Metastatic gastrinoma in the breast mimicking primary solid papillary carcinoma.

We report a case of metastatic gastrinoma to the breast morphologically mimicking solid papillary carcinoma of the breast. A 59-year-old woman presented with a hypoechoic right breast mass that histologically revealed solid nests of small monotonous tumor cells, fibrovascular cores, and round to oval nuclei with fine chromatin and small nucleoli. Immunohistochemistry demonstrated chromogranin and synaptophysin positivity. Tumor prognostic markers showed weak positivity for estrogen receptor and negativity for progesterone receptor. Although an initial diagnosis of solid papillary carcinoma was rendered, subsequent identification of the patient's clinical history of pancreatic gastrinoma and an additional immunohistochemical stain for gastrin supported a diagnosis of metastatic gastrinoma. We report this rare case to increase awareness of metastatic neuroendocrine tumors in the breast. Multiple breast lesions and lack of expression of estrogen/progesterone hormone receptors should prompt careful review of the patient's clinical history to rule out metastatic neuroendocrine disease.

[Chromogranin A and neuroendocrine tumors]. / Cromogranina A y tumores neuroendocrinos.

Chromogranin A (CgA) is the most abundant granin in gastroenteropancreatic neuroendocrine tumors (GEP-NETs). As a tumor marker is moderately sensitive and nonspecific. Despite the limitations of testing methods, which require careful interpretation, especially in the case of gastrinomas, patients treated with somatostatin analogues, and poorly differentiated tumors, it is the best tumor marker in GEP-NETs and may be of value in other tumors with neuroendocrine differentiation. CgA may be used as a marker in blood or tissue samples through immunohistochemical techniques. CgA levels correlate with tumor burden and extension and may be used for diagnosis and monitoring of GEP-NETs, especially midgut carcinoids and endocrine pancreatic tumors. It is also useful as a prognostic marker for detection of recurrence and monitoring of response to different treatments.

Chromogranin A assay in clinical practice.

Chromogranins belong to the family of secretory chromogranin and secretogranin proteins. They are found in secretory vesicles throughout the neuroendocrine system. Chromogranin A (CgA) is the main component. CgA acts as a prohormone submitted to processes of degradation through which active peptides are generated. CgA has auto, para and endocrine functions. It is widely used as an immunohistochemical marker. Despite the lack of international standardization, and the lack of an accurate definition of the diagnostic cut-off levels, some CgA assays are reliable. Numerous studies have suggested that CgA determination may be of interest for the diagnosis and the follow-up of various endocrine tumors. Plasma levels of this general marker are proportional to tumor mass. The localization of the primitive tumor, the presence of associated hormonal secretions and possible renal failure and/or hypergastrinemia must be taken into consideration for proper interpretation of CgA levels. New clinical indications are emerging for the evaluation of stress in intensive care units and the assessment of cardiovascular risk. New assays estimating the concentration of active peptides are under development.

Lymphatic vessel endothelial hyaluronan receptor 1 immunocytochemical staining for pancreatic islets and pancreatic endocrine tumors.

OBJECTIVES: Immunocytochemical staining for lymphatic vessel endothelial hyaluronan receptor 1 (LYVE-1) is able to recognize lymphatic vessel endothelium and pancreatic endocrine cells (PETs). Pancreatic endocrine tumors were studied for LYVE-1 immunocytochemical staining compared with normal pancreatic islets to detect possible presence of LYVE-1 in PETs. METHODS: Twenty-five cases of primary and metastatic PETs were immunocytochemically stained for LYVE-1, including insulinomas, glucagonomas, somatostatinoma, pancreatic polypeptidomas, gastrinomas, and nonfunctioning tumors. With routinely formalin-fixed and paraffin-embedded tissues, LYVE-1 immunostaining was performed with polyclonal goat antihuman LYVE-1. RESULTS: All normal pancreatic islet cells were positive for LYVE-1, whereas 2 cases of 25 PETs, 1 each of gastrinoma and nonfunctioning tumor, were positive for LYVE-1, retaining immunocytochemical reactivity of islet cells. CONCLUSIONS: Normal pancreatic islets were positive for LYVE-1, whereas only 2 of 25 PETs were positive, suggesting that most PETs lost LYVE-1 or contained below detectable levels of LYVE-1. The presence of LYVE-1 in pancreatic islets and in some PETs may suggest structure-function relationship of LYVE-1/lymphatic vessel in hormone synthesis and secretion.

Indium-111 pentetreotide uptake by pancreatic polypeptide cell hyperplasia: potential pitfall in somatostatin receptor scintigraphy.

A 71-year-old man with duodenal gastrin cell tumor was being evaluated for residual/metastatic disease. Somatostatin receptor scintigraphy (SRS) identified a 2-cm area of focal uptake within the head of the pancreas, consistent with a pancreatic neuroendocrine tumor. Pathological examination did not reveal any malignancy within the pancreas. Instead, the pancreatic head showed pancreatic polypeptide cell hyperplasia. Strong and diffuse immunoreactivity to somatostatin receptor 2A antibody by immunoperoxidase staining confirmed that the lesion correlated with the site of radioactive tracer (Indium-111 pentetreotide) uptake seen on SRS. The current report therefore presents pancreatic polypeptide cell hyperplasia as a new pitfall in SRS.

Cytologic diagnosis of pancreatic endocrine tumors by endoscopic ultrasound-guided fine-needle aspiration: a review.

Precise localization and diagnosis of pancreatic endocrine tumors (PETs) is important, because pancreatic PETs have different clinical and biological behavior and treatment modalities than do exocrine pancreatic tumors. In contrast to the much more common exocrine adenocarcinomas, cytologic studies of PET are relatively rare and many cytopathologists lack experience with the cytomorphologic features of these tumors.During the last 10 yr, endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA) has matured into an accurate, highly sensitive, and cost-effective modality for the preoperative localization of pancreatic PETs. This has resulted in an increased number of PETs first sampled as cytology specimens. This manuscript focuses on the cytomorphologic features most suggestive of pancreatic PETs, differential diagnosis, and diagnostic pitfalls of PETs. The technical development of EUS-guided FNA and the ancillary studies for pancreatic PETs are also reviewed. The data summarized in this review indicate that EUS-FNA is a valuable method in the recognition of pancreatic PETs and in most cases cytopathologists could reach a correct diagnosis of these tumors, including their hormone producing capability on aspirated cytologic material.

Expression of somatostatin receptor types 1-5 in 81 cases of gastrointestinal and pancreatic endocrine tumors. A correlative immunohistochemical and reverse-transcriptase polymerase chain reaction analysis.

Somatostatin receptors (SSTRs) have been extensively mapped in human tumors by means of autoradiography, reverse-transcriptase polymerase chain reaction (RT-PCR), in situ hybridization (ISH) and immunohistochemistry (IHC). We analyzed the SSTR type 1-5 expression by means of RT-PCR and/or IHC in a series of 81 functioning and non-functioning gastroenteropancreatic (GEP) endocrine tumors and related normal tissues. Moreover, we compared the results with clinical, pathological and hormonal features. Forty-six cases (13 intestinal and 33 pancreatic) were studied for SSTR 1-5 expression using RT-PCR, IHC with antibodies to SSTR types 2, 3, 5 and ISH for SSTR2 mRNA. The vast majority of tumors expressed SSTR types 1, 2, 3 and 5, while SSTR4 was detected in a small minority. Due to the good correlation between RT-PCR and IHC data on SSTR types 2, 3, and 5, thirty-five additional GEP endocrine tumors were studied with IHC alone. Pancreatic insulinomas had an heterogeneous SSTR expression, while 100% of somatostatinomas expressed SSTR5 and 100% gastrinomas and glucagonomas expressed SSTR2. Pre-operative biopsy material showed an overlapping immunoreactivity with that of surgical specimens, suggesting that the SSTR status can be detected in the diagnostic work-up. It is concluded that SSTRs 1-5 are heterogeneously expressed in GEP endocrine tumors and that IHC is a reliable tool to detect SSTR types 2, 3 and 5 in surgical and biopsy specimens.

Human gastrinoma cells express calcium-sensing receptor.

The intravenous calcium injection test has been reported to be useful for the diagnosis of gastrinoma. However, the mechanism underlying calcium-evoked gastrin release is not fully understood. We investigated the mechanism of calcium-stimulated gastrin release from gastrinoma cells in vitro with a particular focus on the calcium-sensing receptor (CaR). Human gastrinoma cells were taken from mechanically minced gastrinoma tissues obtained at surgery. In the perifusion system, high [Ca2+]o induced gastrin release from gastrinoma cells. As [Ca2+]o increased, [Ca2+]i rapidly increased, as monitored by fluorometry. The response was not inhibited by nifedipine, a blocker of the voltage-dependent calcium channel. Reverse transcriptase-polymerase chain reaction and subsequent Southern blot hybridization revealed the presence of the CaR gene in human gastrinoma tissues. Moreover, the expression of CaR in gastrinoma tissues was confirmed by immunohistochemistry. Our results demonstrated that CaR was expressed in human gastrinoma cells and could be involved in the mechanism of calcium-evoked gastrin release.

[111In-pentetreotide in the study of tumors expressing somatostatin receptors]. / 111In-pentetreótida en el estudio de tumores que expresan receptores para la somatostatina.

UNLABELLED: The usefulness of 111In-pentetreotide scintigraphy in the diagnosis of neuroendocrine tumors was evaluated. MATERIAL AND METHOD: 39 patients with clinical or biochemical suspicion of neuroendocrine tumors were studied. 29 patients underwent surgery and diagnosis was confirmed by anatomopathology findings (7 mixed tumors, 6 carcinoids, 3 insulinomas, 3 pheocromocytomas, 2 glucagonomas, 2 medullary thyroid carcinomas, 1 gastrinoma and 5 metastatic lesions). In 10 patients who did not undergo surgery, the diagnostic criteria were based on a 6 month follow-up. 111In-pentetreotide scintigraphy and computed tomography (CT) were performed in all of the patients. RESULTS: The scintigraphy correctly detected 58% of the primary tumors while the CT was positive in only 45% of the cases. Both techniques detected metastasis in 5 patients. CONCLUSION: 111In-pentetreotide scintigraphy is a useful technique to diagnose those tumors that have somatostatin receptors. In our series, the sensitivity of this method was higher than the CT.

Immunohistochemical detection of somatostatin receptor subtypes sst1 and sst2A in human somatostatin receptor positive tumors.

Although in situ hybridization has been used to examine the distribution of messenger RNA for somatostatin receptor subtypes (sst) in human tumors, the cellular localization of sst1 and sst2A receptors has not been reported. In this study, we describe the cellular localization of human sst1 and sst2A receptor proteins in both cryostat- and paraffin-embedded sections of 25 human tumor tissues using two recently developed polyclonal antibodies. Six somatostatin (SS) receptor (SSR) positive tumors (two gastrinomas, three carcinoids, one pheochromocytoma) and one SSR negative tumor (renal cell carcinoma), selected by positive and negative SSR autoradiography, respectively, were studied by both immunohistochemistry and Western blot analysis. The six SSR positive tumors expressed sst2A, while 4 of 5 expressed sst1 as well. The SSR negative tumor did not express either sst1 or sst2A. Western blot analysis of wheat germ agglutinin purified membrane proteins confirmed the presence of the sst1 and sst2A glycosylated receptors. The paraffin-embedded sections gave best information with respect to the subcellular localization. Sst1 immunoreactivity was observed both on the membrane and in the cytoplasm, while sst2A showed predominantly membrane-associated immunoreactivity. This subcellular distribution of sst1 or sst2A receptors was confirmed in paraffin-embedded sections of 8 additional intestinal carcinoids, 5 gastrinomas and 5 pheochromocytomas. Sst1 receptors were detected in 7 out of 8 carcinoids, in all gastrinomas, and in 4 out of 5 pheochromocytomas, while 6 out of 8 carcinoids, all gastrinomas, and 3 out of 5 pheochromocytomas expressed sst2A receptors. In conclusion, sst1 and sst2A receptors show a differential subcellular localization in human SSR positive tumors. The use of SSR subtype selective antibodies to detect the subcellular distribution of SSR subtypes in individual tumor cells is an important step forward to understand more about the pathophysiological role of the different SSR subtypes in human tumors.

Mixed ductal-endocrine carcinoma of the pancreas presenting as gastrinoma with Zollinger-Ellison syndrome: an autopsy case with a 24-year survival period.

We report an autopsy case of mixed ductal-endocrine carcinoma of the pancreas presenting as gastrinoma with Zollinger-Ellison syndrome. A 38-year-old Japanese male was found to have Zollinger-Ellison syndrome and pancreatic gastrinoma, and gastrectomy and resection of the pancreatic tumor were performed. However, hypergastrinemia persisted, and the patient died of disseminated carcinomatosis at 62 years of age, 24 years after the onset of Zollinger-Ellison syndrome. At autopsy, the main tumor was present in the residual pancreas, and metastases were noted in many organs. In the pancreas and other organs, ductal and endocrine carcinoma areas were mixed and there was a gradual transition between the two. No acinar differentiation was noted. The ductal elements were positive for mucins and carcinoembryonic antigen but negative for neuroendocrine markers, while endocrine elements were positive for chromogranin A and synaptophysin and to a lesser extent for gastrin, but negative for mucins and carcinoembryonic antigen. The ductal elements comprised about 30% of the tumor cells, and endocrine elements 70%. According to the revised World Health Organization classification, our case was diagnosed as mixed ductal-endocrine carcinoma. Our case is rare because the tumor manifested as gastrinoma with Zollinger-Ellison syndrome and the patient survived for 24 years. To the best of our knowledge, no such case has been reported. Our case suggests that pancreatic endocrine tumors may evolve into mixed ductal-endocrine carcinomas.

Quantitative analysis of gastrin mRNA and peptide in normal and cancerous human pancreas.

Gastrin has been shown to stimulate growth of human pancreatic cancer, and does so in an autocrine fashion. In this study, a relationship between gastrin mRNA, peptide, and gastrin receptors were studied in a variety of human pancreatic tissues. Low levels of gastrin mRNA were detected in normal human pancreas by quantitative reverse transcription polymerase chain reaction, but gastrin peptide was not present using radioimmunoassay. Pancreatic adenocarcinoma cells and tissues had 34- to 530-fold higher gastrin mRNA and peptide levels than normal pancreas. Gastrin mRNA and peptide levels were 8,000- and 15,000-fold, respectively, greater in a pancreatic islet cell gastrinoma tumor than in normal pancreas. In comparison to age-matched controls, fasting gastrin plasma levels were 2-fold higher in patients with pancreatic adenocarcinoma and 131-fold greater in subjects with gastrinomas. Receptor binding assays revealed that pancreatic cancer cells had a binding capacity 200-fold greater than gastrinoma tumors, and 10-fold greater than normal pancreas; no differences in K(d) values were recorded between specimens. In contrast to the normal pancreas and gastrinoma tumor, the aggressive behavior of pancreatic adenocarcinoma may be attributed to the autocrine production of gastrin and to the presence of its growth-related receptor.

Somatostatin receptor subtype gene expression in human endocrine gastroentero-pancreatic tumours.

Somatostatin and its analogues are now of current use in the management of endocrine gastroentero-pancreatic (GEP) tumours for the purpose of inhibiting hormone hypersecretion, carrying scintigraphy imaging and attempting to slow down tumour growth. Recent molecular studies have revealed the existence of up to five membrane somatostatin receptor subtypes termed SSTR1-5. However, whether or not scintigraphy imaging and tumour characteristics are correlated with specific subtype(s) remains unclear. SSTR1-5 messenger RNA (mRNA) transcripts were investigated in 38 endocrine GEP tumours (32 islet cell tumours, six carcinoid) using reverse transcriptase polymerase chain reaction (RT-PCR), and their distribution was analysed with respect to tumour characteristics and scintigraphy imaging. SSTR2, SSTR5 and SSTR4 were detected in most cases of endocrine GEP tumours (92%, 84%, and 82% respectively), but SSTR1 and SSTR3 were less frequently observed (66% and 50% respectively). No clear-cut correlation was found between tumour characteristics and subtype mRNA distribution. Moreover, no differences in mRNA subtype distribution were found between the 17 tumours detected by scintigraphy and the four tumours not detected by this method. Somatostatin receptor mRNA subtypes are widely expressed in endocrine GEP tumours, but their distribution is not correlated with tumour characteristics or scintigraphy positivity.

Positive somatostatin receptor scintigraphy correlates with the presence of somatostatin receptor subtype 2.

Somatostatin receptor scintigraphy (SRS) is positive in approximately 75% of all patients with neuroendocrine gastroenteropancreatic tumours. This study aimed to identify specific somatostatin receptor (sstr) subtypes, which are responsible for the in vivo binding of the widely used somatostatin analogue, octreotide in human neuroendocrine gastroenteropancreatic tumours. Twelve patients underwent SRS with radiolabelled octreotide. After surgical resection, tumour tissues were analysed in vitro for somatostatin and octreotide binding sites by autoradiography. In addition, for the first time, sstr subtype mRNA expression was examined by semiquantitative reverse transcription polymerase chain reaction (RT-PCR). Tumour tissues from all SRS positive patients were positive by autoradiography. Semiquantitative RT-PCR revealed most prominently sstr2 expression in scintigraphically positive tumours. Two SRS negative tumours contained in vitro octreotide binding sites as well as high levels of sstr1 and sstr2 mRNAs. Positive SRS is mainly due to sstr2. sstr1, 3, 4, and probably 5 are less important for in vivo octreotide binding. False negative scintigraphic results seem to be influenced by factors independent of the expression of specific sstr.

Pancreatic polypeptide immunoreactivity in sporadic gastrinoma: relationship to intraabdominal location.

Sporadic gastrinoma is a pancreatic endocrine tumor whose ontogeny is unknown. The anatomic area where the vast majority of sporadic gastrinomas is found (pancreatic head region) corresponds topographically to the area traversed embryologically by the ventral pancreatic bud. Pancreatic polypeptide (PP), a 36-amino acid hormone, is secreted by pancreatic endocrine cells derived almost exclusively from the ventral pancreatic bud and is proposed as a marker for ventral bud derivation. Based on these observations we postulate that sporadic gastrinomas, found around the head of the pancreas, are derived from ventral bud tissue and should display a high incidence of PP immunoreactivity. Overall, we found PP immunoreactivity in 7 of 14 (50%) gastrinomas. Of those tumors located to the right of the superior mesenteric artery (SMA) (around the head of the pancreas), seven of nine (78%) contained PP, whereas no gastrinoma to the left of the SMA (n = 5) contained PP (p = 0.021; Fisher exact test). Only one other pancreatic endocrine or exocrine tumor, a glucagonoma located to the left of the SMA, stained positively for PP. We conclude that sporadic gastrinomas found around the head of the pancreas (to the right of the SMA) have a high incidence of PP immunoreactivity. These findings are consistent with our hypothesis that sporadic gastrinomas are derived from the ventral pancreatic bud.

Use of an isotopic somatostatin receptor probe to image gut endocrine tumors.

OBJECTIVES: To evaluate the effectiveness of indium In 111 pentetate (diethylenetriaminepentaacetic acid [DTPA]-D-Phe-labeled octreotide scintigraphy in the localization of gastroenteropancreatic neuroendocrine lesions, and to identify covert lesions, determine multicentricity, define the distribution of metastases, confirm complete removal of tumor postoperatively, and evaluate the efficacy of therapeutic embolization. DESIGN: Unmasked comparison. SETTING: Tertiary care referral center. PATIENTS: We studied 28 patients over a 12-month period. Biochemical evidence of a gastroenteropancreatic tumor was present in 13 patients. Octreoscan 111 was employed in four patients with an ambiguous biochemical diagnosis of gastroenteropancreatic tumor. Postoperative examination to document complete tumor removal was undertaken in seven patients. In one patient, Octreoscan 111 was used to evaluate the efficacy of therapeutic embolization. INTERVENTION: [111In]DTPA-D-Phe-octreotide scintigraphy. MAIN OUTCOME MEASURE: Identification of somatostatin receptor-bearing neuroendocrine tumors. RESULTS: Intravenous administration of [111In]DTPA-D-Phe-octreotide followed by whole-body gamma camera scintigraphy resulted in the localization of gastroenteropancreatic neuroendocrine tumors with 75% sensitivity, 100% specificity, 100% positive predictive value, 63% negative predictive value, and 82% overall accuracy. CONCLUSIONS: While Octreoscan 111 has been shown to localize the majority of amine precursor uptake and decarboxylation system (APUD) cell tumors as well as various other somatostatin-positive tumors, this technique may also be useful in a number of other circumstances. These include prediction of tumors that will respond to octreotide therapy, identification of covert metastases, intraoperative identification of tumors, and postoperative surveillance. Use of an alternative isotope may provide a vehicle for the administration of local therapeutic radiation to tumor cells. The precise efficacy of Octreoscan 111 in the identification of lesions smaller than 3 cm with low-density somatostatin-2 receptor expression remains to be determined.

Identification of a new chromogranin B fragment (314-365) in endocrine tumors.

A rabbit antiserum was raised against the fragment (350-365) of human chromogranin B corresponding to the C-terminal end of a putative proteolytic fragment generated by the cleavage of a dibasic doublet located in position 366-367 of the precursor. A radioimmunoassay was developed. Chromatographic analysis of 10 endocrine tumor extracts (one liver metastasis of a gastrinoma, one liver metastasis of a medullary carcinoma of the thyroid, one VIPoma, one insulinoma, one nonsecreting pancreatic endocrine tumor, one local recurrence of a gut carcinoid, two pituitary gonadotropinoma, and two non-secreting pituitary adenomas) revealed the presence of two forms of immunoreactive material. The most abundant form had an apparent molecular weight of 4500 and was purified to homogeneity by successive reverse-phase HPLC chromatographies and partially sequenced. The N-terminal sequence of the peptide, established by automated Edman degradation, was A-S-E-E-E-P-E-Y-G-E-E-I-K-G-Y-P-V-Q and corresponded to the 314-332 sequence of human chromogranin B. Taking into account the specificity of the antiserum used for peptide identification, we deduced that the purified peptide was chromogranin B(314-365) and represented a new form generated by limited proteolysis of chromogranin B.

Different splice variants of CD44 are expressed in gastrinomas but not in other subtypes of endocrine pancreatic tumors.

Endocrine pancreatic tumors are neuroendocrine neoplasms with malignant potential and give rise to varied clinical syndromes due to excessive secretion of multiple hormones. In this study 22 endocrine pancreatic tumors and 11 carcinoid tumors were examined for the expression of CD44 using a monoclonal antibody. CD44 gene activity of 11 endocrine pancreatic tumor tissues and five carcinoid tumor tissues was also studied by amplifying messenger RNA with the polymerase chain reaction followed by electrophoresis and blot hybridization. Strong immunoreactivity was detected on all gastrinomas examined (P < 0.001), and in two non-functioning endocrine pancreatic tumors. Such immunoreactivity was not observed in other subtypes of endocrine pancreatic tumors. In the normal human pancreas, the acinar portion and ductal epithelial cells stained strongly positive but pancreatic islet cells did not show any significant immunostaining. Furthermore, in endocrine pancreatic tumors with metastatic disease, CD44-positive tumors had a tendency to metastasize to lymph nodes (P = 0.005), as compared with CD44-negative tumors which were locally invasive or metastasized to the liver. Although, in this limited material and short follow-up, we were not able to show any statistical significance, patients with CD44-negative endocrine pancreatic tumors had prolonged survival time compared with patients with CD44-positive tumors (73% versus 59% at 5 years; P = 0.7). Of 10 carcinoid tumors examined, all three foregut carcinoids and one midgut carcinoid stained strongly positive, whereas all other midgut carcinoids were negative. Analysis of CD44 splice variants showed that in all five gastrinomas there was overproduction of alternatively spliced larger molecular variants as compared with other types of endocrine pancreatic tumors and carcinoid tumors. The band pattern from one case of carcinoid tumor with a fulminant clinical course was similar to that of gastrinomas, whereas other carcinoid tumors expressed the epithelial form of CD44. The earlier identified splice variants which confer metastatic behavior on a pancreatic tumor cell line were not expressed in neuroendocrine tumors. Our data indicate that CD44 expression in endocrine pancreatic tumors correlates with the ability to give rise to lymph node metastases and may play a vital role in determining the fate of metastasizing cells. Moreover, because gastrin is not detectable in the normal human pancreas, the pancreatic ductal cell positivity for CD44 strengthened the ductal origin concept of gastrinomas. The band pattern of CD44 splice variants suggests that the previously described splice variants conferring metastatic behavior do not accompany metastatic activity of neuroendocrine tumors.