Molecular imaging in the diagnostic process of neuroendocrine tumors: a systematic review on unknown primary origin and suspected NETs.

Abstract

Background: Neuroendocrine tumors (NETs) are a heterogeneous group of neoplasms often arising in the gastrointestinal tract, pancreas, and lungs. Diagnosing NETs can be challenging due to their small size, slow growth, and low metabolic activity, especially in cases of unknown primary origin (CUP-NETs). Timely identification of the primary lesion is critical for tailored treatment planning, including peptide receptor radionuclide therapy (PRRT), surgery, or somatostatin analogue administration.

Objective: This systematic review evaluates the diagnostic performance and clinical impact of molecular imaging-especially PET/CT with radiolabeled somatostatin analogs-in patients with suspected NETs or CUP-NETs.

Methods: Following PRISMA guidelines, a comprehensive literature search was conducted in PubMed and Scopus. Data on study design, tracers used, detection rates, and impact on patient management were extracted. Quality assessment was performed using CASP tools.

Results: PET/CT using [68 Ga]Ga-DOTA-peptides (DOTATATE, DOTATOC, DOTANOC) consistently outperformed conventional SPECT imaging with [111In]Pentetreotide (Octreoscan), achieving primary tumor detection rates of 38%-83% versus < 10% with SPECT. In suspected NETs, PET/CT demonstrated high sensitivity (up to 95%) and specificity ( > 85%), with changes in clinical management reported in up to 33% of cases. Dual-tracer imaging ([68 Ga]Ga-DOTATATE and [18F]FDG) and newer tracers such as [18F]DOPA and [18F]AlF-NOTA-octreotide ([18F]-OC) provided added diagnostic value, particularly in aggressive or rare subtypes (e.g. paragangliomas, insulinomas). Detection of unknown primaries commonly led to curative surgical options or more targeted therapies.

Conclusions: Molecular imaging-particularly [⁶⁸Ga]Ga-DOTA-peptide PET/CT-represents an essential component in the diagnostic evaluation of undefined/suspected NETs. It not only improves diagnostic accuracy but also has a profound impact on clinical decision-making. Its integration into diagnostic algorithms should be strongly considered for both known and suspected cases of neuroendocrine neoplasia.