Comparative Evaluation of 68Ga-FAPI and 68Ga-DOTATATE PET/CT and PET/MRI in Thyroid Eye Disease

This is a prospective, single-center cohort study planned to enroll approximately 50 adults aged 18 to 75 years. The thyroid eye disease cohort will include approximately 30 participants diagnosed according to European Group on Graves' Orbitopathy criteria. The control cohort will include approximately 20 participants without a history of thyroid-related or orbital disease who undergo clinically indicated FAPI or somatostatin receptor PET imaging for another reason; orbital PET/MR acquisition will be added to the existing examination without additional radiation exposure.

Participants in the thyroid eye disease cohort will undergo 68Ga-FAPI and 68Ga-DOTATATE PET/CT and PET/MR imaging within one week. For each tracer, imaging will begin approximately 50 minutes after intravenous administration of 3 to 5 mCi. PET/CT will include a CT acquisition for attenuation correction and anatomic localization followed by three-dimensional PET acquisition of the head for approximately 5 minutes per bed position. PET/MR will include simultaneous PET acquisition and routine MR sequences, including T1-weighted, T2-weighted, and FLAIR imaging. PET images will be reconstructed using an ordered-subset expectation maximization algorithm and fused with CT or MR images.

Nuclear medicine physicians will evaluate images without access to clinical data. Each eye will be visually classified as positive when any extraocular muscle demonstrates uptake greater than background and negative otherwise. For participant-level analysis, a participant will be considered positive when at least one eye is positive. Quantitative analysis will measure maximum standardized uptake value in the superior, medial, lateral, and inferior rectus muscles. A 1-cm region of interest in occipital soft tissue will be used as background. The target-to-background ratio will be calculated as extraocular muscle SUVmax divided by background SUVmax.

Clinical information collected from electronic medical records will include age, sex, disease duration, clinical diagnosis, treatment plan, and Clinical Activity Score. Imaging measures will be compared between groups and modalities and will be examined for associations with Clinical Activity Score and extraocular muscle thickness. Receiver operating characteristic analysis will be used to evaluate diagnostic accuracy.