Utilizing Quantitative Susceptibility Mapping to Differentiate Primary Lateral Sclerosis from Progressive Supranuclear Palsy: A Case Report

Abstract

We report a patient who presented clinically as progressive supranuclear palsy (PSP) but was pathologically diagnosed as primary lateral sclerosis (PLS) with magnetic resonance imaging (MRI) with a quantitative susceptibility mapping (QSM) protocol. A 70-year-old man was clinically diagnosed with PSP due to early falls and unresponsiveness to levodopa therapy. Postmortem pathological examination revealed mild loss of Betz cells, gliosis, and transactive response DNA binding protein of 43 kDa (TDP-43)-positive inclusions in the motor cortex, leading to the pathological diagnosis of PLS. To explore methods for differentiating PLS from PSP, ante-mortem QSM images were visually and quantitatively assessed for abnormal increases in magnetic susceptibility in the motor cortex. Prussian blue and Luxol fast blue combined with periodic acid-Schiff staining were also performed to understand the source of the susceptibility increases. QSM showed clear hyperintense signals in the motor cortex. Magnetic susceptibility in the motor cortex was higher in the PLS patient (Z = 4.7, p < 0.001) compared to normal controls and pathologically diagnosed PSP patients. Pathological examination of the region showed intracortical myelin loss as well as iron deposition. Underlying pathologies for the increased magnetic susceptibility include not only iron deposition but also intracortical myelin. Our case suggests that QSM is a potential tool to differentiate PLS from PSP, providing insights for accurate diagnosis and enhancing clinical decision-making.

INTRODUCTION

Primary lateral sclerosis (PLS) is a rare motor neuron disorder characterized by selective degeneration of upper motor neurons¹. PLS accounts for approximately 1–4% of patients with motor neuron disease. In addition to the slow progression, a distinguishing feature of PLS from amyotrophic lateral sclerosis (ALS) is the preservation of lower motor neurons. Pathological examination reveals degeneration of the primary motor cortex with preservation of the lower motor neurons. Transactive response DNA binding protein of 43 kDa (TDP-43)-positive inclusions can be observed in neurons of the motor cortex². In contrast, progressive supranuclear palsy (PSP), characterized by parkinsonism, early falls and oculomotor dysfunction, is a neurodegenerative disease caused by tau accumulation. There are various clinical subtypes of PSP, including those that present with some features of PLS, making clinical differentiation from PLS difficult³.

Imaging biomarkers have been explored to differentiate PLS from other diseases such as PSP. The motor band sign is one of the potential biomarkers for ALS and PLS patients, seen as a low signal in the motor cortex on T2*-weighted or susceptibility-weighted imaging (SWI)⁴. Still, this sign should be used cautiously because other neurodegenerative diseases, such as Parkinson’s disease, can have this sign⁴. Quantitative susceptibility mapping (QSM) is a relatively new MRI technique that estimates the magnetic susceptibility of biological tissues and is sensitive to paramagnetic iron or diamagnetic myelin⁵. A few studies have used QSM to compare susceptibility in the motor cortex between PLS, ALS, and controls^{6, 7}. However, to our knowledge, no studies have compared the motor cortex susceptibility between PLS and PSP. In addition, the underlying pathology of this increased susceptibility in PLS is not fully understood.

Here, we present a pathologically diagnosed PLS patient who underwent antemortem QSM scanning. To examine the utility of QSM to differentiate PLS and PSP, we evaluated the patient’s QSM image visually and quantitatively and performed a pathological examination of the motor cortex.

CLINICAL SUMMARY

A 70-year-old man, recruited by the Neurodegenerative Research Group (NRG) from the Department of Neurology, Mayo Clinic, developed an imbalance and experienced frequent falls four years prior to his death. He also noticed gastrocnemius muscle cramps and tightness. Subsequently, he began using a cane for ambulation; however, he continued to fall frequently requiring a walker two years after the onset. Treatment with carbidopa/levodopa was without benefit. Neurological examination three years after the onset revealed the development of spastic dysarthria, spasticity of his lower extremities, and pathologically brisk deep tendon reflexes in all limbs. No restrictions of eye movements were observed. The possibility of PLS was also considered; however, the severity of falls was out of the degree of spasticity and he was diagnosed with possible PSP-postural instability based on the early falls and poor levodopa response⁸.

Two years before death, the patient underwent a standardized magnetic resonance imaging (MRI) protocol using a 3 Tesla scanner, including a 3D multi-echo gradient echo sequence⁹. Figure 1A shows the T2* weighted image (the last echo image), SWI, and QSM from the gradient echo sequence. SWI was reconstructed from the last echo magnitude and phase images¹⁰, and QSM was reconstructed from all echo magnitude and phase images as previously reported⁹. As shown in Figure 1A, SWI showed hypointense, and QSM showed hyperintense in the motor cortex (arrows), while the T2* weighted image did not show clear signal changes. Regarding discriminating between motor and sulcal signals, QSM showed hypointense in the sulcus and hyperintense in the motor cortex. This contrast made it easy to visually discriminate between the two. Figure 1B shows the motor cortex susceptibility in the PLS patient, sixty-seven cognitively normal controls (ages ranging from 47 to 86), and three pathologically diagnosed PSP patients (59, 77, and 80 years old). The clinical subtypes of the PSP patients were classified as PSP-parkinsonism, PSP-corticobasal syndrome, and PSP-Richardson syndrome, respectively. Motor cortex susceptibility was estimated from QSM images based on the fully automated algorithm previously reported⁹ and defined as the 95th percentile in the precentral gyrus referenced to that in the adjacent postcentral gyrus. As shown in Figure 1B, the estimated motor cortex susceptibility in the PLS patient (Z = 4.7, p < 0.001) was significantly higher than that in controls and PSP patients.

Figure 1. QSM images and magnetic susceptibility in the motor cortex.

(A) T2* weighted image, SWI, and QSM images in the PLS patient. Arrows indicate increased magnetic susceptibility in the motor cortex. (B) Estimated motor cortex susceptibility in controls, PSP, and PLS patients. Motor cortex susceptibilities were converted to z-score using the mean and standard deviation of susceptibilities in 67 controls.

PATHOLOGICAL FINDINGS

After death, the patient underwent a systematic and standardized neuropathologic assessment, including hematoxylin and eosin staining and immunohistochemistry for phosphorylated TDP-43, phosphorylated tau antibodies, and ionized calcium binding adaptor molecule 1 (IBA-1)^11–13. Mild loss of Betz cells and gliosis were observed in the motor cortex on hematoxylin and eosin. TDP-43 immunohistochemistry revealed neuronal cytoplasmic inclusions and more frequent glial inclusions in the motor cortex (Figure 2A) and glial inclusions in the pyramidal tract. IBA-1 immunohistochemistry revealed increased microglial activation in the pyramidal tract (Figure 2B). The lower motor neurons in the medulla and cervicomedullary junction had no significant pathology. Pathological diagnosis of PLS was established based on the predominance of upper motor neuron pathology and corticospinal tract degeneration¹⁴. Immunohistochemistry for CP13 antibody (mouse monoclonal; 1:1000; from the late Dr. Peter Davies; Feinstein Institute for Medical Research) showed no tau pathology suggestive of PSP.

Figure 2. Neuropathological findings in the PLS case.

(A) Immunohistochemistry of the motor cortex from the PLS patient using a phosphorylated TDP-43 antibody (p409/410). (B) Immunohistochemistry of the pyramidal tract in the medulla from the PLS case using an ionized calcium binding adaptor molecule 1 (IBA-1) antibody. Scale bar: 25 μm.

To investigate the underlying pathology of the increased QSM, we conducted Luxol fast blue myelin stain and Prussian blue iron stain (Figure 3). We compared this patient with three patients with PSP, three age and sex-matched normal control subjects without neurodegenerative disease (age at death: 65, 70, and 70 years old, respectively). The PSP patients were the same individuals used in the QSM analysis. Luxol fast blue stain revealed mild myelin loss in the white matter subjacent to the motor cortex and inter-cortical myelin loss in the motor cortex in the PLS patient compared to the PSP patients and normal controls. The Prussian blue stain detected more iron deposits in the motor cortex in the PLS patient compared to the PSP patients and normal controls.

Figure 3. Myelin and iron stains of the motor cortex.

(A-D) Luxol fast blue with periodic acid-Schiff stain of the motor cortex of the PLS patient (A), a PSP patient (B), and a normal control subject (C). Comparison of %positive area of Luxol fast blue in the motor cortex among the PLS patient, three PSP patients, and three normal control subjects (D).

(E-H) Prussian blue stain of the motor cortex of the PLS patient (E), a PSP patient (F), and a normal control subject (G). Comparison of %positive area of Prussian blue in the motor cortex among the PLS patient, three PSP patients, and three normal control subjects (H).

GM: gray matter, WM: white matter, Scale bar: 50 μm.

DISCUSSION

In this study, we visually and quantitatively evaluated the QSM image to detect signal changes in the motor cortex in a case pathologically diagnosed as PLS. While the T2* weighted image failed to detect signal changes, SWI and QSM showed hypointense and hyperintense signals in the motor cortex, respectively. Estimated motor cortex susceptibility discriminated the PLS patient from controls and PSP patients. Furthermore, the underlying pathology of the region revealed the cause of the susceptibility increases. Our results suggest that the QSM may be helpful in clinical practice to differentiate PLS from PSP by detecting disease-specific increases in magnetic susceptibility in the motor cortex.

Our results suggest that SWI and QSM are superior to T2* weighted image in detecting the motor band sign in PLS, which is consistent with the previous studies^{6, 7}. Another previous study reported that SWI showed a low signal intensity in the precentral gyrus in globular glial tauopathy¹⁵. In addition, this case suggests that QSM may be apparent in discriminating between motor and sulcal signals, although a firm conclusion cannot be made because sulcal signals in QSM depend on various factors, such as acquisition parameters and reconstruction algorithm.

Another advantage of QSM is its quantitative nature, where quantitative susceptibility values can be obtained in any region of interest besides image contrast⁵. Quantitative values allow differentiation between paramagnetic and diamagnetic susceptibilities and may improve the detection of subtle changes in susceptibility⁵. This study demonstrated that motor cortex susceptibility distinguishes PLS patients from PSP patients. These quantitative values may be helpful in clinical practice, as they aid diagnosis by avoiding oversights and providing objective information.

Previous studies proposed that high QSM signals in PLS or ALS were caused by iron accumulation^{7, 16}. Iron is known to be involved in oxidative stress, inflammation, and ferroptosis in several neurodegenerative diseases¹⁷. The present study suggests that not only iron deposition, but also intercortical myelin loss might contribute to a prominent increase in magnetic susceptibility, given that myelin has negative effects on estimated susceptibility¹⁸.

In conclusion, our case suggests that visual and quantitative QSM evaluation can sensitively detect signal changes in the motor cortex, indicating its potential utility in differentiating between PLS and other diseases such as PSP.

Abbreviations:

PLS

primary lateral sclerosis

ALS

amyotrophic lateral sclerosis

TDP-43

transactive response DNA binding protein of 43 kDa

PSP

progressive supranuclear palsy

SWI

susceptibility-weighted imaging

QSM

quantitative susceptibility mapping

MRI

magnetic resonance imaging

NRG

Neurodegenerative Research Group