We aimed to compare the diagnostic performance of 68Ga-FAPI-04 (FAPI) in comparison to 68Ga-DOTATATE (SSTR) PET/CT for patients presenting with recurrent medullary thyroid carcinoma (MTC). Sixteen MTC patients with elevated calcitonin levels (>150 pg/mL) underwent FAPI and SSTR PET/CT. Two nuclear medicine physicians evaluated all images, categorizing lesions into locoregional metastases, mediastinal lymph nodes (LNs), liver, and bone metastases. SUVmax and tumor-to-background ratio were recorded. PET modalities were compared using the McNemar test. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of FAPI and SSTR PET were calculated. The cohort comprised 16 patients (50% female; mean age 50 ± 17 years). Median calcitonin and CEA levels were 6234 pg/mL and 17.3 ng/mL, respectively. In patient-based analysis, SSTR exhibited higher diagnostic sensitivity compared with FAPI (88% vs 81%), resulting a statistically significant difference (P = 0.004). Mean SUVmax and tumor-to-background ratio values were 10.3 and 5.35 for FAPI, and 9.7 and 11.9 for SSTR PET, respectively. In lesion-based analyses, FAPI demonstrated higher accuracy than SSTR for cervical LNs (91.9% vs 50%), mediastinal LNs (94.9% vs 54.4%), and liver metastases (57.4% vs 7.3%), respectively. Notably, 31% of patients (n = 5) with FAP-expressing liver lesions showed no uptake on SSTR imaging. MRI confirmed liver metastases in 3 of these patients; however, 2 FAP-expressing lesions were confirmed as hemangiomas. False-positive findings of DOTA primarily included reactive LNs and bone hemangiomas. FAPI PET presents promising outcomes in detecting metastases in recurrent MTC patients. Although its diagnostic performance matches SSTR on a per-patient basis, FAPI PET exhibits superior sensitivity and accuracy in lesion-based analyses, notably for liver and bone metastases.
Read full abstract