90-100% of lesional cells - Strongly correlates with ß-catenin activation - Exon 3 B-catenin deletions and some point mutations (D32-S37) lead to this pattern |
| Diffuse heterogeneous | - Moderate to strong cytoplasmic staining in 50-90% of lesional cells - Associated with ß-catenin activation in most cases - Exon 3 B-catenin mutations (T41, S45) lead to this pattern - Small number of ß-catenin mutation in exons 7 and 8 can show this pattern (these are not considered high risk) |
| Indeterminate | - Moderate to strong cytoplasmic staining in close to 50% of lesional cells, difficult to determine diffuse versus patchy pattern of staining - Difficulties are more commonly encountered in biopsies |
| Patchy, perivascular, peripheral enhancement | - Considered negative for ß-catenin activation - B-catenin mutation in exons 7 and 8 can show this pattern (these are not considered high risk) - Minority of S45 exon 3 B-catenin mutation show this pattern |
ANS POR CRIPT
| Clinically atypical | Diagnosis |
|---|---|
| Biopsy | Well-differentiated hepatocellular neoplasm - Histologic features most consistent with HCA - High risk of HCC related to clinical setting |
| Resection | Adenoma |
| Morphologically atypical | Diagnosis |
| Biopsy | Apical hepatocellular neoplasm |
| Resection | Depending on extent and degree of atypia: - Adenoma - Atypical - HCC |
Abbreviations: HCA, hepatocellular adenoma; HCC, hepatocellular carcinoma
ACCEPTED MANUSCRIPT
A
B
C
D
ACCEPTED MANUSCRIPT
A
B
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
A
B
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
A
B
ACCEPTED MANU
ACCEPTED MANUSCRIPT
A
B
ACCEPTED MANUSCRIPT
A
B
ACCEPTED MANU