RT Journal Article SR Electronic T1 Kinetic modeling of [18F]FES PET in patients with metastatic ER+ breast cancer JF Journal of Nuclear Medicine JO J Nucl Med FD Society of Nuclear Medicine SP 547 OP 547 VO 61 IS supplement 1 A1 Iqbal, Ramsha A1 Menke-van der Houven van Oordt, Catharina A1 Oprea-Lager, Daniela A1 Schuit, Robert A1 Windhorst, Albert A1 Boellaard, Ronald A1 Yaqub, Mohammed YR 2020 UL http://jnm.snmjournals.org/content/61/supplement_1/547.abstract AB 547Introduction: 16α-[18F]-fluoro-17β-estradiol ([18F]FES) is a positron emission tomography (PET) tracer, developed for in vivo visualization of the estrogen receptor (ER). [18F]FES PET appears to be a powerful imaging tool for diagnosis, staging and prediction of response to endocrine therapy. Currently, standardized uptake values (SUV) derived from static [18F]FES PET scans at 60 min post-injection are used for predicting response to endocrine therapy. However, prior to its use in clinical practice validation against the gold standard for quantification (i.e. full kinetic modeling with blood sampling and metabolite correction) is essential. The aim of this study was to describe the optimal kinetic model for the quantification of [18F]FES uptake and to assess the validity of simplified quantitative parameters of [18F]FES uptake in patients with metastatic ER+ breast cancer. Methods: Eight patients with metastatic ER+ breast cancer were included in this prospective study (sub study of a phase I trial: NCT03455270), consisting of a 50-min dynamic [18F]FES PET/CT scan which started directly after intravenous administration of 200 MBq [18F]FES (±15%). Volumes of interest (VOIs) were defined on PET scans for tumor lesions (40% iso-contour of max using the last frames) and aorta ascendens (early frames) for extraction of tumor time-activity curves (TACs) and image derived input functions (IDIFs), respectively. Input functions were corrected for calibration, plasma-to-blood ratio and metabolites (determined with radio-high performance liquid chromatography) using discrete collected venous samples over various time points. Single tissue (1T2k), two-tissue irreversible (2T3k) and -reversible (2T4k) plasma input compartment models, all including a blood volume fraction fit parameters, were evaluated. Tested simplified measures consisted of SUV, tumor to whole-blood and metabolite corrected plasma ratio (TBR) calculated over various time intervals. Results: A total of 52 tumor lesions (bone: 31, lymph nodes: 13, subcutaneous fat: 7, lung: 1,) could be analyzed in all patients. The Akaike criterion and visual analysis of the fits preferred 2T3k model (40% preference). Three lesions from one patient were excluded due to a small k3 rate (<0.02). The net influx rate (2T3k Ki) correlated well with SUV and whole blood and metabolite corrected TBRs at 20-40 min (R2 = 0.77, 0.86, 0.82, respectively) and 30-50 min post-injection (R2 = 0.82, 0.89, 0.85, respectively). Conclusions: The 2T3k model is the most optimal model to describe the kinetics of the [18F]FES tracer in tumor lesions. The simplified measures SUV and TBR can be used for reliably assessing tracer uptake, preferably at the latest evaluated uptake interval. Yet, TBR using whole blood showed highest correlations with full quantitative results. Key words: [18F]FES, PET, estrogen receptor, breast cancer, kinetic modeling