Development and Validation of Automated Magnetic Resonance Parkinsonism Index 2.0 to Distinguish Progressive Supranuclear Palsy‐Parkinsonism From Parkinson's Disease

Abstract

Background Differentiating progressive supranuclear palsy‐parkinsonism (PSP‐P) from Parkinson's disease (PD) is clinically challenging. Objective This study aimed to develop an automated Magnetic Resonance Parkinsonism Index 2.0 (MRPI 2.0) algorithm to distinguish PSP‐P from PD and to validate its diagnostic performance in two large independent cohorts. Methods We enrolled 676 participants: a training cohort (n = 346; 43 PSP‐P, 194 PD, and 109 control subjects) from our center and an independent testing cohort (n = 330; 62 PSP‐P, 171 PD, and 97 control subjects) from an international research group. We developed a new in‐house algorithm for MRPI 2.0 calculation and assessed its performance in distinguishing PSP‐P from PD and control subjects in both cohorts using receiver operating characteristic curves. Results The automated MRPI 2.0 showed excellent performance in differentiating patients with PSP‐P from patients with PD and control subjects both in the training cohort (area under the receiver operating characteristic curve [AUC] = 0.93 [95% confidence interval, 0.89–0.98] and AUC = 0.97 [0.93–1.00], respectively) and in the international testing cohort (PSP‐P versus PD, AUC = 0.92 [0.87–0.97]; PSP‐P versus controls, AUC = 0.94 [0.90–0.98]), suggesting the generalizability of the results. The automated MRPI 2.0 also accurately distinguished between PSP‐P and PD in the early stage of the diseases (AUC = 0.91 [0.84–0.97]). A strong correlation (r = 0.91, P < 0.001) was found between automated and manual MRPI 2.0 values. Conclusions Our study provides an automated, validated, and generalizable magnetic resonance biomarker to distinguish PSP‐P from PD. The use of the automated MRPI 2.0 algorithm rather than manual measurements could be important to standardize measures in patients with PSP‐P across centers, with a positive impact on multicenter studies and clinical trials involving patients from different geographic regions. © 2022 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society

Magnetic Resonance Planimetry in the Differential Diagnosis between Parkinson’s Disease and Progressive Supranuclear Palsy

“…These data suggest that the MRPI can be used in distinguishing PSP-RS from PD, while the MRPI 2.0 is more suitable for differentiating PSP-P from PD. Very recently, an automated version of MRPI 2.0 has been developed [ 68 ] to standardize measurements across centers, and the automated MRPI 2.0 was tested with excellent results in two independent large cohorts of PSP-P and PD patients, validating this new biomarker (AUC of 0.93, CI: 0.89–0.98 in a single-center Italian cohort and AUC of 0.92, CI: 0.87.0.96 in an independent multicenter international cohort) [ 68 ]. In this study [ 68 ], the automated algorithm was strongly correlated (r = 0.91, p < 0.001) with the manual measurements performed by an expert rater and showed perfect reproducibility (inter-rater correlation coefficient, ICC = 1) when the automated process was repeated twice starting from the raw 3D T1-weighted images.…”

“…Very recently, an automated version of MRPI 2.0 has been developed [ 68 ] to standardize measurements across centers, and the automated MRPI 2.0 was tested with excellent results in two independent large cohorts of PSP-P and PD patients, validating this new biomarker (AUC of 0.93, CI: 0.89–0.98 in a single-center Italian cohort and AUC of 0.92, CI: 0.87.0.96 in an independent multicenter international cohort) [ 68 ]. In this study [ 68 ], the automated algorithm was strongly correlated (r = 0.91, p < 0.001) with the manual measurements performed by an expert rater and showed perfect reproducibility (inter-rater correlation coefficient, ICC = 1) when the automated process was repeated twice starting from the raw 3D T1-weighted images. The high reproducibility guaranteed by the automated software improves the reliability of this biomarker and makes it suitable for longitudinal studies aiming at evaluating the disease progression of PSP through repeated measurements over time.…”

“…The optimal cut-off values of automated MRPI in distinguishing PSP-RS from PD and of MRPI 2.0 in distinguishing PSP-P from PD derived from the largest international cohorts are 13.88 for MRPI (89% sensitivity, 95% specificity, 94% accuracy) and 2.70 for MRPI 2.0 (86% sensitivity, 92% specificity, 90% accuracy). The cut-off of 13.88 for automated MRPI (PSP-RS vs. PD) was originally identified in a study by Quattrone et al [ 62 ] and validated in a recent international study by Nigro et al [ 61 ]; the cut-off of 2.70 for automated MRPI 2.0 (PSP-P vs. PD) was the optimal cut-off calculated in an international cohort of 56 PSP-P, and 166 PD enrolled in a recent study [ 68 ]. The MRPI cut-off value derived from the ROC analysis corresponds well to the 95th percentile MRPI value obtained in a very large cohort of nearly 1000 control subjects in an independent study [ 57 ], demonstrating that MRPI values higher than 13.88 reflects a degree of the midbrain and SCP atrophy usually not found in the normal population.…”

“…At the bottom, there is the probability of having PSP-RS rather than PD for each automated MRPI value ( C ) and the probability of having PSP-P rather than PD for each automated MRPI 2.0 value ( D ), obtained using logistic regression models balancing the number of PD and PSP patients. Data for the PSP-RS versus PD comparisons were obtained in the international cohort enrolled in the study by Nigro et al, 2020 [ 61 ]; data for the PSP-P versus PD comparisons were obtained in the international cohort enrolled in the study by Quattrone et al, 2022 [ 68 ].…”

Is MRPI 2.0 More Useful than MRPI and M/P Ratio in Differential Diagnosis of PSP-P with Other Atypical Parkinsonisms?

“…In PSP-P, the factors best assessed by neuroimaging have yet to be fully explored. MRI parameters are commonly used to perform differential diagnosis of PSP-P and PD in the early stages, but less is known in the context of the comparative analysis of PSP-P and MSA [ 5 , 6 ]. Among the features commonly associated with PSP-P are the evaluation of magnetic resonance parkinsonism index (MRPI) and its upgraded edition, MRPI 2.0.…”