Chungseok Oh1, Chang-Yeop Jeon2, Jincheol Seo2, Hyeong-Geol Shin1, Sooyeon Ji1, Seongbeom Park3, Soohwa Song3, Dong Hoon Shin3, Youngjeon Lee2, and Jongho Lee1
1Department of Electrical and Computer Engineering, Seoul National University, Seoul, Korea, Republic of, 2National Primate Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB), Cheongju, Korea, Republic of, 3Heuron Co.,Ltd, Incheon, Korea, Republic of
Synopsis
In this study, a hyperintensity region within
substantia nigra is explored in cynomolgus monkeys over a wide range of ages. This
region is similar to that of humans, which is often called as a swallow tail
sign in healthy subjects. This nigral hyperintensity appears consistently in SMWI,
QSM, and T2*-weighted images, particularly in aged animals.
Introduction
Parkinson’s disease (PD) is one of the major
neurodegenerative diseases. Recently, several studies have suggested that a hyperintense
pattern within substantia nigra of an MR image can be used as an imaging
biomarker for PD.1,2 This nigral hyperintensity
is observed in the susceptibility-related
contrast images including T2*-weighted image1, SWI2, and SMWI3 with the best
visibility in SMWI at 3T.3
So far, most of the studies have focused on the
human brain. However, no study has explored a similar pattern in cynomolgus
monkeys (macaca fascicularis), which may be used as an animal model for PD.4 In this study, we
optimize the scan protocol to observe the nigral hyperintensity for the
small-size brain of the cynomolgus monkey and demonstrate that the nigral
hyperintensity can be identified using 3T MRI. This outcome may help us to
monitor the progression of PD in the animal model. Methods
[Optimizing scan
protocol]
For
optimizing the scan protocol, a total of nine healthy cynomolgus monkeys with
ages from 5 to 18 years were examined at a 3T MRI system (Archieva, Philips) using
a commercial 8-channel human knee coil. The images were acquired using a 3D
multi-echo gradient-recalled echo (GRE) sequence. The tested scan protocols in the
nine monkeys and the final protocol are summarized in Figure 1.
[Image processing]
From the
magnitude images of the multiple echoes, a single magnitude map was generated
by the square root of the sum of squares over echoes. From the phase images of the multiple echoes,
a local field map was generated using Laplacian phase unwrapping and V-SHARP
background field removal method. When the scan was acquired twice, local field
maps and magnitude maps from two scans were averaged after registration.
Afterward, the QSM map was generated from the local field map by iLSQR, and the SMWI
map was generated from the magnitude image and QSM map as suggested.3 The overall image processing is summarized in Figure 2.
[Human subjects]
To compare the nigral hyperintensity patterns between monkeys
and humans, SMWI images were reconstructed from thirteen healthy human subjects
(age: 27 ± 2.9 years).
Out of the thirteen subjects, one subject was examined at a 3T MRI system
(Prisma, Siemens Healthineers) and twelve subjects were from the COSMOS dataset
of the previous study.5 The resolution of the image was 0.8 x 0.8 x 1.2 mm3
for one subject and 1 x 1 x 1 mm3
for twelve subjects, and the scan time was 5 min 3-40s for each scan.
[ROI analysis]
The QSM values in substantia nigra were
measured by manually segmenting the region of interest (ROI) for substantia
nigra both in monkeys and humans. A linear model was applied to analyze a trend
between QSM values in substantia nigra and the ages of the monkeys. Results
Figure 3
shows the nigral hyperintensity images in three consecutive slices from a
cynomolgus monkey. The nigral hyperintensity was observed in SMWI, and the same
pattern was also observed in QSM and T2*-weighted images.
Figure 4 shows the nigral hyperintensity images in
cynomolgus monkeys over a wide range of ages. The images reveal that the nigral
hyperintensity is noticeable in monkeys, while the signature is
best-identified in old animals (≥ 8 years) using SMWI
images. When comparing the hyperintense patterns of the monkeys and humans, the
hyperintense patterns of the aged monkeys (≥ 13 years) are
similar to that of the human.
When the trend
between QSM values in the substantia nigra and ages of the monkeys is explored, a linear trend
is observed, reporting increased susceptibility with aging (Figure 5; y =
0.0042x + 0.011 with R2 = 0.54). In humans, the mean ±
standard deviation of the QSM values of the thirteen healthy subjects was
0.11 ± 0.02. When the QSM values of the monkeys
and humans are compared, the QSM values of the monkeys become closer to that of
the humans as the monkeys get older.Discussion and Conclusion
In summary, we observed
hyperintensity within substantia nigra in healthy cynomolgus monkeys. This
hyperintense pattern was best observed in aged animals using SMWI but also can
be identified in other susceptibility-related images and younger animals. Furthermore,
in the aged animals, the hyperintensity pattern was similar to that of humans. This
pattern is suggested to indicate nigrosome-16 and has
been utilized as a biomarker for PD in humans. Future investigations may explore the change of this
hyperintense pattern in the cynomolgus monkey model of PD.Acknowledgements
This work has been supported by Heuron Co. Ltd. and by the Korea government (MSIT) (No.NRF-2021R1A2B5B03002783)References
1. Kwon, D. H. et
al. Seven-tesla magnetic resonance images of the substantia nigra in
Parkinson disease. Am. Neurol. Assoc. 71, 267–277 (2012).
2. Schwarz, S. T. et al. The
‘swallow tail’ appearance of the healthy nigrosome - A new accurate test of
Parkinson’s disease: A case-control and retrospective cross-sectional MRI study
at 3T. PLoS One 9, (2014).
3. Nam, Y., Gho, S. M., Kim, D. H., Kim, E.
Y. & Lee, J. Imaging of nigrosome 1 in substantia nigra at 3T using
multiecho susceptibility map-weighted imaging (SMWI). J. Magn. Reson.
Imaging 46, 528–536 (2017).
4. Dauer, W. & Przedborski, S.
Parkinson’s Disease: Mechanisms and models. Neuron 39, 889–909
(2003).
5. Yoon, J. et al. Quantitative
susceptibility mapping using deep neural network: QSMnet. Neuroimage 179,
199–206 (2018).
6. Blazejewska, A. I. et al.
Visualization of nigrosome 1 and its loss in PD. Am. Acad. Neurol. 81,
534–540 (2013).