Abstract Purpose Radiolabelled fibroblast activation protein inhibitors (FAPI) are promising molecular imaging tracers for the diagnosis and staging of pancreatobiliary cancer. To enable treatment response assessment with 68 GaGa-FAPI-46, the day-to-day variability of its uptake must be defined. This study aimed to determine the repeatability of semi-quantitative 68 GaGa-FAPI-46 PET/CT measurements in pancreatobiliary cancers. Methods Patients with pathologically confirmed pancreatobiliary cancer (pancreatic ductal adenocarcinoma (PDAC) or cholangiocarcinoma (CCA)) were included. Patients underwent two 68 GaGa-FAPI-46 PET/CT scans, within 7 days, without any treatment before or between scans. Static long axial field-of-view (LAFOV) PET/CT scans were performed 60 min post-injection of 68 GaGa-FAPI-46 (EARL2-compliant reconstruction). Suspected malignant lesions were semi-automatically delineated (local background-adjusted 50% isocontour of the peak standardised uptake value (SUV)) and semi-quantitative measurements were extracted (SUV mean , SUV peak and SUV max ). Tumour-to-background ratios (TBR) were also calculated. Repeatability was assessed using the repeatability coefficient (RC) and the intraclass correlation coefficient (ICC). Results Twelve patients were included (seven PDAC, three perihilar CCA and two intrahepatic CCA). The median injected dose was 216 MBq (interquartile range (IQR) 167–266 MBq) and the median uptake time was 60 min (range 60–68). In total, 70 FAPI-positive lesions were delineated. The RCs of SUV mean , SUV peak and SUV max were 23.7%, 23.9% and 29.8%, respectively. For the blood pool adjusted TBR mean , TBR peak and TBR max , the RC’s were 21.6%, 20.8% and 28.0%, respectively. All ICCs were above 0.98. Conclusion Semi-quantitative measurements of 68 GaGa-FAPI-46 PET/CT have an excellent repeatability and can potentially be used in future studies to assess treatment response in pancreatobiliary cancers.
Building similarity graph...
Analyzing shared references across papers
Loading...
Rutger B. Henrar
Matthijs C. F. Cysouw
Xavier Palard-Novello
European Journal of Nuclear Medicine and Molecular Imaging
Inserm
Radboud University Nijmegen
Vrije Universiteit Amsterdam
Building similarity graph...
Analyzing shared references across papers
Loading...
Henrar et al. (Fri,) studied this question.
www.synapsesocial.com/papers/696c789ceb60fb80d1396d2d — DOI: https://doi.org/10.1007/s00259-025-07738-6