Protocol for the examination of specimens from patients with endocrine tumors of the pancreas, including those with mixed endocrine and acinar cell differentiation: a basis for checklists. Cancer Committee of the College of American Pathologists

Case Report: Giant insulinoma, a very rare tumor causing hypoglycemia

Insulinomas, with an incidence of 4 cases per million individuals per year, remain amongst the most frequent functional neuroendocrine tumors. The usual diameter of insulinomas usually remains under 3 cm of major axis. However, 44 exceptional cases of "giant insulinomas", have been reported worldwide, generally exceeding 9 cm in major axis. In this article, we report the case of a 38-year-old woman whom suffered from chronic hypoglycemia despite treatment with diazoxide. Abdominal CT-scan revealed a 88 x 73 mm mass located at the tail of the pancreas. Following surgical excision, histopathological analysis confirmed G1 neuroendocrine tumor, with focal cytoplasmic expression of insulin in tumor cells. After a 16-month follow-up period, the patient didn't address any specific complaint, and no disease recurrence and/or metastasis were observed. A ⁶⁸Ga-DOTATATE-PET scan was performed 6 months after surgery, which came back normal. Genetic evaluation has not been performed in our patient. The physiopathology of giant insulinomas remain unexplained, however with possible relationship with type 1 multiple endocrine neoplasia, sporadic somatic YY1 mutations and possible transformation of bulky non-functional pancreatic neuroendocrine tumors to a functional phenotype, with slow insulin secretion. While giant insulinomas remain rare in the literature, multicentric genetic analysis of tumor samples might reveal unique features of this rare subtype of neuroendocrine pancreatic tumors. Insulinomas of large size tend to have greater malignancy and higher rates of invasiveness. Careful follow-up, especially for liver and lymph node metastases, must be performed using functional imaging techniques to avoid disease relapse.

[Mixed endocrine tumors]

Mixed endocrine tumors are tumors composed of at least two distinct tumor populations, one of which is endocrine. Because of their rarity and unusual presentation, endocrine mixed tumors raise many problems of diagnosis, management and therapy. Three main types of endocrine mixed tumors are recognized: The existence of these various types has been confirmed by recent molecular studies, even if the same studies have also shown that the histogenesis of a mixed endocrine tumor cannot be predicted from its histological features. Composite tumors are the less rare mixed tumors. The recent WHO classification recommends to restrict the term of composite endocrine tumor to the epithelial tumors containing at least 30% of obviously tumoral endocrine cells; some authors recommend to use higher thresholds, of at least 50%, in order to avoid overdiagnosis. The endocrine component is usually well differentiated, easily identified by its suggestive histological features; the endocrine nature of tumor cells is confirmed by the immunodetection of specific endocrine and neuro-endocrine markers (such as chromogranin A and synaptophysin). In some cases, the endocrine component is poorly differentiated: the demonstration of neuro-endocrine markers is necessary to confirm the diagnosis. Mixed tumors can occur in every anatomical site; they are more frequent in organs containing endocrine cells in the normal state (especially the digestive tract and the pancreas), but they can also be observed in organs devoid of endocrine cells (such as the mammary gland). The management of mixed endocrine tumors must take into account the more aggressive component. Mixed tumors containing a well differentiated endocrine component and an adenocarcinomatous component are to be treated like adenocarcinomas. Mixed tumors containing a poorly differentiated endocrine component must be considered as poorly differentiated endocrine carcinomas.

Prevalence and prognostic significance of acinar cell differentiation in pancreatic endocrine tumors

We have noted that many histologically and immunohistochemically confirmed pancreatic endocrine tumors show immunophenotypic evidence of acinar cell differentiation, but the clinical relevance of this finding is unknown. We performed this study to evaluate the prevalence and prognostic significance of exocrine differentiation by immunohistochemistry in pancreatic endocrine tumors that do not show morphologic features of acinar cell differentiation. Routinely processed tissue sections from 87 pancreatic endocrine tumors were immunohistochemically stained with monoclonal antibodies against acinar (lipase, chymotrypsin, trypsin) and endocrine cell markers (chromogranin A, neuron-specific enolase, synaptophysin, Leu-7) and for the proliferation-associated peptide Ki67. The degree of staining with each marker was graded on a three-tier scale for acinar markers (grade 0, <5%; grade 1, 5-10%; grade 2, 11-25%; and grade 3, >25%) and on a four-tier scale for endocrine markers (grade 0, <5%; grade 1, 5-25%; grade 2, 26-50%; grade 3, 51-75%; and grade 4, >75%), and the results were correlated with clinical outcome (mean follow-up 53 months). Greater than 75% of the tumor cells stained for chromogranin A, neuron-specific enolase, synaptophysin, and Leu-7 in 100%, 96%, 93%, and 27% of cases, respectively. Overall, 66% of tumors stained positively for at least one acinar cell marker, 31% stained for at least two acinar cell markers, and 13% stained for all three acinar cell markers. Forty-seven percent stained for lipase (23 grade 1, 11 grade 2, seven grade 3), 37% for trypsin (22 grade 1, three grade 2, seven grade 3), and 25% stained for chymotrypsin (13 grade 1, five grade 2, four grade 3). No correlation was noted between the presence or extent of expression of any single or combination of acinar cell markers and clinical outcome. However, higher tumor stage correlated with a poor clinical outcome (p = 0.002), and location in the tail of the pancreas was associated with a longer interval to tumor recurrence (p = 0.03). The presence of synaptophysin (p = 0.03) and Leu-7 expression (p = 0.03) correlated significantly with less aggressive clinical behavior. An association was observed between increased Ki67 labeling and poorer clinical outcome, but this was not statistically significant (p >0.05). In conclusion, immunophenotypic evidence of acinar cell differentiation is common in pancreatic endocrine tumors, but this feature does not have any relevance to clinical prognosis. However, in addition to tumor stage, location in the pancreatic tail and the immunohistochemical expression of synaptophysin and/or Leu-7 may be useful prognostic indicators in patients with these lesions.

[Gross tumor volume and clinical target volume in radiotherapy: generalities. Definitions and generalities in anatomic pathology]

In the field of oncology, surgical pathology analysis of biopsies or surgical specimens directly determines the care program of each patient. Without going into too much detail, we report here the description of different characteristics of neoplastic lesions, necessary or useful for the establishment of the care program of the patients. These macroscopic and microscopic analyses should result in the most precise possible diagnosis and should take into account a list of other factors which allow the development of an optimal strategy for therapy, and the establishment of prognosis of the tumor. These characteristics include properties of the tumor such as the size, shape, exact location, and the grade of the lesion, etc. They also concern the relationship of the tumor to certain structures like the basal membranes, the stroma reaction, the lymphatic vessels, the nerves, and so forth. The description of local and locoregional spreading, as well as the analysis of the surgical margins gives essential information as well. Finally, in certain cases, the description of modifications induced by neoadjuvant therapy allows the evaluation of the treatment's efficacy.

Cytokeratin 7 and 20 and thyroid transcription factor 1 can help distinguish pulmonary from gastrointestinal carcinoid and pancreatic endocrine tumors

Expression of cytokeratin (CK) 7 and 20 is commonly used to help distinguish adenocarcinomas from different sites. Thyroid transcription factor 1 (TTF-1) is a 38-kd protein, located primarily in the nucleus of type 2 pneumocytes and clara cells. TTF-1 has been shown to be present in a variety of lung and thyroid tumors and in pulmonary small-cell carcinomas. Carcinoid tumors from the lung and the gastrointestinal (GI) tract are histologically similar and thus are difficult to differentiate from each other based on histologic criteria. Pancreatic endocrine tumors (PET) have a similar histologic appearance to these other tumors. The purpose of this study was to determine the efficacy of differentiating these 3 groups of tumors by their expression of CK7, CK20, and TTF-1. Routinely processed paraffin-embedded tissue sections from 62 carcinoid tumors (lung, 16; gastrointestinal [GI] tract, 46) and 12 PETs were immunohistochemically stained for CK7, CK20, and TTF-1. The degree of expression in each tumor was graded as 1+ (1% to 10% of cells positive), 2+ (11% to 25%), 3+ (26% to 50%), and 4+ (>50%). The data were compared between tumor types and between carcinoid tumors from the various locations in the GI tract (stomach, 8; small intestine, 19; large intestine, 17; appendix, 2). CK7 was expressed in 10 (63%) of 16 pulmonary carcinoid tumors and only 5 (11%) of 46 GI carcinoid tumors (P <.001). Pancreatic endocrine tumors showed CK7 positivity in 6 (50%) of 12 cases, which was similar to the findings in lung carcinoids and significantly higher than in GI carcinoids (P <.01). CK20 was expressed in 0 (0%) of 16 pulmonary carcinoid tumors, in contrast to 24% and 33% of GI carcinoid tumors (P <.05) and PETs (P <.05), respectively. TTF-1 expression was highly specific for pulmonary carcinoid tumors. This peptide was present in 11 (69%) of 16 pulmonary carcinoid tumors and in only 1 (2%) of 46 and 0 (0%) of 12 GI carcinoid tumors (P <.001) and PETs (P <.001), respectively. A CK7(+)/CK20(-)/TTF-1(+) immunopanel result was moderately sensitive (sensitivity, 50%), and highly specific (specificity, 100%), for a diagnosis of pulmonary carcinoid tumor. CK7, CK20, and TTF-1 did not differ significantly between carcinoid tumors located in different sites of the GI tract. However, a trend was observed toward a lower prevalence of CK20 positivity in gastric tumors (P =.06) than in GI carcinoid tumors from the small intestine, colon, or appendix. Expression of CK7 and CK20, and particularly TTF-1, may be useful in distinguishing pulmonary from GI carcinoid tumors and PETs, especially when evaluated as a panel of markers. TTF-1 is highly specific for pulmonary carcinoid tumors.