Objectives: We assessed the reproducibility of the kinetic analysis of 3'-deoxy-3'-[(18)F]fluorothymidine (FLT) positron emission tomography (PET) in A431 human epidermoid carcinoma and murine Lewis lung carcinoma (LLC) tumor models.
Methods: We injected 7.4 MBq of FLT (n=10 for each group) and acquired 2-h dynamic PET images. A second scan was performed 1 day later. We calculated standardized uptake value (SUV), kinetic rate constants, volume of distribution of phosphorylated FLT (V(dm)), net influx constant (K(FLT-CA)) and influx constant by Patlak graphical analysis (K(FLT-PA)). The percent difference between measurements of a parameter was calculated to compare the reproducibilities of different parameters.
Results: FLT phosphorylation was higher in mice with A431 tumors than in mice with LLC tumors (P<.005). Differences in the standard deviations of the percent differences of parameters were statistically significant (P<.001) in each model. In mice with A431 tumors, SUV, V(dm), K(FLT-CA) and K(FLT-PA) had standard deviations of the percent difference of < or = 20%. The most reproducible parameter was K(FLT-PA), although the standard deviation (15.6%) was not statistically different from those of V(dm) (15.8%), K(FLT-CA) (17.5%) and SUV (18.9%). In mice with LLC tumors, K(1), K(1)/k(2) and k(3) had standard deviations of the percent difference of < or = 20%. No macroparameters reflecting a total FLT flux had standard deviations of < or = 20%.
Conclusion: Our results show the reproducibility of the kinetic macroparameters of FLT PET in mouse tumors with high FLT phosphorylation.