To the best of our knowledge, this is the first case of solitary concomitant PET and ductal adenocarcinoma of the pancreas reported in the English literature. Ballas et al[
9] reported a case of mixed exocrine-endocrine tumor of the pancreas showing two distinct populations of tumor cells. Most parts of these two components were separated but focally showed a collision type of growth, with two different components growing into one another. Goh and colleagues[
10] reported three cases of concomitant intraductal papillary mucinous neoplasm (IPMN) of the pancreas and pancreatic endocrine neoplasm. The sizes of the endocrine part were small (2, 5 and 8 mm) and well-circumscribed, isolated and nonfunctional. However, the morphology of all three exocrine components was not indicative of classical invasive ductal carcinoma as the present one.
Many reports have discussed the histogenesis of the association between exocrine and endocrine neoplasms of the pancreas[
2,3,11,12]. Leteurtre et al[
3] reported a case of mixed ductal-endocrine carcinoma of the pancreas and supported the hypothesis of a common endodermal histogenesis for ductal and endocrine cells. Chatelain and colleagues[
2] also suggested that these tumors probably arise from a totipotent endodermal cell distributed in the pancreatic ducts and islets. Marrache et al[
11] presented 6 patients with associated IPMN and PET, and suggested that these tumors may arise from a common progenitor cell or due to differentiation from one cell type to the other. Morikane et al[
12] presented a small glucagonoma of the pancreas with evident ductular and tubular structures and supported the hypothesis that pancreatic endocrine tumors originate in the ductular epithelium. In the report of Cho et al[
13], the tumor cells of a mixed acinar-endocrine carcinoma of the pancreas showed amphicrine differentiation, with one cell type harboring both exocrine and endocrine markers simultaneously, indicating that a close histogenetic relationship exists between pancreatic exocrine and endocrine cells. Pour et al[
14] concluded that pancreatic cancer derives from pancreatic stem cells distributed within ductal trees and islets. According to the previous literature, it seems reasonable to hypothesize that associated exocrine and endocrine neoplasms of the pancreas arise from totipotent pancreatic stem cells, which reside in the pancreatic duct and islets. Under certain stimulation, the stem cells differentiate and give rise to exocrine and endocrine primitive cells. The exocrine primitive cells further differentiate into ductal and acinar cells. According to the proposed differentiation pathway, we hypothesize that the amphicrine type arises from early stem cells that proliferate into a neoplasm before differentiating into endocrine and exocrine cells (Figure ). Thus, the amphicrine type, as in the case reported by Cho et al[
13], contains only one cell type that has both exocrine and endocrine features. When the stem cells begin to differentiate into primitive exocrine and endocrine cells, three types of histological patterns may occur: mixed, collision and solitary concomitant. In the mixed type, the exocrine and endocrine cells are intimately intermixed together. Most reported cases of mixed exocrine and endocrine tumor of the pancreas are examples of this type[
2-7]. The collision type, as in the case reported by Ballas et al[
9], shows the endocrine part at one end and the exocrine part at the other end, with an intermixed central zone. The solitary concomitant type, first presented in our case, shows a single tumor composed of distinctly separate exocrine and endocrine components, with a dividing fibrous band found between the two parts. The last type is a multiple concomitant type that contains isolated exocrine and endocrine tumors simultaneously. The three cases reported by Goh et al[
10] can be classified as the early ones of this latter type.
Because the present tumor contained both pancreatic endocrine tumor and exocrine carcinoma, we tried to consider these two parts together when discussing the prognosis of this patient. No evidence of metastasis was found at abdominal CT, whole body positron emission tomography scan and during surgery. The size of the present endocrine tumor was 1.8 cm. The microscopic features showed scarce mitoses and less than 0.5% Ki-67+ cells. Neither extrapancreatic spread nor vascular invasion was identified. According to criteria of the WHO Committee for Histological Typing of Endocrine Tumors, the present endocrine tumor was classified as a well-differentiated endocrine tumor with benign behavior[
1]. In the previous literature and papers presenting cases of mixed exocrine and endocrine carcinoma, the authors suggested that the prognosis depends mainly on the stage of ductal carcinoma[
2,8,9]. We cannot be certain that the prognosis of our case depends on the ductal carcinoma component because this is not a case of mixed exocrine and endocrine carcinoma of the pancreas as previously presented. Further study with more cases is necessary.
Chatelain et al[
2] presented a case of mixed ductal-pancreatic polypeptide-cell carcinoma of the pancreas. In their study, the endocrine cells were not immunoreactive to p53 and Bcl-2, but the exocrine ductal cells were stained strongly positive, suggesting that over-expression of p53 and Bcl-2 plays a major role in neoplastic progression of the ductal component. We performed p53 and Bcl-2 immunostains in the present case. The endocrine part was negative for p53 and more than 55% of exocrine neoplastic cells showed strong intranuclear accumulation. However, both the endocrine and exocrine parts were negative for Bcl-2. The role of p53 and Bcl-2 in tumor histogenesis still needs further investigation.
Glucagonomas are rare and only represent about 10% of all functioning pancreatic tumors[
15-17]. Cases without typical symptoms usually remain undiagnosed, resulting in long-term medical waste, and mental and physical suffering of the patients[
18,19]. In our case, the patient only presented with DM and body weight loss and did not satisfy the criteria of glucagonoma syndrome. However, it was not certain whether the alpha-cell tumor plays a role in the manifestation of DM or not. In a recent study of pancreatic cancer-associated DM, Pannala et al[
20] concluded that pancreatic cancer is a powerful diagnostic state and new-onset DM is likely induced by the tumor. Under this viewpoint, the adenocarcinoma of our case may be closely related to the poor control of DM, especially because the patient did not develop glucagonoma syndrome. In short, we suggest that, in cases of type II DM occurring in middle-aged patients, the development of uncontrolled hyperglycemia should raise clinical suspicion of glucagonoma and/or pancreatic cancer. In this situation, further survey including pancreatic hormones and tumor marker profile is suggested.
In conclusion, combined pancreatic endocrine and exocrine tumors are rare. We have reported the first case of solitary concomitant pancreatic endocrine tumor and ductal adenocarcinoma. We suggest a classification system of tumor histogenesis for this group of tumors, including amphicrine, mixed, collision, solitary concomitant and multiple concomitant. Further investigation of more cases of this rare entity is still necessary.