| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Pathology and Laboratory Medicine, Section of Pathological Anatomy (C.B., C.A., S.P., G.F.), University of Parma, I-43100 Parma, Italy; Department of Digestive Diseases, 2nd School of Medicine (V.D.C., G.D.F.), and Department of Cell Biotechnology and Hematology (V.D.C.), University La Sapienza, I-00161 Rome, Italy; and Digestive Disease Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health (F.G., R.T.J.), Bethesda, Maryland 20892
Address correspondence and requests for reprints to: Cesare Bordi, M.D., Department of Pathology and Laboratory Medicine, Section of Pathological Anatomy, University of Parma, I-43100 Parma, Italy. E-mail: cesare.bordi{at}unipr.it
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsCase ReportsResultsDiscussionNote Added in ProofReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsCase ReportsResultsDiscussionNote Added in ProofReferences |
|---|
Gastric endocrine tumors of MEN-1 patients originate from the enterochromaffin-like (ECL) cells (6, 7), the main endocrine cells of the oxyntic mucosa involved in the synthesis and secretion of histamine with a key role in the mechanism of acid secretion (8). In keeping with such cytological origin, the ECL cell carcinoids of MEN-1 patients are usually restricted to the gastric regions lined by the oxyntic mucosa (i.e. the corpus and fundus of the stomach). These tumors may occasionally be malignant and pursue a very aggressive course (9).
ECL cells are very sensitive to the trophic stimulus of the antral hormone gastrin (8, 10). Hypergastrinemia, which is almost invariably present in MEN-1 patients with ECL cell carcinoids, is reputed to be the important promoting factor in the development of these tumors in combination with the transforming effect of the inactivation of the oncosuppressor gene MEN-1 at 11q13 (6).
Antral endocrine tumors are not mentioned in previous series of gastric endocrine tumors associated with MEN-1 syndrome (7, 11, 12, 13). This absence is consistent with the fact that ECL cells are not present in the antral mucosa (14), in which the endocrine cell population is basically represented by gastrin-producing G cells, somatostatin-producing D cells, and serotonin-producing EC cells (15).
On the basis of the histopathological screening of endoscopic gastric biopsies from a large series of patients with Zollinger-Ellison syndrome (ZES), we found four cases of carcinoid tumors arising in the antral mucosa, all patients being also affected by MEN-1. Immunohistochemically, they did not express the indigenous antral hormones gastrin, somatostatin, and serotonin. We also used the recently available immunohistochemical probe for the isoform 2 of the human vesicular monoamine transporter (VMAT-2). VMAT-2 regulates the intravesicular accumulation of histamine, the specific product of ECL cells (16, 17, 18), and is present only in this cell type among all endocrine gastrointestinal cells (16). The preservation of VMAT-2 in tissues routinely fixed and paraffin embedded makes this protein the first specific marker available for identification of ECL cells and their tumors in routine histological preparations (19, 20). VMAT-2 immunohistochemistry enabled us to demonstrate that in at least two of our cases the antral neoplasms were ectopic ECL cell carcinoids.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsCase ReportsResultsDiscussionNote Added in ProofReferences |
|---|
All patients admitted to the NIH with a confirmed diagnosis of ZES between July 1996 and November 1999 or evaluated at the La Sapienza University Hospital, Rome, between April 1988 and July 2000 were included in the present study. These patients are involved either in the prospective study of patients with ZES that has been ongoing at the NIH since 1974, as approved by the Clinical Research Committee of the National Institute of Diabetes and Digestive and Kidney Diseases, or in a prospective study of patients with gastric disorders approved by the La Sapienza University ethical committee in adherence to the declaration of Helsinki. All patients underwent an assessment of gastric secretory rates (basal and maximal acid output), multiple fasting gastrin determinations and secretin gastrin provocative testing, and tumor imaging with conventional tumor localization studies [computed tomography scan, magnetic resonance imaging (MRI), ultrasound], and, since 1994, somatostatin receptor scintigraphy with [111In-DTPA-Dphe1]octreotide (21, 22). Somatostatin receptor scintigraphy was performed using single photon emission computed tomography (22, 23).
The ZES was diagnosed as described previously (24). Briefly, a confirmed diagnosis required at least two of the following criteria: 1) an elevated fasting serum gastrin level; 2) a basal acid output of 15 meq/h or greater without previous gastric surgery or 5 meq/h or greater if previous gastric surgery; 3) an increase 200 pg/mL or greater after secretin administration or 395 pg/mL or greater after iv calcium infusion (25); or 4) a positive biopsy for gastrinoma. The presence of MEN-1 was established on the basis of a family or personal history indicative of the syndrome and/or evidence for parathyroid disease (on the basis of determination of total calcium, ionized calcium, plasma parathormone midmolecular immunoradiometric assay) or for pituitary disease (on the basis of determination of PRL, GH, urinary cortisol, and of MRI of sella turcica) (26).
A total of 170 patients with ZES were enrolled in the study. Of them, 46 presented associated MEN-1 whereas 124 showed the sporadic form of the disease.
Upper gastrointestinal (UGI) biopsy and biopsy processing
UGI endoscopy with biopsy was performed on first entrance into the study and, thereafter, on a yearly basis. Ten gastric biopsies (eight body and two antrum) were taken using a videoscope Olympus GIF 2T100 gastroscope (Olympus Corp., Melville, NY) and a Paul Drach Jumbo biopsy forceps without a pin (McGowan Co., Pennsacola, FL). In the gastric body four biopsies were taken from the greater curvature (three at La Sapienza University Hospital) and four biopsies from the lesser curvature (three at La Sapienza University Hospital), as detailed elsewhere (27). In the antrum two biopsies were taken, one on the greater and one on the lesser curvature, 2 cm from the pylorus. This allowed assessment of antral and corporal Helicobacter pylori presence, the distribution of which can be affected by omeprazole use, as well as assessment of extent of gastritis and cellular morphology of each area. In addition, extra biopsies were taken from any abnormal lesion(s) seen endoscopically. All biopsies were oriented on auto-adhesive cellulose nitrate filters (Sartorius AG, Gottingen, Germany). Biopsies were fixed for 3–5 h in Bouin’s solution, rinsed with 70% ethyl alcohol/30% water, and processed as described previously (28).
Histology and immunohistochemistry
Serial 5-µm thick sections of gastric biopsy specimens perpendicular to the mucosal surface were stained with hematoxylin and eosin, for evaluation of the type of mucosa, and with immunostaining for endocrine cells using monoclonal antibodies against: chromogranin A (CgA) (clone LK2H10; BioGenex Laboratories, Inc., San Ramon, CA; working dilution, 1:250); C-terminal pentapeptide of gastrin (clone MA64/18; courtesy of Dr. L. Scopsi, Instituto Nazionale dei Tutori, Milan, Italy; working dilution, 1:20,000); somatostatin (clone MAB354; Chemicon International, Temecula, CA; working dilution, 1:100); polyclonal rabbit antisera against VMAT-2 (Phoenix Pharmaceuticals, Inc., Mountain View, CA; working dilution, 1:1000); and serotonin (Euro-Diagnostica, Arnhem, The Netherlands; working dilution, 1:100), as primary antibodies with overnight incubation at 4 C. Immunostaining was done with the streptavidin-biotin kit (LSAB2; DAKO Corp., Glostrup, Denmark) according to the manifacturer’s specifications with diaminobenzidine tetrahydrochloride as a chromogen substrate and nuclear counterstaining with hematoxylin. According to the established criteria (29), only endocrine cell proliferations having a minimum diameter of at least 0.5 mm were diagnosed as carcinoid tumors.
|
Case Reports |
|---|
|
TopAbstractIntroductionMaterials and MethodsCase ReportsResultsDiscussionNote Added in ProofReferences |
|---|
Case 1
A 49-yr-old white male with a well-established familial history of MEN-1 and a previous history of pituitary and parathyroid surgery for adenomas was admitted at the NIH in September 1988 where ZES was diagnosed. Analysis of the MEN-1 gene showed a nonsense mutation (y 312x) in exon 7. Multiple pancreatic endocrine tumors showing focal immunoreactivity for insulin and somatostatin and negative for gastrin were found at surgery. Multiple duodenal tumors staining positive for gastrin were seen endoscopically. A focus in the right lobe of the liver was revealed by octreotide scan. Laboratory data included fasting serum gastrin at 550 pg/mL (normal, <200), PRL at 21 µg/mL (<11), CgA at 65 ng/mL (<5.6), and normal serotonin, 5-hydroxyindoleacetic acid, urinary free cortisol, and calcium. He was treated with ranitidine for 1 yr, omeprazole for 11 yr, and is currently treated with 60 mg omeprazole twice daily.
Case 2
A 47-yr-old white male with a previous history of ZES and multiple endocrine tumors negative for gastrin removed at surgery from the pancreas, jejunum, and mesentery was first seen at the NIH in September 1990. Multiple gastric carcinoid was diagnosed on endoscopic biopsies performed in 1992 and in subsequent examinations. Despite negative familial history of MEN-1, a genomic mutation in the MEN-1 gene (in exon 8 1280 del G) resulting in a frameshift was detected. Laboratory data included fasting serum gastrin at 2330 pg/mL (normal, <200), CgA at 600 ng/mL (<5.6), serotonin at 558 ng/mL (<260), urinary 5-hydroxyindoleacetic acid at 7 mg/24 h (<6), and normal serum calcium, parathormone, and PRL levels. He was treated with ranitidine for 11 yr, omeprazole for 7 yr, and is currently treated with 20 mg omeprazole twice daily.
Case 3
A 48-yr-old white male with a previous personal history of pituitary and parathyroid surgery for adenomas but no familial history was seen at La Sapienza University Hospital in 1990. A diagnosis of ZES was made. A focus in the liver was revealed by octreotide scan whereas MRI showed findings suggestive of pancreatic tumor in the uncinate process. Laboratory data included fasting serum gastrin at 550 pg/mL (normal, <40) and normal serum levels of parathormone and calcium. He was treated with ranitidine for 8 yr, omeprazole for 4 yr, and is currently treated with 20 mg omeprazole daily. The patient was last seen in March 1994.
Case 4
A 34-yr-old white male was first seen at La Sapienza University Hospital in March 1992 for perforated jejunal ulcer. No familial history was ascertained. A diagnosis of ZES was made, and a pancreatic tumor was removed in 1993 showing immunoreactivity for glucagon and pancreatic polypeptide but not for gastrin and insulin. In December 1998, a 3.5-cm large metastatic endocrine tumor immunoreacting for gastrin was removed from the perijejunal area. The antral carcinoid tumor was found in June 1998, whereas multiple carcinoids of the gastric corpus were diagnosed in May 2000. Laboratory data included fasting serum gastrin at 1250 pg/mL (normal, <40) at entry (raised to 25,000 pg/mL in 1998), CgA at 600 U/L (<30), and elevated levels of insulin at 46.3 mUI/mL (<25), parathormone at 186 pg/mL (<54), and serum calcium, total at 11.75 mg/dL (8.6–10.5) and ionized at 1.95 mol/L (1.21–1.30). The patient refused parathyroid surgery. He was treated with ranitidine for 2 months, omeprazole for 8 yr, and is currently treated with omeprazole 20 mg twice daily.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsCase ReportsResultsDiscussionNote Added in ProofReferences |
|---|
) and ranged in size from 0.5–0.7 mm
(Table 1). Because of such small size
they did not produce mucosal abnormalities and were found by random
biopsies. In case 2, a second carcinoid was seen in antral biopsies
collected 21 months after detection of the first tumor, both neoplasms
measuring 0.7 mm in diameter. A total of nine yearly endoscopies were
performed in these four patients, resulting in a total of 18 antral
biopsies taken.
|
|
).
All tumors presented intense chromogranin A immunostaining, but none of
them showed gastrin, somatostatin, or serotonin immunoreactivity in
contrast with the strong reaction of surrounding extratumoral G, D, and
EC cells, respectively. Intense immunostaining for VMAT-2 involving
most to all tumor cells was seen in two cases, including both tumors of
case 2, whereas no VMAT-2 expression was found in the remaining two
carcinoids. The mucosa surrounding the tumors in all cases showed the
full complement of cells characteristic for the antro-pyloric mucosa
with the typical distribution of G and D cells in the middle layer. No
VMAT-2 immunoreactive cells were seen. Neither transitional mucosa with
admixture of oxyntic glands nor metaplastic changes were found.
|
, the two
patients with VMAT-2-positive antral carcinoids also presented multiple
carcinoid tumors in the body mucosa. In case 4, the latter neoplasms
were diagnosed 22 months after the diagnosis of the antral tumor.
Although most of them proved to be VMAT-2 immunoreactive, 5 of 15
gastric body carcinoids of the two patients showed no VMAT-2
immunostaining. Nonantral tumors were not detected in the two patients
with VMAT-2-negative antral carcinoids. In the whole study, carcinoids
of the oxyntic mucosa were found in 17 (36.9%) of 46 patients with
MEN-1-associated ZES and in none of 124 patients with sporadic ZES.
None of these tumors presented evidence for malignancy.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsCase ReportsResultsDiscussionNote Added in ProofReferences |
|---|
The four cases of antral carcinoids identified in our study accounted for 8.7% of all patients with MEN-1 and ZES investigated, roughly one fourth of the incidence of all gastric carcinoids seen in our patient population (36.9%). However, because our endoscopic sampling protocol of the antral mucosa was based on two biopsy specimens alone, our data may underestimate the actual incidence of these neoplasms in MEN-1 syndrome. Similar results were obtained by us in a recent large study assessing the role of the number of biopsies on the detection rate of gastric body carcinoids in patients with MEN-1/ZES (27). In this study, we demonstrated that the frequency of carcinoids found increased in direct proportion to the number of biopsies taken.
In one patient a second antral carcinoids was found 21 months after the identification of the first one, indicating multifocal tumor development typical of the genetically influenced pathogenesis of MEN-1 tumors (5). LOH analysis of the MEN-1 gene locus at 11q13 in multiple MEN-1 endocrine tumors of the pancreas and duodenum (31) and of the stomach (32) showed different genetic alterations in the individual tumors of a same patient, indicating that each of them is the result of an independent somatic genetic event.
All patients were adult male and presented with association of MEN-1 and ZES. The prevalence of the male gender is at variance with the results of animal studies of hypergastrinemia-driven ECL cell corporal carcinoids and with the findings of patients with ECL cell corporal carcinoids and hypergastrinemic atrophic gastritis, both showing significantly higher frequency in female individuals (6, 33). Our data suggest a different gender distribution between hypergastrinemia-associated corporal and antral carcinoids, a hypothesis requiring confirmation in a larger number of cases.
What factors are contributing to the development of antral carcinoids in these patients are, at present, unknown. Hypergastrinemia is known to be an important factor in the development of corporal ECL cell carcinoids (6, 8, 10), but recent studies report such gastric carcinoids also in MEN-1 patients without hypergastrinemia (9, 34). The prevailing role of the MEN-1 genetic alterations in the development of antral tumors is supported by the observation that these tumors were not identified in a series of 124 cases with sporadic ZES included in the study. Such restriction of antral endocrine tumors to those ZES patients who are also affected by MEN-1 syndrome is similar to that presented by ECL cell tumors of the oxyntic mucosa (Refs. 6, 12 , and 35 and our present series). In contrast, it is at variance with the ubiquitous distribution of duodenal gastrinomas in both sporadic and MEN-1-associated cases of ZES (21).
The antral tumors were all intramucosal and with innocent histological appearance. All presented intense CgA immunostaining but none of them showed gastrin immunoreactivity, in contrast with the strong reaction of surrounding extratumoral G cells. Similarly, somatostatin and serotonin were not expressed in the tumors whereas they were easily demonstrated in extratumoral D and EC cells, respectively. These data indicate that antral carcinoids of MEN-1 patients do not express the phenotype of the most common types of antral endocrine cells. The lack of antral gastrinomas, even at the microscopic level, denotes that antral G cells are not prone to undergo tumor transformation in MEN-1. This is surprising in view of the frequent development of clinically functioning gastrinomas from gastrin-producing cells of the adjacent duodenum in MEN-1 (21, 36).
In two of our patients the antral carcinoids, including both tumors found in case 2, showed diffuse immunoreactivity for VMAT-2. This recently identified vesicular transporter controls histamine accumulation from the cytoplasm into secretory vesicles of the ECL cells of the gastric oxyntic mucosa (18). Owing to its exclusive localization in the ECL cells among all types of endocrine cells of the gastrointestinal tract (16), VMAT-2 has become the first specific immunohistochemical marker for ECL cells suitable for routine histological preparations. On this basis, two recent independent studies have demonstrated the specificity of the marker for tumors composed of ECL cells (19, 20). Our results, therefore, indicate that in two of four patients the antral carcinoids were composed of ECL cells.
The antral tumors in the other two cases appeared to be endocrine in nature because of their CgA immunostaining, but the specific cell of origin could not be identified. The lack of detectable VMAT-2 expression, however, does not exclude their ECL cell composition. In fact, VMAT-2 immunoreactivity was absent in one third of ECL cell carcinoids in the oxyntic mucosa of our patients. It is interesting to note that in both cases of VMAT-2-unreactive antral tumors no concomitant carcinoids were seen in the gastric corpus.
The origin of the antral ECL cell tumors is intriguing. ECL cells are not located in the normal antrum (14), an observation supported by the absence of VMAT-2 immunoreactive cells in the extratumoral antral mucosa shown by our and previous (20) studies. The mucosa surrounding the tumors showed the full complement of antral mucosal cells with no evidence of transitional mucosa. This finding does not support a tumor origin from distal extensions of the oxyntic mucosa. The benign histological features of corpus ECL cell carcinoids, when present, do not support the hypothesis that antral carcinoids of our patients represent intramucosal metastases from the corpus tumors. We suggest, therefore, that antral ECL cell carcinoid represent ectopic tumors possibly deriving from uncommitted endocrine cells under the effect of the genetic defect combined with hypergastrinemia.
The clinical significance of the small antral tumors of our patients is conceivably low. It should be noted, however, that in these patients also the concomitant ECL cell tumors of the gastric body consistently were in a small size range that is indicative of innocent clinical behavior (7). It remains to be ascertained whether antral endocrine tumors may participate in the brisk development of ECL cell tumors that sometimes occurs in MEN-1 patients with or without ZES and that has ominous prognostic evolution (9). In one of these cases multiple tumors up to 0.8 cm in size were found in the antral mucosa whereas the largest endocrine tumor was located in the border region between antrum and corpus (9). This region was also found by Lehy et al. (12) to be the location of the earliest ECL cell tumors appearing in ZES/MEN-1 patients, a temporal sequence similar to that observed for the antral tumor in our case 4.
At present, the diagnosis and management of gastric carcinoids in patients with MEN-1 and ZES, irrespective of their corporal or antral location, are controversial. The percentage of patients who will have gastric carcinoids associated with mucosal abnormalities, without mucosal abnormalities, or whether diagnosis can be increased by endoscopic ultrasound, is unknown. It is apparent that in some patients the carcinoid tumors can have an aggressive course. We currently recommend that all patients with MEN-1 and hypergastrinemia undergo UGI endoscopy, with biopsy of any mucosal abnormality in the antrum and body, and with additional one or two random biopsies in the corpus and one in the antrum. This also allows assessment of H. pylori status and of the qualitative ECL cell growth pattern.
In conclusion, this study demonstrates that the antral mucosa is an additional tissue that may harbor endocrine tumors in MEN-1 syndrome. These tumors apparently do not express the phenotype of normal endocrine cells of the gastric antrum, and in at least two cases they could be identified as ectopic ECL cell carcinoids by VMAT-2 immunostaining.
|
Note Added in Proof |
|---|
|
TopAbstractIntroductionMaterials and MethodsCase ReportsResultsDiscussionNote Added in ProofReferences |
|---|
|
|
Footnotes |
|---|
Received September 27, 2000.
Revised January 23, 2001.
Accepted January 31, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsCase ReportsResultsDiscussionNote Added in ProofReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
