Effect of metreleptin treatment in congenital generalized lipodystrophy: a retrospective analysis from the MENA region

Abstract

Purpose

Congenital generalized lipodystrophy (CGL) is characterized by near-total loss of body fat, leptin deficiency, and severe metabolic complications. Clinical data show that metreleptin improves metabolic control in CGL; however, data from the Middle East and North Africa (MENA) is limited. We assessed the real-world effectiveness and safety of metreleptin in patients with CGL from this region.

Methods

We conducted a multicenter, retrospective chart review of patients with CGL. Baseline, 6 ± 2months (short-term), 2 ± 1years (long-term), and >3 years follow-up data were analyzed. Median changes from baseline were calculated for short- and long-term assessments; >3 years data were reviewed individually. Safety was evaluated up to 90 days after last dose.

Results

Thirty-eight patients were included (82% female; median age at baseline, 6.3 years). Median treatment duration with metreleptin was 25 months (max dose, 10 mg/day). Significant short-term median reductions from baseline were observed for HbA1c (− 1.4%; n = 26), fasting plasma glucose (− 0.5 mmol/L; n = 21); fasting triglycerides (TG; −1.5 mmol/L; n = 26), alanine aminotransferase (− 18 IU/L, n = 27) and aspartate aminotransferase (− 7 IU/L, n = 27) (all p ≤ 0.02). These reductions were significant at long-term follow-up (p < 0.01) except for aspartate aminotransferase. At short-term follow-up, 50.0% of patients achieved  ≥1%  reduction in HbA1c and 65.4% had a  ≥30% reduction in TG; corresponding long-term patient proportions were 46.7% and 61.5%. Most patients with >3 years follow-up showed improvement/maintenance of metabolic parameters. Overall, metreleptin was well tolerated with no new safety signals detected.

Conclusions

Our findings support early intervention with metreleptin for sustained diabetes and hypertriglyceridemia control in CGL.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s40618-026-02977-2.

Introduction

Lipodystrophy syndromes are a group of rare, heterogeneous disorders characterized by a lack of adipose tissue and severe metabolic complications [1, 2]. Adipose tissue loss can affect the whole body (generalized lipodystrophy, GL) or certain body areas (partial lipodystrophy, PL) and can have an acquired or genetic etiology [1, 2]. Therefore, four main types are recognized: familial partial lipodystrophy, acquired partial lipodystrophy, acquired generalized lipodystrophy (AGL), and congenital generalized lipodystrophy (CGL) [1–3]. AGL is often attributed to autoimmune-mediated adipocyte destruction, with adipose tissue loss often occurring in childhood or adolescence [4–6]. In contrast, CGL is a predominantly autosomal recessive disorder associated with pathological variants in genes involved in adipose tissue function, resulting in evident adipose tissue loss usually from birth or during early infancy [7–10].

In CGL, the lack of adipose tissue causes ectopic lipid storage in the liver and muscle. This predisposes patients to metabolic and organ system complications including insulin resistance and refractory diabetes, hypertriglyceridemia, metabolic dysfunction-associated steatohepatitis (MASH), and renal and cardiovascular disease [11–17]. CGL is also associated with leptin deficiency, leading to hyperphagia and the exacerbation of comorbidities [13, 18]. Physical features, including acanthosis nigricans, hirsutism and acromegaloid features, may also occur in some patients with CGL [11, 14].

There are four main CGL subtypes, each associated with pathogenic variants in distinct genes. Variants in the AGPAT2 and the BSCL2 genes (encoding proteins involved in triglyceride synthesis, lipid droplet formation and adipocyte differentiation) cause the most common subtypes, CGL1 and CGL2, respectively [9, 10, 19]. CGL3 and CGL4 are linked to variants in the CAV1 and the CAVIN1/PTRF genes, respectively, which encode proteins involved in intracellular trafficking of fatty acids, lipid droplet assembly, and phospholipid metabolism) [9, 10, 19]. Distinct phenotypes have been documented for each CGL subtype; these include: the loss of metabolically active adipose tissue and preservation of mechanically functioning adipose tissue in CGL1; more severe metabolic disease in CGL2; preservation of mechanical and bone marrow adipose tissue, short stature, and vitamin D resistance in CGL3; and musculoskeletal and gastrointestinal abnormalities, elevated creatine kinase levels, and mild metabolic disease in CGL4 [7, 9, 13].

The standard clinical approach to lipodystrophy includes lifestyle modification (e.g., low-fat diet and exercise) and the use of glucose-lowering, lipid-lowering, and cardiovascular therapies to treat specific comorbidities [1, 20]. Metreleptin, a recombinant analog of human leptin, is the only approved therapy (with diet) to treat the metabolic complications of lipodystrophy in Japan, Europe, Brazil, and Canada (GL and PL) and the USA (GL only) [21–26]. Current multi-society lipodystrophy guidelines recommend metreleptin as a first-line therapy for the metabolic and endocrine abnormalities of GL, supported by outcomes from several clinical and real-world investigations [1, 27–32]. Safety analyses show metreleptin is generally well tolerated in GL, with abdominal pain, hypoglycemia, and nausea the most common adverse events [21, 22, 28].

CGL is often reported in consanguineous families, thus highlighting the importance of evaluating treatment outcomes in regions with high consanguinity, including the Middle East and North Africa (MENA) [14, 33, 34]. Previously, members of our research group evaluated the disease characteristics of metreleptin-naïve patients with CGL from the MENA region [14]. Here, we have conducted a retrospective assessment the clinical and metabolic characteristics of a separate cohort of patients with CGL from this geographical region before and after receiving extended metreleptin treatment. We also report on the safety of metreleptin in these patients.

Methods

Research design

This retrospective, multicenter cohort analysis reviewed medical records of patients with CGL from the MENA region who were treated with metreleptin. Anonymized data were extracted from patient medical records at participating institutions in five countries (Bahrain, Egypt, Oman, Saudi Arabia, and the United Arab Emirates [UAE]).

Patient cohort

CGL was clinically diagnosed by treating physicians, with generalized fat loss as the primary diagnostic indicator. Additional clinical features (e.g., pronounced muscular appearance, acanthosis nigricans, family history and parental consanguinity, evidence of hyperphagia) supported the diagnosis. CGL was subsequently confirmed by genetic testing.

Metreleptin dosing

Metreleptin was self- or parent-administered by daily subcutaneous injection per treating-physician guidance. Initial dosing was based on weight [21]. Doses were adjusted at follow-up visits based on patient response and tolerance. The maximum daily dose was 10 mg/day [21].

Variables collected

Anonymized data from medical records were collected retrospectively. Demographics and clinical characteristics included age, sex, weight, height, body mass index (BMI), lipodystrophy type (e.g., CGL type 1, 2, 3 or 4), age at diagnosis, metreleptin dose, and baseline comorbidities (e.g., hepatic, cardiac, and renal). Metabolic parameters (i.e., glycated hemoglobin [HbA1c], fasting plasma glucose [FPG], fasting triglycerides [TG], alanine aminotransferase [ALT] and aspartate aminotransferase [AST] levels) were collected at baseline (i.e., at last visit prior to metreleptin initiation), at 6 ± 2 months of metreleptin therapy, and at last visit while on metreleptin therapy. Measurements were performed at each participating medical center using standardized methods with appropriate quality control. The upper limits of normal (ULN) for the metabolic parameters were based on previously defined thresholds: ALT (≥40 IU/L), AST (≥40 IU/L), HbA1c (≥6.5%, defined here as a diagnosis of diabetes), FPG (≥7.0 mmol/L), TG ≥ 1.69 mmol/L (≥150 mg/dL, defined here as hypertriglyceridemia) [11, 30, 35]. Treatment-emergent adverse events (TEAE) reported during metreleptin treatment and up to 90 days after the last known metreleptin dose received were also recorded.

Statistical analysis

Analyses were performed using Stata version 15.1 (Stata Corp LLC, Texas, USA). Categorical variables were summarized as counts and proportions for baseline lipodystrophy-associated clinical features and adverse events during metreleptin treatment. Continuous variables were summarized using descriptive statistics (mean ± standard deviation [SD] and median [interquartile range, IQR] values) for baseline characteristics, metreleptin exposure, and changes from baseline for all metabolic parameters.

Only patients with baseline and corresponding follow-up data were included in the statistical analyses presented here. The primary analysis assessed short-term response (6 ± 2 months after metreleptin initiation) from baseline for each metabolic parameter. The long-term response to metreleptin was assessed using follow-up data collected 2 ± 1 years post-metreleptin initiation. As metabolic data were non-normally distributed, the Wilcoxon matched-pairs test was used to compute p-values for group-level short- and long-term changes from baseline post-metreleptin initiation with corresponding 95% confidence intervals. A p-value  <0.05 was considered as statistically significant. Given the variability in follow-up duration, data collected after > 3 years post-treatment were assessed on a per patient basis, according to whether final values decreased or remained below the ULN relative to baseline. The number of evaluable patients with baseline and follow-up data after >3 years was insufficient for statistical testing.

The effect of metreleptin on glycemic control and hypertriglyceridemia was further evaluated by calculating the proportion of patients who achieved reductions of ≥1% in HbA1c and ≥30% in TG from baseline. We also calculated the proportion of individuals with HbA1c ≥6.5% or TG ≥1.69 mmol/L at baseline, short-term, long-term, and >3 years follow-up timepoints [30].

Finally, subgroup analyses assessed short- and long-term changes from baseline in HbA1c and TG in patients with baseline HbA1c  ≥6.5% (subgroup 1), and TG  ≥1.69 mmol/L (subgroup 2).

Results

Baseline patient disposition and demographics

Baseline demographics and clinical characteristics are summarized in Table 1, and reported gene variants are listed in Online Resource Table 1. A total of 38 patients (31 females, 82%; median [IQR] age at baseline, 6.3 [3.4, 15] years) with CGL were included in the final analysis set, with most originating from Saudi Arabia (n = 27, 71%). CGL1 (n = 19, 50%) and CGL2 (n = 14, 37%) were the most common CGL types diagnosed in the cohort, followed by CGL4 (n = 5, 13%). Parental consanguinity was reported for 33 patients (87%). Baseline leptin data were available for 12 patients (32%) and indicated severe leptin deficiency. Recorded data varied across centers, with absolute values reported for seven patients (median [range]: 1.2 ng/ml [0.4–2.1]) and categorical values given for five patients (<0.2 ng/ml, n = 3; <1 ng/ml, n = 1; <3.5 ng/ml, n = 1).

Table 1.

Baseline demographic and clinical characteristics of the patient cohort (N = 38)

Country, n (%)
Bahrain	2 (5.3)
Egypt	2 (5.3)
Oman	5 (13.2)
Saudi Arabia	27 (71.1)
UAE	2 (5.3)
General characteristics
Female, n (%)	31 (81.6)
Age at diagnosis (years), median [IQR]	1.0 [0.3, 6.0]
Age at baseline (years), median [IQR]	6.3 [3.4, 15]
Weight (kg),a median [IQR]	29.1 [16.4, 46.9]
Height (cm),a median [IQR]	130 [104, 158]
BMI (kg/m2),a median [IQR]	16.9 [15.3, 18.6]
BMI Z-score,b median [IQR]	0.11 [− 0.74, 0.85]
Lipodystrophy type, n (%)
CGL1	19 (50.0)
CGL2	14 (36.8)
CGL4	5 (13.2)
Consanguinity, n (%)	33 (86.8)
Initial metreleptin dose (mg/kg)
Mean ± SD	0.07 ± 0.02
Median [IQR]	0.06 [0.06, 0.08]
Hepatic comorbidities, n (%)	34 (89.4)
Hepatomegaly	32 (84.2)
Hepatic steatosis	28 (73.7)
Cardiac comorbidities, n (%)	17 (44.7)
Cardiovascular disease	17 (44.7)
Cardiomyopathy	4 (10.5)
Renal comorbidities, n (%)	15 (39.5)
Metabolic features, median [IQR] or n (%)
HbA1c (%)a	5.9 [4.9, 9.5]
Patients with HbA1c  ≥6.5%	16 (43.2)
FPG (mmol/L)c	5.3 [4.3, 12.2]
Patients with fasting glucose  ≥7.0 mmol/L	11 (35.5)
Fasting TG (mmol/L)d	2.4 [1.2, 4.1]
Patients with TG ≥1.69 mmol/L	24 (70.6)
Patients with TG ≥2.26 mmol/L	18 (52.9)
AST (IU/L)a	41 [26, 69]
Patients with AST  ≥40 IU/L	20 (54.1)
ALT (IU/L)a	50 [32, 100]
Patients with ALT  ≥40 IU/L	23 (62.2)
Background medications, n (%)e
Anti-diabetics	15 (78.9)
Lipid-lowering	8 (40.0)
Anti-hypertensive	4 (20.0)

N = 38 unless otherwise stated. ^an=37; ^bn=35; ^cn=31; ^dn=34; ^en=19

BMI Z-score calculations were not possible for three patients (missing data n = 1; age > 19 n = 2)

ALT, alanine transaminase; AST, aspartate transaminase; BMI, body mass index; CGL, congenital generalized lipodystrophy; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; IQR, interquartile range; IU, international units; SD, standard deviation; TG, triglycerides; UAE, United Arab Emirates

The majority of patients (89%) had hepatic comorbidities at baseline, while cardiovascular (45%) and renal (40%) comorbidities were also common. Forty-three percent of patients had HbA1c  ≥6.5%, 71% had TG  ≥1.69 mmol/L, and 54% and 62% had AST and ALT values  ≥40 IU/L, respectively.

Background medication data were available for 19 patients (50%) in the cohort. Of these, 15 patients were receiving glucose-lowering therapies, and eight patients were receiving lipid-lowering therapies at baseline (Table 1).

Metreleptin exposure and change in medication over follow-up

The median [IQR] duration of metreleptin exposure was 25 [12 to 48] months. The mean (SD) metreleptin dose was 0.07 (0.02) mg/kg/day at treatment initiation, 0.10 (0.10) mg/kg/day at 6-month follow-up and 0.11 (0.04) mg/kg/day at final follow-up. The use of background medications was generally numerically consistent over time. A notable exception was that the proportion of patients receiving insulin was reduced during follow-up, from 53% at baseline (10/19 patients with available data) to 20% at 6 months follow-up (3/15) and 11% at final follow-up (2/18). The proportion of patients on metformin also decreased after 6 months of metreleptin therapy (from 68% [13/19] at baseline to 47% [7/15] at 6 months); however, a reduction was not apparent versus final follow up data (72% [13/18]) (Online Resource Table 2).

Table 2.

Changes in metabolic parameters for the overall patient cohort

Outcome	Follow-up time	n	Pre-metreleptin median [IQR]	Follow-up median [IQR]	Change median (95% CI)	p-value
HbA1c (%)	Short-term	26	6.7 [5.0, 11.2]	5.2 [4.8, 5.9]	–1.4 (–3.9, − 0.3)	< 0.001
	Long-term	15	5.9 [5.2, 12.1]	5.5 [5.2, 8.1]	–0.9 (–4.0, − 0.2)	0.001
FPG (mmol/L)	Short-term	21	5.3 [4.3, 10.5]	4.9 [4.3, 5.3]	–0.5 (–3.0, 0.0)	0.02
	Long-term	10	6.4 [4.6, 15.8]	5.0 [4.7, 6.2]	–1.2 (–87, − 0.1)	0.01
Fasting TG (mmol/L)	Short-term	26	2.8 [1.6, 4.1]	1.1 [0.8, 1.9]	–1.5 (–2.6, − 0.5)	< 0.001
	Long-term	13	2.5 [1.5, 5.7]	1.3 [0.7, 2.1]	–0.8 (–5.2, − 0.4)	0.01
AST (IU/L)	Short-term	27	46 [32, 94]	34 [20, 53]	–7 (–23, − 2)	< 0.001
	Long-term	14	33 [23, 54]	18 [14, 38]	–3 (–18, 2)	0.16
ALT (IU/L)	Short-term	27	64 [32, 127]	32 [20, 55]	–18 (–30, − 8)	< 0.001
	Long-term	14	38 [17, 54]	14 [11, 34]	–10 (–29, − 3)	0.007

Short-term follow-up refers to data recorded at 6 ± 2 months post-metreleptin initiation. Long-term follow-up refers to data recorded at 2 ± 1 years post-metreleptin initiation. ALT, alanine aminotransferase; AST, aspartate aminotransferase; CI, confidence interval; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; IQR, interquartile range; IU, international units; TG, triglycerides

Analysis of metreleptin treatment effects in the overall cohort

HbA1c, FPG, TG, AST and ALT were all significantly reduced after short-term metreleptin use (Table 2; Fig. 1a-e; all p ≤ 0.02). Notably, medians above the ULN at baseline were lowered to below the ULN at short-term follow-up for each metabolic parameter except FPG (Fig. 1b).

Fig. 1.

Metabolic and hepatic parameters at baseline and after short-term metreleptin treatment: HbA1c (a), FPG (b), fasting TG (c), AST (d), and ALT (e). Box and whisker plots show median, IQR, range. Dotted horizontal lines represent ULN (HbA1c: 6.5%; FPG: 7.0 mmol/L; fasting TG: 1.69 mmol/L; AST: 40 IU/L; ALT: 40 IU/L)

Short-term follow-up refers to metabolic readings recorded at 6 ± 2 months post-metreleptin initiation. ALT, alanine transaminase; AST, aspartate transaminase; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; IU, international units; TG, triglycerides; ULN, upper limit of normal

At long-term follow-up, significant metabolic improvements were sustained for HbA1c, FPG, TG and ALT (p ≤ 0.01). AST showed a numerical decrease but did not reach statistical significance (Fig. 2a–e; Table 2).

Fig. 2.

Metabolic and hepatic parameters at baseline and after long-term metreleptin treatment: HbA1c (a), FPG (b), fasting TG (c), AST (d), and ALT (e). Box and whisker plots show median, IQR, range. Dotted horizontal lines represent ULN (HbA1c: 6.5%; FPG: 7.0 mmol/L; fasting TG: 1.69 mmol/L; AST: 40 IU/L; ALT: 40 IU/L). Long-term follow-up refers to metabolic readings recorded at 2 ± 1 years post-metreleptin initiation. ALT, alanine transaminase; AST, aspartate transaminase; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; IU, international units; ns, not significant; TG, triglycerides; ULN, upper limit of normal

In patients with >3 years of follow-up, individual trajectories showed that most patients experienced improvement (i.e., decrease to below ULN) or maintenance (i.e., remained below ULN) in HbA1c (10/15 patients), TG (10/13), FPG (5/7), AST (8/12) and ALT (11/14) (Fig. 3).

Fig. 3.

Levels of metabolic parameters at baseline and at last follow-up in patients with >3 years metreleptin treatment according to CGL type. (a) HbA1c; (b) FPG; (c) fasting TG; (d) AST; (e) ALT. Dotted horizontal lines represent ULN (HbA1c: 6.5%; FPG: 7.0 mmol/L; fasting plasma TG: 1.69 mmol/L; AST: 40 IU/L; ALT: 40 IU/L). ALT, alanine transaminase; AST, aspartate transaminase; CGL, congenital generalized lipodystrophy; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; IU, international units; TG, triglycerides; ULN, upper limit of normal

The effect of metreleptin on glycemic control and hypertriglyceridemia

Around half of evaluable patients achieved a clinically meaningful HbA1c response (≥1% reduction) at short-term (50.0%, 13/26) and long-term follow-up (46.7%, 7/15; Table 3), with around a quarter of patients with >3 years follow-up meeting this threshold (26.7%, 4/15). Consistent with this, the proportion of patients with HbA1c  ≥6.5% also decreased from baseline after short-term (from 50.0% to 15.4%) and long-term (from 46.7% to 26.7%) metreleptin treatment.

Table 3.

Categorical analysis of response to metreleptin treatment in patients with short-term, long-term or >3 years follow-up data

HbA1c, n (%)	Short-term (n = 26)		Long-term (n = 15)		> 3 years follow-up (n = 15)
	Baseline	Follow-up	Baseline	Follow-up	Baseline	Follow-up
≥ 6.5%	13 (50.0)	4 (15.4)	7 (46.7)	4 (26.7)	6 (40.0)	6 (40.0)
≥ 1% reduction from baseline	–	13 (50.0)	–	7 (46.7)	–	4 (26.7)

Fasting TG, n (%)	Short-term (n = 26)		Long-term (n = 13)		> 3 years follow-up (n = 13)
	Baseline	Follow-up	Baseline	Follow-up	Baseline	Follow-up
≥ 1.69 mmol/L	20 (76.9)	8 (30.8)	10 (76.9)	4 (30.8)	10 (76.9)	3 (23.1)
≥ 30% reduction from baseline	–	17 (65.4)	–	8 (61.5)	–	9 (69.2)

Short-term follow-up refers to data recorded at 6 ± 2 months post-metreleptin initiation. Long-term follow-up refers to data recorded at 2 ± 1 years post-metreleptin initiation

HbA1c, glycated hemoglobin; TG, triglycerides

Clinically meaningful TG responses (≥30% reduction) were observed at each follow-up period. The proportion of patients with hypertriglyceridemia (TG  ≥1.69 mmol/L) also decreased from baseline across all follow-up periods (Table 3).

Assessment of short- and long-term effects of metreleptin in patient subgroups with baseline HbA1c  ≥6.5% and TG  ≥1.69 mmol/L

In patients with HbA1c  ≥6.5% (subgroup 1), reductions in HbA1c were greater at short- and long-term follow-up compared with the corresponding values for the overall cohort (Table 4). Reductions in FPG at short-term follow-up were also greater than those in the overall cohort. Long-term changes in FPG were non-significant, possibly due to the small number of patients with evaluable measurements (n = 4).

Table 4.

Changes in HbA1c, FPG, TG, AST, and ALT in patients with baseline HbA1c ≥6.5% [N = 16]

Outcome	Follow-up time	n	Pre-metreleptin median [IQR]	Follow-up median [IQR]	Change median (95% CI)	p-value
HbA1c (%)	Short-term	13	11.1 [8.5, 13.2]	5.6 [5.2, 7.4]	–4.0 (–5.7, − 2.8)	< 0.001
	Long-term	7	12.1 [11.1, 15.0]	8.1 [5.4, 9.0]	–4.0 (–8.2, − 2.6)	0.02
FPG (mmol/L)	Short-term	9	12.2 [6.8, 16.5]	5.1 [4.3, 6.3]	–3.5 (–10.6, − 1.6)	0.01
	Long-term	4	16.5 [10.3, 24.0]	7.0 [4.9, 11.0]	–8.1 (–17.3, − 3.9)	0.13
Fasting TG (mmol/L)	Short-term	12	3.9 [2.1, 6.7]	1.2 [0.9, 2.0]	–2.4 (–4.0, − 1.1)	0.001
	Long-term	4	12.4 [5.2, 18.0]	1.9 [0.9, 5.8]	–8.6 (–15.7, − 3.2)	0.13
AST (IU/L)	Short-term	11	70 [25, 112]	27 [18, 54]	–23 (–77, − 7)	0.003
	Long-term	6	34 [23, 148]	13 [11, 28]	–12 (–227, 16)	0.22
ALT (IU/L)	Short-term	11	94 [50, 147]	32 [20, 73]	–21 (–114, − 8)	0.003
	Long-term	6	49 [29, 155]	12 [11, 21]	–29 (–144, − 8)	0.03

Short-term follow-up refers data recorded at 6 ± 2 months post-metreleptin initiation. Long-term follow-up refers to data recorded at 2 ± 1 years post-metreleptin initiation. Abbreviations: ALT, alanine transaminase; AST, aspartate transaminase; CI, confidence interval; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; IQR, interquartile range; IU, international units; TG, triglycerides

Similarly, short- and long-term median reductions from baseline for TG exceeded those for the overall cohort in patients with hypertriglyceridemia at baseline (TG  ≥1.69 mmol/L; Table 5). Results were consistent when applying a higher triglyceride threshold at baseline (TG  ≥2.26 mmol/L; Online Resource Table 3).

Table 5.

Changes in HbA1c, FPG, fasting TG, AST, and ALT in patients with baseline TG ≥1.69 mmol/L [N = 24]

Outcome	Follow-up time	n	Pre-metreleptin median [IQR]	Follow-up median [IQR]	Change median (95% CI)	p-value
HbA1c (%)	Short-term	20	7.4 [5.1, 10.8]	5.2 [4.8, 6.0]	–2.1 (–4.2, − 0.2)	< 0.001
	Long-term	11	5.9 [5.2, 11.5]	5.4 [4.9, 7.4]	–0.9 (–4.5, − 0.1)	0.005
FPG (mmol/L)	Short-term	15	7.8 [4.4, 14.0]	4.9 [4.3, 5.6]	–2.0 (–6.9, − 0.4)	0.004
	Long-term	7	6.6 [4.4, 15.0]	4.8 [4.3, 5.3]	–1.3 (–8.7, − 0.1)	0.02
Fasting TG (mmol/L)	Short-term	20	3.2 [2.3, 4.6]	1.2 [0.9, 2.0]	–2.1 (–3.0, − 1.1)	< 0.001
	Long-term	10	3.8 [2.0, 9.6]	1.3 [0.7, 2.0]	–2.7 (–9.4, − 0.5)	0.002
AST (IU/L)	Short-term	19	48 [21, 94]	33 [20, 53]	–7 (–36, − 3)	< 0.001
	Long-term	11	28 [17, 40]	18 [14, 37]	–5 (–21, 2)	0.23
ALT (IU/L)	Short-term	19	68 [30, 147]	36 [20, 70]	–18 (–112, − 9)	< 0.001
	Long-term	11	33 [17, 50]	14 [11, 30]	–10 (–31, − 2)	0.01

Short-term follow-up refers to data recorded at 6 ± 2 months post-metreleptin initiation. Long-term follow-up refers to data recorded at 2 ± 1 years post-metreleptin initiation. Abbreviations: ALT, alanine transaminase; AST, aspartate transaminase; CI, confidence interval; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; IQR, interquartile range; IU, international units; TG, triglycerides; ULN, upper limit of normal

Other outcomes

Exploratory clinician-reported hyperphagia assessments were available for 37/38 patients; improvements during follow-up were reported for 21/37 patients (56.8%; Online Resource Table 4).

Safety

Safety data were available for 37/38 patients and are summarized in Online Resource Table 5. Four patients (10.5%) experienced a TEAE while receiving metreleptin or within 90 days after the last dose. Epistaxis and weight loss were the most commonly observed adverse events, each reported in two patients. Since data collection, one patient with CGL1 died from an upper respiratory tract infection complicated by cardiac arrest; the event was assessed as not related to treatment (metreleptin had been discontinued 5 weeks prior).

Discussion

To our knowledge, this is the first real-world retrospective assessment of metreleptin effectiveness and safety in patients with CGL from the MENA region. Metreleptin was associated with early and sustained improvements in glycemic control, TG and liver enzymes.

Consistent with previous CGL studies, our cohort was predominantly female, which may be attributable to females presenting with more severe metabolic and early onset metabolic disease than males [17, 28]. Furthermore, adipose tissue loss may be less recognizable in males given that a muscular habitus is common in men without lipodystrophy [2]. Baseline characteristics show that CGL-related metabolic and organ complications can occur early in childhood. Most patients had hepatic comorbidities (89%) prior to metreleptin initiation, and substantial proportions had cardiovascular or renal comorbidities (45% and 40%). Diabetes and hypertriglyceridemia were common (43% and 71% of patients). Studies in metreleptin-naïve GL populations also reported high rates of diabetes and hypertriglyceridemia, with onset typically in childhood or adolescence [14–17, 30]. Notably, an international chart review of 230 patients with lipodystrophy (81 patients with GL) from Brazil, Turkey, and the USA reported that initial GL symptoms occur at an average age of 9.2 years [11]. In a previous MENA study by members of our research group, organ system abnormalities were frequently documented in children <12 years [14]. These findings highlight the need to manage the metabolic derangements from a young age to reduce the risk of severe comorbidities [1, 17, 20, 27, 35]. In this regard, the multi-society practice guidelines for lipodystrophy recommend metreleptin as a first-line therapy for the metabolic complications of GL, including in pediatric patients [1].

Statistically significant reductions from baseline were observed in the cohort for all metabolic parameters after short-term metreleptin treatment. The US National Institute of Health (NIH) clinical study for metreleptin in GL showed a ≥1% decrease in HbA1c in 35% of patients and a ≥30% decrease in TG in 63% of patients after 4 months of metreleptin treatment [28]; these findings are broadly comparable to the short-term response rates in the cohort presented here (50% for HbA1c; 65% for TG). Evidence from Japan also suggests that therapeutic effects of metreleptin in GL may be seen as early as 1-week post-initiation [36].

Early intervention with metreleptin may support long-term metabolic control in CGL. In this analysis, significant improvements in HbA1c, FPG, TG, and ALT were sustained at long-term follow-up, and most patients with > 3 years of follow-up showed improvements or stabilization for each metabolic parameter relative to baseline. In support of this, a recent retrospective study has shown that patients starting metreleptin before severe metabolic complications develop have better long-term control of diabetes, proteinuria, hypertriglyceridemia, and liver enzyme levels than those who start therapy after severe disease onset [35].

Other real-world studies have demonstrated the long-term effectiveness of metreleptin. A Spanish case series of seven GL patients on metreleptin for up to 63 months reported substantial improvements in glycemic control (HbA1c and FPG) as well as TG and liver enzymes [27]. A study of 28 GL patients from France (n = 28) similarly recorded significant median reductions relative to baseline after ≥15 months metreleptin treatment for HbA1c (from 8.4% to 6.9%) and TG (from 3.6 to 1.9 mmol/L) [30]. The findings from our sub-group analyses of patients with baseline HbA1c ≥6.5% or TG ≥1.69 mmol/L also support greater responses to metreleptin in patients with more severe metabolic disease at baseline [28, 29].

Due to the small sample size of this cohort, particularly for the final follow-up, genotype-specific responses to metreleptin could not be fully explored. However, previous studies have shown that metreleptin therapy is associated with lowering of HbA1c and TGs in CGL types 1–4 [37].

Metreleptin was generally well tolerated in our cohort over the extended treatment period with no new safety signals observed [28, 38]. The reported TEAEs were broadly consistent with the expected pharmacological profile of metreleptin [28]. Some patients with >3 years’ follow-up data had increases in HbA1c and TG, which may reflect decreased drug adherence and/or drug supply issues at the time of last follow-up. It is also plausible that a subset of these patients experienced tachyphylaxis due to the development of anti-drug antibodies to metreleptin with neutralizing activity [39]; however, the development of neutralizing antibodies to metreleptin was not confirmed in our cohort.

There were some limitations in our analysis. As a retrospective observational analysis, it is subject to inherent biases such as incomplete data capture and variability in clinical practice. Follow-up and data collection were not formally standardized, and data were captured at up to two follow-up timepoints per patient. Missing data reduced evaluable sample sizes, particularly for longer-term follow-up assessments. It should be noted that all efforts were made to standardize data prior to analysis. Inter-center differences in clinical management meant that treatment decisions varied across the cohort, while adherence/supply issues or changes in concomitant medications may have influenced outcomes. However, background anti-diabetic and/or lipid-lowering therapies broadly remained stable or decreased during follow-up, making it unlikely that intensification of background therapies significantly influenced the results. Although comparable with other lipodystrophy cohorts, the female predominance in the cohort may limit the generalizability of our findings. Data on formal patient-reported outcomes were not collected as part of this study.

Conclusion

This analysis provides the first real-world evidence on the effectiveness and safety of metreleptin in managing metabolic complications in patients with CGL from the MENA region. Our findings support prior research showing that early and continuous metreleptin therapy helps achieve and maintain metabolic control in CGL. Further studies with larger sample sizes from additional geographical regions may support our findings.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization, Methodology, Validation, Formal Analysis, Investigation, Data Curation, Writing – Original Draft, Writing – Review and Editing, Approval of final draft: Majid Alfadhel, Afaf Alsagheir, Azza Al Shidhani, Ahmed Ali Nahari, Yasser Binafif, Najya Attia, Nasser Aljuhani, Khulood Najdeyah, Haya Alkhayyat, Jamal Al Jubeh, Imad Brema, Najah Douba, Elham Alamiri, Mouza Al Yahyaei, Saleha Babli, Mohammed Alqahtani, Amal Al-Qahtani, Ibrahim Al Alwan, Saif Al Yaarubi, Rebecca J. Brown. Conceptualization, Methodology, Writing – Review and Editing, Approval of final draft: Nadia Hergli.

Data availability

All data supporting the findings of this study are available within the paper and its Supplementary Information.

Declarations

Competing Interests

NH is an employee of Chiesi Farmaceutici. SAY reports acting as a speaker for Chiesi Farmaceutici. RJB has received research support from Chiesi and funding from the Intramural Research Program of the National Institute of Diabetes and Digestive and Kidney Diseases and has also received research support or drug for studies from Regeneron, Marea, and Pfizer. All remaining authors report nothing to disclose.

Ethics approval

Approval of this retrospective analysis was obtained from the ethics committee or institutional review board at participating medical centers, as required. This analysis adhered to the tenets of the Declaration of Helsinki [40]. Except for two patients, local regulations did not require written consent for the use of retrospective anonymized (unidentifiable) data; however, all patients (or their legal guardians) provided written consent for genetic testing. The remaining two patients previously provided written consent to participate in a prospective, open-label interventional studies of metreleptin conducted at the US National Institute of Health (NIH) Clinical Center, Bethesda, Maryland. Both patients subsequently transitioned to commercial metreleptin and continued to receive medical care at the NIH. Anonymized, retrospective data from both patients were included in the current analysis.