Dopamine transporter binding using [11C]PE2I and PET corrected for tissue heterogeneity and partial volume effect

Abstract

1745

Objectives: Dopamine transporter (DAT) imaging with PET and SPECT is a useful tool to quantitatively evaluate the loss of nigrostriatal dopaminergic terminals in Parkinson disease. Accurate quantification of DAT loss is important for early diagnosis and for monitoring disease progression. However, accuracy of quantification can be limited by tissue heterogeneity (TH) and partial volume effect (PVE). We have presented quantitative analysis of dopamine transporter binding using [11C]PE2I and PET in human brain. The aim of this study was to propose a new approach to correct TH and PVE in the putamen as a target tissue and the cerebellum as a reference tissue. Methods: The structure of the putamen consists of gray matter, white matter and extracerebral space, and the tracer concentration of gray matter corrected for TH and/or PVE, CGput(t), is described as follows: CGput(t) = CROI(t)/PGput–α*(PWput/PGput)*CCSO(t), where, CROI(t) is the concentration of the ROI, and PGput and PWput are the fraction of gray matter and white matter, respectively. CCSO(t) is the concentration of substitutive white matter in the ROI, which is obtained from centrum semiovale. The α is transfer constant that makes correlation between the concentration of white matter in the putamen and CCSO(t). [11C]PE2I-PET studies were performed on seven subjects. After co-registering PET and MR images, we performed segmentation analysis of MR images using the Alpert algorithm (Alpert, et al.,1996), in which the signal-intensity histogram was used to segment the voxels into gray and white matters and CSF, then, calculated fraction ratios. The data were analyzed using 2- and 3-CPM and reference tissue models, and the binding parameters were compared before and after correction. Results: The values of binding potential of the putamen obtained by 3-CPM were increased by approximately 24% after correction (p<0.05), and %COV improved from 17% to 7%, and those obtained by reference tissue models were increased by 20-25%. The values of K1/k2 did not change. Conclusions: The TH and PVE correction using MRI segmentation algorithm is useful to obtain reliable values of binding parameters for dopamine transporter mapping studies using [11C]PE2I and PET.