Abstract
Background: Neuroendocrine neoplasms (NEN) are a heterogenous group of tumors. Patients with the multiple endocrine neoplasia type 1 (MEN1) syndrome often manifest with simultaneous functional and non-functional NEN in various endocrine glands. In MEN1, nuclear medicine plays an important role in the diagnostic work-up and localization of NEN. Little is known about the comparative efficacy of (68)Ga-Dotatate PET/CT (DOTA) versus (18)F-FDOPA PET/CT (FDOPA) and both versus non-nuclear medicine imaging (CT and MRI) in the identification of primary and metastatic NEN. Methods: This prospective MEN1 cohort study evaluated 15 germline MEN1 mutation-positive patients. Subjects were imaged using CT, MRI, DOTA and (18)F-FDOPA. Radiological review with a multidisciplinary team was performed for each patient. Results: One-hundred twenty-nine total lesions were identified using any of the four scans. DOTA sensitivity was 69% (89/129; 95% CI 61% to 76%) with a mean standardized uptake value (SUV) of 33.9 ± 30.1, FDOPA sensitivity was 18% (23/129; 95% CI 12% to 25%) with mean SUV of 12.1 ± 15.16. DOTA identified an additional 50 lesions not seen on CT (of which MRI detected 8 lesions with an average size 0.95 cm ± 0.48; 3 pancreatic, 2 duodenal, 2 liver, and 1 lymph node) and identified 55 lesions not seen on MRI (of which CT identified 13 with a mean size of 1.1 cm ± 0.45; 1 lung, 4 pancreatic, 2 duodenal, 1 liver, and 5 lymph nodes). Overall, CT detected 51.2% (66/129) of lesions (mean 0.61 cm ± 0.73; 95% CI 43% to 60%) and MRI detected 39.5% (51/129; 0.47 cm ± 0.71; 95% CI 32% to 48%), and there was no significant difference in the size of lesions detected (p=0.18). Analysis by organ NEN revealed equal sensitivity between FDOPA and DOTA for lung carcinoid, detecting 33% (4/12) of lesions, while CT detected 92% (11/12) of lesions. In the duodenum, DOTA identified 100% (11 /11) of lesions, while FDOPA had poor sensitivity (9%) in this location. Within the pancreas, DOTA has a sensitivity of 81% (31/38), while FDOPA had a sensitivity of 21% (8/38). CT localized 42% (16/38) of pancreatic lesions, of which MRI missed 6. Interestingly, DOTA missed 7 pancreatic lesions all approximately 1cm or larger, which is previously unrecognized (range 0.9 - 1.8cm). Twenty-three liver metastases were detected on anatomic imaging (CT identified 14, while MRI detected 15, with only 9 overlapping lesions). DOTA identified 60% (14/23) of lesions, of which 3 lesions were missed by CT and MRI. However, DOTA was more sensitive in the liver than FDOPA which only detected 2 lesions. FDOPA detected one lesion in the adrenal (0.9cm) that was not seen on DOTA. Conclusion: DOTA imaging proved to be superior to FDOPA, CT and MRI overall in detecting NENs in MEN1, specifically in the duodenum. Pancreatic NEN missed by DOTA may represent higher grade tumors and may benefit from (18)FDG PET/CT imaging.