Abstract
A golden yellow polyp was detected in the gallbladder of a 64-year-old man who presented with epigastric pain. The lesion was composed of clear polygonal cells arranged in a trabecular and glandular pattern. The tumor invaded through the wall into the perimuscular subserosal layer. Immunohistochemical stains showed that neoplastic cells were positive for chromogranin A, synaptophysin, somatostatin, gastrin, and pancreatic polypeptide and negative for glucagon, serotonin, insulin, S100 protein, and inhibin. This tumor resembles the recently described clear cell endocrine tumors of the gallbladder and pancreas that are associated with von Hippel-Lindau disease. Our patient, however, had neither personal nor family history indicative of von Hippel-Lindau disease. Furthermore, published accounts of clear cell endocrine tumors in von Hippel-Lindau disease describe immunoreactivity for inhibin; the current case was negative for the disease. There may be a subtype of clear cell carcinoid tumor not associated with von Hippel-Lindau disease, which is characterized by its lack of immunoreactivity against inhibin.
Carcinoid tumor is a rare lesion in the gallbladder.1 Only 18 examples have been found among 3557 gastrointestinal carcinoid tumors examined during a 19-year period by the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute.2 Clear cell carcinoid tumor of the gallbladder is even more rare; the one example of which we are aware arose in a patient with von Hippel-Lindau disease (VHL).3 That report described histologic features similar to clear cell endocrine pancreatic tumors that are also associated with VHL,4 including immunohistochemical positivity for inhibin.
We report herein a clear cell carcinoid tumor of the gallbladder in a 64-year-old man without VHL. Although the tumor of this particular patient had features that were histologically similar to those described in the setting of VHL, immunohistochemical staining for inhibin was negative.
REPORT OF A CASE
A 64-year-old man presented with acute onset of epigastric pain in March 2002. His medical history was remarkable only for a gastric ulcer diagnosed 3 years previously. His parents, 3 brothers, and 3 sisters had no medical history of VHL. Physical examination indicated no specific problem. Laboratory tests disclosed a nonspecific hepatobiliary disturbance. Ultrasonography of the abdomen showed 2 small gallstones (up to 5 mm in diameter) in the gallbladder with a small polypoid lesion at its neck (5 mm in diameter). The polyp was preoperatively diagnosed as a cholesterol polyp. A laparoscopic cholecystectomy was performed 6 days following his first presentation, after a diagnosis of cholelithiasis with a cholesterol polyp. After the operation, his hepatobiliary disturbance was corrected and he was well in November 2002.
PATHOLOGIC FINDINGS
Macroscopically, the wall of the gallbladder was mildly thickened. There were 2 small gallstones in the lumen. An 8-mm sessile polyp was present at the neck of the gallbladder (Figure 1). The polyp was golden yellow and had an uneven surface.
Histologically, the polyp was composed of trabecular and glandular structures, involving the gallbladder mucosa and the perimuscular subserosal layer (Figure 2). The lining cells had clear, cuboidal-to-low columnar or polygonal cytoplasm and small, round-to-oval nuclei (Figure 3). The original biliary glands were entrapped among the tumor glands (Figure 3, arrow). The lesion had a well-developed capillary network in its stroma. There was little or no desmoplastic stromal response. No aggregates of foamy macrophages were present in the stroma. Neither vascular nor lymphatic invasion was present. The tumor contained some area of pale eosinophilic cells (Figure 4). In some foci, pagetoid spreading was found in the biliary glands (Figure 5), but mitotic figure was rare. Periodic acid–Schiff stains with and without diastase demonstrated neither glycogen nor mucin in the clear cells. The rest of the specimen showed nonspecific cholecystitis.
Immunohistochemical analysis was performed on formalin-fixed, paraffin-embedded tissue using the avidin-biotin peroxidase complex method. Specific antibodies that were used, their sources, their dilution, and the results are summarized in the Table. The tumor cells showed strong cytoplasmic positivity for chromogranin A and synaptophysin (Figure 6, A and B). They were focally positive for gastrin, somatostatin, and pancreatic polypeptide but tested negative for glucagon, serotonin, insulin, and inhibin (Figure 6, C). S100 protein also failed to demonstrate sustentacular cells in the tumor.
COMMENT
Clear cell carcinoid tumor of the gallbladder and clear cell endocrine pancreatic tumor have been described as a distinctive manifestation of VHL.3 Clear cell carcinoid tumors that affect patients without VHL have been discovered in the thymus,5 lung,6 stomach,7,8 and appendix.9,10 The clear appearance of the cytoplasm is not caused by glycogen; several reports5,7–10 have demonstrated excess lipid in the tumor cells. Our case was initially diagnosed as a cholesterol polyp on ultrasound examination, and the tumor showed a golden yellow color macroscopically, which also caused it to resemble a cholesterol polyp. Histologically, however, it did not contain aggregates of foamy macrophages in the stroma, which are characteristic in a cholesterol polyp. Periodic acid–Schiff staining failed to show the presence of glycogen and mucin in the clear cytoplasm of the tumor cell. Thus, we believe the clear cell features in this particular case are due to lipid accumulation and that it is reasonable to suggest that there is a rare sporadic clear cell subtype of carcinoid tumor that can affect several organs, including the gallbladder. Because the whole specimen was processed and embedded in paraffin, we examined neither oil staining nor electron microscopy.
The clear cell carcinoid tumor and clear cell endocrine pancreatic tumor associated with VHL were diffusely immunopositive for inhibin,3,4 but our patient's tumor was negative for inhibin. Accordingly, inhibin might be a marker to differentiate the pancreatic and gallbladder clear cell tumors associated with VHL from sporadic tumors.4 At present, none of the reported cases of sporadic clear cell carcinoid in other organs were examined using immunohistochemical assays with inhibin.5–10 Further investigation about the staining quality of inhibin in additional cases of clear cell carcinoid tumor without VHL will be needed to support this hypothesis.
The differential diagnosis for clear cell tumors of the gallbladder includes adenocarcinoma (especially metastatic renal cell carcinoma) and paraganglioma. The strong positivity for neuroendocrine markers eliminates carcinoma from consideration. Paraganglioma of gallbladder is usually found incidentally and protrudes from the external surface.1 Furthermore, S100 protein usually decorates sustentacular cells in paraganglioma.
The biologic behavior of the clear cell carcinoid tumor of the gallbladder is still unclear, because, to our knowledge, there has been only one case reported, in which follow-up was not mentioned.3 The same is true of the ordinary carcinoid tumors in the gallbladder, because follow-up of the reported cases is short.1 Generally, small carcinoid tumors of the gallbladder were incidentally found in resected gallbladders for gallstones, and the tumors smaller than 1 cm did not show metastases.1 Sporadic clear cell carcinoid tumors of other organs have not shown poor prognosis; in 4 reports5,6,8,9 with follow-up for a total of 8 cases, only one patient had widespread metastases at the time of publication.5 Of course, careful follow-up is mandatory for our patient because the tumor invaded into the subserosal layer.
In conclusion, we describe clear cell carcinoid tumor of the gallbladder not associated with VHL. It has histologic features similar to clear cell carcinoid tumors associated with VHL, except for the absence of inhibin reactivity. Immunostaining for inhibin might be a useful marker for clear cell carcinoid tumors associated with VHL.4
Acknowledgments
We gratefully acknowledge the diagnostic advice of K. Krishnan Unni, MB, BS (Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minn).
References
Author notes
Reprints: Eiichi Konishi, MD, PhD, Department of Laboratory Medicine, Saiseikai Kyoto Hospital, 8 Imazato-Minamihirao, Nagaokakyo, Kyoto, 617–0814 Japan ([email protected])