Assessment of tumour size in PET/CT lung cancer studies: PET- and CT-based methods compared to pathology

Background: Positron emission tomography (PET) may be useful for defining the gross tumour volume for radiation treatment planning and for response monitoring of non-small cell lung cancer (NSCLC) patients. The purpose of this study was to compare tumour sizes obtained from CT- and various more commonly available PET-based tumour delineation methods to pathology findings. Methods: Retrospective non-respiratory gated whole body [F-18]-fluoro-2-deoxy-D-glucose PET/CT studies from 19 NSCLC patients were used. Several (semi-) automatic PET-based tumour delineation methods and manual CT-based delineation were used to assess the maximum tumour diameter. Results: 50%, adaptive 41% threshold-based and contrast-oriented delineation methods showed good agreement with pathology after removing two outliers (R-2=0.82). An absolute SUV threshold of 2.5 also showed a good agreement with pathology after the removal of 5 outliers (R-2:0.79), but showed a significant overestimation in the maximum diameter (19.8 mm, p<0.05). Adaptive 50%, relative threshold level and gradient-based methods did not show any outliers, provided only small, non-significant differences in maximum tumour diameter (<4.7 mm, p>0.10), and showed fair correlation (R-2>0.62) with pathology. Although adaptive 70% threshold-based methods showed underestimation compared to pathology (36%), it provided the best precision (SD: 14%) together with good correlation (R-2= 0.81). Good correlation between CT delineation and pathology was observed (R-2= 0.77). However, CT delineation showed a significant overestimation compared with pathology (3.8 mm, p<0.05). Conclusions: PET-based tumour delineation methods provided tumour sizes in agreement with pathology and may therefore be useful to define the (metabolically most) active part of the tumour for radiotherapy and response monitoring purposes.