Impact of FDG-PET on delineation for radiotherapy in lung cancer
Objectives:
1. To assess the impact of fused diagnostic and planning FDG PET/CT scans on CT voluming of lung cancer
2. To compare inter and intraobserver variability in CT contouring
3. To assess the impact of FDG-PET/CT on variability in contouring between registrars and a reference radiation oncologist
Methods: Five radiation oncologists, five radiation oncology registrars and a radiologist contoured five cases of non-small cell lung cancer. The CT, fused diagnostic PET/CT and fused planning PET/CT scans were used to contour three volumes. The concordance index (volume of intersection/ volume of union) was used to compare each volume to that of the reference radiation oncologist.
Results: Comparison of all volumes demonstrated a significant increase in the concordance index (CI) from 65.1% with CT alone, to 66.4% and 69.9% with addition of diagnostic and planning PET/CTs, respectively (p=0.049). This increase was predominantly due to addition of the planning PET/CT. Subgroup comparison of registrars’ volumes showed a significant increase in CI from 65.8% with CT alone, to 68.0% and 72.3% with addition of diagnostic and planning PET/CTs, respectively (p=0.028). The mean variation at the tumour-mediastinum interface was significantly reduced with addition of fused planning PET/CT images (p=0.034).
Conclusions: The addition of fused PET/CT images reduces interobserver variation in lung cancer delineation particularly with use of a planning PET/CT. Registrars’ concordance with the radiation oncologist improves with the addition of registered PET/CT images. Available PET/CT images should be incorporated into the radiotherapy planning process to allow accurate and reproducible delineation of lung cancers.
