Association between Neuroimaging Scores and Clinical Status in Pediatric Patients Diagnosed with Metachromatic Leukodystrophy

Article information

Ann Child Neurol. 2024;32(4):219-225

Publication date (electronic) : 2024 June 25

doi : https://doi.org/10.26815/acn.2024.00542

Sunho Lee, MD¹, Ji Hoon Na, MD², Young-Mock Lee, MD^,2

¹Department of Pediatrics, CHA Ilsan Medical Center, CHA University, Goyang, Korea

²Department of Pediatrics, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Korea

Corresponding author: Young-Mock Lee, MD Department of Pediatrics, Gangnam Severance Hospital, Yonsei University College of Medicine, 211 Eonju-ro, Gangnam-gu, Seoul 06273, Korea Tel: +82-2-2019-3354 Fax: +82-2-2019-4881 E-mail: ymleemd@yuhs.ac

Received 2024 April 18; Revised 2024 June 3; Accepted 2024 June 3.

Abstract

Introduction

Metachromatic leukodystrophy (MLD) is a lysosomal storage disease caused by a deficiency of the arylsulfatase A (ARSA) enzyme, which is encoded by the ARSA gene. This condition exhibits an autosomal recessive Mendelian inheritance pattern. MLD leads to the progressive destruction of myelin in both the central and peripheral nervous systems and is marked by significant gross motor deterioration [1]. Brain magnetic resonance imaging (MRI) plays a crucial role in the quantitative and qualitative assessment of myelin degeneration in the central nervous system [2,3]. Additionally, the brain MRI scoring system is instrumental in predicting and evaluating clinical features across various disorders [4-6]. Specifically, the demyelination observed in brain lesions correlates with the gross motor function scale, which directly impacts the predicted quality of life in patients with leukodystrophy disorders [7,8]. Therefore, the extent of brain involvement is a critical factor in tracking the natural progression of MLD in patients [9-11]. Despite the existence of established scoring systems for neurodegenerative diseases, implementing these systems swiftly in clinical practice is challenging. A thorough discussion process that integrates radiological findings with clinical symptoms is essential. Consequently, we plan to revise the scoring system after adequate deliberation within the radiology department to facilitate its practical application in clinical settings. The objective of this study was to conduct a comprehensive analysis of brain MRI scoring data alongside clinical function scale assessments in individuals diagnosed with MLD, with the goal of reaching a deeper understanding of the disease trajectory and refining patient care strategies.

Materials and Methods

2. Subgrouping and clinical-imaging scoring system

The patients were grouped according to the primary subtypes of the disease. The clinical manifestations of MLD are divided into three main subtypes, categorized by the age at which the disease onset occurs: late-infantile, juvenile, and adult. In the late-infantile subtype, a decline in neurodevelopmental performance typically presents as hypotonia, ataxia, and developmental delays. For the juvenile subtype, symptoms include emotional or behavioral disturbances, intellectual disability, and impairments in vision and hearing, along with seizures. In the adult subtype, neurological decline is characterized by disorganized thinking, hallucinations, incontinence, spastic quadriplegia, and dystonia [15,16].

The clinical function scale, based on the standardized and validated gross motor function classification for MLD (GMFC-MLD score), was modified for this study. The GMFC-MLD scoring system categorizes patients into levels ranging from 0 to 6, based on their basic abilities to walk, sit, and move: level 0 indicates independent walking, while level 6 indicates a complete lack of locomotion [17]. For our purposes, we condensed the clinical function status into five levels: level 1 signifies walking independently, level 2 denotes walking with assistance, level 3 corresponds to mobility via wheelchair or the ability to sit only, level 4 indicates being bedridden, and level 5 represents brain death. Unlike traditional scales that assess gross motor function, our revised scale includes brain death as an outcome, improving our ability to predict prognosis and track progressive decline by including survival outcomes.

All MRI were obtained using a Discovery MR750 MRI scanner (GE Healthcare, Milwaukee, WI, USA), including axial and sagittal T1- and T2-weighted images. MRI scans from the patients were evaluated using a standardized scoring system for leukodystrophy. Eichler et al. [18] proposed a scoring system for brain MRI imaging of MLD, similar to the established method for adrenoleukodystrophy. The severity of myelin deterioration in MLD was assessed using the established scoring method, which is based on the increased signal intensities of T2-weighted MRI. In this system, a score of 0 indicates no abnormal signal intensity, 1 indicates a faint signal, and 2 indicates the presence of a dense signal. Additionally, the distinction between faint and dense signals lies in the presence of a well-defined margin of high signal intensity (Fig. 1). Furthermore, we simplified and modified several scores based on imaging findings to include both white and grey matter in all brain lesions. We categorized the brainstem separately into the midbrain, pons, and medulla. We also updated the cerebellum parenchyma score to reflect severity: 0 for no signal, 1 for faint, and 2 for dense. Additionally, if cerebral and cerebellar atrophy were present, a score of 2 was added to the overall score. Therefore, the total possible score of the modified scoring system was 40, representing the most severe condition (Table 1). More than two independent radiologists confirmed the MRI data, supporting our scoring system primarily using T1- and T2-weighted images.

Fig. 1.

Brain magnetic resonance imaging scoring of cases: (A) high signal in bilateral periventricular areas (T2, dense), (B) regional high signal in pons, indicated by yellow arrows (T2, faint), and (C) fourth ventricle dilation with cerebellar atrophy (T1, sagittal).

Table 1.

Modified imaging scoring table of brain MRI in pediatric MLD

Results

Thirty-four MRI scans and clinical data from 10 patients (four boys and six girls) were assessed. Among these patients, six had the juvenile type and four had the late-infantile form of MLD. All patients initially presented with delayed developmental milestones compared to normal development, which was the first symptom at onset. Other clinical symptoms included gait disturbances and seizures, with the juvenile type predominantly showing gait disturbances. Leukodystrophy, the most notable feature in the brain MRI scans of MLD patients, was initially observed in eight patients but appeared in the final MRI scans of all patients. Eight patients exhibited decreased arylsulfatase A activity below 0.50 nmol/min/mg protein, along with characteristic brain MRI findings for MLD. The remaining two patients, who did not show decreased enzyme activity, had an ARSA mutation confirmed by direct sequencing (Table 2). The MLPA method confirmed an ARSA mutation in six of the eight patients. Pseudo-deficiency was ruled out in the other two patients, who did not undergo gene testing, as they met the other diagnostic criteria and had characteristic brain MRI findings, decreased enzyme activity, and clinically delayed development [19]. All patients experienced developmental deterioration.

Table 2.

General characteristics of MLD patients

Each patient's initial MRI was conducted within 1 year of the first appearance of symptoms, and the most recent MRI was performed within 10 years of symptom onset. The average duration of follow-up at our clinic was 5 years. The mean score of the last brain MRI (32.2±7.2) was twice that of the initial brain MRI scores (16.2±11.8) across all MLD patients. The average initial and final MRI scores for the juvenile type were higher than those for the late-infantile type (late-infantile type: initial 9.0±6.0, last 21.0±12.3; juvenile type: initial 28.2±9.2, last 34.8±3.4). After adjusting for age, correlation analysis revealed a positive relationship between MRI scores and clinical function levels in all MLD patients (rho=0.59, P=0.002) (Fig. 2). A stronger positive correlation was observed between clinical function and brain MRI scores in the late-infantile type (rho=0.700, P=0.003) compared to the juvenile type (rho=0.513, P=0.029). Patients with high signal intensities in the pons (rho=0.557, P=0.013) and basal ganglia (rho=0.470, P=0.024) demonstrated a strong relationship between clinical function decline and brain MRI scores, unlike those with high signal intensities in the midbrain (rho <0.001, P=1.000). Cerebellar atrophy also exhibited a strong positive correlation (rho=0.533, P=0.005). However, no statistically significant correlation was found between any cerebral and cerebellar lesions, including ventricular dilatation.

Fig. 2.

Correlation of magnetic resonance imaging (MRI) score and clinical status. (A) We fix the age as a covariate and the positive correlation between the MRI score and clinical function status in metachromatic leukodystrophy patients. (B) The scatter plot with trend lines of each subgroups. CFS, clinical function scale.

Discussion

This study investigated the association between clinical function status and brain MRI scores in MLD patients. During an average follow-up period of 5 years, most patients exhibited disease progression and a trend toward clinical regression. Additionally, the juvenile type displayed higher early and late MRI scores compared to the late-infantile type, potentially indicating a delayed diagnosis due to slower disease progression and later onset. Fumagalli et al. [8] presented data from a longitudinal study involving 45 MLD patients, revealing that the total MRI score for the late-infantile type was higher than that for other types, while the initial MRI scores for the adult and juvenile types were higher than those for the late-infantile type. Our simplified scoring system further supported these findings, showing that the average initial brain MRI score was lower than the subsequent MRI scores across all MLD patients, with the initial scores for the juvenile type being higher than those for the late-infantile type.

We found a positive correlation between clinical scores and brain MRI scores, indicating that patients with higher brain MRI scores experienced greater clinical deterioration. Previous studies have established a link between clinical status and brain imaging findings in patients with MLD [20]. However, the earlier scoring table primarily focused on the white matter of brain lesions, including periventricular, ventral, and U-fiber areas [18]. Grey matter volume in MLD patients is also affected by progressive deterioration due to demyelination and neuronal dysfunction [21]. In the previous scoring system, cerebral atrophy could receive a maximum of 2 points, whereas cerebellar atrophy was only assigned 1 point, leading to inconsistencies. We have revised the scoring table to integrate the involvement of both white and grey matter in MLD, facilitating a quick and straightforward interpretation in clinical settings.

Clinical function deterioration is a key symptom of disease progression in MLD patients. The assessment of developmental deterioration was based on the gross motor function scale. The established GMFC-MLD scores range from 0–6, with 6 indicating the most severe impairment. However, this scoring system does not reflect survival status, and the range is too narrow to capture the gradual decline [17]. To facilitate quicker assessments in a clinical context, we simplified the scale. We reclassified scores 0-1 as 2, 3–4 as 3, and 5–6 as 4 to simplify the existing ranges further. This modified scale allowed for a more precise classification of patient function levels, aligning more closely with clinical abilities and the severity of MRI findings. We observed positive correlations between the clinical function score and brain MRI score, particularly in the late-infantile type compared to the juvenile type. We presume that a higher MRI score in MLD patients with the late-infantile type indicates a poorer prognosis, as these patients exhibit early onset severe motor dysfunction.

MRI volumetric data, obtained through a regional methodology, demonstrate a correlation between disease severity and the clinical function status of MLD patients [22]. Our findings indicate that involvement of the pons and basal ganglia, as well as cerebellar atrophy detected using the brain MRI scoring system, are associated with a higher clinical function status. In contrast, midbrain involvement did not show a statistically significant correlation with clinical status. Classic MRI findings in MLD typically reveal bilateral symmetric involvement of the periventricular white matter, corpus callosum, corticospinal tracts, and cerebellum [23,24]. Although there have been a few reports of brainstem involvement in the progression or diagnosis of MLD, these are not common [25,26]. Based on our results, we hypothesize that midbrain lesions are less significantly correlated with clinical characteristics than other brainstem lesions in MLD.

It is necessary to consider various factors that contribute to the onset and progression of diseases, including the methods used for diagnosis. Although the age at symptom onset is a critical factor in the natural progression of leukodystrophy, it is only one of several clinical factors used to predict prognosis [27]. The spectrophotometric enzyme assay plays a vital role in detecting arylsulfatase A deficiency [28]. However, in cases of the juvenile type, which progresses more slowly, our data suggest that DNA sequencing may offer greater diagnostic precision than measuring enzyme activity. Although targeted sequencing enhances the accuracy of disease detection, the high cost of gathering individual genetic data poses a significant challenge for predicting disease progression [29].

The brain MRI scoring system serves as a valuable standard tool for tracking disease progression. It also plays a crucial role in formulating treatment plans for managing progressive neurodegenerative diseases. Treatment challenges have arisen, particularly in the context of bone marrow transplantation or gene therapy for early presymptomatic MLD patients. These early interventions could potentially halt the detrimental progression of neurodegeneration [19]. However, additional data, including diagnostic methods and predictive factors, are necessary to confirm whether this tool is optimal for predicting disease prognosis and identifying the most effective treatment plans. This study is limited by the small number of patients enrolled. However, given that MLD is a rare genetic disorder, there have been few studies focusing exclusively on Korean individuals with MLD. To better assist MLD patients in Korea, it is necessary to collect more data and conduct further studies that utilize the proposed scale and scoring system to analyze prognosis.

Article information Continued