The diagnostic work-up of patients suspected of having neuroendocrine tumours (NETs) has traditionally been a challenging issue. The last two decades have been marked by the application to use in the diagnosis of NETs of 3 newly available diagnostic techniques: endoscopic ultrasonography (EUS), multidetector CT (MDCT), and more recently, positron emission tomography using 68Ga-labelled octreotide analogues (PET). In a prospective study conducted at a single referral centre that compared PET with conventional somatostatin receptor scintigraphy and MDCT in diagnosis, staging and follow-up of patients affected by NET, PET detected more primary and secondary lesions than other methods. Recent studies investigated the clinical impact of PET in the management of patients affected by NET, previously studied by MDCT. The investigators recently reported the results of the investigation of 19 patients suspected of having primary pancreatic NET and studied by PET, MDCT and EUS. The investigators preliminary data suggest that PET may be slightly more sensitive than MDCT in detecting small (<2cm) pancreatic lesions; accuracy of PET and EUS is probably similar. No prospective study has yet been devoted to evaluate the accuracy of PET in the diagnosis and staging of primary duodenal-pancreatic NETs. Furthermore, the clinical impact of the adjunct of PET to the traditional protocols of diagnosis and staging of these tumours waits to be thoroughly evaluated. Thus the appropriate place of PET in the diagnostic algorithm of patients suspected of having duodenal-pancreatic NET remains undefined.

The main aim of this project is to prospectively compare the accuracy of PET and MDCT in the diagnosis and staging of patients suspected of having duodenal-pancreatic NETs. The investigators hypothesised that PET is superior to MDCT in the diagnosis of these neoplasm (the dimension of the study sample is estimated in order to detect a 10% difference). The impact of PET on management plan of affected patients will also be evaluated. As a secondary endpoint of the study, the investigators will compare EUS, PET and MDCT in the diagnosis of primary duodenal-pancreatic NET. The study is designed as a multicentre, prospective, non-randomised clinical trial. All patients will undergo MDCT, PET and EUS in this fixed order.

Inclusion Criteria:

1. Patients affected by proved MEN-I, in whom a neoplasm in the duodenal-pancreatic area is suspected.
2. Patients with clinical diagnosis of carcinoid syndrome.
3. Patients with clinical diagnosis of Zollinger-Ellison syndrome.
4. Patients with insulinoma, as proved by fasting test.
5. Patient with clinical pictures and laboratory findings suggesting other functional or non-functional NET.
6. Patients who had previously undergone surgery, including total and subtotal pancreatectomy, or a duodenotomy, intended as curative for a histologically confirmed NET.
7. Patients who were diagnosed with NET metastasis with unknown primary location of the disease.
8. Patients undergoing diagnostic work-up for a periduodenal or pancreatic lesion incidentally found during abdominal ultrasound (not performed for suspicion of a NET) and with ultrasonographic characteristics (rounded, hypoechoic or egg-eye, well demarcated) suspicious for NET.
9. Patients undergoing diagnostic work-up for a periduodenal or pancreatic lesion incidentally found during TC (not performed for suspicion of a NET) and with radiological characteristics (well demarcated, hypervascular) suspicious for NET.

Exclusion criteria:

1. Patient unwilling, or unable to consent.
2. Pregnancy, or lactation.
3. Age <18 years
4. Known diagnosis of duodenal-pancreatic NET.
5. Patients with concomitant life-threatening disease.
6. Patients who had already undergone PET or EUS, in the last six months. In particular patients should be excluded from the study, when a lesion in the duodenal-pancreatic area, with characteristic suspicious for a NET, is incidentally diagnosed by PET, or EUS, or when a previously unsuspected diagnosis of NET is suggested by EUS-FNA of a pancreatic lesion.
7. Patients who had previously undergone total gastrectomy or pancreatectomy will be included in the study, but they will not undergo EUS.