Search

Noninvasive tumor growth monitoring is of particular interest for the evaluation of experimental glioma therapies. This study investigates the potential of positron emission tomography (PET) using O-(2-[18F-fluoroethyl)-L-tyrosine ([18F]-FET) to determine tumor growth in a murine glioblastoma (GBM) model--including estimation of the biological tumor volume (BTV), which has hitherto not been investigated in the pre-clinical context. Fifteen GBM-bearingmice (GL261) and six controlmice (shams) were investigated during 5 weeks by PET followed by autoradiographic and histological assessments. [18F]-FET PET was quantitated by calculation of maximum and mean standardized uptake values within a universal volume-of-interest (VOI) corrected for healthy background (SUVmax/BG, SUVmean/BG). A partial volume effect correction (PVEC) was applied in comparison to ex vivo autoradiography. BTVs obtained by predefined thresholds for VOI definition (SUV/BG: ≥1.4; ≥1.6; ≥1.8; ≥2.0) were compared to the histologically assessed tumor volume (n = 8). Finally, individual "optimal" thresholds for BTV definition best reflecting the histology were determined. In GBM mice SUVmax/BG and SUVmean/BG clearly increased with time, however at high inter-animal variability. No relevant [18F]-FET uptake was observed in shams. PVEC recovered signal loss of SUVmean/BG assessment in relation to autoradiography. BTV as estimated by predefined thresholds strongly differed from the histology volume. Strikingly, the individual "optimal" thresholds for BTV assessment correlated highly with SUVmax/BG (ρ = 0.97, p < 0.001), allowing SUVmax/BG-based calculation of individual thresholds. The method was verified by a subsequent validation study (n = 15, ρ = 0.88, p < 0.01) leading to extensively higher agreement of BTV estimations when compared to histology in contrast to predefined thresholds. [18F]-FET PET with standard SUV measurements is feasible for glioma imaging in the GBM mouse model. PVEC is beneficial to improve accuracy of [18 F]-FET PET SUV quantification. Although SUVmax/BG and SUVmean/BG increase during the disease course, these parameters do not correlate with the respective tumor size. For the first time, we propose a histology-verified method allowing appropriate individual BTV estimation for volumetric in vivo monitoring of tumor growth with [18]-FET PET and show that standardized thresholds from routine clinical practice seem to be inappropriate for BTV estimation in the GBM mouse model. [ABSTRACT FROM AUTHOR]