62 Kinetic data extraction from [formula omitted]F-FDOPA PET brain acquisition

Abstract

Introduction The indications of Positron Emission Tomography (PET) using radiolabeled amino acids in neurology are growing especially for the diagnosis of brain tumors. The last classification “WHO 2016” (World Health Organization) concerning the gliomas implemented molecular parameters such as the IDH mutation and the 1p/19q co-deletion correlated with clinical observed outcomes. The aim of this study was to assess the diagnostic performance of the 18 F-FDOPA (L-3, 4-Dihydroxy-6- 18 F-fluoro-phenyl-alanine) from dynamic PET recordings. Methods Patients initially referred for newly diagnosed gliomas had been sorted according to the WHO classification into 3 distinct groups: oligodendrogliomas, astrocytomas IDH + and glioblastomas according to the anatomico-pathological results. Dynamic recordings were performed on a Biograph6 True Point (Siemens Healthcare) PET scanner after the injection of 3 MBq of 18F-FDOPA/kg. For each patient, a scan of 30 min was performed immediately after the intravenous injection. The data had been reconstructed in dynamic method with a 2D OSEM iterative algorithm (2 iterations, 21 subsets, Gaussian filtering with a FWMH of 4 mm) and an adapted temporal sampling, allowing a compromise between kinetic information and noise. A circular region of interest (ROI) with a diameter of 2 cm was centered on the pixel of the maximum tumor uptake on the static images (summed of the dynamic images). Then, this ROI was applied to the entire dynamic dataset in order to obtain the time-activity curve (TAC) of the 18 F-FDOPA uptake in the tumor. The slope of the curve was obtained by fitting with a linear regression function the data recorded between 10 and 30 min post-injection. Results Dynamic PET images were reconstructed into 30 time-frames of 1 min each. The TAC obtained with raw data were fitted with the following Eq. (1) SUV ( t ) = a 0 + ( a 1 - a 0 ) × A + ( t a 2 ) p B + ( t a 2 ) q and the mean correlation coefficient R2 was equal to 0,958. The time corresponding to the maximum of the tumor fixation was superior to 20 min after the injection for the oligodendrogliomas and astrocytomas IDH + whereas it was about 6 min for the glioblastomas. The slopes of the TAC were positive for the oligodendrogliomas (+0,3 SUV/h), negative for the astrocytomas IDH+ (-0,8 SUV/h) and more negative for the glioblastomas (-6 SUV/h). Conclusions Dynamic 18 F-FDOPA PET recordings could be used to extract quantitative parameters allowing non-invasive staging of gliomas in patients referred for initial diagnosis.