Ghrelin is a 28 aminoacid acylated peptide related to the oxyntomodulin family of intestinal peptides isolated from the X/A-like neuroendocrine cells of the rat and human stomach (391;392). It is predominantly produced by the stomach, but is also detectable in many other tissues (391-402). It is found in bowel, hypothalamus, pituitary, pancreas (391;403) cosegregating with pancreaticalpha cells but also possibly with pancreatic beta cells,(403;404). Since its discovery barely 5 years ago there have been about 650 publications indicating a profound interest in the newest GEP hormone capable of stimulating growth hormone release by activation of the GH secreotogogue type 1 a (GHS-R1a) receptor. The peptide stimulates GH release in animals and humans by acting at both the pituitary and hypothalamic level (405). It also stimulates the release of ACTH and prolactin as well as gastric acid secretion and intestinal motility (393;399;406;407). Ghrelin is an important regulator of energy balance because it has been demonstrated to increase appetite and food intake and modulate insulin secretion negatively (404;408-410). Consistent with its homology with motilin and oxynto modulin it is not surprising that the peptide stimulates gastric motility and gastric acid secretion (407;411). Ghrelin is the first natural hormone in which a hydroxyl group on one of its serine residues is acylated by n-octanoic acid (391). This acylation is essential for binding to the GHS-R1a receptor, for the GH releasing capacity and also likely for its other actions (391;412;413). Although it has been found to cosegregate with glucagon and insulin by some authors this is not consistent and most would agree that its cell of origin in the pancreas constitutes a new cell type (402;403;414). It seems to exert a tonic inhibitory role on insulin secretion (398;410;415-418) which has been found in animals and humans. Ghrelin significantly increases blood glucose levels by reducing insulin secretion (419) and has been shown to increase insulin resistance when administered systemically in humans (420-422). As a corollary it is suppressed by hyperglycemia and insulin (417;423;424), but may in addition have a direct role on glycogenolysis (419).

Expression of Ghrelin protein and or mRNA has recently been identified in almost all gastric and intestinal carcinoids as well as pancreatic neuroendocrine tumors (425;426). Moreover, there have now been two case reports of Ghrelinomas. In one ghrelin was cosecreted with glucagon in a predominantly glucagon expression syndrome (427) and in another nonfunctioning tumor Ghrelin levels were >12,000 pM (n=300pM), and despite the 50 fold increase in levels the patient had normal serum GH and IGF-1 levels. In this study (428) no attempt was made to distinguish acylated Ghrelin from the non-acylated variety hence all the circulating Ghrelin may have indeed been biologically inert

Corbetta and colleagues in Milan in Italy Corbetta(428) studied 40 consecutive patients referred with newly diagnosed, untreated gastroenteropancreatic neuroendoocrine tumors. They measure the circulating levels of Ghrelin in 16 patients with gastrointestinal Carcinoid (10 with midgut, and 6 gastric carcinoids, 24 patient with pancreatic tumors (8 gastrinomas, 2 insulinomas, 2 vipomas, 1 glucagonoma and 11 nonfunctioning tumors] and compared their findings with 35 healthy controls. The mean Ghrelin concentration in these 42 patients 182.7 + 66.5 pM was not significantly different from the control value of 329+ 32 pM. Furthermore there were no significant differences between gastrointestinal and pancreatic, functioning and non-functioning tumors or those with and without metastases. As indicated above they found 1 patient with astronomic levels that did not influence the metabolic status of the patient, the ghrelin appearing to be biologically inert. This was a 66 year old woman with hypertension and type 2 diabetes for 5 years who developed intestinal sub-occlusion. An octreoscan revealed a pancreatic mass with concomitant hepatic metastases. Plasma samples showed increased chromogranin A (CGA) and PP levels and normal values for gastrin and somatostatin and a dramatic increase in ghrelin levels of > 12,000 pM. Despite the 50 fold increase in ghrelin, IGF-1 and GH levels were normal and there were no clinical features of acromegaly. Gastric acid secretory status and the reason for the subocclusion of bowel were not reported. Cortisol secretion was also normal but no other markers of pituitary hypothalamic function were reported. Laparotomy was performed and the tumor and its metastases were immunohistochemical positive for ghrelin as well as neuron specific enolase and secreotogranin. The patient received chemotherapy with epirubicin, 5-fluoruracil, and dacarbazine which caused a decrease in the ghrelin levels to 6600pM in parallel with a reduction in CGA but we do not learn about change in clinical status. She then received recombinant interferon a2b and ghrelin levels remained stable but the patient died 28 months later. The cause of death was unknown.

Based upon the physiologic effects of Ghrelin one would expect that the clinical features of a ghrelinoma would include the following;

1. Hyperglycemia
2. Insulin deficiency
3. Insulin resistance
4. GH excess and increased IGF-1 levels.
5. Acromegaly
6. Gastric acid hypersecretion
7. Intestinal dysmotility

It seems however for now that ghrelin appears to be another hormone produced in almost all gastroenteropancreatic neuroendocrine tumors, has little if any biological activity, and may be useful as a marker for response to therapy. From the screening point of view it does not seem to offer a great deal since all the neuroendocrine tumors of pancreas or gut presented by Corbetta et al (428) had markedly elevated CGA levels (save for possibly one, and the response to therapy showed a parallel reduction of ghrelin and CGA. One should still hold in reserve the possibility that the predicted tumors will emerge in time. Coincidentally, non secretory tumors of the pancreas were for the most part PP producers. Similar to PP (360), it seems that ghrelin may run the risk of a parallel course with initial excitement as occurred with PP which has since sunk into oblivion because of its lack of a biological effect. The difference of course is that ghrelin has been shown to have many effects when administered as the acylated form. The increase in the endogenous levels in the reported tumors be a variant without biological activity, but retaining the structural eiptopes sufficient to be recognized by the antisera to Ghrelin. Acylation specific antisera will help to resolve part of this quandary.