In order to improve the sensitivity and specificity of pre-operative imaging localization of foci in adult endogenous hyperinsulinemic hypoglycaemia (AHH) we aim to evaluate a novel promising imaging compound targeting the glucagon-like peptide-1 receptor (GLP-1R), 68Ga-NODAGA-exendin 4. With the currently used imaging techniques, only about 70-80% of insulin producing pancreatic neuroendocrine tumors (IPPNET) can be successfully visualized. Therefore, we propose to compare GLP-1R PET imaging to the standard imaging techniques in patients scheduled for surgical removal of the tumour. These highly relevant data will allow us to interpret the benefits of GLP-1R-imaging over other imaging techniques for the diagnosis of IPPNET in AHH patients. Since pre-operative localization of foci in AHH remains challenging and frequently still leading to futile surgery or unnecessary partial pancreatectomy, a more sensitive and specific imaging technique would be of great value.

Adult endogenous hyperinsulinaemic hypoglycaemia The most common form of functional neuroendocrine tumours of the pancreas are insulin producing pancreatic neuroendocrine tumors (IPPNET). These tumors are rare and have an incidence of 1-4 newly diagnosed cases per 1 million per year and are malignant in about 10% of the cases. Another cause of AHH is nesidioblastosis, or adult beta cell hyperplasia. It is difficult to exactly determine the incidence of the disease, but it appears that in approximately 5% of the cases of AHH nesiodioblastosis may be the underlying pathology, while IPPNET are responsible for the majority of the cases. The pathophysiological cause of nesidioblastosis is not well understood, but the rising incidence of the AHH as a consequence of gastric bypass surgery for morbid obesity (although often reversible in these patients) may suggest an association with metabolic and hormonal changes.

For IPPNET, surgical removal of the tumour is the therapy of choice and is considered curative in case of a benign tumour. Optimal preoperative localization of the lesion is warranted in order to reduce morbidity by helping to optimize the surgical procedure. Successful preoperative localization of IPPNET is a challenging problem since approximately 30% of IPPNET cannot be visualized using the conventional imaging techniques CT and/or MRI and endoscopic ultrasound. Selective arterial stimulation with calcium with simultaneous venous sampling (ASVS) has been described to have a sensitivity and specificity of almost 90% in identifying IPPNET. This is, however, an invasive technique which is accompanied by an risk for complications. Functional imaging with somatostatin (sst) receptor scintigraphy (SRS) and SPECT/CT are able to detect less than 50% of benign IPPNET because of low or absent expression of sst receptor subtypes 2 and 5, which bind octreotide with high affinity. PET with 68Ga-labeled sst analogs has a higher sensitivity for smaller lesions than SRS. Also 11C-5-HTP and 18F-DOPA, which are used as PET tracers for the detection of IPPNET in some centers may be more sensitive than SRS and CT with 11C-5-HTP showing the most accurate visualization. However, when compared to intra-operative findings, 2 out of 6 IPPNET could still not be detected preoperatively. Palpation and intra-operative ultrasound will allow identifying the lesion in approximately 70-80% of cases. However, it remains a challenge to find small or multiple tumours in the pancreas and partial pancreatectomy is frequently required, especially if the lesion is located close to the pancreatic duct. Precise preoperative localization of the IPPNET is therefore critical to minimize surgical intervention. If no IPPNET can be identified pre- or perioperatively, the diagnosis of nesidioblastosis may be established by resection and histopathologic evaluation of the pancreatic tail. If nesidioblastosis is present, partial pancreatectomy is required; the challenge is to remove enough tissue in order to avoid hypoglycaemia while keeping enough functional endocrine pancreatic tissue so that the patient does not become diabetic. Currently, with the risk of reoperation being considered lower than the consequences and complications of diabetes, surgeons usually choose a fairly conservative approach.

Imaging of diseased beta cells Visualization of the beta cells by a highly specific radiotracer with which high target-to background ratios can be obtained would benefit preoperative visualization of IPPNET in patients with AHH. Reliable visualization of diseased beta cells would then benefit the optimization of treatment of patients with AHH. An innovative method for imaging of beta cells could allow to optimally guide surgical interventions In addition, this novel approach could lead to a minimization of side-effects from the treatment.

Targeting of the GLP-1 receptor In this study we will compare the sensitivity and specificity of pre-operative imaging of IPPNET by GLP-1R scanning to the current standard imaging techniques.

Inclusion Criteria:

• Biochemically proven endogenous hyperinsulinemic hypoglycemia
• Signed informed consent
• Standard imaging not older than 8 weeks.

Exclusion Criteria:

• Breast feeding
• Pregnancy or the wish to become pregnant within 6 months
• Calculated creatinine clearance below 40ml/min
• Evidence of other malignancy than insulin producing tumors in conventional imaging (suspicious liver, bone and lung lesions)
• Age < 18 years
• No signed informed consent