Endoscopic ultrasound outperforms CT/MRI and somatostatin receptor scintigraphy for pancreatic solid lesion detection in von Hippel-Lindau patients

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: MON 292-305-MEN1, MEN2 & Pheochromocytomas
Basic/Clinical
Monday, June 17, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board MON-302
Sophie J van Asselt*1, Adrienne H Brouwers2, Hendrik M van Dullemen2, Eric J van der Jagt2, Alfons H Bongaerts2, Ido P Kema2, Klaas P Koopmans3, Bernard A Zonnenberg4, Henri JLM Timmers5, Wouter W. de Herder6, Wim J Sluiter2, Elisabeth G de Vries2 and Thera P Links2
1University of Groningen, University Medical Center Groningen, Netherlands, 2University of Groningen, University Medical Center Groningen, Groningen, Netherlands, 3Martini Hospital, Groningen, Netherlands, 4University Medical Center Utrecht,, Utrecht, Netherlands, 5Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 6Erasmus MC, Rotterdam, Netherlands
Background: Patients with von Hippel-Lindau (VHL) disease are prone to develop pancreatic neuroendocrine tumors with malignant potential. In VHL, benign serous cystadenomas also occur in the pancreas, which can mimic neuroendocrine tumors on anatomical imaging. Currently, it is unknown which imaging technique is best for detection of pancreatic neuroendocrine tumors in VHL. In this head-to-head comparison study we aimed to evaluate the value of endoscopic ultrasound (EUS) and of 11C-5-hydroxytryptophan positron emission tomography (11C-5-HTP PET) versus standard imaging for pancreatic solid lesion detection, suspicious of neuroendocrine tumor.

Patients and methods: We conducted a cross sectional study in 22 patients in a tertiary care university medical center. Patients with a germline VHL mutation or with one VHL-manifestation and mutation carrier as 1st grade family member, with screening by abdominal computed tomography (CT) or magnetic resonance imaging (MRI) within 4 months and somatostatin receptor scintigraphy (SRS) within 6 months, were eligible. Patients underwent 11C-5-HTP PET and linear EUS with the possibility to obtain cell material by EUS-guided fine needle aspiration (FNA). For all imaging techniques, positivity of pancreatic solid lesions was calculated at the level of individual patients and of individual lesions. The sum of positives of all imaging served as a composite reference standard. EUS and 11C-5-HTP PET were compared with standard screening using Mc Nemar’s test for comparison.

Results: In 10 patients (45%), 20 pancreatic solid lesions were detected with a median size of 9.0 mm (range 2.9-55). Seventeen solid lesions were detected with EUS (P < 0.05 versus CT/MRI+ SRS) in 10 patients, 3 with 11C-5-HTP PET (P = 0.11 versus CT/MRI+ SRS) in one patient, 3 with SRS in three patients, 9 with CT/MRI in seven patients and 9 with CT/MRI+ SRS in seven patients. Next to solid lesions, EUS found in 18 patients multiple pancreatic cysts with a median of 4 cysts (range 1-30) per patient. FNA was performed in 7 solid lesions with a median of 2 passes (range 2-6). In 2 lesions (29%), tumor cells were detected. The remaining samples did not yield enough cells for a diagnosis.

Conclusion: EUS is superior to CT/MRI+ SRS to detect pancreatic solid lesions in VHL disease. In this setting 11C-5-HTP PET is of no value. For confirmation of diagnosis by EUS-guided FNA, improvement of cell yield is warranted and needs further research.

Nothing to Disclose: SJV, AHB, HMV, EJV, AHB, IPK, KPK, BAZ, HJT, WWD, WJS, EGD, TPL

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm

Sources of Research Support: Dutch Cancer Society Grant (2008-4188)