In Brief

Article Outline

• Copyright

Since the initial presentation of the 2 index cases in 1955 there have been more than 1000 patients reported and in excess of 3300 published articles on the Zollinger-Ellison syndrome (ZES). Although a rare cause of ulcer disease, ZES has impacted not only our understanding and treatment of gastric hypersecretion and neuroendocrine tumors, but also ushered in the field of gastrointestinal endocrinology. The intent of this monograph is to provide the reader with in-depth contemporary knowledge of ZES. The initial section is a clinical pathological correlation in 2 cases that will emphasize the clinical aspects of ZES. In the subsequent sections the reader will be provided a comprehensive review of gastrinoma including: 1) the physicians and scientists who have made seminal contributions to the current understanding of ZES and its diagnosis and treatment, 2) gastrin physiology, 3) clinical presentation, 4) molecular pathogenesis, 5) diagnosis, 6) treatment of sporadic and familial gastrinoma (MEN 1), 7) management of metastatic disease, and 8) staging and prognosis.

The case histories will serve as an introduction to this monograph. They emphasize several important points about the ZES that will be discussed in detail in the following sections, but also emphasize some clinical points that are not readily apparent from an overview of the current literature. Case number 1 was that of the longest surviving ZE patient from the original report. The case spanned nearly 50 years. Case number 2 is a recently diagnosed ZES patient, 14 months since her gastrinoma excision.

In comparing the cases the symptom complex is somewhat different. Case 1 had the classic triad of gastric hypersecretion, primary ulcers of the jejunum, and a non-beta islet cell tumor of the pancreas. Case 2 had a different constellation of symptoms including severe dysphagia with esophageal stricture and gastroesophageal reflux with abdominal pain, weight loss, and diarrhea in the initial presentation. Case 2 demonstrates that the symptom complex has changed and may include, in addition to duodenal ulceration, esophageal complications as well as the triad of abdominal pain, weight loss, and diarrhea. Although she had duodenal ulcers, not all patients do. A high index of suspicion is required to establish the diagnosis in these cases. Some patients with ZES may have the disease masked by proton pump inhibitors (PPIs). In addition, case 2 shows that gastrinomas may elude detection by conventional cross-sectional imaging and emphasizes the value of somatostatin scintigraphy and selective secretin stimulation in gastrinoma localization and the effectiveness of image-guided surgery in permitting complete normalization of hypergastrinemia.

A review of historical considerations gives a perspective of the disease and the significant contributions made by numerous surgeons, physicians, and scientists. We highlight the most significant contributions that have allowed us to understand the diagnosis and treatment over the past 50 years.

Presently, gastrin is thought to be a family of hormones with varying degrees of bioactivity and tissue distribution. Understanding the multiple forms of gastrin requires a discussion of the human gastrin gene and its products. According to the current literature the human gastrin gene consists of 3 exons, 2 of which give rise to coding sequences. Biosynthetic labeling experiments have confirmed that preprogastrin is translated in the endoplasmic reticulum. Preprogastrin is converted to progastrin, which progresses through the Golgi complex to the trans-Golgi network where rapidly following phosphorylation cleavage of progastrin occurs and the products progress to the dense core vesicles of the secretory pathway. Post-translational modifications occur resulting in the final products, G-17 and G-34. Other small fragments of 14- and 6-amino-acid residues may occur in tissue, but are not thought to have biologic significance.

Gastrin is synthesized in the G-cells, open-ended endocrine cells found predominantly in the gastric antrum. G-cells are present in smaller numbers in the duodenal mucosa. Gastrin release is controlled by chemical, neural, or mechanical stimuli. Gastrin release is stimulated by digestive proteins (notably phenylalanine and tryptophan), calcium, epinephrine, achlorhydria, and gastric distention. Gastrin release is inhibited by β-blockade. Vagal parasympathetic control is complex, with both stimulatory and inhibitory actions. Activation of vagal cholinergic reflexes by hypoglycemia or sham-feeding stimulates gastrin release. Atropine blocks gastrin stimulation. Truncal vagotomy causes an increase in basal and food-stimulated gastrin release. Hypergastrinemia occurs after truncal vagotomy. These studies suggest that vagotomy interrupts cholinergic inhibitory pathways. In addition, a variety of peptides have been shown to affect gastrin release. Gastrin release is stimulated by bombesin (gastrin releasing peptide) and is inhibited by somatostatin, secretin, glucagon, gastric inhibitory peptide (GIP), and vasoactive intestinal peptide (VIP).

ZES is usually diagnosed in the fifth decade of life, although it may occur in children and adolescents or the elderly. It is diagnosed between the ages of 20 and 60 in 90% of cases. Patients with multiple endocrine neoplasia (MEN 1) have onset of the disease at a younger age. In sporadic gastrinoma patients only 2% are correctly diagnosed at initial presentation compared with 5% with MEN 1. The latter observation is somewhat surprising because of the high frequency of MEN 1 in ZES, occurring in approximately 25% of cases, but ranging from 15% to 38% in reported series. The most frequent initial diagnosis is idiopathic peptic ulcer disease noted in 75%. In approximately 20% of patients, a variety of diarrheal conditions were reported as the initial diagnosis. The average duration of symptoms before diagnosis was 6 years in our experience. The mean delay in the reported NIH experience was 5.2 years, and this was not different for eras before or after the introduction of H-2 receptor antagonists and PPIs.

Despite numerous publications and widespread awareness of ZES, delay in diagnosis persists. Analysis of reported series indicates several features that should lead the physician to suspect ZES and shorten the delay in diagnosis including: 1) the combination of abdominal pain, diarrhea, and weight loss, 2) recurrent or refractory ulcers, 3) prominent gastric rugal folds (secondary to the trophic effect of gastrin) seen on endoscopy (94% in NIH series) and 4) gastrointestinal symptoms with or without ulcers occurring in an MEN 1 patient. It is recommended that patients in these groups have a fasting serum gastrin determination off PPIs for a minimum of 72 hours and possibly up to 7 days.

Although some advances have occurred in our understanding of the molecular pathogenesis of gastrinoma and other endocrine tumors, there are still many unknown details in the tumorogenesis of these tumors. The goals of investigations in this area are several-fold. The first goal is to be able to identify molecular markers that may identify the subset of 25% to 30% of patients with gastrinoma who have aggressive disease and may be candidates for different treatment regimens. The second goal is to develop novel therapeutic approaches that may take advantage of the molecular abnormalities in these tumors. The literature is organized into 5 major areas concerning the molecular biology of gastrinoma: 1) oncogenes, 2) tumor suppressor genes, 3) growth factors, 4) chromosomal abnormalities, and 5) receptor expression.

One can arrive at the following conclusions about the molecular biology of gastrinoma. There remains much to be learned about the development of these tumors and determinants of biologic behavior. We seem to know more about what factors are not important in biologic determination than those which are. However, synthesizing the information, several positive observations have been made. The molecular pathogenesis of gastrinoma may involve oncogenes—specifically c-myc and HER-2/neu and tumor suppressor gene mutations, MEN 1 gene, DPC4/Smad, p16INK4a. Markers for aggressive biologic behavior of neuroendocrine tumors including gastrinomas include oncogenes; RET proto-oncogene, overexpression of growth factor receptors; epidermal growth factor receptor (EGF-R) and insulin-like growth factor I receptor (IGFIr), and chromosomal abnormalities; chromosome 1q, chromosome 22q, and the X chromosome.

The measurement of fasting serum gastrin is the initial diagnostic test performed in a patient suspected of having ZES. Gastrin is determined by radio-immunoassay that is readily available in commercial laboratories. However, fasting hypergastrinemia alone is insufficient to establish the diagnosis since a number of other medical conditions may cause hypergastrinemia.

The following questions are answered: 1) How common is a normal gastrin in ZES? 2) Is the absolute level of fasting serum gastrin diagnostic of ZES? 3) How common are gastrin levels greater than 100 times the upper limit of normal observed in ZES? 4) Is the fasting serum gastrin greater in MEN 1 patients with gastrinoma? 5) Is there a relationship between the fasting gastrin level and basal acid output (BAO) and maximal acid output (MAO)? 6) Is there a relationship between the gastrin level and clinical characteristics of the patients? 7) Is there a relationship between tumor characteristics such as primary location, tumor size, and extent of disease and the level of serum gastrin?

The measurement of fasting serum gastrin is necessary in patients suspected of ZES. If possible, this should be done with the patient not receiving acid-suppressive medications, particularly PPIs. The absolute level is not diagnostic. All patients should have provocative gastrin testing. In addition, patients with normal fasting gastrin level in the presence of symptoms of ZES should have a gastrin stimulation test. The absolute level of fasting gastrin is associated with BAO, a pancreatic primary location, tumor size, and extent of disease, but is not predictive of gastrinoma associated with MEN 1.

Secretin was reported to cause stimulation of gastrin in ZES by Hansky and colleagues in 1971 and was clearly elucidated by Isenberg and colleagues in 1972. Secretin is administered as an IV push. Minimal side-effects occur and may include flushing and nausea. It is not necessary to discontinue PPIs for this test. Important questions concerning this test have been answered in the literature including: 1) What are the criteria for a positive secretin test? 2) What is the false negative rate for a secretin provocative test in ZES? 3) What is the false positive rate for secretin provocation in non-ZES patients? 4) What dose and method of administration of secretin should be employed? 5) Does the response to secretin correlate with clinical manifestations of ZES? 6) Is there a correlation with the gastrin response to secretin and tumor characteristics? A positive secretin test is defined as a rise in baseline gastrin of more than 110 pg/ml by some others whereas others require an increase of 200 pg/ml or greater.

Other provocative tests discussed include: calcium stimulation test, meal stimulation, and glucagon stimulation.

There has been interest in the measurement of other gastrointestinal hormones in gastrinoma to find additional markers of the disease that may assist with diagnosis and which may correlate with extent of disease. In this regard 3 proteins have been of interest: pancreatic polypeptide, pancreastatin, and chromogranin A.

An algorithm for the diagnosis and localization of gastrinoma is presented. Initially a fasting gastrin should be done. This should be obtained with the patient not receiving PPIs and perhaps not receiving H-2 receptor antagonists as well. In a patient in whom the physician has a strong suspicion of ZES, but has a normal fasting gastrin, then repeating the fasting gastrin level is recommended since gastrin secretion may vary from day to day. In this situation the mean of 3 samples is recommended. If the gastrin is elevated, a gastric pH should performed. If there is acid present, then the patient should have a secretin stimulation test. If the patient has achlorhydria, then an evaluation for atrophic gastritis and pernicious anemia is indicated. In this circumstance, parietal cell antibodies and vitamin B-12 should be obtained. In a patient who cannot discontinue PPIs, a secretin provocative test is indicated. The PPIs do not need to be discontinued for this test. If the secretin test is negative in a patient in whom the physician has a high index of suspicion, a calcium stimulation test may be done. Alternatively, the secretin test may be repeated or empiric imaging may be completed. A negative secretin test is a rare event in ZES. If the secretin test is positive, then initial localization should be done with computed tomography and somatostatin scintigraphy. If these are negative then additional imaging with magnetic resonance imaging and/or endoscopic ultrasound is warranted. If the tumor cannot be localized with these tests, then selective arterial secretin stimulation may be performed.

This monograph concludes with a lengthy section on treatment of ZES. This includes discussion of medical treatment of gastric acid hypersecreton, special considerations for pregnancy, pediatric patients, and the elderly, and surgical treatment of gastrinoma. Subsections include sporadic gastrinoma, duodenal gastrinoma, lymph node primary tumors, pancreatic primary tumors, and ectopic gastrinomas. Surgical treatment for cure in sporadic gastrinoma and MEN 1 as well as management of metastatic disease is discussed at length.

There have been many advances in our understanding of the pathogenesis and treatment of gastrinoma. Future research will undoubtedly contribute new methods of diagnosis, imaging, and treatment. The most important messages to take away from this monograph are the needs to establish the diagnosis early and develop consistent and uniform terminology for the staging of this heterogeneous group of patients. ZES is a rare cause of ulcer disease that requires a high index of suspicion on part of the treating physician to establish the diagnosis. Despite advances in treatment, it is clear that the greatest impact on improved survival and preventing disease progression is early diagnosis. From multiple reports, patients with early stage disease in both sporadic and MEN 1 gastrinoma have longer survival. The symptoms of the disease may be masked by long-term treatment with PPIs. The triad of abdominal pain, diarrhea, and weight loss in the presence of ulcer disease should alert the physician to ZES and prompt consideration of a fasting gastrin determination. To prevent delayed diagnosis, it may be more prudent to consider a fasting gastrin determination off PPIs in all patients who require PPIs for more than 6 weeks. The development of uniform terminology defining extent of resection and stage of disease for gastrinoma will be essential to compare results of treatment. Many of the proposed systems have merit, yet no consensus has been achieved. We predict that the next major advance in gastrinoma will be defining the terminology and language that will allow not only the collation of outcomes from multiple centers but will enhance the ability to perform multi-institutional prospective trials for this rare disease.