Comparative case series demonstrating disease modifying outcome with early ambroxol and enzyme replacement therapy in acute neuronopathic Gaucher disease

INTRODUCTION

Gaucher disease (GD) is a rare autosomal recessive lysosomal storage disorder caused by deficient activity of β-glucocerebrosidase (GCase; EC 3.2.1.45), which hydrolyses glucosylceramide (GluCer) into glucose and ceramide. Biallelic pathogenic variants in GBA1 lead to misfolding of GCase in the endoplasmic reticulum, defective trafficking and targeting to the lysosomes, and progressive accumulation of GlcCer with its deacylated derivate, glucosylsphingosine (GlcSph), in lysosomal macrophages known as Gaucher cells in the reticuloendothelial system, bones, and central nervous system^[1,2]. GD presents as a clinical spectrum ranging from a severe, perinatal-lethal phenotype - often seen in collodion babies - to acute or chronic neuronopathic forms, as well as non-neuronopathic GD, which may involve skeletal and visceral manifestations, Parkinson’s disease, or remain asymptomatic into adulthood^[2]. Although the phenotype is variable, three clinical types have been historically identified: type 1 (non-neuronopathic form, MIM# 230800), characterized by the presence of primary central nervous system involvement; type 2 (acute neuronopathic form, MIM# 230900), marked by early and severe neurological symptoms; and type 3 (chronic neuronopathic form, MIM# 231000), which features later-onset and more slowly progressive neurological involvement. GD type 1 (GD1) is the most common and its manifestations are limited to the skeletal and reticuloendothelial systems with the absence of neurological symptoms^[1]. Although GD1 has typically been described as a non-neuronopathic form, there is increasing evidence of neurological involvement with parkinsonian syndrome and Lewy body dementia in affected patients^[3]. GD types 2 (GD2) and 3 (GD3) are characterized by the presence of primary neurologic disease and may be distinguished by the age of onset and rate of disease progression. Individuals with GD2 have early-onset disease before age two, limited psychomotor development, and death in infancy (median nine months)^[4], or early childhood, by age two to four years^[2]. GD type 3 may present before age two years, but often has a more slowly progressive course, with survival into the third or fourth decade^[5]. Assigning a definitive diagnosis for the neuronopathic GD (nGD) subtypes can be challenging in very young patients. Very rarely, deficiency of GCase activity occurs as a result of biallelic pathogenic variants in the PSAP gene, leading to a deficiency in the GCase activator saposin C (MIM# 610539)^[6] and clinical features characteristic of severe nGD or GD1^[2]. Saposin C is one of four homologous proteins, referred to as saposin A-D, which arise from proteolytic cleavage of the saposin precursor protein, prosaposin^[7]. Affected individuals with saposin C deficiency demonstrate glucosylceramide accumulation and visceromegaly, but have normal in vitro GluCase activity^[2].

The incidence of GD1 is approximately 1 in 60,000 globally, rising to 1 in 800 to 1 in 950 in the Ashkenazi Jewish community^[4]. A Japanese study reported a prevalence of 1:530,000, with approximately 38% of patients having GD1, while around 54% have GD2 or GD3^[5]. The prevalence of GD in the Australian population during an 11-year period from January 2009 through to December 2020 was 1 in 74,000 live births^[8]. Traditionally, the diagnosis of GD can be confirmed by demonstrating reduced GCase activity in peripheral blood leukocytes, followed by the identification of biallelic pathogenic variants in GBA1. The presence of a pseudo-deficiency allele has been described as a possible confounder in the interpretation of enzymatic assay results in newborn screening programs, which have reported false positives based on decreased GCase activity^[9,10]. More recently, GlcSph has been shown to be a key diagnostic biomarker of GD with 100% specificity and sensitivity to distinguish GD patients from unaffected individuals and a variety of related inherited metabolic disorders^[11]. Additionally, its utility as a pharmacodynamic marker in evaluating therapeutic response and its prognostic attribute by differentiating neuronopathic (nGD) from their non-neuronopathic counterparts are valuable. The applicability of GlcSph could be expanded to second-tier testing following multiplexed enzymology in newborn screening^[11,12].

The introduction of enzyme replacement therapy (ERT) in 1991 has revolutionized the management of GD1^[13]. Currently, three recombinant glucocerebrosidase enzyme preparations available include imiglucerase (Cerezyme®), velalglucerase alfa (VPRIV®), and taliglucerase alfa (Elelyso®)^[2]. ERT has been shown to improve hematopoietic reconstitution, resulting in a reduction in liver and spleen volumes and an improvement or stabilization of skeletal abnormalities^[2]. However, ERT alone does not alter the trajectory of neurologic disease in GD, although it can result in significantly enhanced quality of life in GD3 due to improvement in systemic manifestations^[2]. The International Collaborative Gaucher Group consensus recommendation for GD1 includes early initiation of treatment for all symptomatic children and adolescents, while enlisting a “watchful waiting” approach restricted to asymptomatic adults with genotypes associated with mild disease severity^[14].

Substrate reduction therapy (SRT) is a newer therapeutic option aimed at decreasing the amount of substrate precursor (GluCer) synthesized to a level at which the patient’s own residual GCase activity can effectively clear the glycosphingolipids and reduce storage burden^[15]. Miglustat (Zavesca®, Actelion) is the first oral agent for the treatment of individuals with mild to moderate GD1 for whom ERT is not a therapeutic option due to intolerance or poor venous access. Miglustat has not been found to have any effect on neurological symptoms in GD3, despite its ability to cross the blood-brain barrier^[1]. Eliglustat, an alternative glucosylceramide synthetase inhibitor, has been reported to improve hematologic and visceral parameters for GD1 adults with extensive, intermediate, or poor cytochrome P450 2D6 (CYP2D6) metabolizer status (> 90% of patients). Eliglustat is metabolized by CYP2D6 and, to a lesser extent, CYP3A; hence, its dosage is based on the individual’s CYP2D6 metabolizer status^[16,17].

No existing treatment is effective for nGD due to the difficulty in delivering therapies across the blood-brain barrier^[18]. Early diagnosis and supportive treatment are imperative but often delayed due to the rarity of the disease coupled with the lack of awareness among clinicians. Since the last decade, ambroxol has gained the spotlight for its potential as a chaperone treatment of GD, predominantly for neurological and skeletal manifestations^[19,20], while also being effective for visceral and hematological complications. Traditionally, ambroxol has been used as an oral mucolytic drug available over the counter for many years as a cough medicine. It is also a potentiator of surfactant and has been utilized for hyaline membrane disease in neonates. Chaperone therapy (CT) is proposed to correct misfolded GCase enzyme to improve its function and facilitate transport across the blood-brain barrier, which is not possible with ERT or SRT. Ambroxol was identified as a potential CT through screening of an established drug database^[18]. Several case studies with limited patient cohorts have reported its potential to improve and stabilize neurological and skeletal manifestations, with and without ERT^[19-31]. While these case reports highlight promise for GD patients with progressive neurological manifestations, ambroxol has been commenced in all but one of these case studies on patients after neurological and skeletal symptoms have been established. A single report described the commencement of ambroxol on an asymptomatic patient^[24]. In this comparative case series, we describe, to the best of our knowledge, the youngest GD2 patient treated with ERT and ambroxol from the neonatal period, and compare his outcome to later initiation in two other nGD patients from our center, to demonstrate its clinical efficacy in delaying the onset of debilitating neurological symptoms and improving quality of life when used as adjunct therapy with ERT for treatment of visceral manifestations.

DISCUSSION

Ambroxol has gained attention in the last decade for its potential to improve neurological and skeletal manifestations (not possible with ERT alone) due to its ability to cross the blood-brain barrier as demonstrated in the mouse model^[33]. In GD, the mutant enzyme misfolds, accelerating endoplasmic reticulum (ER)-associated degradation, despite the maintenance of functional potential^[20]. As a pharmacological chaperone, ambroxol selectively binds to misfolded GCase in the neutral pH environment of the ER, promoting proper protein folding. It then dissociates from the enzyme in the acidic pH environment of the lysosome, thereby enabling functional recovery^[21]. GCase activity has been shown to be increased in the cerebellum, heart, and spleen of GD mice treated with ambroxol^[34] and in the midbrain, cortex, cerebellum, and striatum in non-human primate models^[35,36]. Furthermore, fibroblast studies have provided insight into genetic variants of GBA1 that have shown promising response to ambroxol, including F213I, N370S, R131C, N188S/G193W, R120W/L444P, R120W/N370S, L444P/L444P, F213I/L444P, P415R/L444P, F213I/RecNcil, D409S/IVS10G>A, although conflicting data for several genotypes have demonstrated varying outcomes in fibroblast studies^{[21,34,37-42]}. Variable outcomes of in vitro studies suggest that alternative intracellular processes intrinsic to Gaucher cells potentiate the effect of ambroxol and highlight the need for further studies to elucidate the precise mechanism of action. Importantly, mutations without confirmed in vitro response have also shown in vivo increases in GCase. More than 65% of 40 nGD patients had responded favorably to high-dose ambroxol therapy, irrespective of genotypes^[29].

The first pilot study, conducted to assess the tolerability and efficacy of ambroxol on 12 symptomatic GD1 patients (mean age: 41.1 years; range 24-63), involved treatment with 150 mg of ambroxol daily for six months. The results showed increased platelet counts and reduced spleen volumes, while hemoglobin levels and liver volumes remained stable^[43]. One patient with pulmonary hypertension showed no clinical worsening during the six-month treatment period but experienced deterioration after discontinuing ambroxol. Among the three patients who chose to continue treatment for an additional six months beyond the initial study period, similar further clinical improvements were observed. A subsequent proof-of-concept study evaluated the efficacy, safety, and tolerability of high-dose ambroxol (25 mg/kg/day or maximum dose 1,300 mg/day) in combination with ERT in five patients with neuronopathic GD, aged 5 to 28 years. The study demonstrated marked improvements in myoclonus, seizures, and pupillary light reflex dysfunction in all patients, with notable recovery of gross motor function in two patients, enabling them to walk again. Ambroxol was well tolerated by all patients, and no adverse effects were reported^[20]. Several case reports since then have described variable benefits of ambroxol for progressive neurological symptoms in Gaucher disease [Table 2 and Table 3]. However, a similar beneficial response was not observed in a pair of twin siblings with an intermediate phenotype between GD2 and GD3, characterized by complex neurological symptoms and parkinsonian features. Ambroxol, administered alongside ERT, was discontinued within one month of initiation due to increased tremors and axial instability noted by the parents. Both siblings subsequently responded to levodopa, showing reduced tremors and improvements in motor skills and cognitive function^[44].

Ambroxol supplementation combined with ERT in two Canadian patients with GD3 led to initial clinical improvements in ambulation in both patients; however, these improvements were later lost in the first patient due to fractures. No notable changes were observed in seizures in either patient; nonetheless, seizure duration was substantially reduced in the second patient (N188S/R463H in GBA), eliminating the need for hospital admissions due to status epilepticus over a three-year period and thereby enhancing her quality of life. The variability in response to ambroxol between the two patients was proposed to be associated with their different genotypes^[23]. This hypothesis was supported by a subsequent study highlighting the benefits of in vitro testing of fibroblasts to determine individualized response prior to in vivo trial with high doses of ambroxol in two patients with GD3 (25 mg/kg/day and 1,300 mg/day, respectively) on ERT^[25]. In a patient with compound heterozygous variants, N188S and a null allele (IVS2 + 1G > A), ambroxol therapy initiated at 14.5 years resulted in a 67% increase in enzymatic activity in cultured fibroblasts from a baseline of 26.5 nmol/mg/h (normal: 86.12 ± 25.00). Correspondingly, in vivo improvements were observed in epilepsy and GD biomarkers. In the second patient, homozygous for the L444P variant, with the residual enzymatic activity of 12.8 nmol/mg/h in cultured fibroblasts, ambroxol started at age 60 was clinically ineffective. The authors suggest that additional factors beyond GBA1 mutations - such as variations in the complex ER network involved in proteostasis - may influence therapeutic response to ambroxol^[25].

In a later study of four patients (14-20 years) with nGD and established myoclonic seizures and intellectual disability, combined therapy with high-dose ambroxol and ERT resulted in the arrest of neurological progression - including significantly decreased seizure frequency and improved neurocognitive function - only after the ambroxol dose was increased to 27 mg/kg/day. At this dose, GluSph levels normalized in all patients, and residual GCcase activity increased to approximately 9%-19% of the mean normal level, as evidenced by both in vitro and in vivo studies^[26]. The threshold enzyme activity required to prevent glucosylceramide accumulation has been suggested to be above 10%-15% of normal control^[45]. In addition, GCase activity was reported to increase significantly in the brains of normal mice receiving 60 mg/kg/day ambroxol, which is equivalent to a human dose of 4.8 mg/kg/day^[46]. However, the precise increment in brain GCase activity required to alleviate glucosylceramide and GluSph accumulation remains unknown. Importantly, ambroxol concentrations of 30 μmol/L or higher have been shown to inhibit certain GCase mutants (R296Q and L483P)^[26].

A recent case detailed by Chu et al. described the use of ambroxol in a Taiwanese female neonate identified through newborn screening with markedly low GCase activity (0.47 μM/h; normal > 1.8 μM/h), significantly elevated chitotriosidase activity (577.62 nmol/mL WB/h; normal < 109.9), and L444P/RecNcil variants in GBA, a genotype predictive of GD2/3^[24]. ERT was introduced at five weeks of age due to thrombocytopenia, borderline splenomegaly, and elevated plasma GluSph, suggesting a severe phenotype. At six months, the patient displayed truncal rigidity and mild horizontal gaze palsy, prompting the initiation of high-dose ambroxol (30 mg/kg/day) at eight months. By 18 months, she showed delayed gross motor milestones but had not experienced developmental regression or seizures. Her clinical phenotype appeared relatively mild, supported by a modest elevation in baseline GluSph (117.8 ng/mL; normal < 3) and a mild response to ambroxol, as indicated by a further reduction in GluSph from 55.53 ng/mL post-ERT to 32.7 ng/mL after six months of ambroxol^[24].

In 2021, an investigator-initiated registry published an observational study on the safety and efficacy of ambroxol in patients with GD and GBA-associated Parkinson’s disease (GBA-PD). Data were collected from 13 countries, comprising 41 patients with a median age of 17 years (range 1.5-74). Of these, 11 had GD1 (four also diagnosed with PD), 27 had nGD, and three were GBA mutation carriers with PD. The median treatment duration was 19 months (range 1-76), with a maximum daily ambroxol dose of 435 mg (range 75-1,485 mg). Clinical benefits, including neurological stability or improvement, increased physical activity, and reduced fatigue, were reported in 25 patients. Adverse effects were observed in 12 patients and included mild bowel discomfort, cough, allergic reactions, mild proteinuria, dizziness, and disease progression^[31]. A recent prospective study described the favorable clinical outcome of a Turkish female neonate with GD2 who presented with a collodion baby phenotype (ichthyosis). She received high-dose ambroxol monotherapy starting at four months of age. ERT was added one year later due to progressive hepatosplenomegaly. By three years of age, she demonstrated age-appropriate neurocognitive and motor development. However, there was no clear benefit regarding visceral manifestations. Notably, cerebrospinal fluid GluSph levels decreased significantly compared to pre-treatment, although GluSph and chitotriosidase levels in the blood increased. In vitro studies of GCase activity in fibroblasts revealed a significant increase.

Including the three nGD patients described in this report and additional cases identified through a literature review from 2016 to 2024 [Table 2 and Table 3], a total of 45 patients have received ambroxol therapy. Treatment was discontinued in five patients due to adverse effects such as increased mucus production and cough. Additionally, two siblings with parkinsonian features experienced worsened tremors and axial instability. In two cases, reimbursement issues led to treatment discontinuation. Overall, approximately 81% of patients responded favorably to ambroxol chaperone therapy. In contrast, two GD2 patients managed at our center who did not receive ambroxol died at three and eight months of age, respectively^[47], consistent with previous reports of infant mortality in GD2 (median age of death: nine months)^[4].

Ascertainment of different GD subtypes has important therapeutic implications. Although current treatment options for nGD are limited due to the challenges of crossing the blood-brain barrier, ambroxol has emerged as a promising pharmacological chaperone treatment. In our case series, early detection accompanied by prompt intervention was crucial in modifying disease progression in Patient A with GD2, leading to reduced morbidity and mortality and improved clinical outcomes and overall quality of life. Early diagnosis of nGD also helps mitigate the prolonged diagnostic odyssey often faced by affected families, facilitates access to multidisciplinary supportive services, and supports informed reproductive decision making. These advantages have driven the development of newborn screening programs for GD, utilizing multiplex analytical techniques such as tandem mass spectrometry or digital microfluidics fluorimetry to measure GCase enzyme activity in DBS, with GluSph analysis serving as a second-tier test^[48]. GD is currently included in newborn screening panels in a few US states, including Illinois^[9], New York^[10], and Missouri^[49], though it has not yet been added to the Recommended Uniform Screening Panel (RUSP). Screening is also performed in Italy^[48], Taiwan^[50], Japan^[51], and mainland China^[52]. Determining the population-specific incidence of GD in Australia is essential to assess the potential benefit of implementing a national newborn screening program. Long-term outcome monitoring will require regular neurological examinations, radiological assessments, and formal developmental evaluations at defined intervals.

CONCLUSION

In summary, our case series expands the experience of Ambroxol therapy in nGD. To the best of our knowledge, we report the youngest GD2 patient who received disease-modifying treatment starting at just two weeks of age. This early intervention led to the complete resolution of respiratory compromise, hepatomegaly, and cytopenias, a significant reduction in GluSph, and a remarkably improved quality of life, with the patient achieving age-appropriate developmental milestones by 4 years of age - without the emergence of neurological symptoms or devastating life-limiting complications to date. In contrast, patients in our series who initiated Ambroxol therapy later in life experienced less substantial benefits in terms of neurologic disease stabilization or mitigation. These findings are consistent with the case published by Chu et al., as well as several other studies, suggesting that Ambroxol therapy may delay or attenuate the progression of neurological symptoms in nGD^[24]. Therefore, early initiation of treatment appears to be critical for achieving optimal outcomes. The collective evidence on the utility of Ambroxol indicates that, when used in combination with ERT, it may offer a promising treatment for nGD. Although randomized clinical trials involving larger cohorts would be valuable to further assess outcomes in early-treated patients identified through sibling studies or newborn screening, we acknowledge that the likelihood of industry-sponsored clinical trials is low, given Ambroxol’s low cost and limited commercial return. As a result, case reports like ours and those of other researchers play a vital role in advancing the understanding of this off-label use of Ambroxol chaperone therapy in GD.