To evaluate alterations in the cortico-spinal tract (CST), middle and superior cerebellar peduncles (MCP and SCP, respectively) by using a probabilistic tractography method in a large sample of patients with neurodegenerative parkinsonisms: Progressive Supranuclear Palsy – Richardson’s Syndrome (PSP-RS), Multiple System Atrophy, Cerebellar and Parkinsonian variants (MSA-C and –P, respectively) patients compared to Parkinson’s disease (PD) patients and healthy controls (HC).We enrolled in the study 90 patients with a clinical diagnosis of neurodegenerative parkinsonism according to current criteria^1-3 and 27 HC, comparable for age and sex. All subjects underwent a standardized brain MR protocol including 25-direction diffusion imaging sequences on a 1.5 T scanner (GE Signa HDx 15). We applied the FSL probabilistic tractography algorithm^4-5 in order to reconstruct the CSTs, the MCPs and SCPs (Figure 1). The connectivity maps were normalized by dividing values of connectivity by the waytotal, a number returned by the tractography algorithm that corresponds to the total number of generated tracts that have not been rejected by target masks criteria. Fractional Anisotropy (FA), Mean Diffusivity (MD), Axial Diffusivity (AD) and Radial Diffusivity (RD) were evaluated. For the CST we also performed an along-tract analysis consisting in the subdivision of the tract into 100 percentiles along the z direction in each subject-specific space, allowing a comparison of the corresponding segments among different subjects. We performed an ANCOVA test followed by post hoc tests to compare FA, MD, AD, RD and tract volume within all the whole tracts. When DTI parameters were compared, sex, age and tract volume normalized by TIV were added as covariates of no interest, while, when comparing the volume of the tract, sex, age and TIV were added as covariates of no interest. For the along-tract analysis, we performed comparisons at each percentile and adding as covariates of no interest sex, age and percentile-specific volume of the tract normalized by TIV when comparing DTI parameters and TIV when comparing the along-tract volume. Comparisons were performed with a permutation-based nonparametric method, correcting for multiple comparisons by controlling the family-wise error rate. Moreover, results were then corrected for multiple comparisons with Bonferroni method (p<0.0038, 13 comparisons).Main demographic and clinical features of the study sample are reported in Figure 2. Whole-tract ANCOVA analysis of CST showed significant differences (p<0.05) among groups in all DTI metrics bilaterally (Table 3, Table 2 supplementary). The results of the post-hoc analysis are reported in Figure 3. The along tract analysis of CST showed a bilateral increase in MD in PSP-RS compared to PD and HC in the part of the tract passing through the corona radiata (Figure 4). We found no volumetric differences in CST segments among groups.The whole-tract ANCOVA analysis of MCPs yielded significant differences (p<0.05) among groups in all DTI metrics bilaterally and in left MCP volume (Figure 3). The results of post-hoc analysis are shown in Figure 3. The whole tract ANCOVA analysis of SCPs showed significant (p<0.05) differences among groups in all DTI metrics and tract volume bilaterally (Figure 3). The results of post-hoc analysis are shown in Figure 3.Our study demonstrated a specific pattern of neurodegeneration, involving CST, MCPs and SCPs. In particular, we demonstrated bilateral neurodegenerative changes in both CSTs and SCPs in our PSP-RS sample, according with previous neuropathological and neuroimaging evidences^6-9. Notably, CST neurodegenerative alterations were limited to the portion of the tract within the corona radiata; this portion of the tract includes neurons belonging to the corticostriatal tract, whose alterations have been related to motor control deficits and behavioural/cognitive dysfunction¹⁰. MSA-C displayed predominant neurodegenerative changes in MCPs, and with a smaller extent to SCPs. This pattern of neurodegeneration is typical of this disorder in which cerebellar dysfunction is usually prominent^6,11-12. The MSA-P group showed significant alterations in MCPs compared to PD and HC: these changes are in line with the clinical and neuropathological description of this MSA variant in which cerebellar alterations are present, compared to PD and HC, and they are usually milder when compared to MSA-C^6,9,12. Atypical parkinsonisms are characterized by disease-specific patterns of white matter alterations, reflecting the neuropathology and the clinical features of the disorder, and these changes can be detected in vivo by using probabilistic tractography. These results contribute to the understanding of the physiopathology of neurodegenerative parkinsonisms, supporting the role of advanced brain MR analysis of targeted regions of interest in the in vivo differential diagnosis among parkinsonisms.