T₂*-weighted contrasts sensitive to iron, such as R₂ mapping¹, susceptibility weighted imaging (SWI)², and quantitative susceptibility mapping^3,4 have been used to evaluate iron deposition in the substantia nigra (SN) in Parkinson’s disease (PD). Much of the iron deposition occurs in the SN pars compacta (SNpc), which contains a dense distribution of neuromelanin generating dopaminergic neurons^5,6. Recent results have found that the SNpc, as seen in neuromelanin (NM) sensitive MRI, and the iron containing SN, seen in T₂*-weighted contrasts and denoted iron SN, are located in different spatial positions⁷. In this abstract, we will re-examine iron sensitive measurements with respect to SNpc, defined by NM MRI.

A cohort of 82 subjects who provided written, IRB-approved, informed consent were studied. Demographic data for the cohort is listed in Table 1.

All data were acquired on a 3 T MRI scanner (Skyra, Siemens Medical Solutions, Erlangen, Germany) using a 20 channel receive only coil. SWI was performed using a 3D gradient echo sequence with the following parameters: TE/TR=20/50 ms, 56 contiguous slices, 384×288 imaging matrix, 229×172 mm² field of view, 1 average, FA=17°, and 50 Hz/pixel receiver bandwidth. Images from an MP-RAGE sequence (TE/TR/inversion time=2.46/1900/900 ms, FA=9°, voxel size=0.94×0.94×0.94 mm³) were used for registration from subject space to common space. The processing pipeline for T2w/SWI and common space registration is shown in Figure 1A.

Standard space NM SN masks from literature⁷ were used to define NM SN regions in this analysis. Iron containing SN volumes were generated by thresholding the T₂-weighted images using the thresholding procedure depicted in Figure 1B. After segmentation, iron SN volumes were transformed into MNI and SN probability maps were generated for both groups by summing the transformed maps over the subjects and normalizing by the total number of subjects in each group. In addition, the overlap between the iron and NM SN volumes were calculated for each subject in subject space.

Iron containing SN volume was slightly larger in the PD group than the control group (PD:624±156.8mm³; Control:570.0±133.5mm³; p=0.03). A comparison of the spatial probability map for the iron SN in control and PD groups in MNI space is shown in Figure 2. In common space, the inferior and medial portions of the iron SN in the PD group are enlarged compared to the control group. Furthermore, the increase in the iron SN in PD invades the NM SN, suggesting that the overlap between iron SN and NM SN can be used as a new PD biomarker.

The map for the probability of overlap between the iron SN and the NM SN in control and PD groups is shown in Figure 3. Lateral and superior areas of the NM SN exhibit a high probability of overlap with the iron SN in the PD group. In the control group, lateral areas on the edge of the NM SN have a higher probability of overlap with the iron SN. The distribution of spatial overlap occurs in the ventral lateral tier and is consistent with histological work⁵.

There is greater overlap between the iron SN and NM SN volumes in the PD population compared to controls (control:0.09±0.05; PD:0.23±0.06; p<10^-15). No relationship was seen between overlap of the iron SN and NM SN volumes with UPDRS–III score (p=0.40;R²=0.02), likely indicating that the overlap is an early marker for PD that changes negligibly after disease onset. In addition, we found increased angular frequency/iron deposition in both NM SN (p=0.001) and overlapped volumes (p<0.001) in the PD group. These results are summarized in Figure 4.

In this study, regions used for the SN were defined using NM SN masks generated from a group of healthy controls. This procedure is beneficial since it allows for retrospective analysis of acquired data. In addition, using NM to derive SN regions in controls allows for the examination of PD related changes in a more standardized and unbiased manner since iron deposition occurs at different rates in PD patients as the size and morphology of the iron SN will depend on this deposition. Individual NM SN masks could be used in lieu of these masks. However, a reduction in NM sensitive contrast in PD is expected, and NM SN ROIs derived from PD patients will be smaller than those derived from the controls.

In conclusion, we have found increased iron deposition in the ventral lateral tier of the SNpc and developed PD specific biomarkers related to this finding.