Variants of Gastric carcinoma
Variants of gastric cancer account for approximately 5% of all carcinomas of the stomach. Some variant carcinomas, such as lymphoepithelioma-like carcinoma, hepatoid carcinoma with or without α-fetoprotein (AFP) production, small cell or neuroendocrine carcinoma, adenosquamous and squamous carcinoma, choriocarcinoma, sarcomatoid carcinoma, parietal cell or oncocytic carcinoma, micropapillary carcinoma, mucoepidermoid carcinoma, chief cell carcinoma, and Paneth cell carcinoma have been described, and recognition of the specific subtype is important due to not only their correlation with a distinct clinical course and prognosis but also differential diagnosis from metastasis from the outside of the stomach and different therapeutic modalities, particularly in the era of targeted treatment. Gastric lymphoepithelioma-like carcinoma and parietal cell or oncocytic carcinoma have been found to have a lower risk of lymph node metastasis and better prognosis. However, other variants such as hepatoid carcinoma, small cell or neuroendocrine carcinoma, adenosquamous and squamous carcinoma, choriocarcinoma, sarcomatoid carcinoma, and micropapillary carcinoma have shown to be associated with a poorer prognosis and a higher risk of metastasis to the lymph node and other organs, compared to conventional intestinal or diffuse type of gastric adenocarcinomas.
In recent years, early detection, endoscopic mucosal resection (EMR) for early gastric cancer, and neoadjuvant therapy have made a remarkable progress in the management and prognosis of gastric cancers. Thus, the prediction of aggressive behavior and accurate risk stratification in the variants of gastric cancer has become more important than ever. The World Health Organization (WHO) classification and the degree of differentiation have not been applied to some variants of gastric carcinoma. Considering that EMR and neoadjuvant therapy are selectively applied to patients with gastric carcinoma based on the tumor size, WHO classification, and degree of differentiation, further specific subclassification for each variant should be discussed to allow treatment to be directed to appropriate patient groups..
Recently, Giuffre et al analyzed the HER2 status in a cohort of rare histologic variants of advanced gastric adenocarcinoma such as hepatoid adenocarcinoma and oncocytic adenocarcinoma . This series demonstrated that one of rare variants of gastric carcinoma, hepatoid adenocarcinoma, has shown an increased HER2 overexpression in up to 42.86% of cases compared to the intestinal (31.25%) and diffuse (3.45%) types of gastric adenocarcinoma. Trastuzumab has been known as an additional useful therapeutic standard option for patients with HER2-positive advanced gastric cancer. Therefore, this result and further studies may bring a significant progress in clinical course and prognosis in patients with aggressive variants of gastric carcinoma such as hepatoid adenocarcinoma. However, future studies are needed to evaluate overexpression of HER2 in other variants of gastric carcinoma. The variants of gastric carcinoma are listed in Table 1.
2. Epstein-Barr virus associated lymphoepithelioma-like carcinoma
Lymphoepithelioma-like carcinoma of the stomach (LELCS), also known as gastric carcinoma with lymphoid stroma, undifferentiated carcinoma with lymphoid stroma or medullary carcinoma, is a unique variant of gastric adenocarcinoma which is highly associated with Epstein-Barr virus (EBV) infection. This variant was described originally by Watanabe et al in 1976 and accounts for approximately 4% of all gastric carcinomas [2-4]. It has recently been emphasized that the role of EBV infection in the carcinogenesis of LELCS. The incidence of EBV-associated gastric adenocarcinoma varies from 1.3% to 20.1% in different areas, with an average of 10% worldwide [5, 6]. The prevalence of EBV infection has been found approximately 75 % of LELCS and 16% of conventional gastric adenocarcinoma in North America by EBV-encoded ribonucleic acids in situ hybridization (EBER-ISH) .
The clinicopathologic and molecular characteristics of EBV-associated LELCSs are quite different from those of conventional gastric carcinoma, such as a male predominance (male to female ratio: 2-3.4:1), predisposition to proximal stomach, frequent association with multiple and remnant gastric cancers, a lower frequency of lymph node metastasis, a relatively favorable prognosis, and aberrant concordant methylation of multiple genes [4, 8-10]. The mean age of diagnosis is 54.8 years, younger than conventional gastric carcinomas . The clinical symptoms of LELCS are similar with conventional gastric adenocarcinoma, and patients with EBV-associated gastric carcinoma are known for elevated antibodies against EBV-related antigens. Levine et al reported that patients with EBV-associated gastric carcinoma have significantly high IgG and IgA antibody titer to EBV viral capsid antigen more than 5 years preceding their first diagnoses .
Watanabe and Yanai et al had reported the specific endoscopic findings of EBV-related LELCs in early and advanced lesions [4, 8]. In their series, approximately 80 % of LELCS appear to be a superficial depressed type, such as IIc, IIc+III, and IIa+IIc, frequently combined with ulcer in an early stage. Endoscopic ultrasonography of this variant demonstrates a hypoechoic submucosal tumor-like protrusion with a large thickness compared to the length in an early lesion. Most LELCSs appear as a fungating mass (Bormann type II) accompanied by marked thickening of the gastric wall in an advanced stage (Fig. 1).
Histologically, LELCS has a well demarcated and pushing border and nests of poorly or undifferentiated tumor cells embedded within a lymphoid stroma. The tumor cells are arranged in syncytia, microalveolar, thin trabecular, or primitive tubular patterns. Also, they contain vesicular to clear nuclei, prominent nucleoli, and abundant eosinophilic polygonal cytoplasm with poorly defined cell borders (Fig 2). Generally, tumor cells grow in a diffuse manner intermixed with lymphocytes and plasma cells, mimicking malignant lymphoma (Schmincke type). However, a sharp demarcation between the tumor nests and lymphocyte stroma that is composed of variable amounts of lymphocytic infiltration, sometimes with lymphoid follicles, is noted (Regaud type) but no desmoplastic reaction is identified. Rarely, giant cells and epithelioid granulomas are observed within the lymphoid stroma [13, 14]. In the early lesion, LELCS shows a “lace pattern”, which consists of columnar arrangement and intercolumnar fusion of neoplastic glands between dense lymphocytic infiltration at the intermediate zone of lamina propria . This tumor is usually accompanied with severe atrophic gastritis in the background but not associated with intestinal metaplasia or Helicobacter pylori infection that are often observed in usual gastric adenocarcinoma .
The epithelial cell component of this variant is positive for keratins. Also HLA-DR expression of tumor cells associated with dense lymphocytic infiltrates can be detected in some cases of EBV-associated LELCS . Increased dendritic cells within a lymphocyte stroma are also seen, which are positive for S100 and CD83. CD8+ T lymphocytes are the predominant cells infiltrating into tumor cell nests in EBV-associated LELCS, many of which express perforin and granzyme B . Florid proliferation of CD8+ T lymphocytes within tumor cells seems to be associated with the host immune response to remove EBV antigen or cellular EBV-induced antigen .
EBV has specific latency expression patterns in different EBV-associated malignancies. EBV-associated LELCS can express viral latent genes and related products, not only EBV-encoded small ribonucleic acids (EBER) but also EBV-determined nuclear antigen type 1, BamHI A region rightward transcripts, and variable latent membrane protein 2A and 2B . Polymerase chain reaction (PCR) to amplify EBV deoxyribonucleic acid fragments and EBER-ISH techniques could be used to detect EBV in tumor tissues; however, as a PCR test for amplifying EBV DNA is very sensitive and may cause a false positivity, and EBERs are always abundantly expressed in nearly 100% of tumor cell tissues (Fig.3), EBER-ISH technique is a gold standard test to detect EBV in paraffin-embedded tissues .
Previous studies of conventional gastric adenocarcinoma have been shown 34-60% of p53 overexpression. Some studies showed the frequency of p53 immunoreactivity, approximately 14%, was much lower in patients with EBV-associated LELCS compared to conventional gastric carcinoma . However, other studies have almost same frequency of p53 expression with conventional gastric adenocarcinoma [23, 24]. The role of p53 mutation in the development of EBV-associated gastric carcinoma is still unclear.
If an accompanying lymphoid stroma contains multiple reactive follicles and intense lymphocytic infiltration in LELCS can be mistaken as a pseudolymphoma or lymphoma. It is important to examine the presence of malignant epithelial cell component carefully. Immunohistochemical markers for the cytokeratins, AE1/AE3 and CAM5.2, should be essential to detect in which only scarce undifferentiated malignant epithelial cells are present.
EBV is a ubiquitous γ-1 herpes virus usually acquired during childhood via salivary transmission, which establishes a life-long persistent infection of B cells in over 90% of adults . EBV has been linked to the pathogenesis of several mesenchymal and epithelial neoplasms including nasopharyngeal carcinoma, B and T cell lymphomas, NK cell malignancies, and a subset of smooth muscle tumors . The role of EBV in gastric carcinoma is considered to be directly oncogenic, but is still largely unknown. The infection of EBV is assumed to be occurred in the very early stage of carcinogenesis . However, there is still a controversy over whether EBV infects gastric epithelial cells before or after the development of invasive gastric carcinoma .
Studies for genetic abnormality for EBV-associated gastric carcinoma are still limited and sparse. Several previous studies had failed to demonstrate any significant chromosomal loss or gain in comparative genome hybridization, changes of DNA copy number and microsatellite clearly. In contrast to inconsistent and negative results in genetic alteration, epigenetic alterations such as promoter hypermethylation, which bring in chromatin remodeling and silencing of tumor-related genes, plays a key role in the carcinogenesis of EBV-associated gastric carcinoma . Kusano et al demonstrated that EBV-associated gastric carcinomas were strongly associated with high CpG-island methylation (CpG-island methylator phenotype- high, CIMP-H) by using methylation-specific polymerase chain method . Other epigenetic series about EBV-associated gastric carcinoma have shown simultaneous methylation of multiple genes, including cell cycle regulation (p14, p15, p16, and cox2), DNA repair and protection (HMLH1, MGMT,GSTP1), cell adherence and metastasis (E-cadherin, TIMP3), angiogenesis (THBS1), apoptosis (DAP-kinase, bcl-2, p73), and signal transduction (APC, PTEN, RASSF1A) [28, 29]. Higher frequencies of hypermethylation of cancer-related genes found in EBV-associated gastric carcinoma suggest a close relationship between EBV infection and aberrant methylation of this variant during its carcinogenesis. Also, these findings suggest that DNA methyltransferase inhibitors such as 5-aza-2΄-deoxycytidine (5-aza-CdR) or trichostatin A (TSA), which can induce the lytic phase of EBV infection in EBV positive gastric cancer cell lines, can be a promising therapeutic agent specially for EBV-associated gastric carcinoma .
Another interesting characteristic of EBV-associated gastric carcinoma is its higher occurrence among gastric remnant carcinomas, ranging from 27% to 42% [31-33]. Chen et al hypothesizes that the injuries of gastric mucosa and/or changes of the microenvironment within the remnant stomach may be involved in the development of EBV-associated gastric carcinoma .
EBV-associated LELCS has known to have a low risk of lymph node metastasis . Interestingly, whether the presence of tumor metastasis within the regional lymph nodes or not, perigastric lymph nodes frequently show reactive hyperplasia that represents a host immune response against LELCS . EBV-associated LELCS shows a very low rate of regional lymph node metastasis especially if the tumor involves in the mucosa and submucosa as an early gastric cancer . The patients with this variant tumor can be good candidates for EMR in regard to a lower frequency of lymph node metastasis and a relative well circumscribed tumor margin that can be easily assessed endoscopically. Some clinical trials to remove EBV-associated LELCS in an early stage have been reported using an endoscopic resection such as EMR or submucosal dissection [36, 37]. The patients treated with endoscopic resection have shown a benign clinical course without recurrence or metastasis. The prognosis of EBV-associated gastric carcinoma has been considered better than conventional gastric carcinoma. Song et al reported that 5-year overall survival rate of EBV-associated gastric carcinoma was 71.4%, and disease-free survival rate was 67.5% compared to 56.1% and 55.2% of those usual gastric carcinoma and suggested the prognosis of EBV-associated gastric carcinoma depends on the patient’s inflammatory response (tumor-infiltrating lymphocyte counts) .
3. Hepatoid adenocarcinoma
Hepatoid adenocarcinoma of the stomach (HACS) was first described by Ishikura et al in 1985 as a subtype of AFP producing gastric carcinoma, which has a histologic similarity with hepatocellular carcinoma and distinct clinicopathological properties . The incidence of this variant ranges from 0.38% to 0.73% of all gastric cancer [39, 40]. The mean age of diagnosis is 64.5 years (range from 49 to 78 years), similar to that of conventional gastric carcinomas. The clinical symptoms are similar to those seen in patients with conventional gastric adenocarcinoma except marked elevated serum level of AFP in most patients. Characteristically, earlier studies demonstrated that HACS frequently combined with vascular invasion, liver metastasis, and a higher incidence of lymph node metastasis which result in a poorer prognosis than other common types of gastric adenocarcinoma [41, 42].
AFP is a serum protein produced by fetal liver, yolk sac cells, and some fetal gastrointestinal cells . HACS is characterized by distinct hepatoid differentiation histologically and the production of liver specific proteins including AFP confirmed by immunohistochemistry. However, not all HACSs produce hepatic specific protein. Approximately 54 % of HACSs express AFP by immunohistochemistry and 63% of patients with HACS show an elevated serum AFP level [39, 44]. In addition, there are some histologic types of carcinoma other than HACS, tubular/papillary adenocarcinomas, poorly differentiated medullary carcinoma, and enteroblastic adenocarcinoma, also can produce and secrete AFP [38, 45-48].
The most commonly found macroscopic type of HACS is 0-IIc (superficial depressed type) in an early lesion . This tumor can occur as a circumscribed mass, Bormann type II (fungating type) or III (ulceroinfiltrative type) as an advanced cancer in the antrum and lower body of the stomach (Fig.4) [44, 49]. Microscopically, this variant contains areas of hepatoid differentiation showing structural mimicry of liver tissue such as sheet-like or trabecular arrangement of tumor cells with sinusoid-like vasculature and bile canaliculus-like structure . This tumor consists of cuboidal or polygonal cells with centrally located nuclei and abundant eosinophilic cytoplasm (Fig.5). However, there is no consensus for quantification of the proportion of hepatocellular differentiated component yet to accept as HACS. The proportion of hepatoid component ranges from 10 to 90% in reported cases of HACS . The glandular component such as well-differentiated tubular or papillary adenocarcinoma frequently intermingles with the hepatoid component. The transition between glandular and hepatoid components of HACS can be gradual or abrupt. Occasionally, bile and periodic acid-Schiff-positive/diastase-resistant hyaline globules may be observed in intracellular and extracellular sites. In addition to the production of liver-specific proteins such as AFP as well as prealbumin, albumin, and transferrin, Proteins Induced by Vitamin K Absence - II (PIVKA-II) and Hep-par1 antigen can be detected in tumor cells [38, 52, 53]. Some HACSs exhibit extensive lymphovascular invasion, extending to veins with tumor emboli, which results in early metastasize to other organs, predominantly to the liver .
Immunohistochemically, the tumor cells are positive for CK8, CK18, CK19, CK20, but negative for CK7. Also this variant can express the canalicular pattern of polyclonal carcioembryonic antigen (CEA), α-1 antitrypsin, and α-1 antichymotrypsin . The immunoreactivity of PIVKA-II and Hep-par1 are variably observed. Glypican-3 is a cell surface heparin sulfate proteoglycan considered to be an oncofetal protein because of its presence in fetal liver and liver tumors (hepatocellular cell carcinoma and hepatoblastoma) . The immunohistochemical staining pattern of glypican-3 is strongly positive in membrane and cytoplasm of hepatoid tumor cells (Fig. 5). Hishinuma et al reported that glypican 3 is more sensitive (100% sensitivity) than either Hep-par1 or AFP but Hep-par1 is more specific as an immunohistochemical marker for HACS . In a series by Kinjo et al showed that AFP has 81.1% of sensitivity and 46.1% of specificity and glypican-3 has 90.5% of sensitivity and 63.2% of specificity . These studies support both AFP and glypican-3 could be useful markers to make a diagnosis of HACS.
Many cytogenetic and molecular studies had been undertaken to investigate pathogenesis and biological behavior of this variant. However, the histogenesis of HACS is still unclear. Kishimoto et al proposed that gastric carcinoma might acquire hepatic differentiation during the tumor progression, “HACS transdifferentiation”. Akiyama et al demonstrated that the hepatoid component exhibited exactly same patterns of chromosome X inactivation, p53 gene mutation, the level of p53 expression, and loss of heterozygosity with conventional adenocarcinomatous component of HACS . These findings suggest a monoclonal origin of both glandular and hepatoid elements of HACS and support “HACS transdifferentiation” as the most accepted histogenesis.
Differential diagnoses of HACS include other similar-appearing tumors (lung, pancreas, esophagus so on) and hepatocellular cell carcinoma. Particularly, the metastatic HACS to the liver may be almost indistinguishable from hepatocellular cell carcinoma due to overlapping clinicopathologic features such as elevated AFP level, hepatoid morphology, and immunoexpressions of AFP, polyclonal CEA, and alpha-1 antitrypsin. Moreover, the hepatoid component of HACS can be more prominent in metastatic lesions to perigastric lymph nodes or liver [45, 58]. The presence of underlying disease such as liver cirrhosis and the presence of primary lesion detected by screening modalities such as endoscopy and abdominal computerized tomography scan would be useful to differentiate metastatic HACS and primary hepatocellular cell carcinoma. Although hepatocellular cell carcinoma arising in non-cirrhotic liver and without known risk factors is rare, this tumor has been reported to appear as a single nodule with pseudo-adenomatous or sclerosing pattern histologically [59, 60].
In recent clinical series by Baek et al, approximately 77% of patients with HACS at presentation were diagnosed as an advanced stage, stage III or IV . Median overall survival and progression free survival of these patients after gastrectomy and/or palliative chemotherapy were 8.03 months (95% CI: 6.59-9.47) and 3.47 months (95% CI: 0.65-6.29), respectively. The incidence of liver metastasis of HACS is significantly higher than that of conventional gastric adenocarcinoma. Liu et al reported the overall incidence of liver metastasis was 75.6% including 8.9% synchronous and 73.2% metachronous liver metastasis in HACSs compared to 11.6 %, including 1.8% synchronous and 9.9% metachronous liver metastasis in conventional gastric adenocarcinoma . The overall 5-year survival rate of HACS was 9% compared to 44% in conventional gastric adenocarcinoma . Most of HACS cases clinically appear to chemoresistant and curative resections are limited due to advanced lesions at diagnosis. Kamata et al demonstrated multiple ATP-binding cassette transporters related with multidrug resistance of tumor such as multidrug resistant-associated proteins 1,2, and 6 were expressed frequently in HACS . This finding suggested that ATP-binding cassette transporters participated in the mechanism of multidrug resistance in HACS.
HER2 gene amplification and protein overexpression have been introduced as the target therapy with anti-HER2 humanized monoclonal antibody (trastuzumab) in various cancers including gastric cancer in an advance stage. Although reported rates of HER2 overexpression have been variable, it accounts for approximately 20 % of all gastric carcinomas. Recently, Giuffre et al reported that a markedly increased HER2 amplification was more frequent in HACS (42.9%) than that seen in tubular gastric adenocarcinomas (31.3%) . This finding shows trastuzumab can be a useful therapeutic standard option not only for patients with advanced gastric cancer but also in aggressive variants like HACS.
4. Neuroendocrine cell carcinoma
Primary small cell carcinoma of the stomach (PSCCS) is an exceedingly aggressive variant of gastric carcinoma with neuroendocrine differentiation. Since Matsusaka et al described this variant first in 1976, less than 230 cases have been reported in the literature . This tumor also has been referred to “oat cell carcinoma” and “atypical carcinoid” and accounts for 0.1 - 0.6% of total gastric cancers [63, 64]. The mean age of presentation is 65 years (range, 42 to 84 years) and it commonly affects males . The clinical presentation is similar to those seen in patients with conventional gastric adenocarcinoma in an advanced stage. PSCCS may rarely secrete hormones such as vasoactive intestinal peptide, neuron-specific enolase, pro-gastrin-releasing peptide, antidiuretic hormone, and adrenocorticotropic hormone [66-68]. A few cases of PSCCS have been reported in association with paraneoplastic syndromes including paraneoplastic neurological syndrome and Cushing syndrome by ectopic production of ACTH [69, 70].
PSCCS tends to metastasize early to regional lymph nodes and liver and extends to surrounding organs including liver, transverse colon, pancreas, and diaphragm at the presentation [71-74]. In the largest retrospective series, Chiba et al demonstrated that PSCCSs have a higher incidence rate of lymphatic invasion (88.9% vs. 56.6%), vascular invasion (75.6% vs. 31.6%), and lymph node metastasis (82.1% vs. 58.8%) compared to those of conventional gastric carcinoma .
Neuroendocrine cell carcinomas of the gastrointestinal tract are usually responded well to the chemotherapy. Although there has been no established chemotherapy regimen for PSCCS due to its rarity, generally it has been recognized that surgical resection alone may not be sufficient treatment and emphasize the importance of adjuvant chemotherapy, especially for advanced disease. Fukuda et al insisted that intensive chemotherapy with or without surgical resection should be recommended for this tumor at any stage . Recently, some studies have reported good response of some chemoregimens routinely used for lung SCC such as etoposide/cisplatin, irinotecan/cisplatin, and S1/cisplatin [65, 68, 77, 78]. The 5-year survival rate of patients with gastric neuroendocrine carcinoma has been reported to be 22.1%- 43.8% and the median survival time is 19 month [68, 79, 80]. However, long-term survival up to 3 years was reported in a patient with aggressive adjuvant chemotherapy .
Macroscopically, the average size of the tumor is approximately 6.3 cm . This tumor is an ulcerative or protruding mass (70% of cases) with frequent invasion to subserosa (Fig.6) [65, 81]. PSCCS can be classified as pure (Fig.7) or mixed (composite) SCC (Fig.8) depending on the proportion of neuroendocrine cell component. However, regardless of the proportion of neuroendocrine carcinoma in total volume of gastric carcinoma, its presence has been correlated with a poor prognosis. Approximately 60% of PSCCS cases are associated with conventional gastric adenocarcinoma, and rare cases with sarcomatoid carcinoma, adenosquamous, and squamous carcinoma variants have been reported [82-85]. Histologically, diagnostic criteria for PSCCS are identical with those for pulmonary neuroendocrine tumors by the WHO/International association for the Study of Lung Cancer . First of all, the SCC histology is defined as having a markedly high nuclear/cytoplasmic ratio and hyperchromatic small nuclei (less than 3 resting lymphocytes) with finely granular chromatin and inconspicuous or rarely conspicuous nucleoli. Frequent nuclear molding might be present. Focal or extensive necrosis can be seen. The typical organoid architecture pattern of low grade neuroendocrine neoplasm (eg. carcinoid) is rarely present in PSCCS. As a high grade tumor, their proliferative rate is high, and all exhibit more than 10 mitoses per 10 high power fields, with a mean of 40 to 50 mitoses . Intestinal metaplasia may be seen in the background of gastric mucosa. In cases with classic histologic features of neuroendocrine tumors, positive staining for neuroendocrine markers is not a requirement for a diagnosis .
The pathogenesis of neuroendocrine carcinoma of the stomach is unknown. The most commonly accepted hypothesis is the presence of a “pluripotent stem cell” that has a potential to grow and differentiate into other cell types producing mucin or keratin . This hypothesis has been supported by some cases of composite neuroendocrine carcinomas with other glandular or squamous cell components [89, 90]. Other suggested theory is that neuroendocrine carcinoma of the stomach arises from neuroendocrine precursor cells in gastric adenocarcinoma during its genetic progression .
Since PSCCS could be occasionally misdiagnosed as poorly differentiated adenocarcinoma or malignant lymphoma due to accompanying crush artifact in small biopsies, Grimelius stain and immunohistochemistry would be useful to differentiate morphologic mimickers. Immunohistochemical staining for neuroendocrine markers, including chromogranin, synaptophysin, and CD56, is usually positive (Fig.9) [81, 92, 93]. Its intensity and distribution can vary, with most examples showing patchy and moderately intense immunoreactivity. Also, most PSCCSs show an immunoreactivity for keratin AE1/AE3 and CEA but not for high molecular cytokeratin CK34βE . Recently, Li et al demonstrated the low molecular weight cytokeratin, CK8 (CAM 5.2), is more commonly expressed in SCC in gastrointestinal tract including PSCCS than is the expression of AE1/AE3 or epithelial membrane antigen, suggesting CK8 as a sensitive marker for SCCs of the gastrointestinal tract . The proliferation index for Ki-67 is high, usually more than 25 % positive nuclei. Rindi et al have reported that angioinvasion, tumor size, clinicopathological type, mitotic count, and Ki-67 proliferation index are predictors of tumor malignancy and patient outcome in neuroendocrine tumors of the stomach .
At the current time, neuroendocrine carcinomas in the gastrointestinal tract are classified into neuroendocrine carcinoma (small and large cell subtypes) and mixed adenoneuroendocrine carcinoma by World Health Organization (WHO) . Large cell neuroendocrine carcinoma of the stomach (LCNECS) is defined as a high grade or poorly differentiated malignant neuroendocrine tumor of non-small cell type. In the largest series by Jiang et al, LCNECSs account for at least 1.5% of all gastric cancer . In this series, the mean age of the patients with LCNECS is 62.7 years (range, 47 to 79 years) and it mostly affects males.
Grossly, the mean size of the tumor is 6.4 cm (range 1.1 to 13.0 cm) and 66% of cases are Bormann type II or III in an advanced stage . The diagnostic criteria for LCNECS is defined as having the following features: a diffuse growth pattern or “neuroendocrine architecture” (organoid, nesting, palisading, rosettes, or trabeculae), monotonous polygonal or round to oval cells with moderate amounts of slightly eosinophilic cytoplasm and ill defined cell border, granular or vesicular nuclei with evenly distributed granular chromatin, and with or without visible nucleoli . Focal lumen formation or focal intracytoplasmic mucin might be seen, and are not feature for exclusion. LCNECSs usually combine with other common adenocarcinoma components. Immunohistochemical evidence of neuroendocrine differentiation is defined as positive staining for one of three neuroendocrine markers, chromogranin, synaptophysin, and CD56, in > 20% of the tumor cells . In a relatively large study to compare LCNECS and PSCCS by Matsui et al, they demonstrated LCNECSs reveal a higher mitotic count, larger polygonal cells, a lower nuclear-cytoplasmic ratio, coarser nuclear chromatin, and more frequent conspicuous nucleoli than PSCCSs .
The main difficulty with the diagnosis of gastric neuroendocrine carcinoma is to distinguish them from poorly differentiated adenocarcinoma with solid growth pattern and malignant lymphoma. Preoperative diagnosis of neuroendocrine carcinoma from gastric endoscopic biopsy specimens would be challenging for the pathologists due to its histologic heterogeneity as well as the propensity of neuroendocrine tumor that mainly proliferates under the mucosal layer [65, 71]. Also, neuroendocrine carcinoma of the stomach cannot be easily recognizable due to less prominent neuroendocrine histologic features in high grade tumor cells. These factors can result in limited and demanding biopsy specimens for making a diagnosis of PSCCS preoperatively. However, careful evaluation for hidden or faint neuroendocrine architecture, cytologic features of neuroendocrine cells, and comparable immunohistochemical profiles would be useful to differentiate two disease entities. Sometimes ultrastructural study can help to demonstrate accumulation of electron-dense core neurosecretory granules measuring 200 nm in diameter in these tumors .
Primary gastric malignant lymphoma can be distinguished by immunoreactivity of CD45 and lack of cytokeratins. SCC of the lung metastatic to the stomach should always be considered in the differential diagnosis of PSCCS. With limited data, staining for the lung marker, thyroid transcription factor-1 (TTF1), seems to be almost always negative in neuroendocrine tumors of the gastrointestinal tract . So TTF1 can be helpful to differentiate a metastatic lesion from SCC of the lung.
Another subtype of neuroendocrine carcinoma is a mixed neuroendocrine carcinoma. Although localized endocrine cell differentiation in benign or malignant glandular neoplasm of the gastrointestinal tract is relatively common, truly mixed glandular-endocrine neoplasms (adenoneuroendocrine carcinoma) are rare. These tumors are composed of both glandular component like conventional adenocarcinoma and recognizable neuroendocrine component of small or large cell type that comprises substantial proportions of the tumor volume with each component at least 30% of the lesion . Most of mixed glandular-neuroendocrine neoplasms of the stomach are malignant tumors arising in the background of atrophic gastritis.
5. Squamous and adenosquamous carcinoma
Primary squamous and adenosquamous carcinoma of the stomach is an extremely rare and aggressive variant of the stomach cancer that accounts for 0.04 to 0.9 % of all gastric carcinoma [101, 102]. Although clinical manifestation of the patients with primary squamous and adenosquamous carcinoma of the stomach is similar with patients with conventional gastric adenocarcinoma, it has been reported a mildly elevated serum level of squamous cell carcinoma antigen and a long history of smoking and alcohol abuse in some patients [103-105]. The mean age of presentation is 64 years (range, 17 to 89 years) and men are affected about five times as often as women . Patients with primary squamous or adenosquamous carcinoma of the stomach frequently present with advanced stage disease (pT3 or T4) with or without metastases or involvement of other organs (stage III or IV) . It has known that these variants do not response well to the routine chemoregimen for conventional gastric carcinoma. There is no established standard adjuvant chemoradiation therapy for patients with primary squamous and adenosquamous carcinoma in the stomach. However, one isolated case suggested that low-dose 5-fluorouracil plus cisplatin would be an effective preoperative chemoregimen to shrink the size of tumors and lower postoperative complications .
Although there is no specific macroscopic feature for primary squamous carcinoma of the stomach, recently Oono et al demonstrated a well demarcated, white superficial depressed area of primary squamous carcinoma as an early lesion which is not stained by 3.0% lugol solution on chromoendoscopy . Main differential diagnosis of primary squamous carcinoma of the stomach is primary squamous carcinoma of the esophagus that involves the proximal stomach. Parks et al have proposed the diagnostic criteria for primary squamous carcinoma of the stomach based on not only histologically definite squamous features but also in regard to clinical findings as follows: (1) the tumor must not be located in the cardia; (2) the tumor must not extend into the esophagus; (3) there should be no evidence of squamous carcinoma in any other part of the body  (Fig.10). Another criteria for primary squamous carcinoma of the stomach by the Japanese Classification of Gastric Carcinoma are as follows: (1) the tumor must consist of only squamous carcinoma without any component of adenocarcinoma (pure squamous component only) (Fig.11) and (2) any tumor near esophagogastric junction must be excluded unless evidence for supporting the tumor originated from the stomach exists .
The origin of primary squamous carcinoma of the stomach is unclear, but four hypotheses concerning its development have been proposed, including (1) nests of ectopic squamous cells in gastric mucosa; (2) squamous metaplasia of the gastric mucosa before malignant transformation; (3) squamous differentiation in a preexisting adenocarcinoma; and (4) multipotential stem cells in the gastric mucosa [110-112].
Focal squamous differentiation in the intestinal-type adenocarcinoma is relatively common. Therefore, Straus et al established a diagnostic criteria for adenosquamous carcinoma of the stomach that the squamous component should be present in more than 25 percent of the resected tumor (Fig.12 and 13) . Its biological behavior is usually determined by the adenocarcinoma component. Also the diagnostic criteria by Parks should be applied for the final diagnosis of primary gastric adenosquamous carcinoma to exclude primary esophageal cancer and metastatic lesion.
Differential diagnosis of primary squamous and adenosquamous carcinoma of the stomach include (1) gastric adenocarcinoma, intestinal type, with squamous differentiation; (2) chronic gastritis or ulcer with squamous metaplasia; (3) esophageal squamous carcinoma arising from the esophagogastric junction; (4) metastatic squamous carcinoma to the stomach. Most of reported cases of primary squamous and adenosquamous carcinoma of the stomach have demonstrated a poorer clinical course and prognosis than conventional gastric adenocarcinoma. The mean survival after surgery is less than 12 months (1-16 months) . Radical surgical excision is the only option for localized disease. For advanced-stage disease, surgery plus adjuvant radio-and/or chemotherapy appears to achieve a better outcome than surgery alone in terms of longer survival, although experience is limited due to the rarity of this variant.
Primary choriocarcinoma of the stomach (PCCS) is a highly aggressive variant of gastric carcinoma that was described for the first time by Davidsohn in 1905 . This variant represents approximately 0.08 % of all the gastric cancers . Pure gastric choriocarcinomas are extremely rare and only less than twenty cases were reported [116-118]. Most of primary gastric choriocarcinomas have been reported as a composite or mixed tumor with a combination of predominant choriocarcinoma component and a variable degree of adenocarcinoma. Transitions between adenocarcinoma and choriocarcinoma component may be clear or not. Yolk sac tumor, small cell carcinoma, and hepatoid carcinoma components may be seen as well [119-122]. In the study by Kobayashi et al, the average age of the patients with PCCS is 62.4 years in men and 54.8 years in women with a male predominance (male: female ratio=2.3:1) . The clinical presentation of this variant is similar to that of gastric adenocarcinoma, however, it is a frequent cause of gastrointestinal bleeding and may have some hormonal effects such as gynecomastia in men, precocious puberty, pregnancy mimicking symptoms including amenorrhea, nausea and vomiting in women [124, 125].
PCCS behaves more like gestational choriocarcinoma because it shows extensive hematogenous dissemination or metastasis by mixed or pure components of choriocarcinoma as opposed to the routine lymphatic spreading of gastric adenocarcinoma. Kobayashi et al analyzed previously reported 53 cases of PCCS and concluded that the most common cause of death in patients with PCCS is hepatic failure caused by liver metastasis followed by massive cancerous hemorrhage . Untreated patients with PCCS have an average survival of several months . Although no standard therapy has been established, complete surgical resection with neoadjuvant chemotherapy for non-gestational choriocarcinoma has been used in most of cases. Noguchi et al have reported a good response with the combination chemotherapy of 5-fluouracil and cisplatin after surgery . However, radiotherapy did not show any improvement for clinical course and outcome . Basically, accurate diagnosis by initial biopsy, curative resection, early and appropriate intervention by chemotherapy, and the absence of combined liver metastasis are favorable prognostic factors for patients with PCCS.
The significance of an elevated serum β-human chorionic gonadotropin (HCG) is controversial. Some studies suggested that it is associated with a shorten survival and poorer prognosis . However, other insisted that it does not have any prognostic significance . β-HCG can be detected in blood or tissue in about 10% of patients with conventional gastric carcinoma; however, many of these tumors do not show any histologic evidence of the presence of choriocarcinoma component [130-132]. Also, the elevation of serum β-HCG and HCG immunoreactivity on tissue in these patients were usually mild. Most of reported cases of PCCS have been accompanied by markedly elevated serum level of β-HCG level up to 53,000 IU/ml at the presentation preoperatively and it significantly declined to the low level or baseline several months after surgeries. So, monitoring serum β-HCG level may have a role in evaluating response to treatment and tumor recurrence.
Macroscopically, this tumor usually occurs as a large exophytic mass with extensive necrosis and hemorrhage. Radiographically, it is a heterogeneous mass with enhanced vascularity and hemorrhage, mimicking a vascular tumor such as cavernous hemangioma . Microscopically, this variant consists of choriocarcinoma and conventional adenocarcinoma with a variable proportion. Choriocarcinoma exhibits a typical biphasic pattern of admixed cytotrophoblasts and syncytiotrophoblasts. Polygonal cytotrophoblasts are located with a central core and surrounded by a peripheral rim of multinucleated syncytiotrophoblasts. Cytotrophoblasts have large, round hyperchromatic nuclei, abnormal nuclear chromatin, irregular nuclear membranes, occasional prominent nucleoli, and dense eosinophilic to amphophilic cytoplasm. Pleomorphic and multinucleated large cells are syncytiotrophoblasts. The viable tumor cells are found mainly at the lesion’s periphery while extensive hemorrhage and necrosis are often in the center of the tumor like gestational choriocarcinoma. Typical and atypical mitotic figures may be frequently found. This tumor can show vascular invasion with tumor thrombi and tumor cells lining vascular spaces occasionally. Adenocarcinoma components would be tubular or papillary type with varying degree of differentiation. This tumor has been frequently misdiagnosed as gastric adenocarcinoma at presentation due to the size of tumor, massive tumor necrosis, and combined adenocarcinoma element. The diagnosis may even more become difficult in a small gastric biopsy from the lesion because it may reveal only scant syncytiotrophoblasts intermixed with recognizable routine adenocarcinoma component. Only 8% of PCCS cases in a pooled analysis of 53 cases were correctly diagnosed as choriocarcinoma in an initial biopsy . Therefore. an extensive and large tissue sampling would be required for a precise diagnosis of PCCS prior to surgery. In addition, if the biopsy contain any suspicious elements that suggest syncytiotrophoblasts, immunohistochemical biomarkers for trophoblastic cells will help confirming the diagnosis. In most instances, choriocarcinomatous cells are strongly positive for cytokeratin, β-HCG, and weakly positive for human placental lactogen. However, PCCS may not express β-HCG in the tumor tissue by immunohistochemistry .
Regarding the pathogenesis of PCCS, several theories have been proposed. The dedifferentiation theory first proposed by Pick in 1926 has been the most accepted explanation for the pathogenesis of choriocarcinoma . Liu et al reported the first interphase cytogenetic study of PCCS, the results of which support this theory that PCCS arises from an alternative differentiation pathway, a dedifferentiation, of primary gastric adenocarcinoma [135, 136]. In this study, PCCSs showed the gain of chromosome 12, which is frequently associated with choriocarcinoma, and other genomic imbalances (gains of function mutations in 2q, 7pq, 8pq, 13q, 17q, 18q, 20pq and deletions in 17p) that are common genomic mutations in conventional gastric adenocarcinoma.
The major differential diagnosis of PCCS is metastatic trophoblastic tumor from other sites, particularly a non-gestational gonadal or extragonadal germ cell tumor from men and intrauterine or extrauterine gestational trophoblastic tumor in reproductive aged women. Extensive radiologic and clinical evaluation is recommended to rule out metastatic trophoblastic tumor from genital tracts in the men and women to make an unequivocal PCCS diagnosis.
7. Sarcomatoid carcinoma
Since Queckenstedt described the existence of the sarcomatous component in the gastric adenocarcinoma in 1904, approximately 50 cases of gastric sarcomatoid carcinoma have been reported . Sarcomatoid carcinoma of the stomach (SCS) - also referred to as carcinosarcoma, malignant mixed mesodermal tumor, spindle cell carcinoma, and pseudosarcoma- is an uncommon variant of gastric carcinoma which is a biphasic neoplasm composed of a mixture of malignant epithelial and mesenchymal components. The various terms in use represent the uncertain histogenesis of this tumor, whether sarcomatoid carcinomas represent a single or two separate entities (carcinoma with sarcomatoid differentiation versus collision tumor consisting of carcinoma and sarcoma) remains controversial; however, some stromal tumor cells of SCS showing epithelial features such as staining for cytokeratin favor a monoclonal origin with divergent transformation.
The mean age of presentation is 45 years (range, 27 to 74 years) and males are predominantly affected (2.3:1), with patients frequently having metastasis to regional lymph nodes and liver at the time of diagnosis [138, 139]. Due to this tumor detecting at advance stage and its rapid growth, most of SCSs are associated with a poor clinical outcome. Sato et al reported a mean survival of 10 months in patients with SCS . The current standard therapy for SCS is partial or total gastrectomy with regional lymph node dissection. The effects of chemotherapy or radiotherapy have not been established . SCS may present synchrononously with conventional adenocarcinoma. However, there is no case of SCS that develops secondarily in patients with prior chemotherapy, which is different from primary sarcomatoid carcinoma arising from other organs. SCS usually occurs in the antral or pyloric region and infiltrates the gastric wall frequently. Macroscopically, it is a polypoid, exophytic, or endophytic mass with ulceration (Fig.14) [139, 140, 142-146]. Histologically, SCS exhibits an adenocarcinoma component with variable differentiation and a high-grade sarcomatous component composed of spindle cells with high cellularity, frequent mitotic counts with atypical forms and pleomorphism (Fig.15). The spindle cells generally have plump, pleomorphic nuclei with a coarsely stippled chromatin pattern and small nucleoli with a haphazard arrangement. The transition between epithelial and sarcomatoid components can be abrupt but two components may intermix with each other. The epithelial component is predominantly composed of adenocarcinoma but rarely adenosquamous and neuroendocrine cell carcinoma can be accompanied [82, 83, 140]. Some cases represent atrophic gastritis and dysplasia in the background gastric epithelium [142, 145].
The proportion of the sarcomatoid component is highly variable (5 % up to 80 %) [83, 139, 147]. In most instances, the pattern typically resembles that seen in high grade spindle sarcomas which lack specific immunohistochemical markers for identifying the line of differentiation [83, 84, 141, 148]. In some cases, the sarcomatous component can be heterologous, with osteosarcomatous, chondrosarcomatous, and rhabdomyosarcomatous or leiomyosarcomatous differentiation [82, 139, 146, 149, 150]. There have been three cases reported showing multidirectional differentiation including rhabdomyosarcoma, osteosarcoma, and chondrosarcoma [139, 147, 148].
The epithelial component is identified by CK, EMA, or Cam 5.2, whereas muscle markers like desmin, myogenin, or myoD1 may confirm rhabdomyoblastic differentiation and when chondrosarcoma or liposarcoma is suspected, S100 protein can be used to confirm the diagnosis. In rare cases such as the epithelial component containing neuroendocrine differentiation, neuroendocrine markers (synaptophysin, chromogranin, and CD56) would be useful.
Differential diagnoses of SCS include other primary gastric sarcomas. If primary gastric sarcoma is suspected based on radiologic and endoscopic findings, multiple and extensive sampling of the lesion should be performed to avoid missing an epithelial component. Another consideration is the extremely rare instance of osseous metaplasia arising in the conventional gastric adenocarcinoma. Unusual bone component in the stomach can be mistaken for a part of SCS element. However, the bone component is histologically benign [151, 152]. Similarly, differentiating SCS from benign, borderline, and malignant spindle cell tumors, such as inflammatory fibroid polyp, calcifying fibrous tumor, gastrointestinal stromal tumor, and plexiform angiomyxoid myofibroblastic tumor need to be considered.
8. Acinar cell carcinoma
Acinar cell carcinoma of the stomach (ACCS) defined to have immunohistochemical evidence of pancreatic exocrine enzyme production, is an uncommon variant of gastric adenocarcinoma that is morphologically resemble to primary pancreatic acinar cell carcinoma. Pure ACCS is extremely rare and only two cases had been reported [153, 154]. Most of reported ACCS are mixed tumors which combine with glandular and/or neuroendocrine carcinoma components.
Microscopically, this variant demonstrates acinar, solid, glandular/microglandular, and trabecular arrangement of cells forming large, densely cellular tumor nodules containing minimal stroma (Fig.16). Tumor cells forming acini are cuboidal or columnar cells similar to size of normal pancreatic acinar cells with basally located nuclei and apical eosinophilic granular cytoplasm (Fig 17). Nuclei are round to oval, with minimal to mild pleomorphism, indistinct nucleoli or occasional single nucleolus. Mitotic activity is variable. The cytoplasm is moderately abundant, eosinophilic, and granular. The non-neoplastic stomach shows chronic gastritis with intestinal metaplasia. ACCS can exhibit different morphologic features such as focal ductal differentiation and well-formed individual ductal elements with mucin production that are uncommon in primary acinar cell carcinoma of the pancreas . Periodic acid-Schiff reaction after diastase digestion and Grimelius stains demonstrate characteristic positive granules within the cytoplasm of the tumor cells. By immunohistochemistry, the tumor cells express pancreatic exocrine enzymes, predominantly trypsin, and variably lipase, α-1 anti-trypsin, α-1 anti-chymotrypsin and chymotrypsin. Among many pancreatic enzymes, trypsin is one of the most common and useful markers for the diagnosis of ACCS. The frequency of trypsin expression in series varied from 71% to 100% (average, 97%), whereas chymotrypsin expression ranged from 39% to 100% (average, 66%) . The antigens α-1 antitrypsin and α-1 chymotrypsin can be positive for some tumor cells but not specific for ACCS [153, 157]. Another important diagnostic feature of acinar cell carcinoma is the presence of zymogen granules, which are large (250 to 1000 μm) electron-dense, homogeneous granules on electron microscopy .
The histogenesis of ACCS is unclear. There are several hypotheses that may explain the origin of this variant. Because pancreatic heterotopia is relatively common in the stomach, origin in heterotopic pancreatic tissue was proposed first. Some ACCSs arising from heterotopic pancreas have been reported [158-163]. Acinar cell metaplasia is a relatively common finding in the gastric mucosa, either as a congenital abnormality, or in association with chronic gastritis [164-167]. ACCS may arise from this metaplastic process. Ambrosini-Spaltro et al reported an ACCS arising in association with pancreatic metaplasia in the gastric mucosa . The possibility of the presence of “pluripotent stem cells” in the gastric mucosa, which have the potential to grow and differentiate to diverse cell types and neoplasms has been suggested by several studies of gastric composite tumors by morphology, ultrastructural examination, immunohistochemistry, and molecular genetics [157, 169, 170]. Fukunaga reported a gastric carcinoma resembling pancreatic mixed acinar-neuroendocrine carcinoma and proposed the possibility of a primitive multipotent cell with the capacity of divergent differentiation to explain acinar and neuroendocrine differentiation in the tumor .
The main differential diagnosis is metastatic lesion from pancreatic neoplasm. Since immunohistochemical expression of this tumor in the stomach and pancreas is exactly same, full clinical and radiographic evaluation would be essential to exclude the possibility of metastasis from pancreas.
The prognosis of primary ACCS is unknown due to the scarcity of reported cases. However, considering 50% of patients with primary pancreatic acinar cell carcinoma having metastases at presentation and a worse prognosis, the recognition of acinar cell component in gastric carcinoma may be important for the patients’ treatment and prognosis.
9. Invasive micropapillary carcinoma
Recently, an unusual variant of gastric adenocarcinoma called “ invasive micropapillary carcinoma of the stomach (IMPCS)” has been described by Shimoda et al in 2008 . In a recent study by Roh et al, IMPCSs were present in up to 0.07% of 1,5254 total or subtotal gastrectomy specimens . The mean age of this variant is 66.2 years (range 36-87 years) with a male predominance (male to female ratio: 2.5-3:1). This rare tumor demonstrates an aggressive behavior associated with a higher incidence of lymphovascular invasion and lymph node metastasis, resulting in a poor prognosis similar to invasive micropapillary carcinomas (IMPCs) of other organs including breast, urinary bladder, ureter, lung, parotid gland, and colon [173-179]. In the reported series to date, patients with IMPCS have an estimated 30% of 5- year overall survival rate compared to those with non-IMPCS having 67% of 5-year survival by Kaplan-Meir method .
Histologically, this variant consists of small tight cell clusters of papillary structure devoid of fibrovascular cores within lacunar spaces mimicking lymphatic or vascular channels (Fig.18). The lacunar spaces are artifactual tissue spaces and not lined by an endothelial cell. Several adjunct markers have been introduced to enhance the recognition of IMPCs. IMPCs in most organ systems are characterized by inverted polarity with MUC1 expression with membranous MUC1 staining facing the stroma . Similarly, epithelial membrane antigen shows reverse polarity expression in IMPCs . However, focal or heterogeneous staining pattern of IMPCs by MUC1 and EMA can be seen. Other immunohistochemical expressions of KL-6, CA125, and HER2/neu are frequently increased in IMPCs of the urinary tract but lack of specificity as an ancillary maker [181-184].
No“ pure” form of IMPCS has been reported and all cases of IMPCS reported have been combined with conventional or papillary gastric adenocarcinoma. In the series of 72 IMPCS cases by Eom et al, most of the combined adenocarcinoma components were characterized as intestinal type (64/72 cases, 88.9%) and papillary adenocarcinoma (43/72 cases, 59.7%), and the remaining were classified as diffuse type (8/72 cases, 11.1%) and tubular adenocarcinoma (21/72 cases, 29.2%) .
Although IMPC has become increasingly well recognized as a distinct and aggressive variant in stomach as well as other organs from recent vigorous studies. Diagnostic criteria for a diagnosis of IMPCS remain imprecise. In a recent study about interobserver reproducibility in the diagnosis of IMPC of the urinary tract, this study recommended a combination of some histologic features included small tumor cell nests within stromal retraction spaces, back-to-back lacunae, multiple tumor nests within each single retraction space are useful to make a diagnosis for “classic” IMPC that may bring a better reproducibility. Additional associated histologic features to be considered including epithelial ring forms, intracytoplasmic vacuolization, elongated epithelial nests (i.e., micropapillae), and peripherally oriented nuclei. These diagnostic criteria of IMPC in the urinary tract based on a combination of several histologic features would be highly useful to make a diagnosis of IMPCs in other organs including stomach. In addition, the threshold for a diagnosis of IMPCs based on the percentage/volume of IMPC component is undetermined. There has been no validated clinicopathologic data that support the threshold of IMPC proportion in association with a clinical outcome in patients with IMPC. One study of IMPCs in the urinary tract showed a trend towards an association between the proportion of IMPC and survival with >50 % IMPC component imparting a relative mortality risk of 2.4, compared to with < 50% IMPC of those . However, Kim et al suggested that the proportion of IMPC component with respect the whole tumor is not related with the prognosis of the patients with IMPC in the colorectum . In the stomach, Roh et al failed to find any significant clinicopathologic differences between a group with ≤ 20% of IMPC component and another group with > 20% of IMPC component . Some studies suggest an arbitrary cutoff of IMPCS as ≥ 5% of IMPC proportion in total tumor volume but ≥ 10% of IMPC proportion in other studies [172, 185]. Approximately 70% of reported cases in the stomach were found that the proportion of IMPC component to the entire tumor ranged from 10% to 70%. In previous published cases of IMPC of the colorectum, the proportion of IMPC component ranged from 5 % to 80 % but was usually less than 30 % of the entire lesion [178, 187, 188]. Comperat et al analyzed 72 cases of IMPC of the urinary bladder and the proportions of IMPC component are: 10% of cases with less than 10% of IMPC component, 47% of cases with 10%-50% of those, and 43% of cases with more than 50% of these . Further studies are needed to have an established criteria for IMPCS showing a good reproducibility among inter-and intraobservers and to evaluate a diagnostic threshold of IMPC proportion in total volume of the tumor that correlate best with the clinical outcome.
The incidence of metastasis of IMPC from other organs to the stomach is uncommon. The possible metastatic lesions of IMPCs from other organs included urinary bladder, breast, lung, and ovary. Lotan et al investigated immunohistochemical markers to identify the primary site and differentiate metastatic lesions of IMPC . They recommended that an immunohistochemical panel consisting of uroplakin, CK20, TTF-1, ER and WT-1, and/or PAX8, and mammaglobin is the best one for accurately classifying the likely primary site of IMPC. In their studies, urothelial IMPC were usually positive for uroplakin and CK20, whereas p63, high molecular weight cytokeratin, and thrombomodulin were less sensitive and specific. Lung IMPC was uniformly TTF-1 positive. Breast IMPC was ER positive, mammaglobin positive, and PAX8/WT-1 negative, while ovarian IMPC was ER positive, mammaglobin negative, and PAX/WT-1 positive. However, no specific marker has been introduced for verifying specifically IMPCS.
The main differential entity of IMPCS is papillary adenocarcinoma or conventional adenocarcinoma with multiple endolymphatic tumor emboli. Morphologically, when a distinction of IMPC within lacunar spaces from lymphovascular tumor emboli may be difficult, immunohistochemical studies including factor VIII,
10. Gastric adenocarcinoma of fundic gland type (chief cell predominant type)
Recently, Ueyama et al proposed gastric adenocarcinoma of fundic gland type for a new entity of gastric carcinoma . Although gastric adenocarcinomas with foveolar and pyloric gland type differentiation are relatively common, only a few cases of gastric adenocarcinomas with fundic gland differentiation have been reported. This variant is a well differentiated adenocarcinoma composed of mixed chief and parietal cells mimicking fundic glands (Fig.20). In addition to histologic similarity of chief and parietal cells, immunohistochemical staining with pepsinogen I (a marker for chief cells) and H+/K+ATPase (a marker for parietal cells) exhibit differentiation toward the chief and parietal cells in gastric adenocarcinoma. In the series by Ueyama et al and Singhi et al, the patients’ age range from 42 to 79 years (average: 65 years) with a relatively equal sex distribution [192, 193].
Gastric adenocarcinoma of fundic gland type is a relatively small tumor, the maximum diameter of tumors range from 0.2 to 2 cm (average 0.6 cm) [192, 193]. Characteristically, gastric adenocarcinomas of fundic gland type are located in areas with oxyntic mucosa, in the upper third of the stomach, fundus and cardia [192, 193]. Macroscopic findings of this variant in early lesion are the superficial depressed type (type 0-IIc) or superficial elevated type (type 0-IIa) . Histologically, this variant is a well-differentiated adenocarcinoma composed of columnar cells admixed with predominantly chief cells, with pale grey-blue, basophilic cytoplasm, and scattered parietal cells, with coarse granular eosinophilic cytoplasm. Both cells exhibit mildly enlarged and hyperchromatic nuclei with slight pleomorphism. Mitotic activity is absent or very low.
Differential diagnosis includes fundic gland polyps that are small benign mucosal polyps that occur in the gastric fundus. Histologically, they are composed of dilated glands lined by oxyntic mucosa without atypia.
11. Parietal cell carcinoma and oncocytic carcinoma
Since Capella et al originally described in 1984 as “gastric parietal cell carcinoma”, less than 30 cases have been reported to date . The reported mean age of the patients with this variant is 70.2 years (range 58-84 years) and it exclusively affects men (M:F ratio=9:1). Histologically, parietal cell carcinomas are usually composed of solid sheets of polygonal cells with round nuclei and abundant, finely granular eosinophilic cytoplasm that stain with phosphotungstic acid-hematoxylin and Luxol Fast Blue [194, 195]. Most of gastric parietal cell carcinomas are combined with well to moderately differentiated tubular or papillary adenocarcinoma. Parietal cell differentiation is confirmed by immunoreactivity for antibodies specific for parietal cell biomarkers H+/K+ ATPase and human milk fat globule-2. Ultrastructurally, the granular and eosinophilic cytoplasm, so call “oncocytic cytoplasm” corresponds to the abundance of mitochondria, intracytoplasmic secretory canaliculi, and cytoplasmic tubulovesicles [196, 197]. A few previous studies suggested that this variant of gastric adenocarcinoma is associated with a better prognosis than conventional gastric adenocarcinoma [194, 198-201]. Robey-Cafferty et al reported a case of parietal cell carcinoma with lymphoma-like morphologic features . Takubo et al introduced ten cases of oncocytic adenocarcinoma, which are morphologically similar to parietal cell carcinoma but are negative for anti-parietal antibodies .
12. Miscellaneous carcinomas
Extremely rare variants of gastric carcinoma have been sporadically reported. Among them are (1) mucoepidermoid carcinoma of the stomach; one case was reported that this variant arose from preexisting ectopic mucous glands of stomach . (2) Paneth cell carcinoma or gastric adenocarcinoma with Paneth cell differentiation; histologically, Paneth cell differentiation is characterized by cytoplasmic distinct coarse eosinophilic granules stained red with periodic acid-Schiff and Masson trichrome reagents and reddish brown with phosphotungstic acid hematoxylin, and electron microscopically by lysozyme in cytoplasmic electron dense granules . Immunohistochemical staining for lysozyme, human defensin-5, and CDX2 is usually positive [204, 205]. and (3) gastric carcinoma with osteoclast-like giant cells; these variants contains a minor component of giant cells that contain 3 to 20 nuclei and are positive for CD68 and vimentin . These findings suggest that giant cells are of monocytic/histiocytic origin and probably represent a host response to the tumor . However, the clinicopathological significance of this variant has not been verified due to its rarity. Salient features of variants gastric adenocarcinoma are listed in Table 2.
||EBV infection related tumor with a dense lymphoid stroma, positive EBER-ISH, aberrant methylation, and a better prognosis|
||Subtype of AFP producing carcinoma resembling hepatocellular carcinoma and showing frequent liver metastasis, and a worse prognosis|
||Aggressive tumor with distinct neuroendocrine cell features and a worse prognosis but good response to chemotherapy|
||Pure or composite tumors with definite squamous features with a very strong male predominance and a worse prognosis|
||Tumors with variable choriocarcinomatous components, elevated β-HCG in the serum, frequent hematogenous spread and a worse prognosis|
||Biphasic neoplasm composed of a mixture of malignant epithelial and mesenchymal component with a poor prognosis|
||Adenocarcinoma resembling pancreatic acinar cells with production of pancreatic exocrine enzyme|
||Adenocarcinoma with micropapillary features showing frequent lymphovascular invasion and lymph node metastasis|
||Well differentiated adenocarcinoma mimicking a fundic gland polyp|
||Well differentiated adenocarcinoma with Parietal cell differentiation and a better diagnosis|
Giuffre, G., et al., HER2 status in unusual histological variants of gastric adenocarcinomas. J Clin Pathol, 2012. 65(3): p. 237-41.
Horiuchi, K., et al., Carcinoma of stomach and breast with lymphoid stroma: localisation of Epstein-Barr virus. J Clin Pathol, 1994. 47(6): p. 538-40.
Corvalan, A., et al., Epstein-Barr virus in gastric carcinoma is associated with location in the cardia and with a diffuse histology: a study in one area of Chile. Int J Cancer, 2001. 94(4): p. 527-30.
Watanabe, H., M. Enjoji, and T. Imai, Gastric carcinoma with lymphoid stroma. Its morphologic characteristics and prognostic correlations. Cancer, 1976. 38(1): p. 232-43.
Akiba, S., et al., Epstein-Barr virus associated gastric carcinoma: epidemiological and clinicopathological features. Cancer Sci, 2008. 99(2): p. 195-201.
Lee, J.H., et al., Clinicopathological and molecular characteristics of Epstein-Barr virus-associated gastric carcinoma: a meta-analysis. J Gastroenterol Hepatol, 2009. 24(3): p. 354-65.
Shibata, D. and L.M. Weiss, Epstein-Barr virus-associated gastric adenocarcinoma. Am J Pathol, 1992. 140(4): p. 769-74.
Yanai, H., et al., Endoscopic and pathologic features of Epstein-Barr virus-associated gastric carcinoma. Gastrointest Endosc, 1997. 45(3): p. 236-42.
Fukayama, M. and T. Ushiku, Epstein-Barr virus-associated gastric carcinoma. Pathol Res Pract, 2011. 207(9): p. 529-37.
Ojima, H., et al., Discrepancy between clinical and pathological lymph node evaluation in Epstein-Barr virus-associated gastric cancers. Anticancer Res, 1996. 16(5B): p. 3081-4.
Song, H.J., et al., Host inflammatory response predicts survival of patients with Epstein-Barr virus-associated gastric carcinoma. Gastroenterology, 2010. 139(1): p. 84-92 e2.
Levine, P.H., et al., Elevated antibody titers to Epstein-Barr virus prior to the diagnosis of Epstein-Barr-virus-associated gastric adenocarcinoma. Int J Cancer, 1995. 60(5): p. 642-4.
Tamura, T., et al., Lymphoepithelioma-Like Carcinoma of the Stomach with Epithelioid Granulomas. Case Rep Gastroenterol, 2010. 4(3): p. 361-368.
Ushiku, T., et al., Gastric carcinoma with osteoclast-like giant cells. Lymphoepithelioma-like carcinoma with Epstein-Barr virus infection is the predominant type. Pathol Int, 2010. 60(8): p. 551-8.
Arikawa, J., et al., Morphological characteristics of Epstein-Barr virus-related early gastric carcinoma: a case-control study. Pathol Int, 1997. 47(6): p. 360-7.
Kaizaki, Y., et al., Atrophic gastritis, Epstein-Barr virus infection, and Epstein-Barr virus-associated gastric carcinoma. Gastric Cancer, 1999. 2(2): p. 101-108.
Chapel, F., et al., Epstein-Barr virus and gastric carcinoma in Western patients: comparison of pathological parameters and p53 expression in EBV-positive and negative tumours. Histopathology, 2000. 36(3): p. 252-61.
Kijima, Y., et al., The comparison of the prognosis between Epstein-Barr virus (EBV)-positive gastric carcinomas and EBV-negative ones. Cancer Lett, 2003. 200(1): p. 33-40.
Saiki, Y., et al., Immunophenotypic characterization of Epstein-Barr virus-associated gastric carcinoma: massive infiltration by proliferating CD8+ T-lymphocytes. Lab Invest, 1996. 75(1): p. 67-76.
Young, L.S. and A.B. Rickinson, Epstein-Barr virus: 40 years on. Nat Rev Cancer, 2004. 4(10): p. 757-68.
Chen, J.N., et al., Epstein-Barr virus-associated gastric carcinoma: a newly defined entity. J Clin Gastroenterol, 2012. 46(4): p. 262-71.
Ojima, H., et al., Infrequent overexpression of p53 protein in Epstein-Barr virus-associated gastric carcinomas. Jpn J Cancer Res, 1997. 88(3): p. 262-6.
Gulley, M.L., et al., Epstein-Barr virus infection is an early event in gastric carcinogenesis and is independent of bcl-2 expression and p53 accumulation. Hum Pathol, 1996. 27(1): p. 20-7.
Ohfuji, S., et al., Low frequency of apoptosis in Epstein-Barr virus-associated gastric carcinoma with lymphoid stroma. Int J Cancer, 1996. 68(6): p. 710-5.
Delecluse, H.J., et al., Epstein Barr virus-associated tumours: an update for the attention of the working pathologist. J Clin Pathol, 2007. 60(12): p. 1358-64.
Esteller, M., Epigenetics in cancer. N Engl J Med, 2008. 358(11): p. 1148-59.
Kusano, M., et al., Genetic, epigenetic, and clinicopathologic features of gastric carcinomas with the CpG island methylator phenotype and an association with Epstein-Barr virus. Cancer, 2006. 106(7): p. 1467-79.
Enomoto, S., et al., Lack of association between CpG island methylator phenotype in human gastric cancers and methylation in their background non-cancerous gastric mucosae. Cancer Sci, 2007. 98(12): p. 1853-61.
Chong, J.M., et al., Global and non-random CpG-island methylation in gastric carcinoma associated with Epstein-Barr virus. Cancer Sci, 2003. 94(1): p. 76-80.
Jung, E.J., et al., Lytic induction and apoptosis of Epstein-Barr virus-associated gastric cancer cell line with epigenetic modifiers and ganciclovir. Cancer Lett, 2007. 247(1): p. 77-83.
Yamamoto, N., et al., Epstein-Barr virus and gastric remnant cancer. Cancer, 1994. 74(3): p. 805-9.
Chang, M.S., et al., Microsatellite instability and Epstein-Barr virus infection in gastric remnant cancers. Pathol Int, 2000. 50(6): p. 486-92.
Baas, I.O., et al., Helicobacter pylori and Epstein-Barr virus infection and the p53 tumour suppressor pathway in gastric stump cancer compared with carcinoma in the non-operated stomach. J Clin Pathol, 1998. 51(9): p. 662-6.
Chen, J.N., et al., Epstein-Barr virus genome polymorphisms of Epstein-Barr virus-associated gastric carcinoma in gastric remnant carcinoma in Guangzhou, southern China, an endemic area of nasopharyngeal carcinoma. Virus Res, 2011. 160(1-2): p. 191-9.
van Beek, J., et al., EBV-positive gastric adenocarcinomas: a distinct clinicopathologic entity with a low frequency of lymph node involvement. J Clin Oncol, 2004. 22(4): p. 664-70.
Lee, H.L., et al., Treatment of Epstein-Barr virus-associated gastric carcinoma with endoscopic submucosal dissection. Gastrointest Endosc, 2011.
Tang, S.J., et al., Endoscopic Mucosal Resection of an Epstein-Barr Virus-Associated Lymphoepithelioma-Like Gastric Carcinoma. Dig Dis Sci, 2012.
Ishikura, H., et al., An AFP-producing gastric carcinoma with features of hepatic differentiation. A case report. Cancer, 1985. 56(4): p. 840-8.
Nagai, E., et al., Hepatoid adenocarcinoma of the stomach. A clinicopathologic and immunohistochemical analysis. Cancer, 1993. 72(6): p. 1827-35.
Gao, Y.B., et al., Preliminary study on the clinical and pathological relevance of gastric hepatoid adenocarcinoma. J Dig Dis, 2007. 8(1): p. 23-8.
Ishikura, H., et al., Gastrointestinal hepatoid adenocarcinoma: venous permeation and mimicry of hepatocellular carcinoma, a report of four cases. Histopathology, 1997. 31(1): p. 47-54.
Chang, Y.C., et al., alpha Fetoprotein producing early gastric cancer with liver metastasis: report of three cases. Gut, 1991. 32(5): p. 542-5.
Gitlin, D., A. Perricelli, and G.M. Gitlin, Synthesis of -fetoprotein by liver, yolk sac, and gastrointestinal tract of the human conceptus. Cancer Res, 1972. 32(5): p. 979-82.
Baek, S.K., et al., Clinicopathologic characteristics and treatment outcomes of hepatoid adenocarcinoma of the stomach, a rare but unique subtype of gastric cancer. BMC Gastroenterol, 2011. 11: p. 56.
Kodama, T., et al., Production of alpha-fetoprotein, normal serum proteins, and human chorionic gonadotropin in stomach cancer: histologic and immunohistochemical analyses of 35 cases. Cancer, 1981. 48(7): p. 1647-55.
Matsunou, H., et al., Alpha-fetoprotein-producing gastric carcinoma with enteroblastic differentiation. Cancer, 1994. 73(3): p. 534-40.
Motoyama, T., et al., alpha-Fetoprotein producing gastric carcinomas: a comparative study of three different subtypes. Acta Pathol Jpn, 1993. 43(11): p. 654-61.
Ooi, A., et al., Alpha-fetoprotein (AFP)-producing gastric carcinoma. Is it hepatoid differentiation? Cancer, 1990. 65(8): p. 1741-7.
Kinjo, T., et al., Histologic and immunohistochemical analyses of alpha-fetoprotein--producing cancer of the stomach. Am J Surg Pathol, 2012. 36(1): p. 56-65.
Liu, X., et al., Analysis of clinicopathologic features and prognostic factors in hepatoid adenocarcinoma of the stomach. Am J Surg Pathol, 2010. 34(10): p. 1465-71.
Hishinuma, M., et al., Hepatocellular oncofetal protein, glypican 3 is a sensitive marker for alpha-fetoprotein-producing gastric carcinoma. Histopathology, 2006. 49(5): p. 479-86.
Terracciano, L.M., et al., Hepatoid adenocarcinoma with liver metastasis mimicking hepatocellular carcinoma: an immunohistochemical and molecular study of eight cases. Am J Surg Pathol, 2003. 27(10): p. 1302-12.
Iso, Y., et al., Solitary AFP- and PIVKA-II-producing hepatoid gastric cancer with giant lymph node metastasis. Hepatogastroenterology, 2005. 52(66): p. 1930-2.
Lu, C.C., et al., Pure hepatoid adenocarcinoma of the stomach with spleen and lymph-node metastases. Am J Surg, 2010. 199(4): p. e42-4.
Filmus, J. and S.B. Selleck, Glypicans: proteoglycans with a surprise. J Clin Invest, 2001. 108(4): p. 497-501.
Kishimoto, T., et al., Hepatoid adenocarcinoma: a new clinicopathological entity and the hypotheses on carcinogenesis. Med Electron Microsc, 2000. 33(2): p. 57-63.
Akiyama, S., et al., Histogenesis of hepatoid adenocarcinoma of the stomach: molecular evidence of identical origin with coexistent tubular adenocarcinoma. Int J Cancer, 2003. 106(4): p. 510-5.
Kang, G.H. and Y.I. Kim, Alpha-fetoprotein-producing gastric carcinoma presenting focal hepatoid differentiation in metastatic lymph nodes. Virchows Arch, 1998. 432(1): p. 85-7.
Trevisani, F., et al., Etiologic factors and clinical presentation of hepatocellular carcinoma. Differences between cirrhotic and noncirrhotic Italian patients. Cancer, 1995. 75(9): p. 2220-32.
Okuda, K., R.L. Peters, and I.W. Simson, Gross anatomic features of hepatocellular carcinoma from three disparate geographic areas. Proposal of new classification. Cancer, 1984. 54(10): p. 2165-73.
Kamata, S., et al., Expression and localization of ATP binding cassette (ABC) family of drug transporters in gastric hepatoid adenocarcinomas. Histopathology, 2008. 52(6): p. 747-54.
Matsusaka, T., H. Watanabe, and M. Enjoji, Oat-cell carcinoma of the stomach. Fukuoka Igaku Zasshi, 1976. 67(2): p. 65-73.
Kim, K.O., et al., Clinical overview of extrapulmonary small cell carcinoma. J Korean Med Sci, 2006. 21(5): p. 833-7.
Matsubayashi, H., et al., Advanced gastric glandular-endocrine cell carcinoma with 1-year survival after gastrectomy. Gastric Cancer, 2000. 3(4): p. 226-233.
Namikawa, T., et al., Primary gastric small cell carcinoma: report of a case and review of the literature. Med Mol Morphol, 2005. 38(4): p. 256-61.
O'Byrne, K.J., et al., Extrapulmonary small cell gastric carcinoma. A case report and review of the literature. Acta Oncol, 1997. 36(1): p. 78-80.
Chejfec, G. and V.E. Gould, Malignant gastric neuroendogrinomas. Ultrastructural and biochemical characterization of their secretory activity. Hum Pathol, 1977. 8(4): p. 433-40.
Okita, N.T., et al., Neuroendocrine tumors of the stomach: chemotherapy with cisplatin plus irinotecan is effective for gastric poorly-differentiated neuroendocrine carcinoma. Gastric Cancer, 2011. 14(2): p. 161-5.
Murakami, H., et al., Paraneoplastic neurological syndrome in a patient with gastric cancer. Gastric Cancer, 2010. 13(3): p. 204-8.
Hirata, Y., et al., Gastric carcinoid with ectopic production of ACTH and beta-MSH. Cancer, 1976. 37(1): p. 377-85.
Matsui, T., et al., A case of small cell carcinoma of the stomach. Hepatogastroenterology, 1997. 44(13): p. 156-60.
Arakawa, A., et al., Small cell carcinoma of the stomach: a case report. Radiat Med, 1997. 15(5): p. 321-5.
Arai, K. and M. Matsuda, Gastric small-cell carcinoma in Japan: a case report and review of the literature. Am J Clin Oncol, 1998. 21(5): p. 458-61.
Matsui, K., et al., Small cell carcinoma of the stomach: a clinicopathologic study of 17 cases. Am J Gastroenterol, 1991. 86(9): p. 1167-75.
Chiba, N., et al., Advanced gastric endocrine cell carcinoma with distant lymph node metastasis: a case report and clinicopathological characteristics of the disease. Gastric Cancer, 2004. 7(2): p. 122-7.
Fukuda, T., et al., Early gastric cancer of the small cell type. Am J Gastroenterol, 1988. 83(10): p. 1176-9.
Hanada, N., et al., [A case of partial response in liver metastatic lesion from gastric endocrine cell carcinoma treated with TS-1]. Gan To Kagaku Ryoho, 2006. 33(8): p. 1143-6.
Tsushima, T., et al., [A case of metastatic gastric endocrine cell carcinoma which could be curably resected after chemotherapy with S-1/CDDP]. Gan To Kagaku Ryoho, 2008. 35(5): p. 817-20.
Rindi, G., et al., ECL cell tumor and poorly differentiated endocrine carcinoma of the stomach: prognostic evaluation by pathological analysis. Gastroenterology, 1999. 116(3): p. 532-42.
Huang, J., et al., Primary small cell carcinoma of the stomach: An experience of two decades (1990-2011) in a Chinese cancer institute. J Surg Oncol, 2012.
Matsui, K., et al., Clinicopathologic features of neuroendocrine carcinomas of the stomach: appraisal of small cell and large cell variants. Arch Pathol Lab Med, 1998. 122(11): p. 1010-7.
Tsuneyama, K., et al., A case report of gastric carcinosarcoma with rhabdomyosarcomatous and neuroendocrinal differentiation. Pathol Res Pract, 1999. 195(2): p. 93-7; discussion 98.
Yamazaki, K., A gastric carcinosarcoma with neuroendocrine cell differentiation and undifferentiated spindle-shaped sarcoma component possibly progressing from the conventional tubular adenocarcinoma; an immunohistochemical and ultrastructural study. Virchows Arch, 2003. 442(1): p. 77-81.
Kuroda, N., et al., Gastric carcinosarcoma with neuroendocrine differentiation as the carcinoma component and leiomyosarcomatous and myofibroblastic differentiation as the sarcomatous component. APMIS, 2006. 114(3): p. 234-8.
Haratake, J., A. Horie, and S. Inoshita, Gastric small cell carcinoma with squamous and neuroendocrine differentiation. Pathology, 1992. 24(2): p. 116-20.
Travis, W., T. Colby, and Y. Shimosato, Histological Classification of Lung and Pleural tumors. 1st ed. WHO/IASLC. 1999, Berlin, Germany: Springer.
Brenner, B., et al., Small-cell carcinomas of the gastrointestinal tract: a review. J Clin Oncol, 2004. 22(13): p. 2730-9.
Bartley, A.N., et al., Neuroendocrine and mucinous differentiation in signet ring cell carcinoma of the stomach: evidence for a common cell of origin in composite tumors. Hum Pathol, 2011. 42(10): p. 1420-9.
Ho, K.J., et al., Small cell carcinoma of the esophagus: evidence for a unified histogenesis. Hum Pathol, 1984. 15(5): p. 460-8.
Hussein, A.M., C.L. Otrakji, and B.T. Hussein, Small cell carcinoma of the stomach. Case report and review of the literature. Dig Dis Sci, 1990. 35(4): p. 513-8.
Nishikura, K., et al., Carcinogenesis of gastric endocrine cell carcinoma: analysis of histopathology and p53 gene alteration. Gastric Cancer, 2003. 6(4): p. 203-9.
Kusayanagi, S., et al., Primary small cell carcinoma of the stomach. J Gastroenterol Hepatol, 2003. 18(6): p. 743-7.
Shpaner, A. and T.E. Yusuf, Primary gastric small-cell neuroendocrine carcinoma. Endoscopy, 2007. 39 Suppl 1: p. E310-1.
Li, A.F., et al., Small cell carcinomas in gastrointestinal tract: immunohistochemical and clinicopathological features. J Clin Pathol, 2010. 63(7): p. 620-5.
Hamilton, S. and L. Aaltonen, WHO Classification of Tumors of the Digestive System. IARC WHO Classification of Tumors. 2010, Lyon: IARC Press.
Jiang, S.X., et al., Gastric large cell neuroendocrine carcinomas: a distinct clinicopathologic entity. Am J Surg Pathol, 2006. 30(8): p. 945-53.
Shia, J., et al., Is nonsmall cell type high-grade neuroendocrine carcinoma of the tubular gastrointestinal tract a distinct disease entity? Am J Surg Pathol, 2008. 32(5): p. 719-31.
Murayama, H., T. Imai, and M. Kikuchi, Solid carcinomas of the stomach. A combined histochemical, light and electron microscopic study. Cancer, 1983. 51(9): p. 1673-81.
Ordonez, N.G., Value of thyroid transcription factor-1 immunostaining in distinguishing small cell lung carcinomas from other small cell carcinomas. Am J Surg Pathol, 2000. 24(9): p. 1217-23.
Solcia, E., G. Kloppel, and L.H. Sobin, Histological Typing of Endocrine Tumors (WHO International Histological Classification of Tumours). 2nd ed. 2000, Berlin, Germany: Springer.
Won, O.H., et al., Squamous cell carcinoma of the stomach. Am J Gastroenterol, 1978. 69(5): p. 594-8.
Hoshi, K., et al., [Specific type stomach neoplasms: report based on the questionnaires of the 40th Stomach Cancer Research Group]. Nihon Gan Chiryo Gakkai Shi, 1983. 18(8): p. 2112-24.
Lim, S.M., et al., A case of synchronous squamous cell carcinoma in the esophagus and stomach. Gut Liver, 2012. 6(1): p. 118-21.
Dursun, M., et al., Primary squamous cell carcinoma of the stomach: a case report and review of the literature. Eur J Gastroenterol Hepatol, 2003. 15(3): p. 329-30.
Hara, J., et al., Exophytic primary squamous cell carcinoma of the stomach. J Gastroenterol, 2004. 39(3): p. 299-300.
Marubashi, S., et al., Primary squamous cell carcinoma of the stomach. Gastric Cancer, 1999. 2(2): p. 136-141.
Oono, Y., et al., Primary gastric squamous cell carcinoma in situ originating from gastric squamous metaplasia. Endoscopy, 2010. 42 Suppl 2: p. E290-1.
Parks, R.E., Squamous neoplasms of the stomach. Am J Roentgenol Radium Ther Nucl Med, 1967. 101(2): p. 447-9.
Japanese classification of gastric carcinoma: 3rd English edition. Gastric Cancer, 2011. 14(2): p. 101-12.
Schmidt, C., et al., Primary squamous cell carcinoma of the stomach. Report of a case and review of literature. Hepatogastroenterology, 2001. 48(40): p. 1033-6.
Straus, R., S. Heschel, and D.J. Fortmann, Primary adenosquamous carcinoma of the stomach. A case report and review. Cancer, 1969. 24(5): p. 985-95.
Mingazzini, P.L., P. Barsotti, and F. Malchiodi Albedi, Adenosquamous carcinoma of the stomach: histological, histochemical and ultrastructural observations. Histopathology, 1983. 7(3): p. 433-43.
Tokuhara, K., et al., Primary squamous cell carcinoma in the gastric remnant. Surg Today, 2012. 42(7): p. 666-9.
Davidsohn, C., Chorioepithelium und Magekrebs, eine seltene Verschmelzung zweier bosartiger Geschwulste. Charite Ann, 1905. 29: p. 12.
Unakami, M., et al., [3 cases of malignant choriocarcinoma originating in the stomach]. Gan No Rinsho, 1982. 28(3): p. 204-10.
Mori, H., et al., Choriocarcinomatous change with immunocytochemically HCG-positive cells in the gastric carcinoma of the males. Virchows Arch A Pathol Anat Histol, 1982. 396(2): p. 141-53.
Saigo, P.E., et al., Primary gastric choriocarcinoma. An immunohistological study. Am J Surg Pathol, 1981. 5(4): p. 333-42.
Jindrak, K., J.F. Bochetto, and L.I. Alpert, Primary gastric choriocarcinoma: case report with review of world literature. Hum Pathol, 1976. 7(5): p. 595-604.
Satake, N., et al., Gastric cancer with choriocarcinoma and yolk sac tumor components: case report. Pathol Int, 2011. 61(3): p. 156-60.
Hirano, Y., et al., Combined choriocarcinoma, neuroendocrine cell carcinoma and tubular adenocarcinoma in the stomach. World J Gastroenterol, 2008. 14(20): p. 3269-72.
Motoyama, T., et al., Coexistence of choriocarcinoma and hepatoid adenocarcinoma in the stomach. Pathol Int, 1994. 44(9): p. 716-21.
Garcia, R.L. and V.S. Ghali, Gastric choriocarcinoma and yolk sac tumor in a man: observations about its possible origin. Hum Pathol, 1985. 16(9): p. 955-8.
Kobayashi, A., et al., Primary gastric choriocarcinoma: two case reports and a pooled analysis of 53 cases. Gastric Cancer, 2005. 8(3): p. 178-85.
Senju, M., et al., [Case of advanced gastric cancer with remarkable gynecomastia and marked elevation of serum human chorionic gonadotropin (HCG)]. Nihon Shokakibyo Gakkai Zasshi, 1983. 80(6): p. 1331-5.
Dye, D.W., R. Broadwater, and L.W. Lamps, Uncommon malignancies: case 2. Gastric choriocarcinoma. J Clin Oncol, 2005. 23(25): p. 6251-3.
Krulewski, T. and L.B. Cohen, Choriocarcinoma of the stomach: pathogenesis and clinical characteristics. Am J Gastroenterol, 1988. 83(10): p. 1172-5.
Noguchi, T., et al., A patient with primary gastric choriocarcinoma who received a correct preoperative diagnosis and achieved prolonged survival. Gastric Cancer, 2002. 5(2): p. 112-7.
Tomita, K. and M. Kuwajima, Chorionic gonadotropin in gastric cancer tissue, especially its relation to the patient's prognosis. Gan No Rinsho, 1981. 27: p. 2.
Yakeishi, Y., M. Mori, and M. Enjoji, Distribution of beta-human chorionic gonadotropin-positive cells in noncancerous gastric mucosa and in malignant gastric tumors. Cancer, 1990. 66(4): p. 695-701.
Walker, L.R. and B. Erler, Gastric cancer in the setting of persistently elevated human chorionic gonadotropin: a case report. Case Rep Obstet Gynecol, 2011. 2011: p. 350318.
Fujimoto, S., et al., The presence of an aberrant type of human chorionic gonadotropin in patients with gastric or colorectal cancer. Jpn J Surg, 1987. 17(5): p. 382-7.
Arai, O., et al., [A case of advanced gastric cancer growing extramurally with gynecomastia and high hCG-beta serum level]. Gan To Kagaku Ryoho, 2010. 37(7): p. 1369-72.
Matsunaga, N., et al., Primary choriocarcinoma of the stomach presenting as gastrointestinal hemorrhage: report of a case. Radiat Med, 1989. 7(5): p. 220-2.
Masuda, R., et al., [A primary choriocarcinoma of the stomach with a review of 45 cases in Japan]. Gan No Rinsho, 1990. 36(9): p. 1025-30.
Pick, L., Uber die chorioepthelahnlich metastsierende from desmagencarcinomas. Klin Wochenscher, 1926. 5: p. 1.
Liu, A.Y., et al., Gastric choriocarcinoma shows characteristics of adenocarcinoma and gestational choriocarcinoma: a comparative genomic hybridization and fluorescence in situ hybridization study. Diagn Mol Pathol, 2001. 10(3): p. 161-5.
Queckenstedt, H., Ueber Karzinsakome Leipzig. 1984.
Tanimura, H. and M. Furuta, Carcinosarcoma of the stomach. Am J Surg, 1967. 113(5): p. 702-9.
Nakayama, Y., et al., Gastric carcinosarcoma (sarcomatoid carcinoma) with rhabdomyoblastic and osteoblastic differentiation. Pathol Int, 1997. 47(8): p. 557-63.
Sato, Y., et al., Gastric carcinosarcoma, coexistence of adenosquamous carcinoma and rhabdomyosarcoma: a case report. Histopathology, 2001. 39(5): p. 543-4.
Ikeda, Y., et al., Gastric carcinosarcoma presenting as a huge epigastric mass. Gastric Cancer, 2007. 10(1): p. 63-8.
Bansal, M., M. Kaneko, and R.E. Gordon, Carcinosarcoma and separate carcinoid tumor of the stomach. A case report with light and electron microscopic studies. Cancer, 1982. 50(9): p. 1876-81.
Kumagai, K., et al., [A case of so-called carcinosarcoma of the stomach]. Gan No Rinsho, 1984. 30(15): p. 1931-6.
Hanada, M., et al., Carcinosarcoma of the stomach. A case report with light microscopic, immunohistochemical, and electron microscopic study. Acta Pathol Jpn, 1985. 35(4): p. 951-9.
Robey-Cafferty, S.S., et al., Sarcomatoid carcinoma of the stomach. A report of three cases with immunohistochemical and ultrastructural observations. Cancer, 1990. 65(7): p. 1601-6.
Minamoto, T., et al., [So-called gastric carcinosarcoma--a case of chondrosarcomatous undifferentiated in the metastatic foci]. Gan No Rinsho, 1984. 30(10): p. 1321-6.
Teramachi, K., et al., Carcinosarcoma (pure endocrine cell carcinoma with sarcoma components) of the stomach. Pathol Int, 2003. 53(8): p. 552-6.
Melato, M., et al., Carcinosarcoma and separate neuroendocrine malignant tumor of a malignancy promoter, the gastric stump. Anticancer Res, 1993. 13(6B): p. 2485-8.
Cho, K.J., et al., Carcinosarcoma of the stomach. A case report with light microscopic, immunohistochemical, and electron microscopic study. APMIS, 1990. 98(11): p. 991-5.
Matsukuma, S., et al., Gastric stump carcinosarcoma with rhabdomyosarcomatous differentiation. Pathol Int, 1997. 47(1): p. 73-7.
Olinici, C.D., et al., Heterotopic bone formation in gastric carcinoma. Case report and discussion of the literature. Rom J Gastroenterol, 2002. 11(4): p. 331-3.
Yasuma, T., et al., Bone formation and calcification in gastric cancer--case report and review of literature. Acta Pathol Jpn, 1973. 23(1): p. 155-72.
Sun, Y. and P.G. Wasserman, Acinar cell carcinoma arising in the stomach: a case report with literature review. Hum Pathol, 2004. 35(2): p. 263-5.
Mizuno, Y., et al., Acinar cell carcinoma arising from an ectopic pancreas. Surg Today, 2007. 37(8): p. 704-7.
Klimstra, D.S., et al., Acinar cell carcinoma of the pancreas. A clinicopathologic study of 28 cases. Am J Surg Pathol, 1992. 16(9): p. 815-37.
Ordonez, N.G., Pancreatic acinar cell carcinoma. Adv Anat Pathol, 2001. 8(3): p. 144-59.
Fukunaga, M., Gastric carcinoma resembling pancreatic mixed acinar-endocrine carcinoma. Hum Pathol, 2002. 33(5): p. 569-73.
Chetty, R. and I. Weinreb, Gastric neuroendocrine carcinoma arising from heterotopic pancreatic tissue. J Clin Pathol, 2004. 57(3): p. 314-7.
Emerson, L., et al., Adenocarcinoma arising in association with gastric heterotopic pancreas: A case report and review of the literature. J Surg Oncol, 2004. 87(1): p. 53-7.
Jeong, H.Y., et al., Adenocarcinoma arising from an ectopic pancreas in the stomach. Endoscopy, 2002. 34(12): p. 1014-7.
Song, D.E., et al., Adenocarcinoma arising in gastric heterotopic pancreas: a case report. J Korean Med Sci, 2004. 19(1): p. 145-8.
Tsapralis, D., et al., Pancreatic intraductal papillary mucinous neoplasm with concomitant heterotopic pancreatic cystic neoplasia of the stomach: a case report and review of the literature. Diagn Pathol, 2010. 5: p. 4.
Osanai, M., et al., Adenocarcinoma arising in gastric heterotopic pancreas: clinicopathological and immunohistochemical study with genetic analysis of a case. Pathol Int, 2001. 51(7): p. 549-54.
Doglioni, C., et al., Pancreatic (acinar) metaplasia of the gastric mucosa. Histology, ultrastructure, immunocytochemistry, and clinicopathologic correlations of 101 cases. Am J Surg Pathol, 1993. 17(11): p. 1134-43.
Faller, G. and T. Kirchner, Immunological and morphogenic basis of gastric mucosa atrophy and metaplasia. Virchows Arch, 2005. 446(1): p. 1-9.
Wang, H.H., et al., Prevalence and significance of pancreatic acinar metaplasia at the gastroesophageal junction. Am J Surg Pathol, 1996. 20(12): p. 1507-10.
Krishnamurthy, S., et al., Pancreatic acinar cell clusters in pediatric gastric mucosa. Am J Surg Pathol, 1998. 22(1): p. 100-5.
Ambrosini-Spaltro, A., et al., Pancreatic-type acinar cell carcinoma of the stomach beneath a focus of pancreatic metaplasia of the gastric mucosa. Hum Pathol, 2009. 40(5): p. 746-9.
Jain, D., et al., Composite glandular and endocrine tumors of the stomach with pancreatic acinar differentiation. Am J Surg Pathol, 2005. 29(11): p. 1524-9.
Kusafuka, K., et al., Pancreatic-type mixed acinar-endocrine carcinoma with alpha-fetoprotein production arising from the stomach: a report of an extremely rare case. Med Mol Morphol, 2009. 42(3): p. 167-74.
Shimoda, M., et al., Primary invasive micropapillary carcinoma of the stomach. Pathol Int, 2008. 58(8): p. 513-7.
Roh, J.H., et al., Micropapillary carcinoma of stomach: a clinicopathologic and immunohistochemical study of 11 cases. Am J Surg Pathol, 2010. 34(8): p. 1139-46.
De la Cruz, C., et al., Invasive micropapillary carcinoma of the breast: clinicopathological and immunohistochemical study. Pathol Int, 2004. 54(2): p. 90-6.
Amin, M.B., et al., Micropapillary variant of transitional cell carcinoma of the urinary bladder. Histologic pattern resembling ovarian papillary serous carcinoma. Am J Surg Pathol, 1994. 18(12): p. 1224-32.
Oh, Y.L. and K.R. Kim, Micropapillary variant of transitional cell carcinoma of the ureter. Pathol Int, 2000. 50(1): p. 52-6.
Amin, M.B., et al., Micropapillary component in lung adenocarcinoma: a distinctive histologic feature with possible prognostic significance. Am J Surg Pathol, 2002. 26(3): p. 358-64.
Michal, M., A. Skalova, and P. Mukensnabl, Micropapillary carcinoma of the parotid gland arising in mucinous cystadenoma. Virchows Arch, 2000. 437(4): p. 465-8.
Verdu, M., et al., Clinicopathological and molecular characterization of colorectal micropapillary carcinoma. Mod Pathol, 2011. 24(5): p. 729-38.
Kim, M.J., et al., Immunohistochemical and clinicopathologic characteristics of invasive ductal carcinoma of breast with micropapillary carcinoma component. Arch Pathol Lab Med, 2005. 129(10): p. 1277-82.
Nassar, H., et al., Pathogenesis of invasive micropapillary carcinoma: role of MUC1 glycoprotein. Mod Pathol, 2004. 17(9): p. 1045-50.
Lee, J.H., et al., The presence of a micropapillary component predicts aggressive behaviour in early and advanced gastric adenocarcinomas. Pathology, 2010. 42(6): p. 560-3.
Sangoi, A.R., et al., Immunohistochemical comparison of MUC1, CA125, and Her2Neu in invasive micropapillary carcinoma of the urinary tract and typical invasive urothelial carcinoma with retraction artifact. Mod Pathol, 2009. 22(5): p. 660-7.
Ohtsuki, Y., et al., KL-6 is another useful marker in assessing a micropapillary pattern in carcinomas of the breast and urinary bladder, but not the colon. Med Mol Morphol, 2009. 42(2): p. 123-7.
Samaratunga, H. and K. Khoo, Micropapillary variant of urothelial carcinoma of the urinary bladder; a clinicopathological and immunohistochemical study. Histopathology, 2004. 45(1): p. 55-64.
Eom, D.W., et al., Gastric micropapillary carcinoma: A distinct subtype with a significantly worse prognosis in TNM stages I and II. Am J Surg Pathol, 2011. 35(1): p. 84-91.
Alvarado-Cabrero, I., et al., Micropapillary carcinoma of the urothelial tract. A clinicopathologic study of 38 cases. Ann Diagn Pathol, 2005. 9(1): p. 1-5.
Kim, M.J., et al., Invasive colorectal micropapillary carcinoma: an aggressive variant of adenocarcinoma. Hum Pathol, 2006. 37(7): p. 809-15.
Haupt, B., et al., Colorectal adenocarcinoma with micropapillary pattern and its association with lymph node metastasis. Mod Pathol, 2007. 20(7): p. 729-33.
Comperat, E., et al., Micropapillary urothelial carcinoma of the urinary bladder: a clinicopathological analysis of 72 cases. Pathology, 2010. 42(7): p. 650-4.
Lotan, T.L., et al., Immunohistochemical panel to identify the primary site of invasive micropapillary carcinoma. Am J Surg Pathol, 2009. 33(7): p. 1037-41.
McQuitty, E., et al., Lymphovascular invasion in micropapillary urothelial carcinoma: a study of 22 cases. Arch Pathol Lab Med, 2012. 136(6): p. 635-9.
Ueyama, H., et al., Gastric adenocarcinoma of fundic gland type (chief cell predominant type): proposal for a new entity of gastric adenocarcinoma. Am J Surg Pathol, 2010. 34(5): p. 609-19.
Singhi, A.D., A.J. Lazenby, and E.A. Montgomery, Gastric adenocarcinoma with chief cell differentiation: a proposal for reclassification as oxyntic gland polyp/adenoma. Am J Surg Pathol, 2012. 36(7): p. 1030-5.
Capella, C., et al., Gastric parietal cell carcinoma--a newly recognized entity: light microscopic and ultrastructural features. Histopathology, 1984. 8(5): p. 813-24.
Takubo, K., et al., Oncocytic adenocarcinoma of the stomach: parietal cell carcinoma. Am J Surg Pathol, 2002. 26(4): p. 458-65.
Yang, G.Y., et al., Parietal cell carcinoma of gastric cardia: immunophenotype and ultrastructure. Ultrastruct Pathol, 2003. 27(2): p. 87-94.
Robey-Cafferty, S.S., J.Y. Ro, and E.G. McKee, Gastric parietal cell carcinoma with an unusual, lymphoma-like histologic appearance: report of a case. Mod Pathol, 1989. 2(5): p. 536-40.
Barbosa, A.J., et al., Parietal cell carcinoma of the stomach and Menetrier's disease. Arq Gastroenterol, 1987. 24(1): p. 36-40.
Byrne, D., M.P. Holley, and A. Cuschieri, Parietal cell carcinoma of the stomach: association with long-term survival after curative resection. Br J Cancer, 1988. 58(1): p. 85-7.
Gaffney, E.F., Favourable prognosis in gastric carcinoma with parietal cell differentiation. Histopathology, 1987. 11(2): p. 217-8.
Hedenbro, J.L., I. Hagerstand, and V. Rychterova, Parietal cell carcinoma. A new differential diagnosis for submucosal gastric tumors. Endoscopy, 1990. 22(1): p. 47-8.
Hayashi, I., et al., Mucoepidermoid carcinoma of the stomach. J Surg Oncol, 1987. 34(2): p. 94-9.
Ooi, A., et al., Predominant Paneth cell differentiation in an intestinal type gastric cancer. Pathol Res Pract, 1991. 187(2-3): p. 220-5.
Lev, R. and T.D. DeNucci, Neoplastic Paneth cells in the stomach. Report of two cases and review of the literature. Arch Pathol Lab Med, 1989. 113(2): p. 129-33.
Inada, K.I., et al., Paneth type gastric cancer cells exhibit expression of human defensin-5. Histopathology, 2005. 47(3): p. 330-1.
Stracca-Pansa, V., et al., Gastric carcinoma with osteoclast-like giant cells. Report of four cases. Am J Clin Pathol, 1995. 103(4): p. 453-9.
Baschinsky, D.Y., W.L. Frankel, and T.H. Niemann, Gastric carcinoma with osteoclast-like giant cells. Am J Gastroenterol, 1999. 94(6): p. 1678-81.