Practical Application of DaTQUANT with Optimal Threshold for Diagnostic Accuracy of Dopamine Transporter SPECT

Evaluation of Parkinsonian Syndromes (PS) with Ioflupane iodine-123 dopamine transporter single photon emission computed tomography (DaT-SPECT), in conjunction with history and clinical examination, aids in diagnosis. FDA-approved, semi-quantitative software, DaTQUANTTM (GE Healthcare, Chicago, IL, USA) is available to assist in interpretation. This study aims to evaluate the optimal variables and thresholds of DaTQUANT to yield the optimal diagnostic accuracy. It is a retrospective review with three different patient populations. DaT-SPECT images from all three study groups were evaluated using DaTQUANTTM software, and both single and multi-variable logistic regression were used to model PS status. The optimal models were chosen via accuracy, sensitivity, and specificity, then evaluated on the other study groups. Among single variable models, the posterior putamen yielded the highest accuracy (84% to 95%), while balancing sensitivity and specificity. Multi-variable models did not substantially improve the accuracy. When the optimal single variable models for each group were used to evaluate the remaining two groups, comparable results were achieved. In typical utilization of DaT-SPECT for differentiation between nigrostriatal degenerative disease (NSDD) and non-NSDD, the posterior putamen was the single variable that yielded the highest accuracy across three different patient populations. The posterior putamen’s recommended thresholds for DaTQUANT are SBR ≤ 1.0, z-score of ≤−1.8 and percent deviation ≤ −0.34.

Evaluating dopamine transporter imaging as an enrichment biomarker in a phase 2 Parkinson’s disease trial

Background Dopamine transporter single-photon emission computed tomography (DaT-SPECT) can quantify the functional integrity of the dopaminergic nerve terminals and has been suggested as an imaging modality to verify the clinical diagnosis of Parkinson’s disease (PD). Depending on the stage of progression, approximately 5–15% of participants clinically diagnosed with idiopathic PD have been observed in previous studies to have normal DaT-SPECT patterns. However, the utility of DaT-SPECT in enhancing early PD participant selection in a global, multicenter clinical trial of a potentially disease-modifying therapy is not well understood. Methods The SPARK clinical trial was a phase 2 trial of cinpanemab, a monoclonal antibody against alpha-synuclein, in participants with early PD. DaT-SPECT was performed at screening to select participants with DaT-SPECT patterns consistent with degenerative parkinsonism. Acquisition was harmonised across 82 sites. Images were reconstructed and qualitatively read at a central laboratory by blinded neuroradiologists for inclusion prior to automated quantitative analysis. Results In total, 482 unique participants were screened between January 2018 and May 2019; 3.8% (15/398) of imaged participants were excluded owing to negative DaT-SPECT findings (i.e., scans without evidence of dopaminergic deficit [SWEDD]). Conclusion A smaller proportion of SPARK participants were excluded owing to SWEDD status upon DaT-SPECT screening than has been reported in prior studies. Further research is needed to understand the reasons for the low SWEDD rate in this study and whether these results are generalisable to future studies. If supported, the radiation risks, imaging costs, and operational burden of DaT-SPECT for enrichment may be mitigated by clinical assessment and other study design aspects. Trial registration ClinicalTrials.gov identifier: NCT03318523. Date submitted: October 19, 2017. First Posted: October 24, 2017.

Is Generalized and Segmental Dystonia Accompanied by Impairments in the Dopaminergic System?

Background: The pathogenesis of dystonia is remarkably diverse. Some types of dystonia, such as DYT5 (DYT-GCH1) and tardive dystonia, are related to dysfunction of the dopaminergic system. Furthermore, on pathological examination, cell loss in the substantia nigra (SN) of patients with dystonia has been reported, suggesting that impaired dopamine production may be involved in DYT5 and in other types of dystonia.Objectives: To investigate functional dopaminergic impairments, we compared patients with dystonia and those with Parkinson's disease (PD) with normal controls using neuromelanin-sensitive magnetic resonance imaging (NM-MRI) and dopamine transporter single photon emission computed tomography (DAT SPECT).Methods: A total of 18, 18, and 27 patients with generalized or segmental dystonia, patients with PD, and healthy controls, respectively, were examined using NM-MRI. The mean area corresponding to NM in the SN (NM-SN) was blindly quantified. DAT SPECT was performed on 17 and eight patients with dystonia and PD, respectively. The imaging data of DAT SPECT were harmonized with the Japanese database using striatum phantom calibration. These imaging data were compared between patients with dystonia or PD and controls from the Japanese database in 256 healthy volunteers using the calibrated specific binding ratio (cSBR). The symptoms of dystonia were evaluated using the Fahn–Marsden Dystonia Rating Scale (FMDRS), and the correlation between the results of imaging data and FMDRS was examined.Results: The mean areas corresponding to NM in the SN (NM-SN) were 31 ± 4.2, 28 ± 3.8, and 43 ± 3.8 pixels in patients with dystonia, PD, and in healthy controls, respectively. The mean cSBRs were 5 ± 0.2, 2.8 ± 0.2, 9.2 (predictive) in patients with dystonia, PD, and in healthy controls, respectively. The NM-SN area (r = −0.49, p < 0.05) and the cSBR (r = −0.54, p < 0.05) were inversely correlated with the FMDRS. There was no significant difference between the dystonia and PD groups regarding NM-SN (p = 0.28). In contrast, the cSBR was lower in patients with PD than in those with dystonia (p < 0.5 × 10−6).Conclusions: Impairments of the dopaminergic system may be involved in developing generalized and segmental dystonia. SN abnormalities in patients with dystonia were supposed to be different from degeneration in PD.

High diagnostic performance of independent alpha-synuclein seed amplification assays for detection of early Parkinson’s disease

Alpha-synuclein seed amplification assays (αSyn-SAAs) are promising diagnostic tools for Parkinson’s disease (PD) and related synucleinopathies. They enable detection of seeding-competent alpha-synuclein aggregates in living patients and have shown high diagnostic accuracy in several PD and other synucleinopathy patient cohorts. However, there has been confusion about αSyn-SAAs for their methodology, nomenclature, and relative accuracies when performed by various laboratories. We compared αSyn-SAA results obtained from three independent laboratories to evaluate reproducibility across methodological variations. We utilized the Parkinson’s Progression Markers Initiative (PPMI) cohort, with DATSCAN data available for comparison, since clinical diagnosis of early de novo PD is critical for neuroprotective trials, which often use dopamine transporter imaging to enrich their cohorts. Blinded cerebrospinal fluid (CSF) samples for a randomly selected subset of PPMI subjects (30 PD, 30 HC, and 20 SWEDD), from both baseline and year 3 collections for the PD and HC groups (140 total CSF samples) were analyzed in parallel by each lab according to their own established and optimized αSyn-SAA protocols. The αSyn-SAA results were remarkably similar across laboratories, displaying high diagnostic performance (sensitivity ranging from 86 to 96% and specificity from 93 to 100%). The assays were also concordant for samples with results that differed from clinical diagnosis, including 2 PD patients determined to be clinically inconsistent with PD at later time points. All three assays also detected 2 SWEDD subjects as αSyn-SAA positive who later developed PD with abnormal DAT-SPECT. These multi-laboratory results confirm the reproducibility and value of αSyn-SAA as diagnostic tools, illustrate reproducibility of the assay in expert hands, and suggest that αSyn-SAA has potential to provide earlier diagnosis with comparable or superior accuracy to existing methods.

Explainable AI to improve acceptance of convolutional neural networks for automatic classification of dopamine transporter SPECT in the diagnosis of clinically uncertain parkinsonian syndromes

Purpose Deep convolutional neural networks (CNN) provide high accuracy for automatic classification of dopamine transporter (DAT) SPECT images. However, CNN are inherently black-box in nature lacking any kind of explanation for their decisions. This limits their acceptance for clinical use. This study tested layer-wise relevance propagation (LRP) to explain CNN-based classification of DAT-SPECT in patients with clinically uncertain parkinsonian syndromes. Methods The study retrospectively included 1296 clinical DAT-SPECT with visual binary interpretation as “normal” or “reduced” by two experienced readers as standard-of-truth. A custom-made CNN was trained with 1008 randomly selected DAT-SPECT. The remaining 288 DAT-SPECT were used to assess classification performance of the CNN and to test LRP for explanation of the CNN-based classification. Results Overall accuracy, sensitivity, and specificity of the CNN were 95.8%, 92.8%, and 98.7%, respectively. LRP provided relevance maps that were easy to interpret in each individual DAT-SPECT. In particular, the putamen in the hemisphere most affected by nigrostriatal degeneration was the most relevant brain region for CNN-based classification in all reduced DAT-SPECT. Some misclassified DAT-SPECT showed an “inconsistent” relevance map more typical for the true class label. Conclusion LRP is useful to provide explanation of CNN-based decisions in individual DAT-SPECT and, therefore, can be recommended to support CNN-based classification of DAT-SPECT in clinical routine. Total computation time of 3 s is compatible with busy clinical workflow. The utility of “inconsistent” relevance maps to identify misclassified cases requires further investigation.

Asymmetric Dopaminergic Dysfunction in Brain-First versus Body-First Parkinson’s Disease Subtypes

Background: We have hypothesized that Parkinson’s disease (PD) comprises two subtypes. Brain-first, where pathogenic α-synuclein initially forms unilaterally in one hemisphere leading to asymmetric nigrostriatal degeneration, and body-first with initial enteric pathology, which spreads through overlapping vagal innervation leading to more symmetric brainstem involvement and hence more symmetric nigrostriatal degeneration. Isolated REM sleep behaviour disorder has been identified as a strong marker of the body-first type. Objective: To analyse striatal asymmetry in [18F]FDOPA PET and [123I]FP-CIT DaT SPECT data from iRBD patients, de novo PD patients with RBD (PD +RBD) and de novo PD patients without RBD (PD - RBD). These groups were defined as prodromal body-first, de novo body-first, and de novo brain-first, respectively. Methods: We included [18F]FDOPA PET scans from 21 iRBD patients, 11 de novo PD +RBD, 22 de novo PD - RBD, and 18 controls subjects. Also, [123I]FP-CIT DaT SPECT data from iRBD and de novo PD patients with unknown RBD status from the PPPMI dataset was analysed. Lowest putamen specific binding ratio and putamen asymmetry index (AI) was defined. Results: Nigrostriatal degeneration was significantly more symmetric in patients with RBD versus patients without RBD or with unknown RBD status in both FDOPA (p = 0.001) and DaT SPECT (p = 0.001) datasets. Conclusion: iRBD subjects and de novo PD +RBD patients present with significantly more symmetric nigrostriatal dopaminergic degeneration compared to de novo PD - RBD patients. The results support the hypothesis that body-first PD is characterized by more symmetric distribution most likely due to more symmetric propagation of pathogenic α-synuclein compared to brain-first PD.

Multi-Atlas MRI-Based Striatum Segmentation for 123I-FP-CIT SPECT (DAT-SPECT) Compared With the Bolt Method and SPECT-Atlas-Based Segmentation Method Toward the Accurate Diagnosis of Parkinson's Disease/Syndrome

Aims: This study aimed to analyze the performance of multi-atlas MRI-based parcellation for 123I-FP-CIT SPECT (DAT-SPECT) in healthy volunteers. The proposed method was compared with the SPECT-atlas-based and Bolt methods. 18F-FE-PE2I-PET (DAT-PET) was used as a reference.Methods: Thirty healthy subjects underwent DAT-SPECT, DAT-PET, and 3D-T1WI-MRI. We calculated the striatum uptake ratio (SUR/SBR), caudate uptake ratio (CUR), and putamen uptake ratio (PUR) for DAT-SPECT using the multi-atlas MRI-based method, SPECT-atlas-based method, and Bolt method. In the multi-atlas MRI-based method, the cerebellum, occipital cortex, and whole-brain were used as reference regions. The correlation of age with DAT-SPECT activity and the correlations of SUR/SBR, CUR, and PUR between DAT-SPECT and DAT-PET were calculated by each of the three methods.Results: The correlation between age and SUR/SBR for DAT-SPECT based on the multi-atlas MRI-based method was comparable to that based on the SPECT-atlas-based method (r = −0.441 to −0.496 vs. −0.488). The highest correlation between DAT-SPECT and DAT-PET was observed using the multi-atlas MRI-based method with the occipital lobe defined as the reference region compared with the SPECT-atlas-based and Bolt methods (SUR, CUR, and PUR: 0.687, 0.723, and 0.676 vs. 0.698, 0.660, and 0.616 vs. 0.655).Conclusion: Multi-atlas MRI-based parcellation with the occipital lobe defined as the reference region was at least comparable to the clinical methods.