Abstract & Overview
Recent works have demonstrated the added value of dynamic amino acid positron emission tomography (PET) for glioma grading and genotyping, biopsy targeting, and recurrence diagnosis. However, most of these studies are exclusively based on hand-crafted qualitative or semi-quantitative dynamic features extracted from the mean time activity curve (TAC) within predefined volumes. Voxelwise dynamic PET data analysis could instead provide a better insight into intra-tumour heterogeneity of gliomas. In this work, we investigate the ability of the widely used principal component analysis (PCA) method to extract meaningful quantitative dynamic features from high-dimensional motion-corrected dynamic [S-methyl-11C]methionine PET data in a first cohort of 20 glioma patients. By means of realistic numerical simulations, we demonstrate the robustness of our methodology to noise. In a second cohort of 13 glioma patients, we compare the resulting parametric maps to these provided by standard one- and two-tissue compartment pharmacokinetic (PK) models. We show that our PCA model outperforms PK models in the identification of intra-tumour uptake dynamics heterogeneity while being much less computationally expensive. Such parametric maps could be valuable to assess tumour aggressiveness locally with applications in treatment planning as well as in the evaluation of tumour progression and response to treatment. This work also provides further encouraging results on the added value of dynamic over static analysis of [S-methyl-11C]methionine PET data in gliomas, as previously demonstrated for O-(2-[18F]fluoroethyl)-L-tyrosine.