Chapter 40 Endocrine Pancreas

Taylor S. Riall, Courtney M. Townsend, Jr.

The pancreas is a digestive organ located in the retroperitoneum (Fig. 40-1) that has endocrine and exocrine function. The endocrine cells are organized in discrete clumps throughout the pancreas, called islets of Langerhans. The primary physiologic function of the endocrine pancreas can be summarized as regulation of body energy, largely through hormonal control of carbohydrate metabolism. The islets secrete hormones directly into the bloodstream in endocrine fashion. Insulin is the hormone of energy storage, while glucagon is the hormone of energy release. Additional pancreatic endocrine hormones, such as somatostatin, play a role in the complex regulation of pancreatic exocrine secretion and digestion.

FIGURE 40-1 A, The pancreas is noted in its retroperitoneal position at the level of the second lumbar vertebra. B, Relationship of the head of the pancreas in the C loop of the duodenum, with the pancreatic duct and common bile duct emptying into the ampulla of Vater. C, On microscopic view, the endocrine cells are located in nests, called islets of Langerhans, which are distributed throughout the pancreas (Trichrome stain, ×10).

The pancreas was first identified by a Greek anatomist and surgeon, Herophilus (335-280 BC). In medieval Persia, in 1025, Avicenna provided the first detailed account of diabetes mellitus in the Canon of Medicine. He described a patient with an abnormal appetite, collapse of sexual function, and the sweet taste of diabetic urine. In 1889, Minkowski and von Mering, who were studying fat absorption in dogs after pancreatectomy, noted that the urine attracted flies. On analysis of the urine, they documented glucosuria and ketonuria. They also noted that surgical removal of the pancreas led to eventual coma and death. In 1869, while a medical student, Paul Langerhans described collections of pale staining cells within the pancreas, the islets that now bear his name (see Fig. 40-1). Eugene Opie was the first to associate diabetes with microscopic hyaline changes in the islets of Langerhans. Frederick Banting and Charles Best in Toronto discovered insulin in 1922. Banting and Best surgically ligated the pancreas of one set of dogs, leading to atrophy of the exocrine pancreas. They then removed and homogenized the pancreas and injected the homogenized extract into a diabetic dog, temporarily reversing this condition; a few injections each day could keep it healthy and free of symptoms. Banting and Best were awarded a Nobel Prize for this work.

Adult human pancreatic islets contain multiple cell types (Table 40-1). Alpha (A) cells secrete glucagon, beta (B) cells secrete insulin, delta (D) cells secrete somatostatin and vasoactive intestinal peptide (VIP), and F cells secrete pancreatic polypeptide (PP). Gastrin-producing cells are normally present in the fetal pancreas only. Islet cell tumors may secrete one or more of these hormones. The resulting syndromes are named for the peptide whose clinical symptoms predominate.

Table 40-1 Endocrine Cells of the Pancreas and Tumor Syndromes

In this chapter, we will cover the histomorphology, embryology, and physiology of the endocrine pancreas. We will briefly highlight new technologies, including autologous islet cell transplantation for chronic pancreatitis and allogeneic islet cell transplantation for type 1 diabetes. Our focus will be on the diagnosis and management of endocrine tumors of the pancreas.

Histomorphology of Islets

In the human fetus, pancreatic islets comprise approximately one third of the pancreatic mass. In the adult pancreas, there are approximately 10⁶ islets in the adult pancreas, accounting for less than 2% of the overall pancreatic mass. The average islet contains approximately 3000 cells and ranges in diameter from 40 to 900 µm. Each pancreatic islet should be considered a microorganism with a complex and definite organization, with only intact islet architecture enabling normal endocrine function.

The islet cell types are not distributed evenly within the islets. B cells constitute approximately 70% of the islet cell mass and are located centrally within the islet.¹ Insulin is the main secretory product of B cells, but they have also been shown to secrete amylin and cholecystokinin (CCK; see Table 40-1).² The A cells, located in the periphery, secrete glucagon and constitute approximately 10% of the islet cell mass. The D cells are evenly distributed throughout the islet and constitute approximately 5% of the islet cell mass. D cells secrete somatostatin and D₂ cells secrete VIP. Also located peripherally, F cells secrete PP. B and D cells are concentrated in the body and tail of the pancreas and F cells are concentrated in the head and uncinate process. This distribution is important clinically, because resection of different parts of the pancreas will have varying endocrine effects.

Pancreatic islets have a rich portal microcirculation that has significance in endocrine to endocrine cell signaling. Afferent arterioles enter the islet in an area of discontinuity in the peripheral, non-B cell mantle of cells. The order of islet cellular perfusion and interaction is from the B cell core outward to the mantle, and the mantle is further subordered with most D cells downstream or distal to most A cells. This allows B cells to inhibit A cell secretion and A cells to stimulate D cell secretion.³

Pancreatic endocrine secretion also regulates pancreatic exocrine secretion through the islet-acinar axis of the pancreas. Insulin stimulates pancreatic exocrine secretion, amino acid transport, and synthesis of protein and enzymes, whereas glucagon acts in a counterregulatory fashion, inhibiting the same processes. The role of somatostatin is controversial. Somatostatin may have a direct inhibitory effect on pancreatic acinar cells, which possess somatostatin receptors. It may also act through an inhibitory effect on islet B cells.

Embryology of the Endocrine Pancreas

During the fifth week of gestation, the pancreas begins forming at the junction of the foregut and midgut. It begins as two endodermal pancreatic buds, the dorsal bud and ventral bud, which eventually fuse to form the pancreas. The acinar cells and islet cells differentiate from the endodermal cells found in the embryonic buds. In humans, the first glucagon-producing cells are seen in 3-week-old embryos and the first islets of endocrine tissue appear at approximately 10 weeks. During this early developmental period, predominantly glucagon-positive islet cells initially appear in the tail of the pancreas. Subsequently, there is a major amplification of endocrine cell numbers, particularly B cells. The mature pancreas consists of the endocrine islets of Langerhans, digestive enzyme-secreting acinar cells contained in clusters of acini, and acinar-draining ducts, with accompanying blood vessels and lymphatics.

Islet cells were initially believed to arise from neural crest cells. Gittes and Rutter ⁴ have studied the patterns of expression of hormonal messenger RNA and concluded that both endocrine and exocrine cells of the pancreas arise from the foregut endoderm, a view that is now generally accepted. Early glucagon-positive endocrine cells convert to nonepithelial cells and lose connection with the lumen and tight junctions. This conversion to a nonepithelial location of endocrine cells has been postulated to entail a change in cell division polarity, from perpendicular to the basement membrane to parallel to the basement membrane. There also appears to be downregulation of pdx1, a key marker of early pancreatic progenitor cells, in these endocrine progenitor cells as they become nonepithelial. This conversion process has been postulated to parallel mesenchymal to mesenchyme transformation.

Endocrine Physiology

The primary function of the endocrine pancreas is regulation of body energy. This is achieved primarily through control of carbohydrate metabolism. Insulin secreted by the endocrine pancreas functions to story energy by decreasing blood glucose levels and increasing glucose transport into cells, except beta cells, hepatocytes, and cells of the central nervous system. Insulin also stimulates protein synthesis and inhibits the breakdown of glycogen and fat stores. Glucagon functions antagonistically to insulin, increasing blood glucose levels through stimulation of glycogenolysis, lipolysis, and gluconeogenesis.

Insulin

Insulin is a 56–amino acid polypeptide with a molecular weight of 6 kDa. It consists of two polypeptide chains (A and B) joined by two disulfide bridges. Although the amino acid sequence varies among species, the locations of the disulfide bridges are highly conserved and are critical for its biologic activity. Insulin is synthesized as a precursor peptide, called proinsulin; the two polypeptide chains are joined by means of connecting peptide (C peptide). In response to pancreatic B cell stimulation by glucose, proinsulin is synthesized in the endoplasmic reticulum and transported to the Golgi complex, where it is cleaved into insulin and the residual C peptide (Fig. 40-2). Insulin is then moved via microtubules into secretory granules, where it is released directly into the bloodstream via exocytosis. C peptide and insulin are secreted in equimolar amounts.

FIGURE 40-2 Diagram of insulin synthesis. Proinsulin, synthesized by the endoplasmic reticulum, is packaged within secretory granules of the beta cell, where it is cleaved to insulin and C-peptide. Equimolar amounts of insulin and C-peptide are secreted into the bloodstream.

(From Andersen DK, Brunicardi FC: Pancreatic anatomy and physiology. In Greenfield LJ, Mulholland MW, Oldham KT, et al [eds]: Surgery: Scientific principles and practice, ed 2, Philadelphia, 1997, Lippincott-Raven, p 869.)

There is significant secretory reserve of insulin within the pancreas. Destruction or removal of 80% of the pancreatic islet cell mass is necessary before endocrine dysfunction becomes clinically apparent in the form of type 1 (insulin-dependent) diabetes. Defects in the synthesis and cleavage of insulin can lead to rare forms of diabetes mellitus, such as Wakayama syndrome and proinsulin syndrome.⁵

The B cell is sensitive to even small changes in glucose concentration and is maximally stimulated at concentrations of 400 to 500 mg/dL. In response to glucose, the endocrine pancreas immediately reacts with a short burst of stored insulin (4 to 6 minutes), followed by a sustained secretion of insulin, which requires active synthesis of the hormone within the islet cell. Insulin has a 7- to 10-minute half-life and is primarily metabolized by the liver. Excess insulin is then slowly metabolized by the liver, kidneys, and skeletal muscles. Brain cells and red blood cells do not take up insulin.

Insulin binds to a specific 300-kDa glycoprotein cell surface receptor, which has been isolated and well characterized. After receptor stimulation, glucose is actively transported across cell membranes throughout the body by 55-kDa membrane-bound glucose transporters. There are several classes of glucose transporters, with varying affinities for glucose. Stimulation of the insulin receptor is dependent on insulin concentration. Insulin resistance, present in type 2 diabetes, can be the result of decreased numbers of receptors or a decreased affinity of receptors for insulin. Sulfonylurea compounds, which act independently of glucose concentration, also stimulate insulin secretion and are used in the treatment of type 2 diabetes, in which the primary defect is peripheral insulin resistance.

Orally administered glucose has a greater effect on insulin secretion than an equivalent amount of glucose administered IV, even though blood glucose levels might be the same. This effect is called the enteroinsular axis and is related to the release of enteric peptide hormones from the proximal gastrointestinal tract, which augment nutrient-induced insulin secretion. These insulinotropic factors, called incretins, act directly on the B cells and include gastric inhibitory polypeptide (GIP), glucagon, glucagon-like peptide-1, CCK, amino acids (arginine, lysine, and leucine), and free fatty acids. Humoral inhibitors of insulin secretion include somatostatin, amylin, leptin, and pancreastatin.

Insulin secretion is also under neuronal control. Vagal (cholinergic) stimulation leads to the release of insulin. Alpha-sympathetic stimulation strongly inhibits insulin release, whereas beta-sympathetic fibers stimulate it. Insulin release is stimulated by the peptidergic nerve release of gastrin-releasing peptide (GRP), CCK, gastrin, enkephalin, and VIP, whereas insulin release is inhibited by neurotensin, substance P, and somatostatin. A loss of pancreatic innervation in the setting of pancreatic transplantation or islet cell transplantation can result in changes in the pattern and quality of insulin secretion.

Glucagon

Glucagon is a 29–amino acid, straight chain polypeptide with a molecular weight of 3.5 kDa. Secreted by the A cells, the primary function of glucagon is to elevate blood glucose levels through stimulation of glycogenolysis and gluconeogenesis in the hepatocytes. Glucagon secretion is tightly controlled by neural, hormonal, and nutrient factors. A and B cells respond primarily to serum glucose concentration, but in a reciprocal fashion. Insulin and glucagon are counterregulatory hormones and function together to maintain glucose homeostasis. Dysfunctional secretion of glucagon may play a role in the elevation of blood glucose levels in diabetes. The diabetes resulting from total pancreatectomy is very brittle and difficult to control because of the lack of endogenous glucagon to balance exogenously administered insulin. Like epinephrine, cortisol, and growth hormone, glucagon is considered a stress hormone because it increases metabolic fuel in the form of glucose during stress. Glucagon secretion is stimulated by sympathetic neural transmitters, epinephrine, and the amino acids arginine and alanine. Insulin and somatostatin have a suppressive effect on glucagon secretion.

Somatostatin

Somatostatin, secreted by islet D cells, is a 14–amino acid polypeptide weighing 1.6 kDa. Although it is logical to think that somatostatin modulates the secretion of other islet hormones, its actual function within the pancreas remains unknown. Although exogenous administration of somatostatin has been shown to inhibit the release of insulin, glucagon, and PP, and to inhibit gastric, pancreatic, and biliary secretion, endogenous somatostatin has not been proven to influence the secretion of other islet hormones directly.

A synthetic octapeptide, octreotide, has been developed, which mimics the pharmacologic action of somatostatin. It has a longer half-life in the serum than endogenous somatostatin and is a more potent inhibitor of growth hormone, glucagon, and insulin secretion than the natural hormone. The potent inhibitory effect of octreotide has been used to treat exocrine and endocrine disorders of the pancreas, including secretory diarrhea, bowel fistulas, pancreatic fistulas, and endocrine hypersecretory syndromes.

Pancreatic Polypeptide

PP is a 36–amino acid, 4.2-kDa polypeptide secreted by the F cells of the pancreatic islet. The physiologic role of PP remains unclear; its clinical usefulness is limited to its role as a marker for other endocrine tumors of the pancreas. Cholinergic innervation predominantly regulates PP secretion. As a result, surgical vagotomy ablates the increased PP response normally observed after meals. In diabetes and normal aging, PP secretion is increased, resulting in increased circulating PP levels. Absence of PP may play a role in the diabetes observed after total pancreatectomy or after chronic atrophic pancreatitis.

Other Peptide Hormones

Other peptides are secreted by the pancreatic islets. These include neuropeptides such as VIP, amylin, galanin, and serotonin. VIP is a 28–amino acid, 3.3-kDa polypeptide that stimulates insulin release and inhibits gastric secretion. It is found not only throughout the gastrointestinal tract, but in the respiratory tract, where it causes vasodilation and bronchodilation. Amylin, a 36–amino acid polypeptide, is secreted by B cells and inhibits the secretion and uptake of insulin. Amylin deposits in the pancreas of patients with type 2 diabetes have been implicated in the pathogenesis of the disease. Pancreastatin is part of a larger ubiquitous molecule, chromogranin A, which inhibits insulin secretion. Gastrin-producing cells are present in the fetal pancreas, but not in the normal adult pancreas. Many additional peptides, including thyrotropin-releasing hormone, glicentin, CCK, peptide YY, GRF, calcitonin gene–related peptide, prolactin, adrenocorticotropic hormone (ACTH), parathyroid hormone–related protein, and ghrelin have been reported in normal islets and in islet cell tumors.

Surgical Treatment of Diabetes

Autologous Islet Cell Transplantation

Autologous islet cell transplantation has a role in patients with severe chronic pancreatitis. Surgical resection of all or part of the pancreas for this disease can significantly improve quality of life by eradicating or reducing intractable pain, allowing the return of a normal appetite with subsequent needed weight gain, and reducing the number of hospital admissions. A major disadvantage of a total pancreatectomy, however, is that it renders the patient a brittle diabetic. Partial resections can also greatly reduce the insulin-secreting capacity of the already compromised pancreas, sometimes necessitating exogenous insulin after surgery. Regardless, even without pancreatic resection a significant number of patients with chronic pancreatitis will progress to develop diabetes or impaired glucose tolerance. Because of the loss of insulin, glucagon, and PP, the type of diabetes that develops in patients with chronic pancreatitis is similar to that following pancreatic resection.

Total or partial pancreatectomy with islet autotransplantation is being offered in several centers in the United States. This option has the potential to treat the symptoms of chronic pancreatitis definitively while preventing the onset of diabetes in certain patients. Other patients remain or become insulin-dependent but retain significant insulin and glucagon secretion and the benefits of endogenous C peptide production, thus making the resulting diabetes easier to control. Our dedicated islet isolation facility and a diagrammatic representation of the process are shown in Figure 40-3. Patients undergo pancreatectomy; the pancreatic tissue is immediately digested with the use of enzyme solutions containing collagenase and neutral proteases and the islet cells are purified. The islet cells are then returned to the patient via infusion into the portal vein. The islet cells engraft in the liver and produce insulin and C peptide and glucose levels are measured to evaluate the function of the transplanted islets.

FIGURE 40-3 A, Dedicated islet isolation facility at our center (University of Texas Medical Branch). B, The screen in the upper right hand corner shows isolated islets stained red from a patient undergoing total pancreatectomy and islet autotransplantation. C, Pancreatic islet autotransplantation. The patient undergoes partial or total pancreatectomy. The pancreatic tissue is immediately digested with using enzyme solutions containing collagenase and neutral proteases and the islet cells are purified. The islet cells are then returned to the patient via infusion into the portal vein.

Depending on center expertise and experience, variable results following pancreatic islet autotransplantation have been reported. Insulin independence is as high as 40% to 50% initially in some patients,⁶^–⁸ but there is a notable decline in islet function over time, with increased insulin requirements in the 10 years following transplantation, with only about 10% of patients remaining insulin-independent. Although insulin-independence is not always achieved, most patients are C peptide positive and have diabetes that is more manageable. In addition, all studies have demonstrated improvement in pain and other symptoms of chronic pancreatitis.^9,¹⁰ Success rates depend on the number of isolated and transplanted islets and transplanted islets as well as the cause of the pancreatic disease. Patients who are not diabetic before autotransplantation and younger patients (especially preadolescents) achieve the best results. The most feared procedure-related complication is thrombosis of the portal vein, occurring in less than 1% of cases.

Immune Therapy, Pancreatic Transplantation, and Islet Allotransplantation

Type 1 diabetes results from autoimmune destruction of pancreatic islets. Immune treatment for type 1 diabetes is currently being investigated. There is a growing body of evidence to suggest that the autoimmunity observed in patients with type 1 diabetes is the result of an imbalance between autoaggressive and regulatory T cell subsets.^11,¹² Vaccination with selected T cell receptor autoantigens has been shown to generate autoantibodies and the autoaggressive T cell clones, which are reacting to beta cells. The induction of a lasting, robust immune response generating autoantigen-specific regulatory T cells provides strong justification for further testing of this therapy for type 1 diabetes.

The current surgical treatment for type 1 diabetes is transplantation of allogeneic islet cell tissue by means of whole organ transplantation or transplantation of isolated islets, usually infused into the portal vein. Pancreatic transplantation was first performed in 1966 by Kelly and colleagues. From 1966 through 2008, over 30,000 pancreas transplantations have been reported to the International Pancreas Transplant Registry, including more than 22,000 from the United States.¹³ Whole organ pancreas transplants restore euglycemia almost immediately following transplantation, and 1-year graft survival rates in the United States have improved to 85% for simultaneous pancreas-kidney transplants, 78% for pancreas after kidney transplants, and 76% for pancreas-only transplants. Recipients experience immediate normal fasting and postprandial glucose levels, and hemoglobin A1c levels return to normal. With the observed decrease in morbidity and mortality, recipients who become insulin-independent report a better quality of life, despite the need for immunosuppression. They also experience stabilization or improvements in retinopathy, nephropathy, neuropathy, and microvascular and macrovascular diseases normally associated with poor glucose control. Therefore, for the select group of patients with labile diabetes who do not respond well to conventional approaches or insulin pumps, whole pancreas transplantation remains the gold standard for treatment.

Allogeneic islet cell transplantation is the other option. Currently, long-term insulin independence remains elusive for patients undergoing allogeneic islet transplantation. The data show that even with patients who receive multiple infusions, few remain normoglycemic over time. Data from the Collaborative Islet Transplant Registry (CITR) have demonstrated that 70% of patients achieve insulin independence within the first year (including patients with multiple infusions) but, by the third year, the percentage of patients who remain euglycemic is closer to 35%.^13a The partial pancreatic endocrine function confers some benefit, with decreased occurrence of severe hypoglycemic events, abatement of hypoglycemic unawareness, persistent C peptide levels, improvement in glycemic control, and stabilization of diabetic complications.

Whole pancreas transplantation procedures currently outnumber islet transplantation procedures. Pancreas transplantation is associated with a higher surgical morbidity, whereas islet transplantation is a less invasive method of achieving insulin independence. However, pancreas transplantation is associated with a higher success rate. In addition, islet cell transplantation requires two donors per recipient to maintain graft function.¹⁴ Stem cell therapy offers the potential of producing an unlimited source of cells, and a growing number of studies have demonstrated successful in vitro differentiation and expansion of embryonic cells of murine and human origin from pancreatic ducts that express insulin and respond to glucose stimulation.

Islet Cell Tumors

Overview and History

Endocrine tumors of the pancreas in the United States are rare, with an estimated incidence of 5 to 10 cases/1 million population annually. These tumors are 1000 to 2000 times more common in autopsy statistics, indicating that most are benign and nonfunctional. Endocrine tumors of the pancreas vary greatly in the mode of onset, severity of symptoms, location, functionality, and malignant potential.¹⁵ The incidence of malignancy in these tumors varies from approximately 10% in insulinomas to almost 100% in glucagonomas and somatostatinomas (see Table 40-1). On hematoxylin and eosin–stained sections, all pancreatic endocrine tumors look similar. Malignancy is determined by the presence or absence of metastases and immunostaining allows for the identification of the endocrine content of the cells (Fig. 40-4). Over time, they may vary significantly in secretion of hormone products and biologic aggressiveness. Although the tumor syndromes are classically attributed to pancreatic islet tumors, tumors are often found in extrapancreatic locations, such as the duodenum and peripancreatic soft tissue. Almost all insulinomas, glucagonomas, and VIPomas arise from the pancreas, whereas most gastrinomas occur in the duodenum. Stomatostatinomas are equally divided between the pancreas and proximal small bowel.

FIGURE 40-4 Pathology of a pancreatic endocrine tumor stained positive for chromogranin, a neuroendocrine tumor marker. The chromogranin is cystoplasmic and stains brown.

(Courtesy Dr. Christine Iacobuzio-Donahue, Johns Hopkins University School of Medicine, Baltimore.)

The morbidity from pancreatic endocrine tumors is a result of both secretion of active gastrointestinal hormones and the malignant potential. Secretion of hormones by functional tumors leads to the characteristic syndromes and physiologic derangements associated with these rare neoplasms. Although multiple hormones may be produced by a single tumor, the syndrome is recognized and named for the clinical signs and symptoms associated with the predominant endocrine agent.

In 1908, Nichols described a pancreatic adenoma consisting of islet cell tissue. In 1935, Whipple and Frantz were the first to report an association between a clinical syndrome and an islet cell tumor. They described hyperinsulinism and the associated symptoms that became known as Whipple’s triad—the appearance during fasting of neuroglycopenic symptoms of hypoglycemia, low blood glucose (<45 mg/dL), and relief of symptoms by the administration of glucose. Over the next 25 years, additional syndromes associated with islet cell tumors were described. In 1942, Becker described a patient with severe dermatitis, anemia, and diabetes who also had an islet cell tumor; McGarvan later identified the cause of the syndrome as glucagon-secreting islet cell carcinoma of the pancreas. In 1955, Zollinger and Ellison described two patients with a fulminant peptic ulcer diathesis, acid hypersecretion, and non–beta islet cell tumors of the pancreas.¹⁶ It was later determined that the secretagogue was gastrin. The first description of watery diarrhea and hypokalemia related to an islet cell tumor was by Priest and Alexander in 1957. In 1958, Verner and Morrison described two patients who died from refractory watery diarrhea and hypokalemia and an associated islet cell tumor. Later, this syndrome was clearly defined when patients with this constellation of symptoms and islet cell tumors were found to have high circulating levels of VIP. The development and refinement of sensitive radioimmunoassay techniques in 1956 allowed for the detection of micromolar concentrations of circulating peptides and greatly contributed to our understanding of these syndromes.

Molecular Genetics of Islet Cell Tumors

Similar to the adenoma-carcinoma sequence in colorectal cancer, tumorigenesis of islet cells and other neuroendocrine cells involves an accumulation of a number of genetic events, including activation of oncogenes and inactivation of tumor suppressor genes (Fig. 40-5). This progression is distinct from that of pancreatic adenocarcinoma. Mutations in the k-ras, p53, dpc4, myc, fos, jun, src, and retinoblastoma (mainly RB1) genes are not seen. Transcriptional silencing is believed to play a role in islet cell tumorigenesis. More than 90% of gastrinomas and nonfunctioning neuroendocrine tumors had homozygous deletions or epigenetic silencing by 5′ CpG island methylation.¹⁷ Loss of heterozygosity (LOH) at chromosome 11q is common in functional pancreatic endocrine tumors, whereas LOH at chromosome 6q is associated with the development of nonfunctional tumors.¹⁸ One third of patients with sporadic pancreatic endocrine tumors have allelic loss on chromosome loci 3p35, 3p27, and 11p13, suggesting that these loci encode tumor suppressor genes critical for the development of endocrine tumors. This allelic loss is associated with malignant clinical disease.¹⁹ More than 90% of these tumors demonstrate silencing of the tumor suppressor gene p16/MTS. Evers and colleagues²⁰ have shown amplification of the proto-oncogene HER-2/neu, but not p53 or ras in gastrinomas. Others have reported an increase only in aggressive tumors.²¹ Studies of insulinomas have shown that the G protein G_s has threefold greater expression in insulinoma when compared with normal islet cells, suggesting that it may be involved in unregulated insulin secretion or tumorigenesis. Activation of the myc oncogene, TGF-α, and ras genes may be early genetic events in insulinoma tumorigenesis. Loss of the sex chromosome (X in women and Y in men) has been identified in pancreatic endocrine tumors and appears to be associated with an aggressive phenotype.