Abstract
1660
Objectives FDG-PET/CT is increasingly used for radiotherapy(RT) planning in lung cancer(NSCLC).For the planning purposes it is necessary to register PET and CT.The influence of the registration method of PET with CT data on PET-based gross tumor volume(P-GTV) definition is not known to date.We compared intra- and interindividual differences of P-GTV between linear and non-linear registered PET and CT.
Methods Sixteen patients with histologically proven NSCLC underwent FDG-PET and three CT acquisitions(expiration-EXP,inspiration-INS,mid-breathhold-MID) in RT position on the same day.All scans were registered with linear(L) and non-linear(NL) procedure resulting in six fused datasets:L-INS,L-EXP,L-MID,NL-INS,NL-EXP and NL-MID.For the delineation of P-GTVs a self developed contrast oriented algorithm based on tumor to background ratio was used.To accelerate the delineation a semiautomatic software prototype (Hermes Medical Solutions) was utilized.
Results The mean P-GTV determined in L-RT-PET-CT was 86.1±106.2 ml, whereas the P-GTV in NL-RT-PET-CT was 93.2±110.1 ml (r=0.98). With regard to the standardized uptake values (SUV) of the tumor there was also an excellent correlation (r=0.98) between linear RT-PET-CT (10.1±4.1) and non-linear RT-PET-CT (10.1±3.9). In intra- and interindividual comparison regarding the different breathing manouvers there were no significant differences with regard to the P-GTVs as well as tumor diameters and SUVs of all data sets showing an excellent correlation (r=0.96-0.99).
Conclusions Our results indicate that there is no difference in the size of FDG-based GTVs after registration with CT regardless of the breathing manouver or registration method used.
- © 2009 by Society of Nuclear Medicine