Gaucher disease, state of the art and perspectives

Abstract

Knowledge about Gaucher disease (GD), considered a model for rare diseases, has considerably increased since its discovery. The pathophysiology of this lysosomal disorder is better known, and specific therapies that can control many aspects of the disease have been developed, particularly for the most common form, Type 1 GD. Yet, in part because of the rarity of GD, but also because of a lack of awareness by physicians, diagnostic delay too often leads to a belated management of patients having accumulated comorbidities. Gaucher cells, the most visible consequence of glucocerebrosidase deficiency, have been known for many years. However, the pathophysiological mechanisms underlying some major lesions, such as bone disease, predisposition to Parkinson's disease in Type 1 GD, or neurological involvement in Type 2 and Type 3 GD, remain poorly understood. Diagnostic, therapeutic, and follow‐up issues associated with these symptoms remain critical to optimize the care of these patients. In this review, clinical characteristics, pathophysiology, diagnosis, treatment, and prognosis of GD are successively considered, highlighting for each of them the remaining challenges. Continued efforts to better understand pathophysiological mechanisms, use of the most modern methods such as artificial intelligence, international collaboration, and development of new therapeutic strategies seem essential for the future of this rare disease.

Abbreviations

ABCC1

ATP‐binding cassette subfamily C member 1

aPTT

activated partial thromboplastin time

CBC

complete blood count

CRP

C‐reactive protein

DBS

dried blood spots

ER

endoplasmic reticulum

ERT

enzyme replacement therapy

GCase

glucocerebrosidase

Gcer

glucocerebroside

GD

Gaucher disease

HDL

high‐density lipoproteins

HSCT

hematopoietic stem cell transplantation

ICGG

International Collaborative Gaucher Group

iPSC

induced pluripotent stem cells

IQR

interquartile range

IWGGD

international working group on Gaucher disease

MGUS

monoclonal gammopathy of undetermined significance

MLPA

multiplex ligation‐probe amplification

MRI

magnetic resonance imaging

MSC

mesenchymal stem cell

NGS

next generation sequencing

PCR

polymerase chain reaction

PT

prothrombin time

ROS

reactive oxygen species

S1P

sphingosine‐1‐phosphate

SRT

substrate reduction therapy

UPR

unfolded protein response

VUS

variant of uncertain significance

Introduction

Gaucher disease (GD) was described by Philippe Gaucher in 1882 on the occasion of his medical doctorate, where he reported the autopsy of a 25‐year‐old woman who had an enlarged spleen without leukemia. Cells in that spleen, since then called “Gaucher cells” [1], were depicted as especially large, with an eccentric nucleus of condensed chromatin and a heterogeneous cytoplasm of “crumpled paper” aspect, also described as wrinkles/striations. These are, in fact, abnormal macrophages, preferential targets of this rare autosomal and recessive genetic disease. Mutation of the gene‐encoding glucocerebrosidase (GCase), GBA1, located on chromosome 1 (1q21), results in a deficient activity of this lysosomal enzyme [2]. GD is thus a lysosomal sphingolipidosis. Lysosomal diseases are congenital, most often linked to a genetic variant impairing the activity of one of the numerous lysosomal enzymes. The overall consequence of these mutations is the cytoplasmic accumulation of substrate(s) normally catabolized by the enzyme involved. The morphology of affected macrophages previously led to the denomination of “overload diseases.” GD is the second most frequent lysosomal disease diagnosed in France, after Fabry disease. Its incidence is about 1/136,000 in the general French population but may reach 1/800 births in the population of Ashkenazi Jews [3]. The median age of first symptoms in the GD French registry [4], which had recorded 457 patients in 2023, is 12 years of age (interquartile range [IQR] 4–32), and the median age of diagnosis is 19 years old (IQR 3–36). GD is a rare disease that has benefited, for more than 30 years, from specific treatment that diversified over time. This unique hindsight in rare diseases allows for good knowledge of therapeutic solutions, yet efficacy remains insufficient, and some major objectives in the management of this condition persist, as described in this review.

Clinical characteristics

GD (ORPHA:355) is characterized by hepato‐splenomegaly, cytopenias, sometimes severe bone lesions, and, in some cases, neurologic lesions. Three main presentations are classically described, respectively, as Type 1, Type 2, and Type 3.

Type 1 GD (MIM #230800, ORPHA:77259) is defined by the absence of any initial neurological involvement. It is the most frequent form of GD (prevalence 90%–95%), the diagnosis of which can be made at any age. Type 1 GD can impair some daily activities but is rarely lethal. Clinical presentation is highly heterogeneous, rarely asymptomatic [5, 6]. The median age at diagnosis is between 10 and 20 years old, depending on reports [5, 6]. Fatigue is frequent (50%), impacting school and socio‐professional life [7]. In children, growth or pubertal retardation can be noted [8]. Although it is a less characteristic sign than in acid sphingomyelinase deficiency (formerly Niemann–Pick disease type B, MIM #607616, ORPHA:77293), it can be noted that high‐density lipoprotein (HDL) cholesterol levels may be reduced in GD [9].

Splenomegaly is present in more than 90% of the patients, sometimes massive and symptomatic. It can be complicated by splenic infarct, but spleen rupture is exceptional. It is important to consider GD when facing a splenomegaly without obvious etiology, especially if it is associated with thrombocytopenia, high ferritin, polyclonal hypergammaglobulinemia, or monoclonal gammopathy of undetermined significance (MGUS). Diagnostic splenectomy should be avoided as it may lead to GD aggravation and is, anyway, not devoid of other risks regardless of the GD context.

About 60%–80% of patients also present with hepatomegaly, rarely evolving toward fibrosis or cirrhosis [5, 6, 10].

Hemorrhages (i.e., spontaneous bruises, epistaxis, dental surgery‐related bleeding), rarely severe, can be present at diagnosis. They are related to thrombocytopenia (60%–90% of the cases), sometimes aggravated by platelet function defect and occasionally coagulation anomalies [5].

Anemia, observed in 30%–50% of the cases, is usually moderate, and leukopenia is rare.

Bone lesions are of variable frequency and severity but may impact quality of life [6, 11, 12]. Bone involvement is most often asymptomatic but may induce chronic diffuse or localized pain [13, 14]. Chronic pain is more frequent in splenectomized or untreated patients. Osteonecrotic lesions can generate localized acute pain crises, sometimes with fever and leukocytosis, thus mimicking osteomyelitis [5, 12], especially in children. They can also be discovered by chance during systematic imaging [8]. In untreated patients, bone crises are reported by 15%–22% of non‐splenectomized patients and 55% of splenectomized patients [4, 15].

For women with severe or untreated GD, pregnancy may increase disease symptomatology and impact both the pregnancy course and childbirth, notably with postpartum hemorrhage [16]. Of note, fertility is not impaired in women with GD.

Contrary to the classical definition of Type 1 GD, some neurological diseases appear to be favored by GD with the appearance of a Parkinsonian syndrome, that is, Parkinson disease or Lewy body dementia. The risk of Parkinsonian syndrome is 10–30 times higher in patients with GD. The phenotype is non‐specific but can be severe with rapid evolution. The risk of developing Parkinson disease and, in this case, its severity could be related to GBA1 gene variants [17, 18, 19].

Type 1 GD is also associated with a moderate but significantly increased risk of lymphoid malignant proliferations, myeloma, and solid cancers, especially hepatocellular and renal carcinomas [20, 21]. However, the risk of developing cancer in patients with GD remains a controversial issue [20].

Other organ involvement is rare but may interest the lungs, heart, or digestive tract. Ophthalmic and orodental lesions are extremely rare.

Type 2 GD, also called “acute neuronopathic GD” (MIM #230900, ORPHA:77260, and MIM #608013, ORPHA:85212 in its perinatal lethal form), is exceptional (<1% of cases) [22]. During the first year of life, patients present a progressive and severe neurological alteration. A suggestive triad associates tonic crises of the nape and trunk (opisthotonos), bulbar signs (notably severe swallowing impairment), and oculomotor paralysis or fixed bilateral strabismus. Psychomotor development is altered. Convulsions, occurring later, are part of a myoclonic epilepsy, resistant to anti‐epileptic therapies. Hepatosplenomegaly is the rule, with thrombocytopenia in 60% of the cases. Ichthyosis can be observed in the earliest forms. These symptoms are completed by dysphagia, stridor by laryngeal spasm, and cachexia. Patients die before the age of 3 of massive aspiration or prolonged apnea. There is no specific treatment for Type 2 GD, which is taken care of by symptomatologic or palliative management [23]. A perinatal lethal form, even more exceptional, also exists that must be confirmed by a biochemical assay in order to provide appropriate genetic counsel.

Also called “juvenile” or “subacute neurological,” Type 3 GD (MIM #231000, ORPHA:77261), which represents 5% of the cases, is associated with the visceral manifestations of Type 1 GD and neurological signs that appear before the age of 20 [24, 25]. An abnormal gaze can be the sole anomaly, usually associated with convergent strabismus. The systematic search of horizontal saccadic eye movements is important, as this symptom does not lead to much complaint from the patient and can be the only neurological symptom, particularly early in the course of the disease. Of note, some patients may experience some difficulty in driving or playing sports due to impairment of rapid tracking of objects. Motor troubles translate into a cerebellar syndrome, tremor, myoclonia (that may look like irregular tremor), dystonia, Parkinsonian, or pyramidal syndrome (with spasticity in severe forms). Epilepsy crises of variable type may co‐exist in the same patient. Progressive myoclonic epilepsy is the most severe form of Type 3 GD. It is a chronic encephalopathy of progressive aggravation that may lead to major handicap, with motor (myoclonia) and cognitive impairment, together with treatment‐resistant epilepsy. Type 3 GD is also frequently, but not systematically, characterized by troubles of intellectual development and learning, with possible autistic symptoms, decreased attention and hyperactivity, and, in adults, cognitive decline. Type 3 GD can finally induce thoracic kyphosis, even in the absence of vertebral compression, corneal opacities, and valvular calcifications in the case of D409H (c.1342G > C) homozygosity.

Given the phenotypic heterogeneity and multi‐organ involvement of GD, assessing its severity must be multifactorial. Although bone and neurological involvement are obvious severity factors, evaluating overall disease burden has been the subject of several studies. The most widely accepted severity score is the DS3 (disease severity scoring system—GD‐DS3) [26], which considers various somatic manifestations (bone involvement, hematological abnormalities, and organomegaly) of Type 1 GD. Although this score is helpful to monitor disease severity and progression and is regularly used in cohort studies, its prognostic value remains moderate. For instance, optimization has been attempted with a more detailed description of bone involvement through a logical grading of lesions [27].

Perspectives and challenges

The heterogeneity of GD clinical presentation often leads to diagnostic delay, linked to the generally poor knowledge of this disease within the medical world. In this context, an international consensus, based on a Delphi methodology, evaluated a panel of clinical and biological signs and two co‐variables (Jewish ascent and family history of GD) to be considered for an early diagnosis of GD. These experts also proposed a score of diagnostic probability based on a ponderation, from 0.5 to 3, of observed clinical and biological parameters. In Type 3 GD, the association of a massive splenomegaly > 3 times the normal size) and gaze palsy are considered the most characteristic. The partial dissociation between the type of GBA1 gene mutation and GD phenotype is another element that contributes to the difficulty in precisely detecting the diagnostic elements useful to appreciate patient prognosis. Furthermore, the role of environmental factors and/or associated genetic modifications in a given patient is still poorly understood. All these factors complicate the diagnosis of a disease both rare and heterogeneous. In this context, it would be useful to establish a cooperative international action in order to obtain large enough cohorts for the analysis of significant subgroups of patients. At the same time, progress is needed to better understand the pathophysiology of GD (see below), including from broader biological analyses, as such an initiative would bring new light. Finally, in the era of unprecedented growth of artificial intelligence applications in the medical world, it is likely that new tools will be developed for the diagnosis [27, 28] or prognosis [29] of such rare diseases [30] as GD, integrating biological data [31] and patient‐reported outcome measures (PROMs) [32].

Pathophysiology

Mutation of the GBA1 gene leads to a deficit in the activity of the GCase, resulting in an accumulation of its substrate, glucosylceramide (GL1) (or glucocerebroside [Gcer]), particularly in macrophages. Gcer is also the substrate of another enzymatic pathway where it is transformed by ceramidase into glucosylsphingosine (also called GL1 or Lyso‐Gb1 or Lyso‐GL1 or LGL1) (Fig. 1a). This deacetylated form of Gcer diffuses in fluids. This pathway is favored in the case of Gcer deficiency, leading to an increase of Lyso‐Gb1 in plasma, making it a specific biomarker of GD. Lyso‐Gb1 is metabolized in the cytoplasm by a GCase (GBA2), active at a neutral pH. This generates sphingosine‐1‐phosphate, expelled from the cell by membrane pumps such as the ATP binding cassette subfamily C member 1 [33].

Fig. 1.

Gaucher disease (GD) schematic pathophysiology. (a) The lysosome is a cell organelle allowing for the degradation of molecules considered worn‐out, issued from cell metabolism, or endocytosed from outside the cell. It contains enzymes active at acidic pH. Deficit in glucocerebrosidase activity results in an accumulation of its substrate, glucosylceramide (or glucocerebroside). Glucosylceramide is also the substrate of another enzymatic pathway where it is transformed by ceramidase in glucosylsphingosine This deacetylated form of glucosylceramide diffuses in liquids. This pathway is favored in case of glucocerebrosidase deficiency, leading to an increase in plasma Lyso‐Gb1, a biomarker specific of GD. Lyso‐Gb1 is metabolized in the cytoplasm by a glucocerebrosidase (GBA2), active at a neutral pH. This generates sphingosine‐1‐phosphate (S1P), the excess of which is excreted by the cell. (b) Increased glucosylceramide gets organized in fibrillar structures that accumulate mainly in Type 2 macrophages. These cells have a specific “crumbled paper” cytoplasmic aspect and are called Gaucher cells. This could be due to the high activity of glucocerebrosidase in these scavenger cells, which produce notably chitotriosidase and CCL18, considered biomarkers of Gaucher cell burden. (c) Gaucher cells infiltrate the bone marrow, spleen, and liver and are considered to be the main contributors to the clinical features of GD, that is, cytopenia, spleen and liver enlargement, and bone lesions. Of note, this is not the case for neurological involvement, as Gaucher cells are not observed in nervous tissue. GCase 1, acid β‐glucosidase or glucocerebrosidase; GCase 2, neutral glucocerebrosidase; GL1, glucosylceramide; GSL, glycosphingolipids; Lyso‐Gb1, glucosylsphingosine.

The peculiar aspect of Gaucher cell cytoplasm, with the typical “crumpled paper” aspect, is related to the presence of Gcer aggregates, forming characteristic fibrillar twisted ribbon structures, clearly visible in electron microscopy [34] (Fig. 1b). Macrophages are rich in lysosomes, organelles originating from the endoplasmic reticulum, delineated by a membrane, and containing acid hydrolases. Lysosomes catabolize, at an acidic pH, macromolecules (proteins, lipids, polysaccharides, and nucleic acids) issued from absorption by the cell of nutriments or antigens, from cellular metabolism, or from the destruction or recycling of other organelles or cell membranes. The specific macrophagic nature of Gaucher cells seems related to the high lysosomal and GCase activity in the monocytic cell lineage, specialized in scavenging. Gcer is a complex glycolipid derived from the degradation of erythrocyte or leukocyte cell membranes, rich in glycosphingolipids. Incidentally, this accumulation enhances resistance to tuberculosis in patients with GD through the anti‐microbial activity of Lyso‐Gb1 [35].

Gaucher cells are related to M2 anti‐inflammatory macrophages through their role in the phagocytosis of aged/abnormal hematopoietic cells (essentially red blood cells) and erythroblast nuclei and secrete specific molecules used as biomarkers, such as chitotriosidase or CCL18 (see the GD biomarkers section). Patients with GD nevertheless present a “pseudo‐inflammatory” status, characterized by modifications of their plasma cytokine profile. The latter would depend on the activation of M1 macrophages in the vicinity of Gaucher cells [36]. Gaucher cell infiltrates, present in the bone marrow, spleen, and liver, are considered to be responsible for clinical symptoms (Fig. 1c). The accumulation of Gcer in bone marrow cells could be the first step toward bone lesions [37]. The involvement of Gaucher cells in the pathophysiology of neurological impairment is less clear, as they have been predominantly observed in perivascular areas in histopathological studies of autopsied brains, occasionally within the parenchyma, and primarily in Types 2 and 3 GD. That these observations mainly concern severe phenotypic forms potentially reflects a selection bias in the patients studied [38]. The role of Gaucher cells in cerebral abnormalities such as neuronal death and gliosis remains poorly understood. Physiologically, within the cell cytoplasm, newly formed GCase in the ER moves to the lysosome lumen through the lysosomal integral membrane protein 2 (LIMP‐2). It is activated by the acidic pH of the lysosome after dissociating from LIMP‐2 (Fig. 2a,b) [34]. LIMP‐2 mutations could modify the phenotype of GD, favoring Type 3.

Fig. 2.

Recent data favoring a more complex pathophysiology. (a) Protein maturation occurs in the endoplasmic reticulum (ER) and the Golgi apparatus. Addressing glucocerebrosidase (GCase) to the lysosome requires a transfer molecule, lysosomal integral membrane protein 2 (LIMP‐2), and the adapters AP‐1 and AP‐3. In the acidic lysosomal compartment, GCase is freed from LIPM‐2 and binds to its co‐factor, saposin C. (b) The lysosome destroys worn‐out cellular molecules and fuses with endosomes to form phagosomes, which digest larger external or internal structures by autophagy. (c) In Gaucher disease (GD), the mutated GCase has reduced activity, leading to the accumulation of glucosylceramide and lysoGL1. (d) Additionally, the mutated GCase presents anomalies in tridimensional folding, favoring its management by chaperones such as Hsp70/90 et Hsp27 that drive it toward the proteasome and participate in its decreased activity. (e) Simultaneously, the accumulation of GCase in the ER, because of its faulty folding, induces a chronic stress of this organelle with three major consequences. First, activation of a repair pathway, UPR (unfolded protein response), which could also be altered in GD, favoring GCase/LIMP‐2 binding, increasing cellular stress. The second is an increase in calcium release by the ER, and the third is an alteration of oxidative metabolism with an overproduction of reactive oxygen species (ROS). (f) Mitochondria are impaired in several ways. The excess calcium alters their membrane potential and the production of ROS, themselves triggering calcium release by the ER, establishing a vicious cycle. Anomalies of inter‐relations between lysosomes and the oxidative metabolism of mitochondria favor an alteration of mitochondria. Because of the dysfunctional autophagy, abnormal, sometimes fragmented, mitochondria are difficult to degrade. (g) Alpha‐synuclein is normally degraded by the proteasome, lysosomes, and mitochondria. In GD, the proteasome is jammed by the elimination of abnormal proteins. Mutated GCase, GL1, and lyso‐GL1 favor the accumulation of α‐synuclein, which organizes in oligomers able to inhibit GCase, increasing the enzymatic deficiency. Accumulating in neurons, they form insoluble aggregates called Lewy bodies. Autophagy and lysosome reformation anomalies decrease the degradation of α‐synuclein molecules, which insert themselves in the membrane of mitochondria, highly involved in the pathophysiology of Parkinson's disease. GL1, glucosylceramide; Lyso‐Gb1, glucosylsphingosine.

Decreased GCase activity is not only the consequence of a modification of its enzymatic domain (qualitative deficiency) but can also result from abnormal addressing to the lysosome or faulty molecular folding (quantitative deficiency) (Fig. 2c,d) [39]. The GBA1 gene mutation indeed leads to an accumulation of altered GCase molecules in the ER, generating a chronic stress triggering the unfolded protein response pathway that manages this protein accumulation [40] (Fig. 2e). Alteration of this pathway in patients with GD would favor the aggregation of GCase and LIMP‐2, increasing the abnormal addressing of GCase toward lysosomes and exacerbating ER stress. Of note, ER stress is also involved in Parkinson's disease [39].

Moreover, in GD models, the stressed ER displays increased calcium release. Mitochondria take up this excess calcium, which leads to an alteration of their membrane potential and the production of reactive oxygen species. The latter also triggers calcium release by the ER, thereby establishing a vicious circle [41]. Excess calcium is also delivered to lysosomes and impacts their interaction with mitochondria, modifying oxidative metabolism (Fig. 2f) [40, 42, 43, 44]. Additionally, mitochondria contain GCase [45]. In GD, mitochondria display metabolism alterations and even fragmentation, favoring their destruction. The latter is normally taken care of by autophagy, consisting of the formation of phagophores around organelles that need to be eliminated. The phagophore then fuses with lysosomes. Alteration of autophagy by abnormal lysosomes therefore also disrupts the elimination of damaged mitochondria and thus cell homeostasis in GD.

GD may result, in exceptional cases without GCase deficiency, from a mutation of the PSAP gene, leading to a deficit in saposin C (a GCase co‐factor), with a clinical pattern similar to Type 3 GD [37]. Lastly, GCase is involved in the metabolism of alpha‐synuclein, which is normally eliminated by the proteasome, lysosomes, and mitochondria. In GD, the proteasome is overwhelmed by protein elimination. The mutated GCase molecules, Gcer, and Lyso‐Gb1 promote the accumulation of alpha‐synuclein molecules, which form oligomers capable of inhibiting the transfer of GCase from the ER to lysosomes, worsening the enzyme deficiency [46, 47]. Accumulating in the cytoplasm of neurons, these oligomers form insoluble aggregates known as Lewy bodies. Autophagy abnormalities and lysosome regeneration impair the degradation of alpha‐synuclein molecules [48], which embed in the mitochondrial membrane, a characteristic also largely involved in the pathophysiology of Parkinson's disease. Additionally, a particularly abundant expression of GCase may exist in dopaminergic neurons [49]. The impact of GCase impairment on the accumulation of alpha‐synuclein is more significant in the case of GBA1 variants (c.1448T > C former L444P or c.1342G > C former D409H) [50] associated with disease severity.

Although the impact of GD on bone tissue is observed in most patients (>80%), its pathophysiology remains ill‐known. Two major hypotheses are considered, related to the bone mass, that is, increased resorption or decreased bone formation [51]. Perturbations of the physiological balance between osteoblasts and osteoclasts have been described in GD. In a murine model of GD, osteopenia is related to a decrease of osteoblast proliferation and differentiation, induced by Gcer and Lyso‐Gb1 [51, 52]. Sphingosine could also be especially toxic for the bone, as GBA2 deletion reverses the GD phenotype, notably in bones. Mesenchymal stem cell (MSC) proliferation is also decreased in GD, as well as the ability of osteoblasts, derived from them, to produce mineral matrix [53]. The few data about GD bone tissue morphology report, moreover, an increase of bone resorption in vivo [51]. The production of osteoclasts in vitro from mononuclear cells obtained from patients with GD is increased compared to that of control mononuclear cells. Several studies ex vivo, using MSC or induced pluripotent stem cells (iPSC) from patients with GD, have shown a deficit in osteogenesis and an alteration of osteoblastic differentiation, notably through a decreased activity of the canonic Wnt pathway and an increase of the negative regulator Dkk‐1 [51, 52, 53, 54]. In summary, it is likely that both hypotheses co‐exist in vivo. GD MSCs have both difficulties in differentiating into osteoblasts and an increased ability to trigger osteoclastogenesis, which would explain some of the bone anomalies of GD. However, the precise relationship between the alteration of GCase and its consequences on bone tissue remains unknown. Besides, the alterations observed do not explain the specific lesions of bone infarct and osteonecrosis [51, 52, 53, 54].

Perspectives and challenges

Even if the broad outlines of the cellular consequences of GBA1 variants have been progressively elucidated, a lot remains to be done to understand which molecular interactions result in the numerous clinical alterations of GD, especially at the osseous and neurological level. GD models [7, 54] have brought some answers through the inactivation of GBA1 ortholog genes in the mouse, zebra fish, medaka fish, or drosophila. Macrophagic, bone, and neurological anomalies may develop in such animals, paving the way for new therapeutic or preventive approaches. In this context, interesting results of cell therapy have already been obtained with iPSC of neuronal cell lines [55]. Broader approaches with analyses of proteomic, transcriptomic, and genomic profiles are lacking that could help identify factors involved in the complex phenotypic expression of the disease.

Diagnosis

The diagnosis of GD is often delayed by several years after the first clinico‐biological manifestations, which may generate complications [56, 57, 58]. Confirmation has long been obtained by evidencing a deficit of GCase activity in a blood sample of from dried blood spots (DBSs) [59]. This allows for the diagnosis, on the same sample, of Niemann‐Pick A/B disease (a deficit in acid sphingomyelinase), the presentation of which can be very similar to that of GD. This technique, rather cheap and easy to perform, could contribute to reduce the delay to reach diagnosis, were it performed more systematically. Dinur et al. have proposed that assaying Lyso‐Gb1 (see below) would allow for more reliable results from DBS [60].

More precise methods, using flow cytometry, allow quantification of the enzymatic activity in specific cell types [6, 61] but are not widely spread for this type of diagnosis and remain unavailable in most laboratories.

In the case of normal GCase activity, although clinical features and biomarkers (such as high levels of chitotriosidase, see below) are in favor of GD, an exceptional deficit in saposin C must be investigated by sequencing the PSAP gene.

After obtaining a firm biological diagnosis, the initial evaluation of GD comprises biomarker assays, GBA genotyping, further biological investigations, and imaging.

GD biomarkers

GD major biomarkers are Lyso‐Gb1, chitotriosidase, and CCL18.

Lyso‐Gb1 is the most recently described marker, clearly elevated in GD, at about 100 times the level observed in healthy controls. Its value correlates to the severity of the disease, that is, thrombocytopenia, liver volume, and bone mineral density [62]. Lyso‐Gb1 is considered to be the most specific biomarker of GD because it is only increased in this disease and could allow to discriminate phenotypic subgroups [63, 64, 65, 66].

As chitotriosidase is mostly produced by Gaucher cells, its assay allows monitoring therapeutic efficacy and might have some prognostic value [67, 68]. However, the baseline activity of chitotriosidase varies among patients, depending on a variant of the CHIT1 gene, which encodes it (duplication of 24 base pairs, c.1049_1072dup24 polymorphism), observed as homozygous in about 6% of the general population and heterozygous in 35%. This variant leads to a total deficiency in chitotriosidase activity in homozygous subjects and a decrease in heterozygous individuals, which impairs the interpretation of enzymatic assays and between‐patient comparisons [69]. Chitotriosidase‐increased activity can also occur in such conditions as other lysosomal disorders (i.e., Niemann–Pick disease), sarcoidosis, or visceral leishmaniasis.

CCL18 is a chemokine produced by macrophages, essentially of M2 type, and dendritic cells. CCL18 favors the recruitment of regulatory T‐cells (Tregs) via the CCR8 receptor [70]. Gaucher cells are responsible for an increase of CCL18 plasma levels, up to 10–50 times compared to healthy controls [71]. They are less variable than those of chitotriosidase, as there is no genetic polymorphism of CLL18. CCL18 kinetics is globally similar to that of chitotriosidase during initial therapy [70], and its assay can be used for follow‐up in the case of chitotriosidase deficiency. However, CCL18 can also increase in chronic inflammatory diseases such as idiopathic pulmonary fibrosis, some cancers, or systemic sclerosis [71].

Of note, the kinetics of these three biomarkers under treatment is generally equivalent, in the current state of knowledge.

Perspectives and challenges

GD biomarkers are interesting tools to confirm therapeutic response and/or compliance, but their predictive performance to estimate disease progression remains unknown. Their large‐scale use in multicentric collaborations is hindered by the variability of techniques and results. Thus, harmonization of assays and/or a procedure allowing to compare laboratory results are prerequisites for the analysis of broad cohorts and potential phenotypic subgroups. A project to this effect is currently underway within the international working group on Gaucher disease (IWGGD) network [https://iwggd.com/]. In practice, explorations can be limited to assaying Lyso‐Gb1, as mentioned above, even on DBS. However, chitotriosidase and CCL18 levels are considered indicators of Gaucher cell mass and could be of particular interest in cases with symptoms directly related to this mass, although knowledge in this area remains incomplete. In this context, CCL18 might be the best choice due to the absence of genetic polymorphism. Finally, identification of new biomarkers with predictive prognostic value remains a challenge in 2025.

Genotyping

Analysis of the GBA1 gene, located on the long arm of Chromosome 1 (1q21), is performed in two steps, first a detection of most frequent variants by polymerase chain reaction, followed by gene sequencing in the case of variant of uncertain significance (VUS). More than 500 variants have been described for this gene. Genotype/Phenotype correlations make it mandatory to determine the genotype, which can provide prognostic information, notably in children, and to establish the risk of appearance of neurological forms of GD. Variant c.1226A > G (previously N370S) [72] excludes the risk of Type 2 or Type 3 GD. Conversely, the c.1448T > C (formerly L444P) variant [4], when homozygous, is usually associated with Type 2 or Type 3 GD. Patients carrying two “null” variants, resulting in a complete absence of enzymatic activity (c.[1448T > C;1483G > C;1497G > C] former RecNciI, c.84dup) do not survive the perinatal period (fetal forms, Type 2 GD). Intra‐familial phenotypic variations are frequent and ill‐explained (possibly through modifier genes). In Ashkenazi Jewish patients with Type 1 GD, variants c.1226A > G (former N370S), c.1448T > C (former L444P), c.84dup (former 84GG), c.115 + 1G > A (former IVS2 + 1G>A), and c.[1448T > C;1483G > C;1497G > C] (former RecNciI) account for more than 96% of the mutated alleles, versus about 60% of total variants in non‐Ashkenazi Caucasian patients. Among Asian patients, neither variant c.1226A > G (former N370S) nor c.84‐85insG are seen, but c.1448T > C (former L444P) and c.754T > A (former F213I) are relatively common. In France, 86% of patients with GD present at least a c.1226A > G (former N370S) variant and 38% at least a c.1448T > C (former L444P) variant. Among them, 21% are homozygous for c.1226A > G (former N370S), 24% are heterozygous composites c.1226A > G (former N370S)/c.1448T>C (former L444P) and 8% are homozygous for c.1448T > C (former L444P) [2, 73].

The IWGGD has published in 2022 precise guidelines for the diagnosis of GD disease, including recommendations for genotyping [74]. Genetic testing can be performed first‐hand as a diagnostic tool, but it is recommended to evidence the enzymatic deficiency first. It is stressed that a GD diagnosis relies on the identification of biallelic pathogenetic variants. Specific algorithms have been published, indicating ways of identifying variants, through Sanger or next‐generation sequencing or even having recourse to RNA multiplex ligation‐probe amplification. Of note, many variants have been described and new VUS should be investigated further to test their pathogenicity. It is also recommended to test the family members of the diagnosed patient and to propose prenatal consulting for genetic counseling, particularly in families with the history of a child affected by a severe form of the disease.

Laboratory investigations

Complete blood count (CBC) often evidences thrombocytopenia (90% of the cases) of variable intensity, liable to be as low as 60 × 10⁹/L platelets. Anemia is less frequent (36% of the cases) and moderate, the hemoglobin level being rarely below 9 g/dL. Nevertheless, low hemoglobin levels should prompt searching for other causes, such as iron or vitamin B12 deficiency or hemolytic anemia [75]. Leukopenia is rare. Cases of GD without thrombocytopenia do exist.

Cytopenias are attributed to splenic sequestration and bone marrow infiltration, but a direct impact of the enzymatic deficiency on immature hematopoietic cells has also been described [76]. Normal CBC can be seen in the case of an antecedent of splenectomy [5, 6]. Bone marrow aspiration, not mandatory to confirm the diagnosis, is often performed in case of isolated thrombocytopenia and/or splenomegaly. It reveals the presence of Gaucher cells, quite evocative when they have a typical morphology, even if present in small numbers [77]. However, it may be difficult to distinguish them from “pseudo‐Gaucher” cells observed in such hematological diseases as multiple myeloma (because of the accumulation of immunoglobulin crystals), Waldenström disease, lymphoproliferative diseases with monoclonal gammopathy, chronic myeloid leukemia, or myelodysplastic syndromes [77, 78, 79, 80, 81]. Atypical mycobacteria infections may also be associated with the presence of similar cells. During the evolution of GD, bone marrow aspiration is legitimate in the case of monoclonal gammopathy or absence of thrombocytopenia improvement under treatment.

Several coagulation anomalies have been described in GD [82], such as prolonged prothrombin time and activated partial thromboplastin time (aPTT), related to factor (F)X, FV, or thrombin deficiency or to liver failure (rare in GD), vitamin K deficiency, genetic condition (von Willebrand disease), or acquired von Willebrand disease [83]. Moreover, low factor XI levels are common in the Ashkenazi population, a characteristic that can also prolong aPTT [84].

However, the relationship with hemorrhagic signs is not obvious, the more so that a platelet function defect is rather frequent [85, 86]. Notably, activation anomalies have been shown to be a cause for platelet dysfunction in patients with GD [87].

Whenever clinical or biological coagulation anomalies are evidenced, seeking specialized counsel is recommended.

Ferritin levels are increased in most patients (>85%), whereas serum iron, transferrin saturation coefficient, and levels of soluble transferrin receptors are normal or decreased [4]. Ferritinemia in GD is not an indication for venesection. Associated hemochromatosis can be investigated if the transferrin saturation coefficient is increased [88].

It is recommended to search for polyclonal hypergammaglobulinemia (25%–91%) and possibly monoclonal gammopathy (1%–35%) because therapy leads to a decrease of the polyclonal hypergammaglobulinemia [20, 21, 89].

Liver tests sometimes disclose cholestasis, rarely transaminitis. HDL‐cholesterol is frequently decreased [9]. C‐reactive protein may increase in case of bone infarct or infectious complication (cholecystitis, rarely osteomyelitis). It is recommended to assay serum calcium and phosphorus as well as vitamin D levels. Indeed, vitamin D deprivation appears to be more frequent in GD than in the general population.

Increases of tartrate‐resistant acid phosphatase and angiotensin conversion enzyme, previously considered GD biomarkers, are now obsolete parameters.

A biochemical prenatal (or preimplantation) assessment of GD can be performed by assaying enzymatic activity in chorial villosities or cultured amniotic cells. This is suggested to couples who had a child with severe Type 1 GD, Type 2, or Type 3 GD, or in view of a possible medical pregnancy termination.

Imaging (Fig. 3)

Fig. 3.

Examples of anomalies observed in imaging. (a) Magnetic resonance imaging (MRI) T1 turbo spin echo (TSE) of a 21‐year‐old untreated patient with Gaucher disease (GD). Left femoral head osteonecrosis with loss of spherocytosis and evolved degenerative remodeling (Ficat classification stage IV). (b) MRI in balanced fast field echo (BFFE). Untreated 29‐year‐old woman. Liver enlargement (20.6 cm), spleen enlargement (25.4 cm). (c) T2 spectral pre‐saturation inversion recovery (SPIR) MRI of a 69‐year‐old splenectomized man with Type 1 Gaucher disease under supplementation regimen. Subcapsular lesion (2.9 cm) of liver segment V (hepatocellular carcinoma). Homogeneous liver. (d) Sagittal scan of an untreated 18‐year‐old woman with Type 3 GD. Thoracic kyphosis, aortic and ascending aorta valvular calcification, and 16 cm splenomegaly. (e) Sagittal MRI of a 63‐year‐old man. Ancient compression fractures of L3, L4, and (f) coronal lower limbs T1 showing sequelae of femoral and tibial bone infarcts for the same patient.

Bone magnetic resonance imaging (MRI) is the reference investigation to appreciate the consequences of GD on bone tissue. Semi‐quantitative approaches using MRI allow for a global quantification of bone marrow infiltration by GD, attributed to Gaucher cells, but remain challenging to standardize [90]. It is nonetheless a valuable (and non‐irradiating) tool to identify bone lesions and monitor their progression, that is, recent edema of aseptic osteonecrosis or bone infarct versus older events. Whole body MRI evaluates at the same time abdominal and bone features. Standard x‐rays may identify, depending on symptoms, localized lesions of lysis or fracture. These x‐rays should not be repeated routinely.

Bone lesions will be detected in more than 80% of the patients and can involve any bone [91]. Several bone lesions may co‐exist, associated with diffuse infiltration, local osteosclerosis, cortical thinning, lytic lesions, or deformities of the femur or humerus. Thus, broadening of the metaphysio‐diaphyseal inferior part of the femur in the shape of an “Erlenmeyer vial” may appear as early as infancy without any functional consequence. This deformity is present, according to various studies, in 20%–80% of adult patients [4, 92].

Bone involvement may less often involve osteonecrosis (15% of the French cohort) [4]. This denomination is preferred to that of avascular necrosis, which tends to erroneously reduce the pathophysiology of these lesions to an ischemic vascular origin [4, 91]. Some authors prefer the wording “bone infarct” in case of osteonecrosis occurring in the metaphysis and diaphysis of long bones or in flat bones, limiting the term aseptic osteonecrosis to epiphyseal lesions [92]. When located in femoral or humeral heads, they may justify prosthetic surgery [5, 6]. Finally, bone involvement can be osteopenia or even osteoporosis, with a risk of long bone fracture related to cortical thinning or spontaneous vertebral compression.

A defect in bone mineralization can be seen as early as 5 years of age [11, 91]. In the cohort from the International Collaborative Gaucher Group (ICGG) registry, 11.6% of children, 41.5% of adolescents, and 21.1% of adults less than 50 years old presented such a defect (Z‐score lower than −2) [93]. After 50 years of age, about 60% of patients with GD show osteopenia (T‐score between −1 and −2.5) and 30% osteoporosis (T‐score ≤−2.5) of the lumbar spine and/or femoral head [5, 94]. By comparison, in the European general population over 50 years of age, the prevalence of osteoporosis is 6.6% in men and 22.1% in women [95]. Osteodensitometry is indicated to assess osteopenia and osteoporosis. The severity of osteopenia could correlate with the genotype, splenomegaly, and hepatomegaly [11, 96].

Abdominal MRI is the best suited investigation to appreciate spleen and liver size and morphology. Organ volume calculated from MRI data is often used in international studies. The spleen sometimes presents a nodular aspect evocative of non‐Hodgkin lymphoma, but these nodules are in fact collections of Gaucher cells or “gaucheromas.” Whenever MRI is not available or feasible (uncontrolled claustrophobia), abdominal sonography can be used.

In the specific case of Type 2 and Type 3 GD, it may be relevant to investigate ocular movements in search of saccadic anomalies and ophthalmoscopy. Auditory evoked and cerebral trunk potentials, as well as brain MRI, can be useful to search for basal ganglia or white matter anomalies. Based on clinical observations, electroencephalograms and neuropsychological tests can also be used. Echocardiography will detect valvular calcifications, pulmonary arterial hypertension, or cardiomyopathy. Finally, in the case of a suspicion of interstitial syndrome, thoracic imaging can be informative.

Perspectives and challenges

Bone involvement other than femoral deformation indicates a phenotype with potentially major functional impact. Precisely detecting and mapping bone infarcts is essential to monitor these patients. Approaches using artificial intelligence are beginning to be used to accurately assess bone involvement in GD and will likely optimize individual evaluation [97, 98].

Treatment

The major difficulty in GD is to treat the disease before the appearance of complications liable to generate invalidating or treatment‐resistant sequelae.

There are currently two GD‐specific therapeutic modalities, substitutive enzymotherapy and substrate reduction [99].

Enzymatic substitution aims at correcting the decreased activity of GCase. Two synthetic recombinant molecules are currently commercialized in Europe: imiglucerase and velaglucerase alpha. A third enzymotherapy, taliglucerase alpha [100], is available in some countries.

The principle of substrate reduction therapy (SRT) is to inhibit the activity of GL1 synthase, the enzyme that produces GL1 from glucose and ceramide. Two SRTs are commercialized: miglustat and eliglustat. A new SRT, liable to cross the blood–brain barrier, venglustat, is currently in clinical development.

In France, the therapeutic indication is validated by experts from a labeled reference center or through multidisciplinary concertation meetings, and 75% of the patients are treated, 65% of them by enzyme replacement therapy (ERT). The decision is discussed with each patient. In 2022, French experts proposed formal criteria for therapeutic indications (Table 1) [5]. In a cohort from the ICGG registry, bone pain was reported by 49% of patients before starting ERT and by only 30% after 1 year of treatment [101]. Pain incidence keeps decreasing after 10 years of treatment. Moreover, 18% of patients in the ICGG registry had reported a bone pain crisis before starting ERT, versus 4.6% after 1 year of treatment. Among the former, 16.7% and 7.4%, respectively, splenectomized or not, presented a crisis within 10 years of therapy [101]. A significant reduction of the risk of pain crisis while taking ERT has recently been confirmed in a prospective, non‐comparative study [13]. Several clinical trials suggest that early initiation of specific therapy limits the risk of bone complications [102, 103]. After 6 years of ERT, bone mineral density increases significantly, especially in children and adults below 30 years of age. Bone mineral density is also improved in older adults, yet, to a lesser magnitude compared to younger patients [94]. Specific treatment by ERT is also able to reduce the prevalence of osteoporosis, as suggested by the study developed in Argentina in a population of 116 patients (median age, 25.8 years) where 10.3% had osteoporosis [104].

Table 1.

Therapeutic indications in Gaucher disease.

Criterion	Therapeutic options
Thrombocytopenia
<50 × 109/L or symptomatic	Treatment
50–100 × 109/L	Case by case in multidisciplinary review
>100 × 109/L	Abstention if no hemorrhage
Symptomatic anemia or ≤10 g/dL hemoglobin	Treatment
Symptomatic splenomegaly	Treatment
Bone involvement a
Painful bone crises	Treatment
Aseptic bone osteonecrosis, bone infarct, pathological, or fragilized bone fractures
Lytic lesions, cortical thinning	Treatment
Osteoporosis
(T‐score ≤−2,5 over 50‐year‐old or Z‐score ≤−3 before 50‐year‐old).	Treatment
Other organ lesion b	Treatment
Type 3 GD	Treatment
Pregnancy	Treatment

^a Bone lesions, except Erlenmeyer ones, are taken into account whatever the moment they appear (before or after GD evaluation for a given patient).

^b Especially interstitial pneumopathy, liver fibrosis, heart lesion.

Theoretically, allogeneic hematopoietic stem cell transplantation (HSCT) should be able to “cure” GD, but the benefit/risk ratio is lower than that of specific therapies, with the exception of patient‐specific Type 3 forms of GD.

HSCT represents a therapeutic option that might definitely restore the production of GCase. This exceptional option is usually proposed for the most severe forms of Type 3 GD, although its efficacy on neurological manifestations is rather modest at best [105, 106]. More recently, HSCT was proposed as a cost‐effective alternative to enzymotherapy in Type 1 GD [107]. However, when specific treatment (ERT or SRT) is available, the benefit/risk ratio of HSCT is usually unfavorable with regard to the risks of induction‐related immunosuppression, long‐term immunosuppressive therapy, and graft‐versus‐host disease. An alternative is to recourse to gene therapy.

Explored since the end of the 1990s, gene therapy would involve the introduction of an unmutated GBA gene in monocytic/macrophagic cells [108, 109]. To do so, two strategies are currently being investigated. The first one involves the collection of CD34⁺ hematopoietic stem cells from the future receiver. Then, the unmutated gene is introduced in these cells, ex vivo, with a viral vector, before reinjecting the manipulated graft. This process avoids the use of immunosuppressors and provided encouraging results in a phase 1/2 trial. The second strategy consists of directly perfusing the patient with a non‐pathogenic vector (lentivirus, adenovirus) carrying the unmutated gene after corticotherapy and immunosuppressive therapy.

Chaperones are small molecules able to modify or stabilize the spatial configuration of proteins, conditioning their functional capacities. They also protect proteins from unwanted links and facilitate their addressing to the lysosome. Chaperones may partially restore intracellular GCase activity, even if the enzyme is mutated, including in the central nervous system. Chaperone therapy is therefore an interesting option for several lysosomal diseases that remain to be confirmed [110, 111]. Among molecules able to exert chaperone activity, an orally administered mucolytic, ambroxol, displays interesting characteristics. It has been shown to increase the residual activity of GCase and is able to cross the blood–brain barrier. With good tolerance, the first clinical results have demonstrated an improvement of neurological symptomatology when this molecule is prescribed at high dose, together with enzymotherapy in Types 2 and 3 GD [112]. Monotherapy with ambroxol can also improve visceral and hematological symptomatology in moderate forms of GD and could constitute an alternative to specific therapy [113].

Perspectives and challenges

What has not yet been identified is the way to single out patients who will progress or who must be treated without waiting for deleterious bone or neurological consequences. In the case of bone lesions, treatments or complementary interventions are necessary. Protection against any bone event still needs to be optimized, as 10% of patients under treatment continue to experience bone infarcts [4]. Besides, drugs crossing the blood–brain barrier would be a great advance for neurological lesions. Another subject of investigation is that of the peculiar tumors that are gaucheromas, probably not explored enough. Finally, the long‐term outcome of patients with GD again requires an international effort of collaboration.

Prognosis

GD‐specific treatments allow, in most cases, the amendment of cytopenias and organomegaly and decrease the impact on bones, yet without suppressing the risk of acute bone complication in all patients, and are also likely to improve patient quality of life.

However, evolution can be unsatisfactory, in spite of specific therapies, because of aggressive and invalidating bone disease, development of a Parkinsonian syndrome, malignant hematological disease, or another cancer, the relative risk of all appearing to be increased. As mentioned above, though, there are biases and controversies about the risk of cancer. Moreover, these risks must be re‐evaluated, taking into account the long‐term consequences of splenectomy, broadly performed before the era of enzymotherapy. With the early specific therapy currently applied, the life expectancy of patients with Type 1 GD is nearing that of the general population [3].

During Type 3 GD, neurologic degradation is progressive, and specific therapies are barely efficient. Life expectancy is usually shortened, but some patients are poorly symptomatic and have no excess mortality. By contrast, Type 2 GD rarely allows for more than 3 years of life.

Perspectives and challenges

Although the lifespan of patients with Type 1 GD under treatment is close to that of the general population, prognostic factors are lacking. Bone involvement is a major functional issue, and 10% of treated patients continue to experience infarcts despite specific treatment. Furthermore, early treatment prevents the risk of severe bone disease, yet some patients never develop any bone involvement in their lifetime. Therefore, having biomarkers for bone risk would be a significant development. Here again, new approaches are being explored.

The current challenge is likely that of the dichotomy of similarities between GD and Parkinson's disease with GBA1 variants. A major therapeutic advance would be to have available therapy that could be initiated early and that would be able to cross the blood–brain barrier, restore the enzyme conformation, and/or increase enzymatic activity and/or limit substrate accumulation. Another option resides in gene therapy that could be initiated before symptoms related to the destruction of dopaminergic neurons occur, both for patients with GD and those carrying a GBA1 variant at risk of Parkinson's disease. These strategies are already being addressed for GD [29], including for better assessment of bone risk [114].

Finally, more broadly, the use of PROMs appears to be an essential strategy to better describe and understand the impact of GD on the daily life of patients, particularly to adapt therapeutic education and orient research aimed at improving quality of life [115, 116, 117].

Conclusion

Type 1 GD is a model among rare diseases, owing to the progress achieved in knowledge about pathophysiology, treatment with ERT, then with a second efficient substrate inhibitor, and long‐term follow‐up. Great challenges remain.

The first challenge is probably to alleviate diagnostic delay, an issue shared by all rare diseases that would benefit from educational initiatives. Integrating rare diseases in etiologic algorithms of the most common key symptoms would help clinicians, even non‐specialists.

A second one relates to understanding the phenotype of the various clinico‐biologic manifestations of GD. This remains crucial, in spite of the important progress demonstrating that it goes beyond the sole influence of Gaucher cells. Better understanding of Type 3 neurologic involvement, of bone lesions and factors favoring it, and of the relationship between GD and Parkinsonian syndromes would allow for the emergence of new therapeutic strategies.

The latter is the third challenge. Despite the efforts of research teams and pharmaceutical companies to try and find molecules crossing the blood–brain barrier, results are currently disappointing, in spite of the small hope brought by chaperones. The parallel advances in gene therapy led to the development of this approach for Type 3 GD as well as for Type 1 GD. Stakes are high also regarding treatment for bone lesions and Parkinson's disease. All these issues are hampered by the relatively small numbers in patient cohorts.

Finally, there is also room for progress regarding biomarkers. Better understanding between known markers and phenotypes would help optimize their use. Again, a large international cohort is lacking, hindered by the lack of standardization and high variability between studies. Experts in GD have been working collaboratively in a network for a long time. The IWGGD serves as the academic network for GD, bringing together multidisciplinary experts involved in the management and research of this condition, as well as representatives of patient associations, united internationally under the International Gaucher Alliance, itself being a complementary driving force of the proposal. Thus, the IWGGD addresses various aspects (medical, biological, societal, etc.) that can contribute to improving patient care. For example, the IWGGD has published its diagnosis guidelines regarding geographical differences in the access to diagnostic services [73] and has initiated such a standardization strategy for biomarker assays that should favor data collection of real‐life information, allowing to better study phenotypic groups. Some resources could also be found in the registries of pharmaceutical companies. Besides, predictive biomarkers of neurological or bone evolution are still lacking.

Nevertheless, research on GD remains quite active and should allow progressively meeting these challenges.

Conflict of interest statement

Fabrice Camou: Consultancies with Sanofi, Takeda, AstraZeneca, Gilead, Pfizer; Scientific Advisory Boards for Sanofi, Pfizer. Marc G. Berger: Scientific presentations for Sanofi‐Genzyme and Takeda.

Camou F, Berger MG. Gaucher disease, state of the art and perspectives. J Intern Med. 2025;298:155–172.