Clinical, Radiological, and Genetic Profiles of Eight Patients with Combined Dystonic Manifestation of Type-III GM1 Gangliosidosis: A Video Case Series from India

Introduction

GM1 gangliosidosis is an uncommon autosomal recessive lysosomal storage disorder resulting from biallelic pathogenic variants in the GLB1 gene, which encodes the lysosomal enzyme β-galactosidase. This enzyme is essential for degrading GM1 gangliosides and related glycoconjugates. Its deficiency leads to progressive intracellular accumulation, particularly in neurons, causing widespread neurodegeneration. Based on the clinical phenotype, GM1 gangliosidosis has been classified into three types: type-I or infantile form with onset between birth and 6 months, rapidly progressive with hypotonia, severe central nervous system (CNS) degeneration and death by 1–2 years of age; type-II including late infantile or juvenile form with onset usually between 7 months and 3 years, lag in motor and cognitive development, and slower progression; and type-III or adult/chronic variant with later onset between 3 to 30 years, a progressive movement disorder [1].

Pathogenic variants in the GLB1 gene on chromosome 3p21.33 produce different phenotypes depending on the region affected. Mutations in exons 2, 6, 15, and 16 are typically associated with neurodegenerative features of GM1 gangliosidosis, while mutations near the 3’ end result in Morquio B disease, which mainly affects the skeleton [2]. Variants in the core protein region tend to result in early infantile presentations, whereas those affecting surface regions manifest later (types II and III).

Patients with Type-III GM1 gangliosidosis typically retain 5–10% of enzyme activity [1]. Juvenile and adult-onset forms often present with complex movement disorders. Patients typically develop generalized dystonia with prominent oromandibular, lingual, and facial involvement, along with slurred or spastic speech, spasticity, and progressive gait difficulties. Cognitive dysfunction may vary and is not universally present in early stages. A combination of progressive dystonia involving bulbar and appendicular muscles, along with corticospinal signs, can serve as key diagnostic indicators [3].

The adoption of next-generation sequencing (NGS) has enhanced diagnostic precision, particularly in cases of atypical or late-onset presentations. However, underdiagnosis remains common in developing nations due to limited awareness and constrained access to molecular testing. In this case series, we describe eight genetically confirmed patients with Type-III GM1 gangliosidosis from a tertiary care center in India. We aim to delineate the clinical and imaging features, highlight recurrent genetic variants, and emphasize the importance of considering GM1 gangliosidosis in young-onset, cranial-predominant, generalized, combined dystonia.

Methods

This is a retrospective, descriptive case series conducted at the Department of Neurology, National Institute of Mental Health and Neurosciences (NIMHANS), India. The study included patients with genetically confirmed GM1 gangliosidosis, managed by the authors, between January 2019 and May 2025. The available demographics, clinical history, examination details, and investigations, including imaging, electrophysiological, and genetic data, were documented. Additionally, patient videos were reviewed when available. The data were expressed using descriptive statistics. The waiver was obtained from the National Institute of Mental Health and Neurosciences Institute Ethics Committee (No. NIMH/DO/IEC (BS & NS DIV)/2024-25) as the study involved retrospective data and the data were anonymised. Written informed consent was taken from the patients for the video recording and for publication in print or online.

Results

Eight patients (6 males) with genetically confirmed GM1 gangliosidosis were identified and recruited. The clinical, investigations, treatment, and follow-up details of all eight patients are summarized in Table 1. MRI imaging findings of seven patients are provided in Figure 1.

Figure 1

Discussion

Our cohort delineates the Type-III (adult/chronic) GM1 gangliosidosis phenotype, characterized by an adolescent-to-adult onset and a striking movement disorder profile dominated by generalized dystonia, in which oromandibular and cranio-cervical involvement is early and prominent. Parkinsonian features (bradykinesia/rigidity, hypomimia) and corticospinal signs occurred variably. MRI consistently demonstrated selective basal-ganglia involvement, notably the “wishbone” susceptibility pattern in the globus pallidus with posterior putaminal T2/FLAIR change. Genetic testing frequently identified heterozygous c.1325G>A;p.Arg442Gln, along with another heterozygous variant. Clinically, response to symptomatic therapy was variable but generally mild, mirroring prior literature [1, 6, 7, 8, 9]. The uniqueness of this series lies in the convergence of a highly stereotyped oromandibular–cranio-cervical dystonia phenotype with reproducible “wishbone” imaging and a recurrent R442Q background within an Indian centre, features that together strengthen bedside recognition and genotype-phenotype inference in adult GM1 gangliosidosis.

Type-III GM1 gangliosidosis typically manifests as a slowly progressive combined dystonia (often cranio-cervical/oromandibular with generalization) along with corticospinal signs, parkinsonism, ataxia, and dysarthria, but with minimal or no visceromegaly or coarse facies. The facial dyskinesia, especially the lower facial involvement, is an interesting feature of type-III GM1 gangliosidosis and has been described as facial dystonia or facial grimacing in the previous reports [7, 10, 11]. Although clinically it appears a bit fast for a dystonic movement, one of the patients underwent surface electromyography of the angle of the mouth and had a burst duration of around 1–2 seconds, suggestive of dystonic spasms. Even though cerebellar signs are reported in a few cases of type-III GM1 gangliosidosis [5, 12], it usually not frequent in type-3 [7, 10, 13] compared to the other types [14, 15, 16]. This may either be due to selective involvement of the brain structures in different subtypes or due to dystonia marring the underlying cerebellar features. Disability accrues over decades and is driven primarily by motor burden; cognition is often normal or only mildly impaired for many years [1, 3, 6, 7]. In contrast, type-I (infantile) begins within the first six months and features rapid neuroregression, visceromegaly, cherry-red spots, dysostosis multiplex, cardiomyopathy, and early lethality [1, 17]. Type II (late-infantile/juvenile) presents later with slower neuroregression, atypical seizures, variable skeletal disease without organomegaly [1, 17]. These differences parallel higher residual enzyme activity seen in late-onset Type-III disease [1, 7].

Radiologically, in Type-III, susceptibility-weighted sequences often reveal the “wishbone” pattern of pallidal mineralization with posterior putaminal involvement, a highly suggestive sign for adult GM1 gangliosidosis when aligned with the clinical picture [5, 9]. By comparison, type-I more frequently shows diffuse/delayed myelination, thalamic/brainstem T2 hyperintensities, and early cerebral atrophy. Type-II tends to display patchy white-matter abnormalities and milder atrophy rather than selective pallidal mineralization [1]. Emphasizing this contrast facilitates earlier recognition of adult GM1 gangliosidosis among movement disorder presentations.

Clinically and radiologically, Type-III GM1 gangliosidosis can closely mimic neurodegeneration with brain iron accumulation phenotype. Pantothenate kinase-associated neurodegeneration can have similar phenotype with pallidal mineralization, but in addition has central pallidal hyperintensity resulting in eye-of tiger appearance of T2 MRI sequence. The splitting of pallidal mineralization can mimic mitochondrial membrane protein-associated neurodegeneration; however, it lacks the posterior putaminal atrophy noted in type-III GM1 gangliosidosis. Facial-faucial-finger myoclonus seen in Kufor-Rakeb syndrome can mimic the rapid facial dystonia of type-III GM1 gangliosidosis, but the MRI finding helps in differentiating the two.

Electrophysiological investigations such as EEG, NCS, EMG, and EP are usually normal or nonspecific in Type-III GM1 gangliosidosis. Routine haematology/biochemistry is typically unrevealing. Bone-marrow vacuolated cells or foamy histiocytes, typical of infantile disease are uncommon in late-onset GM1 gangliosidosis, mirroring the paucity of systemic storage signs in adults [1, 7]. These negative or nonspecific ancillary tests, when juxtaposed with the characteristic clinical–imaging dyad, help narrow the differential.

Across the GM1 gangliosidosis spectrum, phenotype correlates with residual β-galactosidase activity: infantile forms generally exhibit less than 1% activity, juvenile forms around 1–5%, and adult forms retain higher residual activity, enabling slower substrate accumulation and a later onset [1, 2, 17, 18]. Many late-onset alleles are missense variants that destabilize folding or lysosomal trafficking, or modestly impair catalysis rather than abolish it. Structural studies mapping disease variants onto the (β/α)8 TIM-barrel catalytic domain and adjacent β-domains reveal that changes outside catalytic residues often preserve measurable activity and partial lysosomal targeting—molecular features congruent with the Type-III phenotype [2, 17, 18].

The c.1325G>A;p.Arg442Gln or R442Q variant is repeatedly linked with late-onset/Type-III GM1 gangliosidosis subtype (Table 2) [2, 18, 19, 20, 21]. Its absence from a large Indian multicenter GLB1 survey suggests regional heterogeneity of variant spectra [22]. In a well-characterized adult compound-heterozygote (p.T329A/p.R442Q), fibroblast studies demonstrated residual β-galactosidase activity that increased with galactose, supporting a misfolding/stability defect rather than catalytic ablation [23]. Longitudinal experience indicates that substrate-reduction therapy with miglustat may stabilize or improve disease in juvenile/adult GM1 gangliosidosis—including this lineage—underscoring that R442Q aligns with residual-activity, potentially therapy-amenable biology. Collectively, R442Q sits among late-onset alleles with plausible chaperone/SRT responsiveness.

Currently, no disease-modifying therapies are approved for GM1 gangliosidosis. Care remains multidisciplinary and supportive symptomatic therapy. All patients in our cohort were managed with symptomatic therapy with variable subjective improvement. However, promising therapies are in development, particularly for late-onset forms. Approaches under investigation include gene therapy, substrate reduction therapy, pharmacological chaperones to stabilize misfolded enzyme proteins, and hematopoietic stem cell transplantation. The presence of residual enzyme activity and slower disease progression in Type-III GM1 gangliosidosis makes these patients ideal candidates for future interventions. Substrate-reduction therapy with miglustat has shown stabilization or improvement in the juvenile–adult GM1 gangliosidosis series [17]. Pharmacologic chaperoning (for example, galactose and iminosugar compounds such as NOEV) can augment residual β-galactosidase activity in an allele-specific fashion, including R442Q compound-heterozygous contexts. Gene-therapy programs using AAV-mediated GLB1 delivery are progressing in pediatric cohorts; although adult/Type-III data remain limited, the shared pathobiology supports future applicability as delivery and dosing strategies evolve. Due to the lack of established benefit of deep-brain stimulation (DBS) in GM1 gangliosidosis and complex dystonia manifestation, we did not offer DBS for any of the patients.

Our study contributes valuable real-world data on a rare condition, particularly from an underrepresented region. It highlights a consistent clinical and imaging phenotype, reinforcing the significance of recurrent GLB1 mutations in the Indian context.

Limitations include the retrospective nature of data collection and the limited availability of long-term follow-up.

Conclusion

This case series highlights the distinct clinical, radiological, and genetic features of late-onset GM1 gangliosidosis in an Indian cohort. Our findings reinforce that Type-III GM1 gangliosidosis should be considered a key differential diagnosis in young individuals presenting with progressive generalized dystonia, particularly when accompanied by predominant oromandibular and facial involvement, and corticospinal signs. The consistent presence of the “wishbone pattern” on MRI and the recurrent detection of the p.Arg442Gln GLB1 variant further strengthen the diagnostic framework.

Ethics and Consent

The waiver was obtained from the Ethics Committee of the National Institute of Mental Health and Neurosciences Institute [No. NIMH/DO/IEC (BS & NS DIV)/2024-25]. Patients’ details were anonymized to maintain patient privacy and informed consent was obtained for the study, video recording, dissemination, and publication. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Competing Interests

The authors have no competing interests to declare.

Author Contributions

1. Research project: A. Conception, B. Organization, C. Execution;

2. Statistical Analysis: A. Design, B. Execution, C. Review and Critique;

3. Manuscript Preparation: A. Writing of the first draft, B. Review and Critique;

   SR: 1A, 1B, 1C, 2A, 2B, 3A

   CA: 1A, 1B, 1C, 2A, 2B, 3B

   VVH: 1A, 1B, 1C, 2B, 3B

   SP: 1C, 2C, 3B

   PP: 1C, 2C, 3B

   NK: 1C, 2C, 3B

   BM: 1C, 2C, 3B

   JS: 1A, 2C, 3B

   RY: 1A, 1B, 1C, 2C, 3B

   PKP: 1A, 1B, 1C, 2C, 3B

Subhajit Roy, Cheshta Arora and Vikram V. Holla authors equal contributions.