Gaucher disease (GD; ORPHA: 355) is a rare lysosomal disorder characterized by an abnormal accumulation of lipids in the cells of the phagocytic system. GD is caused by mutations in the GBA1 gene that results in decreased catalytic activity of the enzyme glucosylceramide-β-glucosidase (glucocerebrosidase) that leads to accumulation of glucosylceramide in the lysosomes of macrophage cells. The inheritance pattern is autosomal recessive and more than 250 mutations (mostly single nucleotide substitutions) responsible for the enzymatic deficiency have been identified in the human GBA1 gene (chromosome 1q22).1 There is a clinical classification of GD based on the presence and severity of neurological involvement. Infantile GD (type 2) is very rare and causes death in the first 2 years of life. In GD type 3 the neurological affectation is less severe and not always fatal and often improves with bone marrow transplantation and enzyme replacement therapy. The most common form is type 1, in which the central nervous system is not affected and the major symptoms are enlargement of spleen and liver, infiltration of the bone marrow cells by storage cells, and bone disease.1

The pathophysiology of bone manifestations in GD is not fully understood, but involvement of mineralized bone and bone marrow has been reported. The accumulation of Gaucher cells (macrophages filled with glucocerebroside) in the bone marrow appears to be the initial step in the pathological process.1

This accumulation causes expansion of the red marrow, which displaces the yellow marrow. As a result, yellow marrow is progressively pushed to the periphery, leading to vascular compression and occlusion, which are responsible for bone infarctions.2,3

The buildup of glucocerebroside also triggers an inflammatory cascade through macrophage activation, resulting in the release of various inflammatory mediators such as interleukins IL-1, IL-6, and tumor necrosis factor alpha (TNF-α). These affect osteoclast and osteoblast activity.1

Consequently, the complex interactions between bone marrow, immune cells, and bone cells, combined with altered vascularization and the action of cytokines, contribute to the skeletal manifestations observed in GD.2,4

Specific therapies for GD have been shown to reduce bone marrow infiltration and the frequency of bone crises in patients with GD. Nevertheless, skeletal manifestations often respond more slowly than visceral or hematologic symptoms, and improvements in bone mineral density (BMD) may remain incomplete.5,6 In light of these findings, we decided to conduct a study aimed at describing bone involvement in the Tunisian registry of GD patients and assessing specific therapies for GD impact on bone structural changes.

Bone involvement, observed in 65% of our cohort, was the third most common clinical feature of GD, yet it was rarely the primary indication for initiating specific therapies, such as enzyme ERT or SRT. Instead, treatment was typically prompted by visceral or hematologic manifestations, such as hepatosplenomegaly or cytopenias, with bone complications often identified later through systematic screening. Delayed initiation of ERT or SRT, primarily due to late diagnosis and limited access to treatment in our North African setting, likely contributed to suboptimal bone outcomes, as skeletal manifestations are known to respond more slowly than visceral or hematologic symptoms.5,6 These findings underscore the importance of early and systematic bone disease screening to optimize therapeutic impact on skeletal health. In this context, we present descriptive data on bone involvement derived from the Tunisian national Gaucher disease registry, highlighting its frequency and characteristics in our cohort.

The present case series of 74 Tunisian patients with GD provided the following findings: (i) bone radiological deformities were frequent (76%) even in patients without any clinical symptoms; (ii) primary bone lesions in our series included Erlenmeyer flask deformity of the femur (5%) and lytic lesions (3%); (iii) decreased bone density was described in 24 GD patients (32%) and specific disease treatment demonstrated a trend toward improved bone mineral density, although formal statistical testing was not feasible due to the small sample size and heterogeneity of treated subgroups.

In a literature review of GD bone involvement, Mikosch and Hughes classified these manifestations into three clusters based on the underlying mechanisms.3 The first cluster represents patients with primary bone lesions due to cytokine expression changes or increased local pressure from glucocerebroside accumulation in the bone marrow. The second one represents patients with secondary bone lesions that result from complex pathological mechanisms mainly related to vascular damage and increased local pressure. The third one represents patients with tertiary bone lesions due to chronic bone change.3,4

In our cohort, we observed a significant discrepancy between clinical findings and imaging results: 55% of patients had at least one radiological bone lesion, while only 23% reported bone pain.

This mismatch has also been reported in previous cohort studies,7,8 highlighting the importance of systematic bone disease screening, not only at diagnosis, but also throughout follow-up to detect asymptomatic lesions and enable early intervention.

MRI appears to be the gold standard for evaluating structural bone abnormalities in GD, whereas DXA scanning remains essential for assessing BMD. In settings where these are not available, standard X-rays and bone scintigraphy may serve as alternative tools.1

Bone marrow infiltration, a triggering factor for bone abnormalities in GD, was present in 70% of our patients on bone MRI. Erlenmeyer flask deformity was less frequently observed in our cohort compared to other registries, such as the French registry (20.9%)9 and the International Collaborative Gaucher Group (ICGG) Registry (40%).7 However, it is important to note that only 27% of our patients underwent femoral X-rays, which may have led to an underestimation of the true prevalence of this deformity.

According to the literature, this characteristic deformity is thought to result from the loss of metaphyseal concavity and abnormal widening of the metaphysis, likely due to disrupted bone remodeling processes involving cytokine dysregulation. While most commonly seen in the femur, as documented in our patients, it may also affect other long bones, particularly the tibia.3

That said, clinicians should always consider other differential diagnoses for this radiological finding, including infantile osteopetrosis, renal tubular acidosis, otopalatodigital syndrome, and Melnick–Needles syndrome.10

X-rays allowed screening for lytic bone lesions as well that have been previously reported in the French Registry (9.4%)9 and the ICGG Registry (8%).7 Yet, it remains crucial to rule out more common differential causes in this context, especially malignancies as it was the case in our cohort.

Our findings regarding BMD are in line with the literature with osteopenia occurring in between 26 and 42% of patients7,11 and osteoporosis in around 12%.11

A previous study suggested risk factors for bone density loss the N370/84GG genotype, a history of splenectomy and a history of hepatomegaly.12 Such correlations could not be assessed due to the small size of patients screened for each mutation. A case of diffuse osteosclerosis of the lumbar spine and pelvis was noted in one female patient. No established link between diffuse osteosclerosis and GD has been found in the literature, suggesting that this finding is likely unrelated to GD.

Osteonecrosis, a secondary bone lesion, was seen in 22% of our patients which is also consistent with data from the ICGG Registry (25%)7 and the Spanish cohort (17%).11

No tertiary bone lesions were observed in our patients, likely because most were evaluated at initial consultation, before chronic changes developed.

Other skeletal abnormalities were reported in the literature but not in our series include: bone fractures,8 cortical erosion (7%)9 and sequelae of bone infarctions (14%).9

N370S was the most frequent GBA mutation in both the total cohort and the bone involvement patients subgroup, which is similar to internal studies findings.1,7 It has been shown that this mutation is associated with bone damage.13 In our cohort, the small number of patients who underwent genetic assessment did not allow such study.

Although diffuse osteosclerosis and monoclonal gammopathies are not primary features of Gaucher disease bone involvement, clinicians should remain aware of differential diagnoses when evaluating bone abnormalities. In particular, conditions such as multiple myeloma or other metabolic bone disorders should be considered, especially in patients presenting with atypical imaging findings or abnormal serum protein electrophoresis. Rarely, Gaucher disease may coexist with monoclonal gammopathy or multiple myeloma, highlighting the need for careful diagnostic evaluation.14

In our study, ERT with velaglucerase alfa demonstrated trends toward improvement in clinical symptoms, specifically bone pain, while both velaglucerase alfa and eliglustat showed trends toward improvement in BMD parameters. However, the limited number of treated patients and the descriptive nature of the analysis precluded formal statistical evaluation.

These observations align with the notion that primary bone manifestations in Gaucher disease may be partially reversible, in contrast to secondary or tertiary lesions.3 Comparative literature supports these findings, with several studies reporting enhancements in BMD metrics, particularly lumbar and femoral Z-scores, under ERT and substrate reduction therapy.

For eliglustat, a phase 2 open-label trial by Lukina et al. involving treatment-naïve patients with type 1 Gaucher disease showed a mean increase in lumbar spine Z-score of +0.69 (from baseline −1.81 to −1.12) and femoral neck Z-score of +0.31 (from baseline −1.46 to −1.15) after 13 months of therapy, with statistically significant improvements (p<0.001).15 An 8-year extension of this trial reported further gains, with mean lumbar spine Z-score increases of +1.61 and femoral neck Z-score increases of +1.02 from baseline, achieving normalization in many patients.16

Regarding velaglucerase alfa, Zimran et al. described achievement of therapeutic goals in treatment-naïve patients by 4 years in phase 3 trials, including mean lumbar spine Z-score improvements of +1.2 and femoral neck Z-score improvements of +0.8, with sustained benefits even after dose reduction.17 Similarly, Elstein et al. observed continuous BMD gains over 69 months, with mean lumbar spine Z-score increases of +1.46 (p<0.001) and femoral neck Z-score increases of +0.93 (p<0.001), highlighting long-term efficacy.18

For imiglucerase, Cappellini et al. analyzed long-term outcomes in Italian patients with type 1 or 3 Gaucher disease from the International Collaborative Gaucher Group Registry, noting mean lumbar spine Z-score improvements of +1.1 and femoral neck Z-score improvements of +0.7 over extended follow-up periods (up to 20 years), with greater gains in those without prior splenectomy.19 Sims et al. reported comparable results in a 48-month cohort study, with mean lumbar spine Z-score increases of +0.6 (p<0.001) and femoral neck Z-score increases of +0.4 (p<0.001) in patients with skeletal manifestations.20

These studies collectively underscore the potential for BMD recovery with targeted therapies, though response varies by treatment duration, patient baseline characteristics, and access to early intervention, as observed in our North African cohort.

Our study has several limitations. First, its retrospective design did not allow for standardized data collection. Second, bone assessments were not systematically standardized across all patients, leading to heterogeneous and incomplete data. Lastly, and due to the limited resources, not all patients received specific therapies which also made it more difficult to assess its real value in regard to bone damage.

This study provided unique insights into bone involvement in GD within an underrepresented North African cohort, where consanguinity and delayed diagnosis are prevalent, contributing to the global understanding of this rare lysosomal storage disorder.

Bone involvement was frequent in our cohort. Bone density showed a trend towards improvement in patients treated with velaglucerase alfa and eliglustat although formal statistical testing was not feasible due to small and heterogeneous subgroups. It is also worth noting the late start substitutive treatment in our patients due to diagnosis and access to therapies delay. Even though it is rare, rheumatologists must always keep in mind this disease especially when other diseases are ruled out, especially in a context of consanguinity and a history of cytopenias.

HA: methodology, data curation, formal analysis, & writing – editing and original draft; ZM: conceptualization, editing, reviewing; NBY: data curation formal analysis, editing; HB: supervision, reviewing; MZ, CA, AO, ST, SK, ABC, KB: data curation; MSA: project administration, validation, resources; SBH: project administration, validation, reviewing; TL: reviewing & validation.

No funds were needed to conduct this study.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.