FPCIT SPECT in parkinsonism predicts survival: a data-driven analysis

Introduction: Dopamine transporter SPECT is an established method to investigate nigrostriatal integrity in case of clinically uncertain parkinsonism. The present study investigated whether a completely data-driven analysis of FPCIT SPECT is able to stratify patients according to mortality risk. This was tested in a clinical patient sample. Patients from our clinical registry were included if they had received FPCIT SPECT between 10/2008 and 06/2016 for diagnosis of parkinsonism and if their vital status (alive or date of death, if applicable) could be determined in 07/2017. Atlas-based regions of interest were used to extract individual specific binding ratios (SBR; occipital cortex as reference) of the whole striatum and striatal subregions (caudate nucleus, anterior and posterior putamen) of both hemispheres. SBR of the whole striatum, its asymmetry (asymmetry index, AI; absolute value), and the rostrocaudal gradient of striatal binding (C/pP: caudate SBR divided by posterior putamen SBR) were used as input for hierarchical clustering of patients into subgroups. A Cox proportional hazards model was computed in order to test whether survival (adjusted for age) was different between the subgroups. Vital status could be determined in 518/566 patients. 77/518 patients had died within the median follow-up duration of 3.3 years (95% C.I. 3.1 to 3.7). The subgroups identified by hierarchical clustering were characterized by relatively (1) high SBR, low AI, and low C/pP,(2) low striatal SBR, high AI, and high C/pP, and (3) low striatal SBR, high AI, and low C/pP. Mortality in group 3 was twice as high as that of group 2 (HR = 1.9 [95% C.I. 1.1 to 3.3], p = 0.029) and group 1 (HR = 2.2 [95% C.I. 1.3 to 3.8], p = 0.003). In patients with parkinsonism, data-driven analysis of FPCIT SPECT identified a subgroup of patients with significantly increased mortality during follow-up. This high-risk group showed an FPCIT signature suggestive of atypical Parkinsonian syndromes, whereas the other groups had either normal scans or showed an FPCIT profile suggestive of Parkinson’s disease. This suggests that FPCIT SPECT might not only serve as a diagnostic tool but also provide valuable prognostic information (“red flag”, e.g. triggering additional examinations like FDG PET for differential diagnosis of parkinsonism or intensified clinical follow-up).