Comparison of tumor volumes and heterogeneity parameters derived from MRGlu maps and static SUV images in 18F-FDG PET

International audienceObjectives: Metabolically active tumor volume (MATV) and tumor activity distribution heterogeneity parameters derived from 18F-FDG PET images were recently shown to have an impact in patient management. In all of these studies, static FDG-PET images were exclusively considered. However, images of the glucose metabolic rate (MRGlu) obtained from dynamic PET acquisitions may contain more accurate information. The objective of this study was to compare MATV and tumor heterogeneity parameters extracted from these two different types of images. Methods: 20 patients with diagnosed non-small cell lung cancer (NSCLC) were enrolled in this retrospective study. All of these patients underwent a dynamic 18F-FDG PET scan which was acquired prior to therapy. Primary tumor delineation was performed using adaptive thresholding and a fuzzy locally adaptive Bayesian algorithm (FLAB). Image derived parameters considered in this study included MATV, heterogeneity (based on textural features and cumulative histogram), coefficient of covariance and tumor lesion glycolysis. Pearson's correlation coefficient and Wilcoxon test were used in the comparison. P-values