GM1 Gangliosidosis (Infantile, Juvenile and Adult Forms)

Autosomal recessive lysosomal disease due to deficiency of acid β-galactosidase, encoded by GLB1, with a continuous spectrum ranging from multisystem infantile neurodegeneration to chronic dystonia of the adolescent and the adult

Literature and regulatory information updated: 31 October 2026.

0. Initial Reference Table

1. Introduction

1.1. From «pseudo-Hurler» to β-galactosidase deficiency

GM1 gangliosidosis was recognised as an entity before its biochemical defect was known. In 1951 Caffey described infants whom he interpreted as having «gargoylism» of prenatal onset and who, according to the OMIM review, probably corresponded to the first cases [2]. In 1964 Landing and colleagues published the first definitive description, bringing together patients until then labelled «Hurler variant», «pseudo-Hurler disease» or «Tay-Sachs disease with visceral involvement»: severe cerebral degeneration with death in the first two years of life, ganglioside accumulation in neurones, liver and spleen histiocytes and glomerular epithelium, and skeletal deformities similar to those of Hurler disease [2]. In 1965, O'Brien and colleagues identified the stored lipid as a GM1 ganglioside, distinct from the GM2 of Tay-Sachs disease, and proposed the designation «generalised gangliosidosis» [2]. The term «pseudo-Hurler» from this period is not to be confused with «pseudo-Hurler polydystrophy», the historical name of mucolipidosis III, a disease due to deficiency of N-acetylglucosamine-1-phosphotransferase without mucopolysacchariduria [18].

The founding paper of the biochemical understanding is that of Okada and O'Brien, published in Science in 1968. In the tissues (liver, spleen, kidney and brain) of two patients with generalised gangliosidosis, β-galactosidase activity was reduced 10- to 30-fold, both with the artificial substrate p-nitrophenyl-β-D-galactopyranoside and with GM1 ganglioside labelled with carbon 14 in the terminal galactose; the authors concluded that this deficiency was the primary enzymatic defect of the disease [19, 20]. In the following years, complementation studies in cell hybrids and studies of immunoreactive material showed that the infantile, juvenile and adult forms resulted from alterations at the same locus, despite contradictory early complementation results that O'Brien and Norden explained by interactions between mutant monomers [2, 5].

1.2. Molecular milestones

The β-galactosidase gene was assigned to chromosome 3 by somatic cell hybridisation in 1978–1979 and localised by in situ hybridisation to 3p21.33 in 1993; the current designation is 3p22.3 [6]. A locus initially called «β-galactosidase-2» on chromosome 22 later proved to be that of protective protein/cathepsin A (CTSA), on chromosome 20, whose deficiency causes galactosialidosis [6]. Spranger's book on bone dysplasias still gives the historical location 3p21-pter [21]. Oshima and colleagues cloned the complementary DNA in 1988 from human placenta; Morreau and colleagues demonstrated in 1989 the alternative processing that gives rise to the elastin-binding protein [6]. Yoshida and Nishimoto independently identified in 1991 the first pathogenic variants in Japanese patients with the three clinical forms, with an inverse correlation between severity and residual enzyme activity [6]. In 2012 the crystal structure of the human enzyme was published, which made it possible to place the variants of GM1 gangliosidosis and of mucopolysaccharidosis IVB in their domains [22].

1.3. Nomenclature, classification and definition of the entities

The classification used by GeneReviews divides GM1 gangliosidosis into type I (infantile, onset before 12 months), type II (subdivided into late infantile, with onset between 1 and 3 years, and juvenile, between 3 and 10 years) and type III (chronic or adult, between 10 and 30 years), and acknowledges that this is a continuum divided arbitrarily [1, 23]. Saudubray uses the numbering 1, 2a, 2b and 3 and places the onset of type 2a between 7 months and 2–3 years [3]; OMIM places type II between 7 months and 3 years and type III between 3 and 30 years [4, 5]. The French cohort of Laur and colleagues applied different cut-offs: type 1 with onset up to 6 months, 2a from 7 months to 2 years, 2b from 2 to 3 years and type 3 after 3 years [10]. These divergences have a practical consequence: the same child may be classified differently depending on the reference, and trial inclusion criteria use their own definitions (for example, onset before 6 months for type 1 in the intracisternal gene therapy trial) [13]. Lang and colleagues proposed criteria for subtyping type 1 from 154 published cases [24].

The same enzyme deficiency causes a second disease, mucopolysaccharidosis IVB (Morquio B syndrome), in which skeletal dysplasia due to keratan sulfate accumulation dominates and the central nervous system is usually spared [1, 3]. Mucopolysaccharidosis IVB is covered here only as an allelic form, as a source of mixed phenotypes and as a differential diagnosis. Galactosialidosis, a combined secondary deficiency of β-galactosidase and neuraminidase due to a cathepsin A defect, and the GM2 gangliosidoses are part of the differential diagnosis. The old names «beta-galactosidase-1 deficiency» and «beta-galactosidosis» remain useful for searching the older literature [1].

2. Molecular Genetics and Pathophysiology

2.1. The GLB1 gene and its products

GLB1 spans about 62.5 kb at 3p22.3 and has 16 exons [6]. The MANE transcript is NM_000404.4, which encodes the protein NP_000395.3 [7]. The main 2.5 kb transcript encodes the lysosomal form of the enzyme, with 677 amino acids, a calculated mass of 75 kD, a 23-residue signal peptide and seven potential N-glycosylation sites [6]. The enzyme is synthesised as an 85 kDa precursor, glycosylated after translation into an 88 kDa form and processed into the mature 64 kDa lysosomal enzyme [25]. A minor 2.0 kb transcript, lacking exons 3, 4 and 6, encodes a 546-amino-acid protein without enzymatic activity (S-Gal), which is the elastin-binding protein (EBP); this protein binds tropoelastin in the endosomal compartment, acts as a chaperone for its secretion and for elastic fibre assembly and is anchored at the cell surface by association with neuraminidase and cathepsin A [6, 26]. The enzyme has a TIM barrel catalytic domain followed by two β domains [22].

2.2. Enzyme function and substrates

Acid β-galactosidase (EC 3.2.1.23) removes terminal galactose residues in β linkage from glycoconjugates. It is needed to degrade GM1 ganglioside, other galactose-containing glycosphingolipids, oligosaccharides derived from glycoproteins and keratan sulfate [3, 26]. In the ganglioside degradation pathway, removal of the terminal galactose converts GM1 into GM2, which hexosaminidase A then converts into GM3 [27]. In the lysosome, the enzyme works in a multienzyme complex with neuraminidase 1 (NEU1), protective protein/cathepsin A (CTSA) and N-acetylgalactosamine-6-sulfate sulfatase (GALNS); this complex explains why galactosialidosis, in which the lack of cathepsin A destabilises β-galactosidase and neuraminidase, has a similar phenotype [3, 6]. Galactosylceramidase (GALC), deficient in Krabbe disease, is a genetically distinct β-galactosidase [6]. As the enzyme also removes galactose from glycoproteins, Saudubray includes GM1 gangliosidosis among the lysosomal diseases of protein deglycosylation, alongside the mucolipidoses and the oligosaccharidoses [28].

2.3. From enzyme deficiency to disease

Loss of function leads to accumulation of three groups of substrates: GM1 ganglioside (massive storage in the brain), oligosaccharides derived from glycoproteins (liver and urine) and keratan sulfate (liver and spleen) [3]. The lyso compound of GM1 is increased and may have pathogenic significance [3]. Degeneration of the central nervous system is the dominant feature because it is there that ganglioside synthesis is greatest [1]. In areas of intense neuronal vacuolation there is neuronal death and demyelination, accompanied by astrogliosis and microgliosis [25]. For this reason, the most severe forms combine features of neuronal lipidosis, mucopolysaccharidosis and oligosaccharidosis [3]. Urinary excretion of keratan sulfate is lower in GM1 gangliosidosis than in mucopolysaccharidosis IVB [3]. In murine models, GM1 accumulated in the mitochondria-associated endoplasmic reticulum membranes alters calcium transfer between the two organelles, opens the mitochondrial permeability transition pore and activates neuronal apoptosis [6]. Some variants interfere with phosphorylation of the precursor, which is then secreted instead of being routed to the lysosome [2], and others retain the misfolded protein in the endoplasmic reticulum [6, 29].

2.4. Residual activity and phenotype

There is an inverse relationship between residual activity and the amount of substrate stored in neural tissue [1]. In Yoshida's Japanese series, activity was 0.65–1.58% of control in the infantile form, 2.65–3.05% in patients with the R201C variant and late infantile or juvenile forms, and 4.24–7.28% in the adult form [6]. GeneReviews summarises activity in leucocytes or fibroblasts as negligible in type I, about 1–5% in the late infantile form, 3–10% in the juvenile form, 5–10% in type III and 2–12% in mucopolysaccharidosis IVB, and stresses that the correlation is general but does not allow the type to be predicted in an individual patient, because the assays use artificial substrates [1, 26]. The substrate specificity of the mutant enzyme also counts: variants that mainly affect the keratan sulfate binding pocket produce mucopolysaccharidosis IVB with little GM1 accumulation [1, 26].

2.5. Variant spectrum and nomenclature

Saudubray refers to more than 160 pathogenic variants and the mini-review by Nicoli and colleagues counted 261 in 2021, with a predominance of missense and nonsense variants and clustering in exons 2, 6, 15 and 16 [3, 23]. In 2008, Brunetti-Pierri and Scaglia counted up to 102 [25]. Sequencing detects more than 99% of the recorded pathogenic variants and targeted deletion and duplication analysis less than 1% [1]; these figures are proportions of variants detectable in a database and are not equivalent to the clinical sensitivity of the test. The following table presents recurrent variants, normalised in VariantValidator against the MANE transcript.

The version of GeneReviews consulted writes the Japanese variant c.152T>C as p.Ile15Thr; normalisation confirms p.(Ile51Thr), in agreement with the designation I51T in OMIM [1, 5, 7]. In India, a series of 46 patients identified 33 variants, 20 of them novel, with NM_000404.4:c.75+2dup (historical designation c.75+2insT) in 14% and c.1010T>C p.(Leu337Pro) (L337P) in 10% of alleles, 41 homozygous patients (25 children of consanguineous parents) and no genotype-phenotype correlation [6, 7, 33]. In Italy, Caciotti and colleagues identified 27 variants in 21 patients with GM1 gangliosidosis and 4 with Morquio B, including a juvenile patient homozygous for a variant described in Morquio B and a complex allele p.R68W+p.R109W in trans with p.R148C [34]. Against NM_000404.4, these variants correspond to c.202C>T p.(Arg68Trp), classified as pathogenic in ClinVar, c.325C>T p.(Arg109Trp), classified as benign or likely benign, and c.442C>T p.(Arg148Cys), pathogenic or likely pathogenic [33].

2.6. Pseudodeficiency

Gort and colleagues described a clinically unaffected father, from a Basque family, with reduced β-galactosidase activity due to compound heterozygosity for his son's truncating variant and for p.R595W, which retains 33–59% of activity; p.R595W was found in 3.2% of Basque controls and in 0.8% of non-Basque controls [6]. Against the MANE transcript, this variant is NM_000404.4:c.1783C>T p.(Arg595Trp), recorded in ClinVar as of uncertain significance [33]. This finding means that low activity in an asymptomatic individual must be interpreted together with the genotype, particularly in family and carrier studies.

2.7. Genotype-phenotype correlations and mixed phenotypes

The correlations are imperfect because of allelic heterogeneity [1]. In the structure of the enzyme, variants located at the surface are associated with less severe phenotypes (type III), and those in the protein core or the active site with type I [1]. In the German and Austrian series of Arash-Kaps, most genotypes showed a clear correlation with the phenotype [35], whereas in the Indian series there was no correlation [6]. A variant typical of mucopolysaccharidosis IVB in compound heterozygosity with another GLB1 variant may give GM1 gangliosidosis type II or III, or a mixed phenotype with dysostosis and neurodegeneration [1]. Paschke and colleagues found W273L in 14 of 15 European patients with Morquio B and the Morquio phenotype also in compound heterozygotes for W273L and alleles typical of GM1 gangliosidosis [6]. In the Brazilian series from Campinas, all juvenile and adult patients were compound heterozygotes and p.T500A, the most frequent allele (8 families), was associated with a phenotype with a Morquio B component [32]. Within the same family, siblings may differ in severity and age of onset [1], and between families with the same variants concordance is lower [32]. In the Japanese adult form, the age of onset and the course varied widely among I51T homozygotes, which points to additional genetic or environmental factors [5].

2.8. Elastin-binding protein and cardiac involvement

Variants located in the region shared by the lysosomal enzyme and the EBP abolish expression of the EBP. Hinek and colleagues showed that the Morquio B variants and those of infantile GM1 gangliosidosis with cardiac involvement, homozygous in Italian patients studied by Morrone, impair tropoelastin secretion and elastic fibre assembly, unlike missense variants that affect only the lysosomal enzyme; these diseases have been described as «secondary elastinopathies» [2, 6].

2.9. Inheritance and allelic disorders

Inheritance is autosomal recessive in all forms, with no second inheritance pattern described in the sources consulted [1, 2, 4, 5]. ClinGen has two «definitive» gene-disease validity assertions, for GM1 gangliosidosis and for mucopolysaccharidosis IVB, but no actionability assertion [26]. There are no phenotypes associated with germline GLB1 variants other than GM1 gangliosidosis and mucopolysaccharidosis IVB [1]. A case of maternal uniparental isodisomy of chromosome 3 showed that the disease can arise with only one carrier parent [36]. In crossed animal models, the dose of the β-galactosidase gene modified the phenotype of galactosylceramidase deficiency, which suggests a modifying role of GLB1 in another lysosomal disease [6].

3. Clinical Spectrum and Manifestations by System

3.1. Overview by type

The following table, adapted from GeneReviews, summarises the findings that separate the types; the cohort figures appear in the subsections.

Source: GeneReviews, Table 3 [1].

3.2. Type I (infantile)

Prenatal and neonatal period. The disease may present before birth with non-immune fetal hydrops (6% in a series cited by GeneReviews), intrauterine growth restriction (1%) and placental vacuolation [1, 3]. At birth there may be coarse facies, facial oedema, pitting oedema of the hands and feet, macroglossia, thick gums, hypotonia and feeding difficulty [37, 38]. A photograph of a newborn in the atlas of bone dysplasias shows thickened eyelids, upslanting palpebral fissures, depressed nasal bridge, anteverted nostrils, long philtrum and gingival hyperplasia [39]. Extensive congenital dermal melanocytosis, with dorsal and ventral distribution and progressive, is a cutaneous sign of lysosomal disease (differential diagnosis with the mosaic forms of phakomatosis pigmentovascularis): in a review of 39 cases, 11 had GM1 gangliosidosis, and six Indian infants with type I had patches from birth [1, 2]. Diffuse angiokeratomas before 10 months were described in one case [2]. In Giugliani's 1985 series in Porto Alegre, fetal losses and macrosomia appeared to be more frequent [2].

Nervous system. Infants are often hypotonic in the first weeks, with poor head control; development stops between 3 and 6 months; the exaggerated startle gives way to spasticity and rapid regression, with blindness, deafness, decerebrate rigidity, seizures, dysphagia and excessive secretions by the end of the first year [1, 3]. In the French cohort of 41 patients with type 1, the first symptoms were hypotonia (63%), dyspnoea (17%) and nystagmus (15%) [10]. In the prospective Minnesota study (8 infants with GM1 gangliosidosis), hypotonia appeared before 6 months, severe motor impairment in the first year, dysphagia and placement of a feeding tube before 18 months, and neurodevelopmental tests reached the floor of the scale between 20 and 28 months [40]. The meta-analysis of 154 published cases showed multiorgan dysfunction and regression concentrated between 6 and 18 months [24]. Saudubray indicates seizures in 10% of cases [3], but in the Minnesota study most infants with GM1 gangliosidosis had epilepsy by 18 months, according to the discussion by Fiori and colleagues [41]; the difference reflects distinct populations and eras. Infantile GM1 gangliosidosis is among the lysosomal diseases with prominent epilepsy in infancy [38].

Eyes. The macular cherry-red spot is present in about half of cases and rarely before 6 months; visual failure appears after a few months, often with pendular nystagmus [3]. Optical coherence tomography may show hyperreflectivity of the ganglion cell layer [41].

Viscera and heart. Hepatosplenomegaly is present in 85%, typically in the first 7.5–10 months, and may be absent at the time of diagnosis [1]. Cardiomyopathy, hypertrophic or dilated, occurs in about one third [3]; there may be cardiomegaly, hypertension and mitral valve thickening [1, 2]. GM1 gangliosidosis is among the lysosomal diseases with storage cardiomyopathy [38].

Face and skeleton. Dysmorphism is present in 87%: coarse facies, frontal bossing, depressed nasal bridge with a broad tip, long philtrum, large low-set ears, gingival hypertrophy, macroglossia, thick skin and hirsutism [1, 3]. Radiological changes are constant in typical cases and minimal in cases with psychomotor deterioration alone: subperiosteal bone formation at birth, widening of the diaphyses with tapering of the ends and, around 6 months, Hurler-like changes with thoracolumbar vertebral beaking and proximal pointing of the four lateral metacarpals [3, 39]. Kyphosis, scoliosis, hip dislocation and osteoporosis are part of the picture [1]. In the Minnesota study, vertebral beaking, kyphosis and scoliosis distinguished infantile GM1 gangliosidosis from the GM2 gangliosidoses [40].

General laboratory findings. Aspartate aminotransferase (AST) is raised with normal alanine aminotransferase (ALT), serum chitotriosidase is increased and the blood film shows vacuolated lymphocytes and eosinophils with abnormal granulation [1]. Persistently raised alkaline phosphatase has been described as an early sign in infants still without neurological signs [41, 42].

Survival. GeneReviews indicates death at 2–3 years, often from aspiration pneumonia [1]. The mean age at diagnosis in Lang's meta-analysis was 8.7 months and the mean age at death 18.9 months [24]. In the RETRIEVE study, median survival of the 60 patients with onset before 24 months and retrospective data was 19.0 months (95% confidence interval [CI] 18.0–22.0) [43]. In France, mean survival in type 1 was 23 months (95% CI 7–39) [10]. The natural history review by the NIH group places death between 2 and 4 years, often from aspiration pneumonia, and stresses that the skeletal dysplasia, already present at diagnosis, is progressive and contributes to restrictive lung disease [44]. These figures describe the disease without disease-modifying treatment, with variable supportive care.

3.3. Type II (late infantile and juvenile)

Late infantile (IIa). Onset is between 1 and 3 years in GeneReviews, with arrest and regression of development, corneal clouding, motor abnormalities and diffuse, progressive cerebral atrophy [1]. The first signs may be instability when sitting or standing, hypotonia, gait abnormalities, pyramidal signs or dysphagia [3]. In France, the initial manifestations of type 2a were psychomotor regression (9 of 11) and seizures (3 of 11), and mean survival was 9.1 years (95% CI 4.5–13.5) [10]. GeneReviews gives a life expectancy of up to 5–10 years [1]. An Italian case with onset at 17 months progressed with blindness and spasticity in the third year, episodes of fever and hypertension at 5 years, renal failure and death at 6.5 years [4].

Juvenile (IIb). Onset is between 3 and 10 years, with ataxia, progressive loss of gross and fine motor function, dysarthria and dystonia; cognitive decline is not always present and vision is generally normal [1, 3]. Survival extends to the second decade or early adulthood in GeneReviews, and to the third decade in the NIH natural history review, which describes stagnation of motor and cognitive development followed by slow, continuous regression, with slurred speech and difficulty walking as the first symptoms [1, 44].

Minnesota cohort (juvenile and late infantile forms). In the prospective natural history study at the University of Minnesota, 13 patients with GM1 gangliosidosis (8 with the classic juvenile form and 5 with the late infantile form) were compared with 6 patients with juvenile or late infantile Tay-Sachs disease. All met the milestones of the first year; the first symptoms that led to the search for a diagnosis were gait and balance abnormalities, frequent falls and speech difficulties. The first symptom appeared between 18 and 54 months in the juvenile form and between 13 and 20 months in the late infantile form, and the median age at diagnosis was 81 and 18 months, respectively. Two of the five late infantile patients never walked unaided and none achieved sphincter control, which most of the juvenile patients achieved and then lost. In the juvenile form, the median age at loss of independent walking was 132 months and the median for loss of speech was not reached. Epilepsy was present in 3 of 4 late infantile patients assessed (median onset at 36 months) and in only 3 of 8 juvenile patients (onset between 3 and 11 years). No cherry-red spot was found in any of the patients with GM1 gangliosidosis examined, strabismus preceded the diagnosis in three patients in the study, and kyphosis, scoliosis and hip dysplasia were frequent in GM1 gangliosidosis and absent in Tay-Sachs disease. A feeding tube was placed in 2 of 8 juvenile patients (between 11.0 and 14.4 years) and in most late infantile patients (median 4.0 years). Median survival was 7.7 years in the late infantile form and was not reached in the juvenile form; of the deaths, five were from respiratory failure associated with aspiration pneumonia [45].

Prospective NIH cohorts. In a first study of six late infantile and nine juvenile patients, deterioration of expressive language and of gait was faster in the late infantile form; 14 scans of six late infantile patients showed progressive atrophy of the cerebrum and cerebellum, 26 scans of nine juvenile patients showed atrophy of more variable extent and progression, and quantitative spectroscopy showed a greater N-acetylaspartate deficit in the late infantile form, correlated with gait and expressive language [46]. In the 10-year study of 41 patients with type II, hearing was normal or near-normal and there was no cherry-red spot or hepatosplenomegaly; some older children with juvenile onset developed thickening of the aortic and/or mitral valves; cerebral atrophy was progressive and more marked in the late infantile form; on magnetic resonance spectroscopy, the increase in myo-inositol and the N-acetylaspartate deficit worsened and correlated with the Vineland scale, with faster progression in the late infantile form [11]. Volumetry of 56 scans from 24 patients showed global atrophy in both forms and more marked atrophy of the cerebellum, thalamus and corpus callosum in the late infantile form [47]. Brain age estimated by magnetic resonance increased by 12.25 years per chronological year in late infantile patients and by 1.57 in juvenile patients, against 0.83 in controls [48].

Skeleton in type II. In 13 late infantile and 21 juvenile patients at the NIH, the mean time between symptoms and diagnosis was 1.9 and 6.3 years. All late infantile patients had odontoid process hypoplasia and pear-shaped vertebrae, findings that were rare in juvenile patients (0% and 14%); juvenile patients had irregular vertebral end plates (15/21), central indentation of the end plates (10/21) and squared, flattened vertebrae (10/21). Bone mineral density was reduced (Z-score of −2.1 in the lumbar spine, −2.2 at the femoral neck and −1.8 at the total hip), with no fractures [49]. Avascular necrosis of the femoral heads, kyphosis, scoliosis, acetabular dysplasia and coxa valga have also been described [1].

Course. Neurological decline is accompanied by progressive dysphagia with a risk of aspiration; aspiration pneumonia may require ventilatory support, and one autopsy described a patient with long survival on ventilation [1]. The prospective PRONTO study (NCT05109793) enrolled, between February 2022 and September 2023, at 13 centres in six countries, 31 participants aged 2 to 20 years with genetically confirmed GM1 or GM2 gangliosidosis, neurological onset at 12 months or later and impaired gait or speech, followed for up to 24 months [31]. The 15 participants with GM1 (7 from France, 3 from Italy, 3 from the United States, 1 from Germany and 1 from the United Kingdom) were on average 12.3 years old, diagnosed at a mean age of 7.0 years, with a mean score of 26.5 on the Scale for the Assessment and Rating of Ataxia (SARA, from 0 to 40), 6.4 signs on the Inventory of Non-Ataxia Signs (INAS, from 0 to 16) and a history of epileptic seizures in 7 (46.7%), with a mean onset at 7.9 years [31]. In the combined GM1 and GM2 group, at 24 months the SARA increased by a mean of 2.6 points (standard deviation 5.6; n = 27), the functional SARA by 1.8 points (n = 28) and the INAS by 1.1 points (n = 22); enteral feeding rose from 13.8% (4/29) to 32.1% (9/28) and the Vineland-3 adaptive composite fell by 5.4 points (n = 14), mainly in daily living skills [31]. Progression was greater in GM2 on almost all instruments; in GM1, the INAS changed by only +0.2 points (n = 12), paresis was the non-ataxia sign that worsened most and there were no deaths in 24 months, whereas two participants with GM2 died [31]. Five of the 29 participants assessed with the SARA (three with GM1) already had the maximum score at baseline, which limits measurement of progression; excluding those patients, the mean increase in the SARA was 3.2 points (n = 22) [31]. Individual trajectories were heterogeneous, the seizure data in the electronic diary were incomplete and, with a descriptive design without imputation of missing data and funded by the sponsor of a drug under trial, the figures serve as a natural history reference and not as a population estimate [31].

3.4. Type III (chronic or adult)

Type III begins in late childhood, adolescence or adulthood. Generalised dystonia with gait instability and speech disturbance is the dominant manifestation, followed by extrapyramidal signs, including akinetic-rigid parkinsonism, which in the Japanese population appeared within a short time in 64% of patients and frequently lead to the mistaken diagnosis of Parkinson disease [44]; intellectual impairment is frequent in late stages and behavioural and psychiatric disturbances may appear [1, 3]. There is no hepatosplenomegaly or coarse facies, and skeletal changes (short stature, kyphosis, hip dysplasia, scoliosis) were found in 95% [1]. Saudubray describes an absence of ocular changes, whereas GeneReviews allows for corneal clouding in some patients [1, 3].

The series describe the variability. In 16 Japanese patients from 10 families, with onset between 3 and 30 years, the gait and speech abnormalities resulted from persistent hypertonia, with dystonic posturing, facial grimacing and parkinsonism [5]. Three Japanese siblings with generalised dystonia had symmetrical lesions confined to the putamen [5]. In a patient with dystonia from the age of 4, autopsy showed neuronal storage predominantly in the basal ganglia [5]. In India, eight patients with a median onset at 6 years (3–18) and presentation at 23 years had generalised dystonia with early oromandibular-craniocervical involvement, dysarthria in all, corticospinal signs in six, parkinsonism in two and frequent kyphoscoliosis [12]. In Giugliani's Brazilian series (10 patients, median age at diagnosis 11.5 years), dystonia and swallowing problems predominated; there was cognitive impairment and behavioural or psychiatric disturbance in all 9 patients assessed, dysarthria in 8 of 9, and none had parkinsonism or spinal cord compression [50]. In the Campinas series (4 juvenile and 8 adult patients), six patients began with skeletal deformities and six with neurological signs; all had short stature, dysostosis multiplex and dysarthria, 10 of 12 extrapyramidal signs, and all the women needed hip arthroplasty for severe dysplasia [32]. Prognosis depends on the degree of neurological impairment, and life is shorter than that of relatives [1]. An exaggerated auditory startle rarely occurs [1].

3.5. Mucopolysaccharidosis IVB and mixed phenotypes

Mucopolysaccharidosis IVB has axial and appendicular dysostosis multiplex, short stature, kyphoscoliosis, joint laxity, platyspondyly and odontoid hypoplasia, without central nervous system dysfunction, hepatosplenomegaly or coarse facies [1]. In a survey of 30 patients or carers, 84% had difficulty walking, 96% chronic pain, the mean was 3 operations per person (80% on the hip) and about half lived independently [51]. Saudubray notes that some patients with mucopolysaccharidosis IVB have slowly progressive central nervous system disease [52]. Smith's book considers forms A and B of Morquio syndrome clinically indistinguishable and recalls that other GLB1 variants cause generalised gangliosidosis type I [53].

4. Clinical Recognition and Warning Signs

4.1. When to suspect, by age

Fetus and newborn. Non-immune fetal hydrops, increased nuchal translucency and generalised oedema with coarse facies lead to GM1 gangliosidosis being included among the metabolic diseases to be sought; faced with generalised oedema of unexplained cause, the Oxford Desk Reference recommends requesting leucocyte enzymes [37, 54]. The combination of hydrops with a cherry-red spot points to GM1 gangliosidosis, galactosialidosis or sialidosis [38]. Macrocephaly and a cherry-red spot are other signs of the GM1 and GM2 gangliosidoses in the infant with white matter disease [55]. Coarse facies present at birth narrows the differential diagnosis to galactosialidosis, mucolipidosis II, GM1 gangliosidosis, sialidosis type 2 and mucopolysaccharidosis VII [38].

Infant. The signs that should prompt consideration of the disease are hypotonia with arrest of development between 3 and 6 months, oedema of the hands and feet or a turgid consistency of the limbs, coarse facies, macroglossia and gingival hypertrophy, hepatosplenomegaly, cherry-red spot, extensive dermal melanocytosis and vertebral beaking on radiographs [1, 38, 41]. Hepatosplenomegaly with signs of storage between 1 and 12 months leads to the differential diagnosis of GM1 gangliosidosis, mucolipidosis II, sialidosis II, Niemann-Pick A and Sandhoff disease [38]. In the German and Austrian series, the diagnostic delay was 6±2.6 months in the infantile form [35]. Persistently raised alkaline phosphatase, raised AST with normal ALT and increased chitotriosidase are simple biochemical clues [35, 41, 42].

Child. Motor and language regression from 1–3 years, with ataxia, dysarthria, dystonia or seizures, without visceromegaly or a cherry-red spot, should prompt consideration of type II [1, 11]. In the German and Austrian series, coarse facies, cherry-red spot and visceromegaly occurred only in the infantile form, 17 of 22 patients had dystonia and 14 dysphagia; the diagnostic delay was 2.6±3.79 years in the late infantile form and 14±3.48 years in the juvenile form [35]. Skeletal changes (odontoid hypoplasia, pear-shaped or flattened vertebrae) are early clues [49]. In the Minnesota cohort, the complaints that prompted the search for a diagnosis were mainly gait and speech abnormalities, and strabismus sometimes preceded the diagnosis [45]. In the child with ataxia, the Oxford Desk Reference includes leucocyte enzymes, which cover the gangliosidoses and galactosialidosis, among the second-line biochemical tests [56].

Adolescent and adult. Generalised dystonia with oromandibular involvement, dysarthria, dysphagia and kyphoscoliosis, especially with hip or vertebral dysplasia and short stature, and a scan with hyperintensity of the posterior putamen and pallidal hypointensity in a «wishbone» pattern on susceptibility-weighted imaging, point to type III [1, 12, 57]. Saudubray includes GM1 gangliosidosis among the metabolic diseases of the adult with dystonia and parkinsonism and with putaminal lesions [57], and the Oxford Desk Reference proposes leucocyte enzymes as a second-line test in dystonia [58].

4.2. Warning signs and urgent situations

The events requiring prompt action are aspiration and aspiration pneumonia, refractory seizures, heart failure from cardiomyopathy or valve disease, atlantoaxial instability or cervical cord compression (particularly before anaesthesia), hip dislocation in the patient who cannot communicate (unexplained crying, refusal to be moved) and unplanned anaesthesia [1, 49]. The presence of anti-GM1 antibodies in serum does not indicate GM1 gangliosidosis: it is associated with immune neuropathies such as multifocal motor neuropathy and Guillain-Barré syndrome [1].

4.3. Newborn screening

The sources consulted do not identify any national newborn screening programme for GM1 gangliosidosis. Enzyme activity measured by tandem mass spectrometry in dried blood spots from seven patients was far below that of random newborns, and the glycan biomarkers dp5 and A2G2 were markedly raised [1, 59]. The Atlas of Inherited Metabolic Diseases notes, in relation to mucopolysaccharidosis IV, that the activity of these lysosomal enzymes can be measured by mass spectrometry in blood spots and incorporated into newborn screening programmes, citing the application to mucopolysaccharidosis IVA [29]. The ScreenPlus project, in the United States, is evaluating newborn screening for several diseases under investigation [13].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnostic criterion

In a patient with suggestive findings, the diagnosis of a GLB1-related disorder is established by identification of biallelic pathogenic or likely pathogenic variants in GLB1 or by demonstration of significantly reduced β-galactosidase activity in peripheral blood leucocytes or in fibroblasts [1]. Either method suffices, but the two are often used together, the second to confirm the first or to explore the genotype-phenotype correlation [1, 26]. Identification of two variants of uncertain significance, or of one pathogenic variant and one of uncertain significance, neither establishes nor excludes the diagnosis [1]. Saudubray recommends that an enzymatic diagnosis be completed by identification of the variants and that a diagnosis made first through the gene be completed by a functional assay whenever possible [3].

5.2. Orientation tests

Urinary oligosaccharides. These are a good orientation test: in the early infantile form excretion is massive and has a pathognomonic profile (oligosaccharides with terminal galactose), and in the predominantly neurodegenerative forms it may be much lower [1, 3]. The characteristic profile can also be obtained in dried urine on paper by ultra-high-performance liquid chromatography coupled to tandem mass spectrometry (UHPLC-MS/MS) [1, 41]. In the German and Austrian series, oligosaccharides were abnormal in 10 of 19 patients [35].

Urinary glycosaminoglycans. In GM1 gangliosidosis there is a distinctive pattern that suggests but does not establish the diagnosis, keratan sulfate is usually increased, and glycosaminoglycan screening may be falsely negative [1, 3]. Urinary keratan sulfate may be diagnostic of mucopolysaccharidosis IV but does not distinguish IVA from IVB [26]. In Fiori's case, glycosaminoglycans were slightly increased in two samples and normal in a third [41], and in Menkovic's case keratan sulfate was markedly raised [42].

Blood biomarkers and cytology. Raised AST (13 of 20 patients) and increased chitotriosidase (12 of 15) in the Arash-Kaps series have been proposed as indicators for clinical screening [1, 35]. Persistently raised alkaline phosphatase has been described as an early sign in the infantile form [41, 42]. Vacuolated lymphocytes in peripheral blood and foamy histiocytes in the bone marrow support the suspicion [3, 60]. The oligosaccharide H3N2b, measured by liquid chromatography with tandem mass spectrometry, separated patients from controls with cut-offs of 6.2 ng/mL in plasma (sensitivity and specificity of 100%), 0.65 ng/μg of creatinine in urine (99.3% and 100%) and 5.1 ng/mL in cerebrospinal fluid (98.9% and 100%); it correlated with severity and decreased after gene therapy, but comparison with other lysosomal diseases was limited [61]. Quantification of GM1 in lymphocytes by flow cytometry correlated with clinical severity [62]. For the gene therapy trials, fluorimetric β-galactosidase assays in cerebrospinal fluid and serum were validated, with sensitivity limits of 0.05 and 0.20 nmol/mL/3 h and dependence on rapid processing and freezing of the samples [63]. These new biomarkers are research tools and do not replace enzymatic or genetic testing.

5.3. Enzyme testing

Acid β-galactosidase activity is measured in leucocytes or in dried blood spots with an artificial fluorogenic substrate (4-methylumbelliferyl-β-D-galactopyranoside) or with p-nitrophenyl-β-galactoside [3, 29]. Samples for leucocyte enzymes must reach the laboratory within a few hours, and the differential diagnosis should be discussed with the laboratory beforehand [64]. Reference values depend on the laboratory: in Fiori's case, activity was 11.2 nmol/h/mg with a normal range of 84–292 [41], and in one of Fischetto's patients 6.9 nM/mg/h with a normal range of 100–600 [30]. Faced with β-galactosidase deficiency, neuraminidase should be measured to exclude galactosialidosis [3]; in galactosialidosis β-galactosidase falls to about 15% and neuraminidase 1 to less than 1% of normal, so a partial β-galactosidase deficiency warrants measurement of cathepsin A and neuraminidase 1 in fibroblasts [1]. Enzyme activity may not predict carrier status [1], and pseudodeficiency due to p.R595W may lower activity in healthy individuals [6]. A reduction of β-galactosidase in skin and other tissues also occurs, secondary to the primary defect, in mucopolysaccharidoses II and I, so activity alone in one tissue does not suffice in the face of a mucopolysaccharidosis phenotype [65].

5.4. Molecular testing

The first option is sequencing of GLB1, followed by targeted deletion and duplication analysis when only one or no variant is found; a lysosomal disease panel is an equivalent alternative [1]. Exome or genome sequencing is appropriate when the phenotype cannot be distinguished from other neurodegenerative or lysosomal diseases [1]; in Menkovic's case, the diagnosis came from clinical exome sequencing [42]. Sequencing detects more than 99% of the recorded variants and deletions or duplications account for less than 1% [1]. A negative result does not exclude the disease: deep intronic variants may escape detection, as suggested by a patient with Morquio B in whom only one variant was found and a splicing defect excluding exons 2–7 was demonstrated [6]. In PRONTO, in one of the 14 genotyped patients with GM1 only a single heterozygous GLB1 variant was identified, and the diagnosis was confirmed by reduced β-galactosidase activity, which shows the need to integrate biochemical and molecular data [31]. Uniparental isodisomy may simulate homozygosity with only one carrier parent [36]. The report should state the gene, the transcript and its version (NM_000404.4), the HGVS nomenclature, the zygosity, the class according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) and the need for parental studies to confirm the phase; in the NIH study, reclassification of 37 variants by the ACMG criteria raised the class of 6 and led to the submission of 4 new ones [11].

5.5. Imaging

Skeleton. The skeletal survey shows dysostosis multiplex: J-shaped sella turcica, oar-shaped ribs, ovoid or hook-shaped vertebrae, constriction of the iliac bones, coxa valga, irregular modelling of the long bones, proximal pointing of metacarpals 2–5 and, in the infant with type I, periosteal cloaking [39]. The skeletal dysplasia tables of the Oxford Desk Reference include GLB1 in the group with valgus, contractures and periosteal cloaking [66]. Repeating the radiographs is of value: in Menkovic's case, the survey was normal at birth and abnormal at 3 months [42]. In type II, odontoid hypoplasia and pear-shaped vertebrae favour the late infantile form and squared, flattened vertebrae with central indentation the juvenile form [1, 49]. In type III there are beaked or flattened vertebrae, hip dysplasia, osteopenia, J-shaped sella turcica, scoliosis, kyphosis, pectus carinatum and odontoid hypoplasia [1, 50].

Brain magnetic resonance imaging. In type I there is increased ventricular volume, atrophy of the cerebellar white matter, the caudate, the putamen and the corpus callosum [1]. In a retrospective study of 24 patients from several centres, hypomyelination with preservation of the posterior limb of the internal capsule dominated in the infantile form; white matter changes were present in almost all infantile and late infantile patients and absent in juvenile and adult patients; T2 hypointensity of the thalamus, predominantly in the anterior and lateral nuclei, was constant in all subtypes; internal hypertrophy of the occipitomastoid sutures appeared in more than half of the early-onset cases; and the juvenile and adult forms had T2 hyperintensity of the dorsal putamen and hypointensity of the globus pallidus corresponding to mineralisation [67]. In type III, the combination of posterior putaminal hyperintensity and the «wishbone» pallidal pattern was constant in eight Indian patients [1, 12]. In a patient with the late infantile form illustrated in GeneReviews, atrophy was minimal at 3 years and 6 months and progressive, with extensive iron accumulation in the globus pallidus, at 6 years and 10 months [1]. T2 hypointensity of the basal ganglia is also reported in adult GM1 gangliosidosis [38].

5.6. Practical algorithm

5.7. Prenatal and preimplantation diagnosis

When the familial variants are known, prenatal diagnosis and preimplantation genetic testing are possible [1]. In a Russian family with a son affected by compound heterozygosity for NM_000404.4:c.699del p.(Gln234ArgfsTer20) (described as c.699delG) and c.809A>C p.(Tyr270Ser) [7, 68], preimplantation testing by direct analysis of the variants and of flanking microsatellite haplotypes in 8 biopsied embryos identified 4 carriers, 2 affected and 2 without variants; an embryo without variants was transferred, first-trimester prenatal diagnosis confirmed the result and an unaffected child was born [68]. In mucopolysaccharidosis IVB, diagnosis in chorionic villi and preimplantation diagnosis have been achieved [21]. Non-immune fetal hydrops should lead to GM1 gangliosidosis being included in fetal metabolic studies [1, 54].

6. Differential Diagnosis

6.1. Type I (infantile)

Source: GeneReviews, Table 4, supplemented by Saudubray and the Atlas [1, 20, 38]. AR, autosomal recessive; AD, autosomal dominant.

6.2. Type II (late infantile and juvenile)

Source: GeneReviews, Table 5, and Oxford Desk Reference [1, 57, 60].

6.3. Type III (chronic or adult)

Source: GeneReviews, Table 6, Saudubray (tables of metabolic diseases of the adult with dystonia and putaminal lesions) and Oxford Desk Reference [1, 57, 60]. XL, X-linked.

6.4. Main differential diagnoses

The differential diagnosis in the infant centres on the other lysosomal diseases with a cherry-red spot and with coarse facies. The GM2 gangliosidoses have no skeletal disease or coarse facies, and median survival with early onset was 44.0 months in RETRIEVE, against 19.0 months in GM1 gangliosidosis [1, 43]. Galactosialidosis reproduces the picture, because it shares the multienzyme complex, but has combined neuraminidase deficiency and sialylated oligosaccharides in urine; the early infantile form runs with hydrops, telangiectasia, heart failure, proteinuria and death at about 7 months, and the juvenile or adult form, frequent in Japan, with myoclonus, seizures, bilateral cherry-red spot and angiokeratomas; a CTSA variant (p.G86V) has been described in Portuguese patients [20, 69]. The mucopolysaccharidoses (I, II, VII) and the mucolipidoses share dysostosis multiplex and coarse facies, but have a different glycosaminoglycan profile and distinct enzymes [39, 70, 71]. On radiology in the infant, GM1 gangliosidosis type I and galactosialidosis are very similar to mucolipidosis II and are excluded only by enzyme studies; the cherry-red spot, rapid neurological degeneration and hepatomegaly are present in GM1 gangliosidosis and absent, or less marked, in mucolipidosis II, in which joint contractures are more severe [72]. Outside the lysosomal group, GM1 gangliosidosis is among the inborn errors that cause progressive encephalopathy and enter the differential diagnosis of mitochondrial diseases such as MELAS (mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes) [73]. In the child and the adult, what separates GM1 gangliosidosis from the ataxias, the dystonias and early-onset parkinsonism is the association with skeletal dysplasia, short stature and the magnetic resonance pattern [1, 12]. Anti-GM1 antibodies do not belong to this differential diagnosis [1].

7. Treatments and Risk Management Strategies

7.1. Framework and sources of the recommendations

As of 31 October 2026 there is no approved disease-modifying treatment for GM1 gangliosidosis in any jurisdiction, nor any clinical practice guideline from a scientific society devoted to this disease [1, 3, 26, 74]. The management recommendations available are those of GeneReviews, which result from expert consensus, supplemented by the data from the natural history cohorts [1, 26]. GeneReviews recommends that care be coordinated by specialists in biochemical genetics, cardiology, orthopaedics and neurology, with therapists familiar with these diseases, that early and continued intervention be aimed at comfort, mobility and participation at school and in society, and that surgery be performed in centres with surgeons and anaesthetists experienced in lysosomal diseases [1, 26]. Palliative care is a central part of the plan in the infantile and late infantile forms [69].

7.2. Assessments after diagnosis

Source: GeneReviews, Tables 7, 8 and 9 [1].

7.3. Symptomatic treatment by problem

Seizures. Antiepileptic drugs are chosen and adjusted by an experienced neurologist according to the general rules; in the infantile forms seizures are often progressive and refractory, no antiepileptic has shown specific efficacy, complete control is rare and excessive sedation should be avoided [1]. An EEG is warranted in the face of acute change in mental state, sudden decline or abnormal movements [1].

Tone, dystonia and mobility. Physiotherapy, occupational therapy, stretching, splints and positioning devices; for hypertonia and dystonia, GeneReviews mentions baclofen, tizanidine, botulinum toxin, antiparkinsonian drugs and orthopaedic procedures, without doses specific to this disease [1]. In the Indian type III series, symptomatic treatments had a modest and variable benefit [12]. In a 24-year-old woman with type III and progressive generalised dystonia refractory to drugs, bilateral deep brain stimulation of the internal globus pallidus improved the dystonia by 20% in a double-masked comparison with and without stimulation after 1 year, with functional benefit but without altering disease progression; this is an isolated report [44, 75]. Augmentative and alternative communication and speech therapy, with a reduced rate of speech, help with dysarthria [1].

Feeding and nutrition. Feeding therapy, thickening or cooling of food and a nasogastric tube or gastrostomy maintain hydration and calories; gastrostomy increases longevity but does not preserve development, and the threshold for a swallowing study should be low [1]. In PRONTO, the proportion of children with type II or GM2 on enteral feeding rose from 13.8% to 32.1% over 24 months [31]. Constipation is treated with softeners, prokinetics, osmotic laxatives and stimulants [1].

Secretions and aspiration. Gastrostomy, a vibrating vest and pulmonary hygiene reduce the risk; sialorrhoea may be treated with topical anticholinergics or botulinum toxin in the salivary glands, bearing in mind that the toxin may diffuse into the bulbar muscles and worsen dysphagia [1]. In the prospective Minnesota study, chest physiotherapy was associated with longer survival in the infantile form (p=0.0056), an observational association without randomisation [40].

Heart. Cardiac care follows the usual lines for cardiomyopathy and valve disease [1]. Fiori's case illustrates the use of furosemide and captopril in mitral regurgitation due to a dysplastic valve [41].

Eyes. Keratoplasty is an option in severe corneal clouding with visual impairment (atypical in type II), with temporary results [1].

Skeleton. Orthopaedic treatment follows assessment of new abnormalities and of neurological signs of compression. Low bone mineral density and the frequent odontoid hypoplasia, particularly in the juvenile and late infantile forms, influence surgical decisions [1, 49]. Severe hip dysplasia led to arthroplasty in all the women in the Campinas series and in a North American patient at the age of 33 [5, 32].

Mental health (type III). Cognitive behavioural therapy is the preferred approach; tricyclic antidepressants, antipsychotics and electroconvulsive therapy should be avoided when possible, as they may worsen the neurological disease [1]. Vocational counselling and an individual education plan complete the approach [1].

Family. Home nursing, carer respite and social support are part of the plan [1]. In PRONTO, the short version of the Burden Scale for Family Caregivers (BSFC-s, from 0 to 30) had at baseline a mean of 14.2 among carers of children with GM1 (n = 10) and of 19.3 in GM2 (n = 7), values indicating high burden according to the authors, with small changes over 24 months; carers reported effects on physical strength, health and finances [31].

7.4. Haematopoietic cell transplantation

In a presymptomatic patient with juvenile GM1 gangliosidosis, bone marrow engraftment normalised leucocyte β-galactosidase activity but did not improve the long-term clinical outcome [76]. In the canine model, early allogeneic transplantation also brought no benefit despite engraftment [6]. The Atlas reports, for the mucopolysaccharidoses in general, limited success and an unfavourable neurological outcome of cord blood transplantation [70]. With these data, transplantation is not a treatment for GM1 gangliosidosis outside research.

7.5. Miglustat (substrate reduction therapy; off-label use)

Mechanism and rationale. Miglustat inhibits glucosylceramide synthase, the first enzyme of glycosphingolipid synthesis, and reduces the production of gangliosides upstream of the block [30, 77]. In patients' cells, miglustat reduced the GM1 content [62].

Evidence in GM1 gangliosidosis. The data are case series without a control group. Deodato and colleagues treated three Italian patients (two juvenile with activity ≤5% and one adult with about 7%) with 600 mg per day from the ages of 10, 17 and 28 years; all improved gradually, the juvenile patients walked a few metres unaided again, the MD-CRS score of one patient went from 4 to 2 and the distance on the 6-minute walk test of the adult from 338 to 475 m [16]. Fischetto and colleagues treated four children with type 2 (mean age 55 months) in three Italian hospitals: the drug was safe and relatively well tolerated, with stabilisation or slowing of neurological progression in three [30]. In the Minnesota study, the Syner-G regimen (miglustat with a ketogenic, low-carbohydrate diet) in tube-fed infants was associated with longer survival in the infantile form (p=0.025) [40], but the Syner-G trial (NCT02030015, 16 children with GM1 or GM2) was terminated with the reason «efficacy», with no participants completing it and with 13 deaths, and the outcomes were not analysed [13]. In the juvenile and late infantile cohort from the same centre, 7 of 19 children with GM1 or GM2 gangliosidoses received miglustat, dosed by body surface area according to the paediatric regimen for NPC, combined with a ketogenic or low-carbohydrate diet; the diet was intended to reduce the gastrointestinal effects of inhibition of the intestinal disaccharidases and, possibly, to improve seizure control. There were slight, transient improvements in some neurodevelopmental scores, without significant overall clinical change, in a sample too small for conclusions [45]. In GM2 gangliosidosis, miglustat did not improve neurological impairment in the infantile, juvenile or late-onset forms, although some patients had temporary stabilisation [44]. Saudubray and GeneReviews refer to Syner-G as a trial and not as an established treatment [1, 3].

Doses. There is no guideline dose for GM1 gangliosidosis. The published series used the dosage for Niemann-Pick type C disease (NPC), an analogous condition treated with the same drug, adjusted to body surface area [30]; the dose below is therefore an extrapolation, outside the label indications.

Adverse effects and monitoring. Diarrhoea and other gastrointestinal effects occur in more than 80% of patients, through inhibition of the intestinal disaccharidases, and respond to reduction of sucrose and lactose, taking the drug between meals, loperamide or temporary dose reduction; tremor occurred in 58% in the NPC trial; cases of Crohn disease, peripheral neuropathy (baseline and repeated neurological assessment), a slight fall in platelets and reduced growth in children, which should be monitored, have been described; in men, contraception should be maintained during treatment and for the following 3 months, because the drug affects spermatogenesis in the rat; the SmPC recommends reassessing the benefit in NPC every 6 months and deciding on continuation after at least 1 year [78]. In the Fischetto series there was occasional diarrhoea and abdominal pain, severe hyporexia that required the drug to be temporarily stopped in one child and weight loss from 42.6 to 38.9 kg in the first 4 months in an adolescent [30]. In the Minnesota study, gastrointestinal effects limited its use [40].

Regulatory status.

7.6. Gene therapy

Phase 1–2 trial with intravenous AAV9-GLB1 (NHGRI). This is the only trial of a disease-modifying treatment with results published in a peer-reviewed journal. Between August 2019 and October 2022, nine children with type II (five boys, 2.7 to 7.3 years; four late infantile in the low-dose group) received a single intravenous infusion of AAV9 carrying the GLB1 complementary DNA, at a low dose (1.5×10^13 vector genomes/kg, five participants) or a high dose (4.5×10^13 vg/kg, four participants), with immunosuppression by rapamycin (sirolimus) and rituximab started 3 weeks beforehand, methylprednisolone 1–2 hours before and oral prednisone afterwards [14]. Exclusion criteria were a Vineland-3 composite score below 40 and anti-AAV9 antibodies above 1:50 [14]. Over 3 years there were 124 adverse events, 30 considered related (8 gastrointestinal, 21 laboratory changes of an inflammatory nature and 1 tachycardia), and 5 serious adverse events, among them vomiting with hospitalisation attributed to the treatment; AST and ALT rose in all and returned to baseline at a median of 18 months; all developed anti-AAV9 neutralising antibodies within the following 2 weeks [14]. In all, β-galactosidase in cerebrospinal fluid rose and GM1 fell. Expressive communication and gross motor function remained stable, while fine motor function and receptive communication declined; the clinical global impression of improvement (CGI-I, from 1 to 7) had a median of 3 (minimal improvement) at 2 years and of 4 (no change) at 3 years, whereas in historical controls this score increases; magnetic resonance suggested a lower rate of atrophy and favourable changes in myelination [14]. With nine participants, no concurrent control group and comparison with historical data, the efficacy results are preliminary. H3N2b decreased in plasma, urine and cerebrospinal fluid of the treated patients, in parallel with enzyme activity [61]. The registration also provides for type I cohorts (1.5×10^13 and 4.5×10^13 vg/kg) and a dose of 7.5×10^13 vg/kg in type II, and recruitment continues, with primary completion estimated in 2030 [13].

Expanded access (preprint, September 2026). The NHGRI group described, in a manuscript not yet peer-reviewed, a girl with the juvenile form, between the ages of 6 and 10 years, who before treatment had developed seizures, dysarthria and loss of walking during the natural history study. She received a single intravenous infusion of AAV9-GLB1 at 1.5×10^13 vg/kg under expanded access, with 5 years of safety follow-up and 3 of efficacy. The authors report resolution of dysphagia, more interaction, walking with support, gains in domains of adaptive functioning and absence of seizures without antiepileptics from 18 months; in cerebrospinal fluid, GM1 and H3N2b fell by 50% and β-galactosidase reached normal activity; brain and thalamic volume increased and ventriculomegaly improved at 3 years [83]. This is a single, open-label case without peer review.

Other programmes.

GeneReviews and Saudubray mention these trials as planned or ongoing [1, 3]. For the trials with administration into the central nervous system there were no results published in a journal up to the reference date in the sources consulted.

Imagine-1 (PBGM01): interim data released by the sponsor. The results of Imagine-1 were communicated by Passage Bio in press releases and presentations filed with the United States Securities and Exchange Commission (SEC); they have not been published in a peer-reviewed journal, and the figures that follow have that status. In February 2023, with six patients from cohorts 1 to 3 and a data cut-off in December 2022, the sponsor described stabilisation of a magnetic resonance severity score (cerebral and cerebellar atrophy, white matter changes and signal changes in the basal ganglia and hippocampi) in all treated patients, over 6 to 12 months, and a fall in the urinary substrate Dp5, an exploratory biomarker of peripheral β-galactosidase activity, in the two patients with the late infantile form who received the higher dose and in the patients with raised baseline values treated with the low dose [88]. In August 2023, the sponsor presented the data from cohorts 1 to 4: eight children, four with early onset (symptoms from birth to 5 months) and four with late onset (12 to 14 months), aged 6 to 31 months at enrolment, treated with dose 1 (3.3×10^10 genome copies/g of estimated brain weight) or dose 2 (1.1×10^11 genome copies/g), with low-dose corticosteroid therapy for 4 weeks followed by tapering; with follow-up of 8 to 28 months and a data cut-off on 26 June 2023, there were no treatment-related serious adverse events, the related events were mild to moderate, no patient required intervention for altered liver function or adjustment of immunosuppression, nerve conduction studies showed no dorsal root ganglion toxicity and the intracisternal injection had no complications [84]. With dose 1, β-galactosidase activity in cerebrospinal fluid increased 1.5- to 4.8-fold by day 30, without reaching the mean level of the healthy adult, and GM1 ganglioside in cerebrospinal fluid did not decrease; in one patient with a modest enzyme increase, the rise in GM1 accompanied clinical worsening. With dose 2, activity increased 4.7- to 16.1-fold by day 30 in 3 of 4 children, above the mean of healthy adults, with the effect maintained up to 12 months, and GM1 in cerebrospinal fluid fell to the normal adult level after one year [84]. For survival, the sponsor compared the six treated patients with the infantile form with Lang's meta-analysis (154 infants with onset before 12 months, mean survival 18.9 months, no survivors beyond 35 months) and reported that the three patients who had passed 20 months of age were alive [24, 84]; this is a comparison with historical controls, with three patients at risk at that point and without statistical analysis. The first patient at dose 3 (2.2×10^11 genome copies/g) was treated in July 2023, and the amended protocol lowered the age of enrolment for the early-onset forms to 1 to 12 months and the upper limit for the late-onset forms from 36 to 24 months, additionally requiring the ability to sit unsupported [84]. In December 2023, the company suspended enrolment of new patients and stopped clinical development to reduce expenditure; in the same annual report it states that the application for authorisation of the trial to the FDA was initially under clinical hold, later lifted, and that preliminary toxicology results in non-human primates showed toxicity of the trigeminal and dorsal root ganglia [85]. The programme was licensed to GEMMA Biotherapeutics by an agreement of 31 July 2024 [86, 87]. A 2026 biomarker review reports that the patients treated in this programme had normalisation of lysosphingolipids in cerebrospinal fluid with early signs of clinical benefit [89]; in the sources consulted, the Imagine-1 results remain unpublished in a peer-reviewed journal.

7.7. Small molecules in development

Nizubaglustat (AZ-3102). This is an oral azasugar that crosses the blood-brain barrier, developed by Azafaros [15]. The NAVIGATE programme includes a randomised (2:1), double-masked, placebo-controlled phase 3 trial of 18 months, in patients with GM1 or GM2 gangliosidosis (Tay-Sachs, Sandhoff, AB variant) with neurological onset between 1 and 10 years, age 4 years or older and a SARA between 3 and 30; the primary endpoints are the change in total SARA and in functional SARA at 18 months, and the participating countries include Portugal and Brazil (NCT07082543) [13]. On 23 July 2026, the company announced complete recruitment, with at least 75 patients at 25 centres in 13 countries, and anticipated main results in early 2028 [15]. According to the same announcement, the drug has rare paediatric disease, orphan drug and Fast Track designations from the FDA, an EMA orphan designation for the GM1 and GM2 gangliosidoses and an Innovation Passport from the UK agency [15]; these designations are not equivalent to a marketing authorisation. A long-term open-label extension (NCT07885917) had not yet started recruiting [13]. According to the discussion in PRONTO, the early clinical studies of nizubaglustat (ACTRN12621000576820 and NCT05758922) showed favourable safety and tolerability, reduction of plasma glucosylceramide and signs of brain penetration with pharmacodynamic activity in cerebrospinal fluid [31]. PRONTO, funded by Azafaros, was designed to provide natural history cohorts usable as an external control in trials, and the SARA and functional SARA, the primary endpoints of NAVIGATE, were also its main endpoints [31].

Venglustat. The phase 3 AMETHIST trial (NCT04221451), in late-onset GM2, with a secondary group that admitted patients with GM1 gangliosidosis, was terminated in 2024 for «absence of positive trends in the clinical endpoints» [13, 23]. This is a negative result of relevance to the substrate reduction strategy.

Preclinical research. Between 2025 and 2026, studies in animal models were published with sinbaglustat in the mouse [90] and with an AAV vector expressing an enzyme able to cross the blood-brain barrier [91]. Gene therapy with autologous haematopoietic stem cells transduced by a lentiviral vector restored enzyme activity, reduced brain storage, attenuated the neuromuscular phenotype and increased survival in the murine model [92], and allosteric modulators of β-galactosidase reduced GM1 accumulation in fibroblasts from a canine model, more than the competitive chaperone N-nonyl-deoxygalactonojirimycin under the conditions tested, as yet without in vivo studies [93]. Pharmacological chaperones had been studied earlier: NOEV increased brain enzyme activity in a murine model with the R201C variant [6], 4-epi-isofagomine derivatives increased the activity of the mutant enzyme in patients' fibroblasts [29] and a deoxygalactonojirimycin derivative (NB-DGJ) promoted maturation and activity of the p.D151Y enzyme in fibroblasts from a child from the Emirates with type I [2]. None of these results has demonstrated clinical translation.

7.8. Trials: limits of interpretation

The gene therapy results come from open-label, single-arm trials compared with historical controls, whose trajectory depends on the definition of subtype and on the era [14, 43]. RETRIEVE showed that survival in the GM1 and GM2 gangliosidoses and in early-onset Gaucher disease type 2 differs enough to advise against grouping them in trials [43], and the same heterogeneity prevents indirect comparisons between trials with different populations and routes of administration. In PRONTO, 17.2% of participants had the maximum SARA score at entry, and the authors recommend that this ceiling effect be taken into account in trial inclusion criteria and statistical plans [31].

7.9. Agents and circumstances to avoid

Unplanned anaesthesia should be avoided, particularly in patients with skeletal involvement (vertebral anomalies, short neck, large head, atlantoaxial instability from odontoid hypoplasia); psychotropic drugs that may worsen the neurological disease, by analogy with the worsening described in adults with Tay-Sachs; in type I, positions that increase the risk of aspiration and antiepileptic doses that cause excessive sedation; and, in types II and III, circumstances that increase the risk of falling [1].

7.10. Anaesthesia and perioperative care

All GLB1-related disorders may carry anaesthetic risks [1]. Preoperative assessment includes previous anaesthetics, the airway, the heart and respiratory function; fibreoptic bronchoscopy and smaller endotracheal tubes may be needed; cervical radiographs in flexion and extension should be reviewed before anaesthesia, with cervical stabilisation when there is instability, and the procedure should be conducted by an experienced paediatric anaesthetist [1, 49]. In Fiori's case, magnetic resonance showed thickening of the soft tissues of the oropharynx and nasopharynx with moderate narrowing, a finding of relevance to the airway [41].

7.11. Pregnancy, fertility and transition

There are no studies of fertility in the mild forms [1], and the sources consulted do not describe pregnancies in women with GM1 gangliosidosis type III. The offspring of a patient are obligate carriers [1]. Miglustat requires male contraception during treatment and for the following 3 months [78]. Transition of juvenile and adult patients to adult care should maintain the multidisciplinary team (neurology, orthopaedics, cardiology, rehabilitation and psychiatry) and surveillance of anaesthetic risk and of the hips [1, 32].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks

Inheritance is autosomal recessive. The parents of a patient are, as a rule, obligate heterozygotes, and their molecular study is recommended to confirm that the two variants are in trans and to define the recurrence risk [1]. If only one parent is a carrier, the possibilities are a de novo variant in the proband, post-zygotic mosaicism in a parent or uniparental isodisomy [1]; the case of maternal uniparental isodisomy of chromosome 3 described by King and colleagues showed that, in that scenario, the recurrence risk is much lower than 25% [36]. Heterozygous carriers are asymptomatic and have no increased risk of disease [1].

Patients with severe forms have no offspring [1].

8.2. Carrier and family testing

Carrier status is determined by molecular testing when the familial variants are known; enzyme activity may not predict carrier status [1], and pseudodeficiency alleles such as p.R595W may lower activity without disease [6]. Testing the paediatric-age siblings of a patient with a late-onset form allows presymptomatic diagnosis, with benefit for skeletal and anaesthetic surveillance; the gene therapy trials admitted presymptomatic patients with confirmed type I variants [13].

8.3. Prenatal diagnosis and preimplantation genetic testing

With the familial variants identified, prenatal diagnosis in chorionic villi or amniocytes and preimplantation genetic testing are possible [1, 21]. Zagaynova's report documents preimplantation testing with direct analysis of the variants and of microsatellite haplotypes, confirmed by first-trimester prenatal diagnosis [68]. Counselling before pregnancy is the appropriate time to discuss these options [1]. A fetal gene therapy trial is registered but has not yet started recruiting [13].

8.4. Content of counselling

Counselling explains that the classification into types is approximate and that residual activity and genotype give guidance, but do not accurately predict the course in an individual patient [1, 23]. It should include the progressive nature of the disease, the absence of approved treatment, the ongoing trials and their inclusion criteria, the resources supporting families and the planning of palliative care in the infantile forms [1, 13, 69].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Global frequency and what each figure measures

GeneReviews estimates the prevalence of GM1 gangliosidosis of all types at 1:100,000 to 1:300,000, from a reference review, and states that the infantile form is the most common [1]. GeneGuidance stores 1.0 per 100,000, which corresponds to the upper limit of that range and does not include mucopolysaccharidosis IVB [26]. Orphanet records a validated birth prevalence of 0.75 per 100,000 in Europe and of 0.34 per 100,000 in Sweden, and considers the worldwide point prevalence unknown [9]. In France, the incidence estimated from all cases diagnosed between 1998 and 2019 was 1/210,000 [10]. Caciotti and colleagues cite a worldwide incidence of about 1:100,000 to 1:200,000 live births for GM1 gangliosidosis and mucopolysaccharidosis IVB together [34]. Mucopolysaccharidosis IVB has a prevalence of 1:250,000 to 1:1,000,000 and accounts for less than 5% of mucopolysaccharidosis IV phenotypes [1]; the cardiogenetics textbook estimates the incidence of mucopolysaccharidoses IV A and B together at 1:200,000–1:300,000 live births [94]. More than 200 patients with GM1 gangliosidosis and about 60 with mucopolysaccharidosis IVB had been described up to 2021 [1]. These figures mix birth prevalence, incidence in hospital series and estimates from reviews, and are not directly comparable.

9.2. Populations with a founder effect

9.3. United States and Europe

The prospective data from the United States come mainly from the NIH natural history study (15 patients with type II in the first imaging study and 41 followed for up to 10 years in the more recent study) and from the Minnesota programme (8 infants with the infantile form and 13 children with the late infantile and juvenile forms) [11, 40, 45, 46]. In Europe, the published cohorts include the French one of 61 patients [10], the German and Austrian one of 22 [35] and the Italian genetic and miglustat treatment series [16, 30, 34]. RETRIEVE and PRONTO brought together centres from several European countries, from the United States and from Brazil [31, 43].

9.4. Brazil and Portugal

In Brazil, GM1 gangliosidosis was, in Giugliani's 1985 series, the most frequently diagnosed inborn error of metabolism in the Paediatrics Department in Porto Alegre [2]. The estimated frequency is 1:17,000 [1], and the duplication c.1577dup is frequent in the infantile form [1]. The series from Campinas and Minas Gerais and the multicentre series from Porto Alegre describe the juvenile and adult forms, with a predominance of dystonia, dysphagia and skeletal dysplasia [32, 50]. Brazil took part in RETRIEVE and PRONTO and takes part in the PBGM01 and nizubaglustat trials [13, 31, 43]. In Portugal, the sources consulted provide no frequency data for GM1 gangliosidosis; the Atlas describes a CTSA variant in Portuguese patients with galactosialidosis, a differential diagnosis [20], and Saudubray notes the B1 variant of Tay-Sachs disease with a high incidence in northern Portugal [3]. Portugal is among the countries in the phase 3 trial with nizubaglustat [13]. The absence of Portuguese data in the sources consulted does not mean an absence of cases.

9.5. Special populations

In adults with type III, diagnosis is often late and confused with dystonias or parkinsonism of another cause [12, 35]. Children with type II have normal or near-normal hearing, which distinguishes them from the infantile forms [11]. Parental consanguinity was frequent in the Indian series and is documented in cases from the United Arab Emirates and Italy [2, 6, 41].

9.6. Reference networks and patient associations

The relevant European reference network is MetabERN, for hereditary metabolic diseases (https://metab.ern-net.eu/) [17]. The patient associations cited by GeneReviews include the Cure GM1 Foundation (https://www.curegm1.org), the National Tay-Sachs & Allied Diseases Association (https://www.ntsad.org), Metabolic Support UK, the MPS Society in the United Kingdom, the National MPS Society in the United States and the Canadian MPS Society; MedlinePlus has an information page on the disease [1].

10. Illustrative Clinical Case

Fiori and colleagues, of the Vittore Buzzi Children's Hospital in Milan, described in 2026 an early diagnosis of infantile GM1 gangliosidosis [41]. The patient was a girl born at term from a dichorionic twin pregnancy, the fourth child of first-cousin parents, with normal prenatal ultrasound scans and a birth weight on the 5th centile; her twin brother and two older siblings were healthy. At 14 days she was taken to the emergency department for mild oedema of the hands and feet; the echocardiogram showed left atrial dilatation with mitral regurgitation, and there was hypoalbuminaemia that normalised within a few days. At 30 days the mitral regurgitation had worsened, with overload of the left chambers, and furosemide was started (0.5 mg/kg/day, then 1.5 mg/kg/day because the oedema returned).

At 3 months, by then at the reference centre, she was in good general condition, with adequate weight gain, a normal neurological examination, no facial dysmorphism and no organomegaly on ultrasound; palpation revealed a turgid consistency of the forearms, legs and thighs. Alkaline phosphatase was persistently raised (1640, 1590 and 1744 U/L; reference 130–518) with the rest of the bone metabolism work-up and the transaminases normal, and mild hypoalbuminaemia persisted with no renal cause. The combination of turgid limbs and hyperphosphatasaemia led the authors to suspect a lysosomal disease. Fundoscopy showed a bilateral cherry-red spot and optical coherence tomography showed hyperreflectivity of the ganglion cell layer (Figure 1, https://pmc.ncbi.nlm.nih.gov/articles/PMC12963784/figure/jmd270075-fig-0001/); the radiograph showed beaking of the second lumbar vertebra (Figure 2, https://pmc.ncbi.nlm.nih.gov/articles/PMC12963784/figure/jmd270075-fig-0002/); brain magnetic resonance revealed mild widening of the frontal cerebrospinal fluid spaces and thickening of the soft tissues of the oropharynx and nasopharynx with moderate narrowing (Figure 3, https://pmc.ncbi.nlm.nih.gov/articles/PMC12963784/figure/jmd270075-fig-0003/). The echocardiogram showed a dysplastic, thickened mitral valve, and the child started captopril (0.5 mg/kg/day). Auditory evoked potentials were normal.

On biochemical testing, urinary glycosaminoglycans were slightly raised in two samples and normal in a third, chitotriosidase was normal, the oligosaccharide profile in a dried blood spot was consistent with GM1 gangliosidosis and leucocyte β-galactosidase was 11.2 nmol/h/mg (reference 84–292). Sequencing of GLB1 showed homozygosity for NM_000404.4:c.176G>A p.(Arg59His), classified as pathogenic, and the diagnosis of infantile GM1 gangliosidosis was made at 3 months. From 6 months there was progressive neurological deterioration with rapid regression of motor milestones and, at 8 months, dysphagia; the child was 10 months old at the date of publication [41].

The case shows that the infantile form may begin with non-neurological signs (oedema, mitral valve disease, hyperphosphatasaemia) and that neurological regression appeared within the interval described in the cohorts, between 6 and 18 months [24]. The variant p.(Arg59His) is the one that GeneReviews associates with Roma populations and with Brazil [1]. The cardiac drugs used are individual symptomatic treatment and not a protocol for the disease. The figures were consulted from the published legends.

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis by type

These data describe the disease without disease-modifying treatment. The differences between sources arise from distinct subtype definitions (onset before 6, 12 or 24 months) and from different eras of supportive care [10, 24, 43]. In the infantile form, gastrostomy prolongs life but does not preserve development [1], and chest physiotherapy was associated with longer survival in an observational study [40]. In the late infantile form, cerebral atrophy and functional loss progress faster than in the juvenile form [11, 45–48].

11.2. Surveillance

GeneReviews summarises annual surveillance of quality of life (neurologist, physiotherapist and dietitian), of seizure risk, of cervical stability and of the risk of hip dislocation, and surveillance every 1–3 years with ECG, echocardiogram and ophthalmological examination [1]. Surveillance in mucopolysaccharidosis IVB follows that of Morquio syndrome. Low bone mineral density in type II, with no fractures described, warrants taking bone fragility into account in surgical decisions [1, 49].

12. Summary and Final Considerations

GM1 gangliosidosis is an autosomal recessive lysosomal disease caused by biallelic variants of GLB1, which encodes acid β-galactosidase. The lack of the enzyme causes accumulation of GM1 ganglioside in neurones and of oligosaccharides and keratan sulfate in the viscera and skeleton, and the same enzyme, when it mainly loses the ability to degrade keratan sulfate, causes mucopolysaccharidosis IVB. The clinical spectrum is continuous and its division into types uses cut-offs that vary between sources. Type I begins before 12 months, with hypotonia, oedema, coarse facies, hepatosplenomegaly, cherry-red spot, dysostosis multiplex and cardiomyopathy, and median survival is around 19 months. Type II, late infantile or juvenile, runs with motor and language regression, ataxia, dystonia and skeletal dysplasia, without visceromegaly or a cherry-red spot, and with survival ranging from the first decade, in the late infantile form, to young adulthood, in the juvenile form. Type III presents with generalised dystonia with oromandibular involvement, dysarthria, dysphagia and vertebral and hip dysplasia, with a distinctive magnetic resonance pattern in the putamen and globus pallidus.

Diagnosis rests on β-galactosidase activity in leucocytes or fibroblasts and on sequencing of GLB1, after orientation tests such as urinary oligosaccharides; exclusion of galactosialidosis by measuring neuraminidase, interpretation of pseudodeficiency and normalisation of the variants against transcript NM_000404.4 are part of the process. Raised alkaline phosphatase and AST, chitotriosidase and H3N2b are clues or biomarkers under evaluation. There is no approved disease-modifying treatment and no scientific society guideline; care is symptomatic and multidisciplinary, with attention to swallowing, seizures, the heart, the cervical spine, the hips and anaesthesia. Miglustat is used off-label, with doses extrapolated from Niemann-Pick type C disease and evidence limited to case series. Intravenous gene therapy with AAV9-GLB1 showed, in nine children with type II, biochemical and imaging effects and signs of stabilisation with a safety profile that requires immunosuppression and hepatic monitoring, and a phase 3 trial with nizubaglustat will have results in 2028. The epidemiology shows populations with a founder effect, among them Malta, the Roma communities, Brazil and Japan, which warrant earlier clinical suspicion in these settings.

13. References

1. Regier DS, Tifft CJ, Rothermel CE. GLB1-Related Disorders. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Initially published 17 Oct 2013; last updated 22 Apr 2021. https://www.ncbi.nlm.nih.gov/books/NBK164500/
2. OMIM. #230500 GM1-gangliosidosis, type I; GM1G1. Johns Hopkins University. https://omim.org/entry/230500
3. Vanier MT, Caillaud C, Levade T. Disorders of sphingolipid synthesis, sphingolipidoses, Niemann-Pick disease type C and neuronal ceroid lipofuscinoses. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 40, section 40.2 Disorders of Lysosomal Sphingolipid Degradation — Sphingolipidoses (40.2.3 GM1 Gangliosidosis; 40.2.4 GM2 Gangliosidoses). doi:10.1007/978-3-662-63123-2_40
4. OMIM. #230600 GM1-gangliosidosis, type II; GM1G2. https://omim.org/entry/230600
5. OMIM. #230650 GM1-gangliosidosis, type III; GM1G3. https://omim.org/entry/230650
6. OMIM. *611458 Galactosidase, beta-1; GLB1. https://omim.org/entry/611458
7. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-8. doi:10.1002/humu.23348. PMID: 28967166. Accessed on 31 Oct 2026 (GRCh38, MANE NM_000404.4). https://variantvalidator.org
8. OMIM. #253010 Mucopolysaccharidosis, type IVB (Morquio). https://omim.org/entry/253010
9. Orphanet. GM1 gangliosidosis (ORPHA:354). https://www.orpha.net/en/disease/detail/354
10. Laur D, Pichard S, Bekri S, Caillaud C, Froissart R, Levade T, et al. Natural history of GM1 gangliosidosis-Retrospective cohort study of 61 French patients from 1998 to 2019. J Inherit Metab Dis. 2023;46(5):972-981. doi:10.1002/jimd.12646. PMID: 37381921.
11. D'Souza P, Farmer C, Johnston JM, Han ST, Adams D, Hartman AL, et al. GM1 gangliosidosis type II: Results of a 10-year prospective study. Genet Med. 2024;26(7):101144. doi:10.1016/j.gim.2024.101144. PMID: 38641994; PMC11348282.
12. Roy S, Arora C, Holla VV, Prasad S, Phulpagar P, Kamble N, et al. Clinical, Radiological, and Genetic Profiles of Eight Patients with Combined Dystonic Manifestation of Type-III GM1 Gangliosidosis: A Video Case Series from India. Tremor Other Hyperkinet Mov (N Y). 2026;16:9. doi:10.5334/tohm.1152. PMID: 41694796; PMC12904113.
13. ClinicalTrials.gov. Search «GM1 gangliosidosis» and registrations NCT03952637, NCT04713475, NCT04273269, NCT02030015, NCT07082543, NCT07054515, NCT07885917, NCT07479953, NCT04221451. Bethesda: National Library of Medicine. https://clinicaltrials.gov/
14. Lewis CJ, D'Souza P, Johnston JM, Acosta MT, Farmer C, Baker EH, et al. AAV9 Gene Therapy in Type II GM1 Gangliosidosis - A Phase 1-2 Trial. N Engl J Med. 2026;394(12):1184-1194. doi:10.1056/NEJMoa2510935. PMID: 41665410; PMC13215599.
15. Azafaros. Azafaros completes enrollment of Phase 3 (NAVIGATE) study evaluating nizubaglustat for the treatment of patients with GM1/GM2 gangliosidoses [press release]. Leiden; 23 Jul 2026. https://forbion.com/media/0iidtams/final-azafaros-pr_ph3-gmx-recruitment-completed.pdf
16. Deodato F, Procopio E, Rampazzo A, Taurisano R, Donati MA, Dionisi-Vici C, et al. The treatment of juvenile/adult GM1-gangliosidosis with Miglustat may reverse disease progression. Metab Brain Dis. 2017;32(5):1529-1536. doi:10.1007/s11011-017-0044-y. PMID: 28577204.
17. MetabERN — European Reference Network for Hereditary Metabolic Disorders. https://metab.ern-net.eu/
18. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 615–6 (Chapter 83, Mucolipidosis II and III/I-cell disease and pseudo-Hurler polydystrophy).
19. Okada S, O'Brien JS. Generalized gangliosidosis: beta-galactosidase deficiency. Science. 1968;160(3831):1002-4. doi:10.1126/science.160.3831.1002. PMID: 5647842.
20. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 733–40 (Chapter 97, Galactosialidosis).
21. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 117–23 (5.2 Mucopolysaccharidosis IV).
22. Ohto U, Usui K, Ochi T, Yuki K, Satow Y, Shimizu T. Crystal structure of human β-galactosidase: structural basis of Gm1 gangliosidosis and morquio B diseases. J Biol Chem. 2012;287(3):1801-12. doi:10.1074/jbc.M111.293795. PMID: 22128166; PMC3265862.
23. Nicoli ER, Annunziata I, d'Azzo A, Platt FM, Tifft CJ, Stepien KM. GM1 Gangliosidosis-A Mini-Review. Front Genet. 2021;12:734878. doi:10.3389/fgene.2021.734878. PMID: 34539759; PMC8446533.
24. Lang FM, Korner P, Harnett M, Karunakara A, Tifft CJ. The natural history of Type 1 infantile GM1 gangliosidosis: A literature-based meta-analysis. Mol Genet Metab. 2020;129(3):228-235. doi:10.1016/j.ymgme.2019.12.012. PMID: 31937438; PMC7093236.
25. Brunetti-Pierri N, Scaglia F. GM1 gangliosidosis: review of clinical, molecular, and therapeutic aspects. Mol Genet Metab. 2008;94(4):391-396. doi:10.1016/j.ymgme.2008.04.012. PMID: 18524657.
26. GeneGuidance. GLB1 — curation of the gene and associated conditions (version of 2026-09-26).
27. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 662–9 (Chapter 88, Tay-Sachs disease/hexosaminidase A deficiency).
28. Jaeken J, Morava E. Congenital disorders of glycosylation, dolichol and glycosylphosphatidylinositol metabolism. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. doi:10.1007/978-3-662-63123-2_43. Chapter 43, section 43.5 Congenital Disorders of Deglycosylation (43.5.2 Lysosomal Storage Disorders).
29. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 590–9 (Chapter 80, Morquio syndrome/mucopolysaccharidosis type IV/keratan sulfaturia).
30. Fischetto R, Palladino V, Mancardi MM, Giacomini T, Palladino S, Gaeta A, et al. Substrate reduction therapy with Miglustat in pediatric patients with GM1 type 2 gangliosidosis delays neurological involvement: A multicenter experience. Mol Genet Genomic Med. 2020;8(10):e1371. doi:10.1002/mgg3.1371. PMID: 32779865; PMC7549581.
31. López de Frutos L, Hahn A, Dhamija R, Héron B, Fay A, Schneider SA, et al. A prospective, longitudinal, natural history study of neurological disease trajectory in children with late-infantile or juvenile-onset GM1 gangliosidosis or GM2 gangliosidoses: the PRONTO study. Mol Genet Metab. 2026;149(3):110291. doi:10.1016/j.ymgme.2026.110291. PMID: 42896439.
32. Kannebley JS, Silveira-Moriyama L, Bastos LO, Steiner CE. Clinical Findings and Natural History in Ten Unrelated Families with Juvenile and Adult GM1 Gangliosidosis. JIMD Rep. 2015;24:115-22. doi:10.1007/8904_2015_451. PMID: 26108645; PMC4582032.
33. National Center for Biotechnology Information. ClinVar — GLB1 records against NM_000404.4: c.202C>T (VCV000000944), c.325C>T (VCV000129154), c.442C>T (VCV000092907), c.1010T>C (VCV001251981), c.1783C>T (VCV000550288). Accessed on 31 Oct 2026. https://www.ncbi.nlm.nih.gov/clinvar/
34. Caciotti A, Garman SC, Rivera-Colón Y, Procopio E, Catarzi S, Ferri L, et al. GM1 gangliosidosis and Morquio B disease: an update on genetic alterations and clinical findings. Biochim Biophys Acta. 2011;1812(7):782-90. doi:10.1016/j.bbadis.2011.03.018. PMID: 21497194; PMC3210552.
35. Arash-Kaps L, Komlosi K, Seegräber M, Diederich S, Paschke E, Amraoui Y, et al. The Clinical and Molecular Spectrum of GM1 Gangliosidosis. J Pediatr. 2019;215:152-157.e3. doi:10.1016/j.jpeds.2019.08.016. PMID: 31761138.
36. King JE, Dexter A, Gadi I, Zvereff V, Martin M, Bloom M, et al. Maternal uniparental isodisomy causing autosomal recessive GM1 gangliosidosis: a clinical report. J Genet Couns. 2014;23(5):734-41. doi:10.1007/s10897-014-9720-9. PMID: 24777551.
37. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 260–5 (Oedema, generalized, or puffy extremities).
38. Saudubray JM, García-Cazorla Á. Clinical approach to inborn errors of metabolism in paediatrics. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1. doi:10.1007/978-3-662-63123-2_1
39. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 109–16 (5.1 Dysostosis Multiplex).
40. Jarnes Utz JR, Kim S, King K, Ziegler R, Schema L, Redtree ES, et al. Infantile gangliosidoses: Mapping a timeline of clinical changes. Mol Genet Metab. 2017;121(2):170-179. doi:10.1016/j.ymgme.2017.04.011. PMID: 28476546; PMC5727905.
41. Fiori L, Turzi M, Tagi VM, Asnaghi L, Bonaventura E, Tonduti D, et al. Alkaline Phosphatase and Infantile GM1 Gangliosidosis: A Simple Biomarker for a Complex Disease? JIMD Rep. 2026;67(2):e70075. doi:10.1002/jmd2.70075. PMID: 41800148; PMC12963784.
42. Menkovic I, Williams M, Makhijani N, Wei R, Young SP, El-Gharbawy A, et al. Persistent elevations of alkaline phosphatase as an early indicator of GM1 gangliosidosis. Mol Genet Metab Rep. 2025;42:101191. doi:10.1016/j.ymgmr.2025.101191. PMID: 39897471; PMC11786200.
43. Héron B, Batzios S, Mengel E, Giugliani R, Patterson M, Gautschi M, et al. A natural history study of pediatric patients with early onset of GM1 gangliosidosis, GM2 gangliosidoses, or gaucher disease type 2 (RETRIEVE). Orphanet J Rare Dis. 2024;19(1):459. doi:10.1186/s13023-024-03409-1. PMID: 39639297; PMC11619657.
44. Regier DS, Proia RL, D'Azzo A, Tifft CJ. The GM1 and GM2 gangliosidoses: natural history and progress toward therapy. Pediatr Endocrinol Rev. 2016;13 Suppl 1:663-73. PMID: 27491214; PMC8186028.
45. King KE, Kim S, Whitley CB, Jarnes-Utz JR. The juvenile gangliosidoses: a timeline of clinical change. Mol Genet Metab Rep. 2020;25:100676. doi:10.1016/j.ymgmr.2020.100676. PMID: 33240792; PMC7674119.
46. Regier DS, Kwon HJ, Johnston J, Golas G, Yang S, Wiggs E, et al. MRI/MRS as a surrogate marker for clinical progression in GM1 gangliosidosis. Am J Med Genet A. 2016;170(3):634-44. doi:10.1002/ajmg.a.37468. PMID: 26646981.
47. Kolstad J, Zoppo C, Johnston JM, D'Souza P, Kühn AL, Vardar Z, et al. Natural history progression of MRI brain volumetrics in type II late-infantile and juvenile GM1 gangliosidosis patients. Mol Genet Metab. 2025;144(3):109025. doi:10.1016/j.ymgme.2025.109025. PMID: 39874851; PMC11875878.
48. Lewis CJ, Chipman SI, D'Souza P, Johnston JM, Yousef MH, Gahl WA, et al. Brain aging in Type II GM1 gangliosidosis. Commun Med (Lond). 2026;6(1):568. doi:10.1038/s43856-026-01745-w. PMID: 42410243; PMC13720018.
49. Ferreira CR, Regier DS, Yoon R, Pan KS, Johnston JM, Yang S, et al. The skeletal phenotype of intermediate GM1 gangliosidosis: Clinical, radiographic and densitometric features, and implications for clinical monitoring and intervention. Bone. 2020;131:115142. doi:10.1016/j.bone.2019.115142. PMID: 31704340; PMC6937522.
50. Giugliani L, Steiner CE, Kim CA, Lourenço CM, Santos MLSF, de Souza CFM, et al. Clinical findings in Brazilian patients with adult GM1 gangliosidosis. JIMD Rep. 2019;49(1):96-106. doi:10.1002/jmd2.12067. PMID: 31497487; PMC6718113.
51. Bleier M, Yuskiv N, Priest T, Moisa Popurs MA, Stockler-Ipsiroglu S. Morquio B patient/caregiver survey: First insight into the natural course of a rare GLB1 related condition. Mol Genet Metab Rep. 2018;16:57-63. doi:10.1016/j.ymgmr.2018.06.006. PMID: 30094186; PMC6072644.
52. Jones S, Wijburg FA. Glycosaminoglycans and oligosaccharides disorders: glycosaminoglycans synthesis defects, mucopolysaccharidoses, oligosaccharidoses and sialic acid disorders. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 41, section 41.2 Mucopolysaccharidoses (41.2.1 Clinical Presentation). doi:10.1007/978-3-662-63123-2_41
53. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 652–5 (Morquio syndrome).
54. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 768–9 (Oedema, nuchal translucency and cystic hygroma).
55. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 208–11 (Leukodystrophy/leukoencephalopathy).
56. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 74–7 (Ataxic child).
57. Mochel F, Sedel F. Inborn errors of metabolism in adults: a diagnostic approach to neurological and psychiatric presentations. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 2. doi:10.1007/978-3-662-63123-2_2
58. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 138–41 (Dystonia).
59. Su P, Khaledi H, Waggoner C, Gelb MH. Detection of GM1-gangliosidosis in newborn dried blood spots by enzyme activity and biomarker assays using tandem mass spectrometry. J Inherit Metab Dis. 2021;44(1):264-271. doi:10.1002/jimd.12269. PMID: 32506457.
60. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 128–31 (Developmental regression).
61. Kell P, Mishra S, D'Souza P, Tifft CJ, Dietzen DJ, Lasio LD, et al. Diagnostic and therapeutic applications of the glycan biomarker H3N2b in GM1 Gangliosidosis. Mol Genet Metab. 2026;148(3):110159. doi:10.1016/j.ymgme.2026.110159. PMID: 42160923; PMC13271843.
62. Tonin R, Caciotti A, Procopio E, Fischetto R, Deodato F, Mancardi MM, et al. Pre-diagnosing and managing patients with GM1 gangliosidosis and related disorders by the evaluation of GM1 ganglioside content. Sci Rep. 2019;9(1):17684. doi:10.1038/s41598-019-53995-5. PMID: 31776384; PMC6881353.
63. Quadrini KJ, Vrentas C, Duke C, Wilson C, Hinderer CJ, Weinstein DA, et al. Validation of high-sensitivity assays to quantitate cerebrospinal fluid and serum β-galactosidase activity in patients with GM1-gangliosidosis. Mol Ther Methods Clin Dev. 2024;32(3):101318. doi:10.1016/j.omtm.2024.101318. PMID: 39282076; PMC11401230.
64. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 120–1 (Corneal clouding).
65. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 577 (Chapter 78, Hunter disease/mucopolysaccharidosis type II/iduronate sulfatase deficiency).
66. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 853–5 (Skeletal dysplasia charts).
67. Coppola F, Prasad S, Morana G, Whitehead MT, Zakzouk R, Manteghinejad A, et al. Neuroimaging Spectrum of GM1 Gangliosidosis with Description of Novel Imaging Signs. AJNR Am J Neuroradiol. 2026;47(2):489-495. doi:10.3174/ajnr.A8973. PMID: 40841163; PMC12867068.
68. Zagaynova VA, Nasykhova YA, Tonyan ZN, Danilova MM, Dvoynova NM, Lazareva TE, et al. Case report: Preimplantation genetic testing for infantile GM1 gangliosidosis. Front Genet. 2024;15:1344051. doi:10.3389/fgene.2024.1344051. PMID: 38404665; PMC10884188.
69. Jones S, Wijburg FA. Glycosaminoglycans and oligosaccharides disorders: glycosaminoglycans synthesis defects, mucopolysaccharidoses, oligosaccharidoses and sialic acid disorders. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 41, section 41.3 Oligosaccharidoses and Mucolipidoses. doi:10.1007/978-3-662-63123-2_41
70. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 555–7 (Chapter 75, Introduction to mucopolysaccharidoses).
71. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 106–9 (Coarse facial features).
72. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 124–8 (5.3 Mucolipidosis II).
73. Ribeiro HPB, Mont'Alverne AG, Ribeiro EM. MELAS (miopatia mitocondrial, encefalopatia, acidose lática e episódios semelhantes a acidente vascular cerebral). In: Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 137–41.
74. Rha AK, Maguire AS, Martin DR. GM1 Gangliosidosis: Mechanisms and Management. Appl Clin Genet. 2021;14:209-233. doi:10.2147/TACG.S206076. PMID: 33859490; PMC8044076.
75. Roze E, Navarro S, Cornu P, Welter ML, Vidailhet M. Deep brain stimulation of the globus pallidus for generalized dystonia in GM1 type 3 gangliosidosis: technical case report. Neurosurgery. 2006;59(6):E1340. doi:10.1227/01.NEU.0000245620.24603.1B. PMID: 17277672.
76. Shield JP, Stone J, Steward CG. Bone marrow transplantation correcting beta-galactosidase activity does not influence neurological outcome in juvenile GM1-gangliosidosis. J Inherit Metab Dis. 2005;28(5):797-8. doi:10.1007/s10545-005-0089-7. PMID: 16151914.
77. Morris AAM, Jones S. Medications used in the treatment of inborn errors of metabolism. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 45. doi:10.1007/978-3-662-63123-2_45
78. European Medicines Agency. Zavesca (miglustat): EPAR — Product Information (Summary of Product Characteristics). Amsterdam: EMA; 1st authorisation 20 Nov 2002, renewal 8 Dec 2012. https://www.ema.europa.eu/en/documents/product-information/zavesca-epar-product-information_en.pdf
79. U.S. Food and Drug Administration. Zavesca (miglustat) capsules — prescribing information; NDA 021348 (original approval 31 Jul 2003). Revised 04/2023. https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=021348
80. INFARMED. Infomed — search «Miglustat» (Miglustato Dipharma 100 mg; Miglustato Gen.Orph 100 mg; Opfolda 65 mg). https://extranet.infarmed.pt/INFOMED-fo/
81. Agência Nacional de Vigilância Sanitária. Open data on registered medicines: ZAVESCA (miglustate), registration 112363431, Janssen-Cilag. https://dados.anvisa.gov.br/dados/DADOS_ABERTOS_MEDICAMENTOS.csv
82. U.S. Food and Drug Administration. Opfolda (miglustat) capsules — prescribing information; NDA 215211. Revised 11/2024. https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=215211
83. Lewis CJ, Akmal H, D'Souza P, Johnston JM, Ashraf S, Vezina G, et al. Expanded access use of intravenous gene transfer with AAV9-GLB1 in a juvenile GM1 gangliosidosis participant [preprint]. medRxiv. 10 Sep 2026. doi:10.64898/2026.09.04.26361573.
84. Passage Bio. Passage Bio announces promising interim clinical data from first eight patients with GM1 gangliosidosis in Imagine-1 study [press release and presentation; SEC Form 8-K, Exhibits 99.2 and 99.3]. Philadelphia; 7 Aug 2023. https://www.sec.gov/Archives/edgar/data/1787297/000155837023013486/pasg-20230807xex99d2.htm
85. Passage Bio. Annual report (Form 10-K) for the fiscal year ended December 31, 2023. U.S. Securities and Exchange Commission; 4 Mar 2024. https://www.sec.gov/Archives/edgar/data/1787297/000155837024002310/pasg-20231231x10k.htm
86. Passage Bio. Passage Bio out-licenses three pediatric gene therapy programs to GEMMA Biotherapeutics and enters new research collaboration [press release; SEC Form 8-K, Exhibit 99.1]. 1 Aug 2024. https://www.sec.gov/Archives/edgar/data/0001787297/000110465924084615/tm2420570d1_ex99-1.htm
87. Passage Bio. Annual report (Form 10-K) for the fiscal year ended December 31, 2025. U.S. Securities and Exchange Commission; 3 Mar 2026. https://www.sec.gov/Archives/edgar/data/1787297/000110465926022549/pasg-20251231x10k.htm
88. Passage Bio. Passage Bio presents additional interim data from Imagine-1 study for GM1 gangliosidosis at 19th Annual WORLDSymposium 2023 [press release; SEC Form 8-K, Exhibit 99.1]. Philadelphia; 24 Feb 2023. https://www.sec.gov/Archives/edgar/data/1787297/000155837023001964/pasg-20230224xex99d1.htm
89. Casazza K, Giugliani R, Regier DS, Jarnes J. From molecule to meaning: neuronopathic biomarkers and clinical relevance in GM1. J Inherit Metab Dis. 2026;49(1):e70134. doi:10.1002/jimd.70134. PMID: 41500845; PMC12779496.
90. Jubran L, Wannemacher R, Baumgärtner W, Felmy F, Steiner MA, Leitzen E, et al. Impact of Sinbaglustat on Neurons of the Medial Nucleus of the Trapezoid Body in a Murine Model of Human G(M1)-Gangliosidosis. J Clin Med. 2026;15(6):2249. doi:10.3390/jcm15062249. PMID: 41899173; PMC13026635.
91. Matsushima SK, Shimada Y, Kinoshita M, Nagashima T, Okamoto S, Iizuka S, et al. Adeno-associated virus expressing a blood-brain barrier-penetrating enzyme improves GM1 gangliosidosis in a preclinical model. J Clin Invest. 2025;135(12):e180724. doi:10.1172/JCI180724. PMID: 40198143; PMC12165788.
92. Bucciarelli L, Fabris C, Accardo M, Biffi A, Poletti V. Lentiviral hematopoietic stem cell gene therapy ameliorates GM1-gangliosidosis in mice. Mol Ther. 2026;34(3):1854-66. doi:10.1016/j.ymthe.2025.12.007. PMID: 41376163; PMC12974154.
93. Pérez-Carmona N, Cubero E, Ruano A, Pons-Vizcarra M, Delgado A, Trapero A, et al. Development of small-molecule allosteric modulators of beta-galactosidase (β-Gal) for the treatment of GM1 gangliosidosis and Morquio B. Int J Mol Sci. 2026;27(8):3631. doi:10.3390/ijms27083631. PMID: 42074270; PMC13115887.
94. Linhart A. Metabolic cardiomyopathy. In: Baars HF, van der Smagt JJ, Doevendans PAFM, van Tintelen JP, editors. Clinical Cardiogenetics. 3rd ed. Cham: Springer; 2020. p. 151–66. doi:10.1007/978-3-030-45457-9_9