Terminology and Classification

The terminology for neuroendocrine neoplasms has always been problematic. For well-differentiated tumors of the GI tract, the term carcinoid tumor has been widely used for more than a century, although some authorities advocate restricting the term to serotonin-producing NETs. Pancreatic tumors used to be designated as islet cell tumors. The preferred term,¹ which is endorsed by the World Health Organization (WHO) 2010 classification,² is well-differentiated neuroendocrine tumor (WDNET). WDNETs must be distinguished from poorly differentiated neuroendocrine carcinomas (PDNECs), such as small cell carcinoma and large cell neuroendocrine carcinoma. Designation as a WDNET is independent of the stage of the neoplasm, which represents a change from the WHO classifications of 2000 and 2004, which advocated the term well-differentiated neuroendocrine carcinoma for WDNETs with metastases.

Except for incipient neoplasms (i.e., tumorlets), all WDNETs should be considered malignant neoplasms. Low-grade and early-stage tumors often follow a protracted or benign clinical course after initial resection. Until recently, some authorities had classified these low-grade, early-stage lesions as benign (i.e., adenomas), citing their low incidence of recurrent or metastatic behavior, and this approach had been incorporated in an earlier WHO classification of pancreatic neuroendocrine tumors (PanNETs).³ Other investigators have maintained that even the rare occurrence of metastasis is an adequate sign of malignant potential, and they have proposed classifying the entire spectrum of NETs as carcinoma.^4–6

As experience with the long-term behavior of these tumors has improved, it has become clear that WDNETs may be more aggressive than previously believed. In any case, they remain clinically, pathologically, and genetically distinct from PDNECs, and current recommendations restrict the use of the term carcinoma to the poorly differentiated group (not to metastasis or dissemination of WDNET) to avoid confusing these fundamentally distinct families of neuroendocrine neoplasms.

In reporting WDNETs of the GI tract (i.e., carcinoids) and pancreas (i.e., islet cell tumors), the preferred diagnostic expression is “WDNET (carcinoid tumor)” for the former and “WD pancreatic NET” for the latter. The types included in this group according to the WHO 2010 classification are NET grade 1 (Ki67 proliferation index less than 3% and mitotic index less than 2 mitoses per 10 high-power fields [HPFs]) and NET grade 2 (Ki67 index of 3% to 20% or 2 to 20 mitoses/10 HPFs).

Some morphologically well-differentiated NETs fall into the grade 3 category by virtue of their Ki67 and mitotic indices. Based on preliminary evidence,⁷ it appears that these rare examples are more closely related to WDNETs than to PDNECs. As such, the grade 3 category defined by the cutoff of 20% Ki67 or 20 mitoses per 10 HPFs includes two distinct but occasionally overlapping entities: highly proliferative WDNETs and true PDNECs of small cell and large cell neuroendocrine types similar to those in the lung. The latter group typically has a very high Ki67-labeling index (diffusely above 50%), very aggressive behavior, and favorable but short-lived responses to platinum-based chemotherapy. One study demonstrated that grade 3 neuroendocrine neoplasms with a Ki67 index in the 20% to 55% range had a better survival rate than those with a Ki67 index greater than 55%, but the less proliferative group did not respond to platinum-based chemotherapy.⁸

Further refinement of the grading cutoff points is needed. Data are being accumulated to allow better stratification based on the biologic, genetic, and therapeutic subgroups of WDNETs and PDNECs. The current approach has been endorsed by the WHO 2010 guidelines,² College of American Pathologists (CAP),⁹ and American Joint Committee on Cancer (AJCC),¹⁰ and all NETs are graded and staged separately as for any other cancer type.

The most widely used staging systems for GI NETs are those from the CAP,⁹ AJCC,¹⁰ and European Neuroendocrine Tumor Society (ENETS).^11,12 The prognostic value of these systems is still being verified. In reporting NETs, it is important to specify which staging system is used.

Neuroendocrine Tumor Differentiation

Neuroendocrine tumor differentiation manifests in various ways in the GI and pancreatobiliary tracts (Box 29.1). Many of the characteristics of these lesions are explored in this chapter, and more detailed discussions can be found in other chapters.

Early neuroendocrine cell proliferations can be seen throughout the GI tract. This phenomenon is best recognized in the stomach, where a spectrum of ECL cell proliferations occurs in the setting of hypergastrinemia and is usually compensatory for atrophic gastritis-related hypochlorhydria. In this condition, the trophic effects of gastrin lead to a spectrum of ECL cell proliferations, ranging from diffuse, linear, or nodular hyperplasia (Fig. 29.1) to dysplasia (i.e., in situ [Tis]) and to full-blown WDNETs. It is difficult to draw sharp lines between these processes, but criteria (albeit arbitrary) have been proposed. For microscopic proliferations, if there is nodular growth of ECL cells larger than 150 µm or a conglomeration of nodules, signs of microinfiltration, and new stroma, the lesion may be reported as “dysplasia or Tis.” These proliferations also can be designated “ECL cell proliferation with micronodule formation,” with further characteristics such as the size detailed in a comment. It is essential to look for signs of atrophy in the background and to investigate the clinical setting to determine the biologic significance of the lesion. The type 1 and 2 WDNETs that arise in this setting appear to be biologically different and more indolent than sporadic WDNETs (see Site-Specific Features, Stomach).

Another example of an incipient NET occurs in the pancreas. Multiple NETs may occur in patients with multiple endocrine neoplasia type I (MEN I) syndrome, but they can also occur without any syndromic background. Numerous microscopic nodules that appear to be precursor lesions often develop in these patients. Larger ones (<0.5 cm) are referred to as microadenomas (Fig. 29.2). They are often distinguished by morphology different from the background islets, a fibrous band around them, and clonal labeling for one of the pancreatic hormones; normal islets typically are shown to express multiple hormones in an established distribution by immunohistochemistry. As in the gastric ECL cell proliferations, it is often difficult to appreciate where the hyperplasia ends and true autonomous proliferation of the neoplasm begins.

Incidental minute foci of neuroendocrine cell clusters encountered in the wall of the appendix are probably examples of the same proliferative phenomenon (Fig. 29.3). There are no established guidelines for the terminology or classification of these lesions, and there are no known underlying medical or genetic conditions that predispose to the proliferation. Patchy, small clusters that are less than a millimeter in the greatest dimension are typically reported as an “incidental neuroendocrine cell proliferation,” and a comment regarding their presumed benign nature should be provided.

Pathologic Features

Grossly, WDNETs are usually well demarcated and grow with pushing borders. Rarely, PanNETs, especially early-stage and indolent tumors, may have capsule-like fibrous tissue surrounding the tumor. WDNETs are typically cellular, stroma-poor tumors, and cut sections reveal a fleshy, homogenous appearance (Fig. 29.5). The vascularity may lead to darker colors and hemorrhagic foci, especially if the tumor has been manipulated. In formalin-fixed specimens, the lesions become yellow to white. Necrosis and mucosal ulceration can be seen in more advanced cases and are typically signs of aggressive behavior. Some, particularly ileal NETs, can be associated with abundant sclerosis.

In the GI tract, WDNETs often form mucosa-covered, broad-based, polypoid lesions, with the bulk of the lesion in the submucosa and muscularis. PanNETs often protrude into the peripancreatic soft tissues, although rare examples may arise from the pancreatic duct wall and appear as sclerotic lesions constricting the duct, causing secondary dilation of upstream ducts, and mimicking intraductal papillary mucinous neoplasms (IPMNs).^21,22

WDNETs typically are solid tumors. However, in the pancreas, as many as 10% manifest as a cystic mass caused by central degeneration, leaving a cuff of histologically conventional NET clinging to the cyst wall.²³

Microscopically, WDNETs are composed of uniform, round cells, with a moderate amount of cytoplasm and a coarsely granular, salt-and-pepper pattern of chromatin (Fig. 29.6). The latter finding is probably the most specific diagnostic feature of these tumors. Cytoplasmic granules may be evident and are especially prominent in midgut examples, along with melanin or lipofuscin pigments (Fig. 29.7). The cells grow in nests, acini, rosettes, ribbons, festoons, and trabeculae (Fig. 29.8). Gland formation by the tumor cells is common. Delicate vasculature is another hallmark, especially in cases with a prominent nested pattern (Fig. 29.9). Artifactual clefting around the nests is common, particularly in intestinal examples.

Morphologic Variants

In most of the numerous morphologic variants of WDNETs, at least one of the three classic morphologic characteristics (i.e., cellular monotony, cytoplasmic abundance, and salt-and-pepper chromatin pattern) is retained. Some WDNETs have prominent clear cytoplasm, a hallmark of VHL-associated PanNETs,²⁴ although it is not entirely specific. In some clear cell NETs, the cytoplasm contains abundant lipid, producing a microvesicular cytoplasm and creating a picture reminiscent of adrenocortical cells. These lipid-rich variants (Fig. 29.10) often lack the nuclear features characteristic of NETs, potentially leading to misdiagnosis.²⁵ In other NETs, the cytosolic contents, mostly intermediate filaments,²⁶ may push the nucleus to the periphery and create a rhabdoid or plasmacytoid appearance (Fig. 29.11), mimicking signet ring cells, but intracellular mucin is lacking.

Some NETs, especially in the pancreas and occasionally in the stomach or rectum, have oncocytic features characterized by abundant acidophilic granular cytoplasm and single, prominent, eccentric nucleoli. When metastatic to liver, oncocytic NETs may be mistaken for hepatocellular carcinomas (Fig. 29.12). Some cases have a high nucleus-to-cytoplasm ratio that imparts a small, blue cell appearance. These cases often have a more diffuse growth pattern, which further accentuates the concern about a high-grade neoplasm, especially based on small biopsies. The Ki67 labeling index is helpful in establishing the well-differentiated nature of these samples.

The endocrine atypia characteristic of normal endocrine organs may be seen in WDNETs as large, bizarre, pleomorphic nuclei with smudgy chromatin (Fig. 29.13).²⁷ This pattern mimics the degenerative atypia seen in ancient schwannomas (i.e., rare, encapsulated, neural sheath tumor variants) and symplastic leiomyomas (i.e., atypical, bizarre, smooth muscle tumor variants). Pleomorphism in NETs can lead to their misdiagnosis as more aggressive neoplasms, but this finding has no prognostic importance unless accompanied by an elevated proliferative index rate.²⁷

Stromal alterations may add variety to the morphologic spectrum of WDNETs. In some cases, these already hypervascular tumors become massively congested or hemorrhagic, creating a peliotic appearance. In others, particularly in the stomach, the stroma can exhibit myxoid features (Fig. 29.14). Although most WDNETs are fundamentally stroma-poor tumors, they can exhibit stromal sclerosis, especially the ileal NETs and PanNETs, which secrete serotonin (Fig. 29.15).²⁸

Rosette or glandular formations can be prominent in some WDNETs. In ampullary somatostatinomas, gland formation is so characteristic that the name glandular psammomatous NET of the duodenum has been applied (Fig. 29.16).^18,29 They often entrap the ampullary ductules, which with the glands formed by the tumor can be mistaken for an adenocarcinoma. A WDNET variant in the appendix (i.e., tubular carcinoid) is also characterized by gland formation, and considering that neuroendocrine marker expression may be limited in this variant, it can be difficult to distinguish it from an adenocarcinoma. Ileal WDNETs also tend to have glands, which are often prominent in the superficial or mucosal component of the tumor and are often located at the periphery of the individual nests.

Unlike the pulmonary WDNETs, spindle cell morphology is uncommon in GI NETs and PanNETs. Occasionally, gastric WDNETs may have vague spindle cell morphology. Appendiceal tumors characterized by goblet cells form a distinct category (discussed later).

Immunohistochemistry and Electron Microscopy

WDNETs are defined by the presence of neuroendocrine granules, which usually can be highlighted by three widely available neuroendocrine markers: chromogranin, synaptophysin, and CD56. Chromogranin is the most specific, but its sensitivity is lower. Rectal and a subset of appendiceal NETs (i.e., tubular types) can be devoid of chromogranin A, which is the target of most available chromogranin antibodies. Synaptophysin is very sensitive but less specific, with a variety of mimics showing potential expression of this marker. CD56 is even less specific. These markers may not be needed for the diagnosis of WDNETs because the morphology is often distinctive enough. They may, however, become necessary for some of the morphologic variants and for the diagnosis of PDNECs (discussed later).

NETs show epithelial differentiation, and they express keratins in most cases. However, the type and degree of keratin positivity may vary. Although staining for a wide spectrum keratins is commonly positive, cytokeratin (CK) 7 and CK20 expression is relatively uncommon. CK19 occasionally is expressed in PanNETs and has been used as a marker for adverse outcomes.

Nuclear transcription factors involved in the embryonic development of site-specific neuroendocrine cells have been used to determine the primary site of a WDNET,³⁰ but they are not specific. For example, CDX2 is commonly expressed in the NETs of the GI tract (Fig. 29.17, A). The islet 1 transcription factor (ISL1) is expressed commonly in the PanNETs but is also common in rectal WDNETs (see Fig. 29-17, B).^31,32 Pancreatic duodenal homeobox 1 (PDX1) is a marker of PanNETs at primary and metastatic sites,³³ but the specificity of this marker is unknown. Thyroid transcription factor 1 (TTF-1) expression is uncommon in pancreatic and GI WDNETs, and its sensitivity for pulmonary carcinoid tumors is limited. TTF-1 is expressed in small cell carcinomas of any organ. The PAX8 transcription factor is commonly expressed in pancreatic and rectal NETs, but it is typically absent in ileal NETs. As in pancreatic islet cells, which express progesterone receptors and CD99 (presumably resulting from cross reactivity),³⁴ PanNETs can also express these markers.

There are immunohistochemical caveats to consider. Ileal and rectal specimens can express prostate-specific acid phosphatase. Carcinoembryonic antigen (CEA) is expressed in 60% of WDNETs, and CA 19-9 positivity may also be seen. Staining for S100 can be positive in some WDNETs, especially in the appendix and some ampullary somatostatinomas.

The value of immunohistochemical analysis of hormone production is not clear. Correlation with functional activity is highly imperfect. Many tumors have multihormonal activity, and they may alter the predominant secretion over time.

Although somatostatin receptor scintigraphy is widely used in the clinical setting, immunohistochemical labeling for the somatostatin receptor is not typically performed in the United States. In Europe, it is more widely used in selected clinical conditions.^35,36