Subregional patterns of preferential striatal DAT loss differ in PD, MSA and PSP

75 Objectives Parkinson disease (PD), multiple system atrophy (MSA) and progressive supranuclear palsy (PSP) are all known to affect dopaminergic neurons and often present with similar clinical features. Dopamine transporter (DAT) imaging can demonstrate presynaptic dopaminergic neuronal loss in PD, MSA and PSP. However, differentiating atypical parkinsonism from PD is often difficult. In this study, we investigated differences in striatal subregional DAT loss in PD, MSA and PSP, as well as the diagnostic value of 11C-CFT PET imaging in the differential diagnosis of parkinsonism. Methods Brain 11C-CFT PET imaging was performed in 219 patients with parkinsonian disorders. The PET images were spatially normalized and a standardized set of regions of interest (ROIs) were defined bilaterally for the caudate, anterior putamen (AP), posterior putamen(PP), ventral putamen (VP), dorsal putamen (DP) and occipital cortex with ScAnVP software. The nondisplaceable binding potential (BPND; defined as the ratio of (striatum-occipital)/occipital radioactivity counts) of each striatal subregion and intersubregional ratio (ISR; defined as the ratio of the BPND of one subregion to that of another subregion) were calculated. After the imaging, these patients were assessed by blinded movement disorders specialists for a mean of 2.3 years before a final clinical diagnosis was made. The accuracy of DAT imaging-based classification was assessed by comparison with the final clinical diagnosis. Results 133 patients had a final clinical diagnosis of PD, 55 MSA, and 31 PSP with severity matched. The BPND of caudate in the PSP group was significantly lower than those in the PD and MSA groups. The ISR of each subregional putamen/caudate in the PSP group was significantly higher than those in the PD and MSA groups. At the cutoff value of 1.16, the AP/C ISR had 87%sensitivity and 93% specificity for differentiating PSP from PD. The classifications were also accurate for PP/C (0.63; 80% sensitivity, 88% specificity), VP/C (1.13; 87% sensitivity, 92% specificity) and DP/C (0.87; 87% sensitivity, 90% specificity) ISRs. At the cutoff value of 1.23, the AP/C ISR had 81% sensitivity and 87% specificity for differentiating PSP from MSA. The classifications were also accurate for PP/C (0.78; 68% sensitivity, 84% specificity), VP/C (1.15; 84% sensitivity, 86% specificity) and DP/C (0.98; 81% sensitivity, 82% specificity) ISRs. The BPND of DP and the ISRs of DP/non-DP subregions in the MSA group were significantly higher than those in the PD group. At the cutoff value of 0.81, the DP/AP ISR had 76% sensitivity and 82% specificity for differentiating MSA from PD. The classifications were also accurate for DP/C (0.76; 71% sensitivity, 75% specificity) and DP/VP (0.87; 76% sensitivity, 76% specificity) ISRs. Conclusions PD, MSA and PSP have different preferential subregional decreases in striatal DAT binding when examined by 11C-CFT PET imaging. The ISRs could provide valuable information to distinguish parkinsonian disorders, even in earlier stages of disease.