Influence of diet and body weight in treatment-resistant acquired partial lipodystrophy after hematopoietic stem cell transplantation and its potential for metabolic improvement

Abstract

Lipodystrophy is a rare disease characterized by various metabolic complications resulting from the complete or partial loss of adipose tissues and abnormal fat accumulation. Acquired lipodystrophy may occur due to certain drugs, autoimmunity or for unknown reasons. Recently, cases of acquired lipodystrophy after hematopoietic stem cell transplantation (HSCT) have been reported. Leptin administration, used recently to treat generalized lipodystrophy, effectively controlled metabolic complications; however, few reports demonstrated the effectiveness of leptin for acquired partial lipodystrophy. In this report, we present the case of a 17-year-old woman who developed insulin resistance, hypertriglyceridemia, and fatty liver after HSCT. Due to her thin gluteal fat and low blood adiponectin levels, her metabolic abnormalities were attributed to partial lipodystrophy. While both leptin and pemafibrate administration partially attenuated metabolic abnormalities, its effects were relatively limited, probably because the serum leptin levels were maintained, which is not likely in generalized lipodystrophy. Nevertheless, after she developed adjustment disorder and experienced weight loss, along with decreased food intake, her metabolic markers significantly improved. This case suggests the modest effect of leptin and permafibrate in partial lipodystrophy after HSCT, highlighting the importance of diet therapy in metreleptin treatment for acquired partial lipodystrophy.

Introduction

Lipodystrophy is defined as the complete or partial loss of adipose tissue, and includes genetic and acquired diseases [1, 2]. In congenital generalized lipodystrophy, genes related to lipid synthesis, such as AGPAT2 and BSCL2, are affected, leading to a near absence of adipose tissue. On the other hand, in familial partial lipodystrophy, is characterized by the loss of fat from the limbs with abnormal fat accumulation in the trunk and visceral regions, mutations in the LMNA gene commonly occur [3]. Another implicated cause of acquired lipodystrophy is medication, such as HIV treatment [4] and immune checkpoint inhibitors [5].

Reportedly, patients who undergo hematopoietic stem cell transplantation (HSCT) develop impaired glucose tolerance and hyperlipidemia [6–8]. Metabolic complications reportedly occurred in a childhood HSCT survivor exhibiting lipodystrophy and central fat accumulation [9, 10]. More importantly, Adachi et al. described the clinical features of lipodystrophy after HSCT, which resembled Dunnigan-type familial partial lipodystrophy [11].

Regardless of the type of lipodystrophy, patients often develop various levels of insulin resistance and hyperlipidemia and/or diabetes, likely due to reduced levels of adipocytokines, including leptin and adiponectin [12, 13]. Leptin administration effectively attenuated metabolic complications in patients with lipodystrophy [12]. Some predictors of treatment responses include generalized rather than partial lipodystrophy, higher triglyceride (> 500 mg/dL in partial lipodystrophy) and lower leptin (< 4 ng/mL) levels at baseline. In partial lipodystrophy after HSCT, the average serum leptin level is 12.6 ng/mL, and the effects of leptin administration are limited [11]. Nevertheless, two previous case reports described the successful treatment of uncontrollable metabolic complications with metreleptin, a recombinant leptin analog, in partial lipodystrophy after HSCT [14, 15].

Case report

A 17-year-old woman contracted acute myeloblastic leukemia with maturation at the age of 6 years. She underwent high-dose cyclophosphamide and total body irradiation, followed by allogeneic HSCT. One year later, she developed chronic graft-versus-host disease (GVHD) and was prescribed higher doses of prednisolone than the typical post-HSCT dose. After 1 year of treatment, the GVHD was successfully cured, and steroid treatment was discontinued.

Approximately 5–6 years after HSCT, she developed hypertriglyceridemia and hypercholesterolemia. Her pediatrician prescribed 15 mg/day of simvastatin; however, dyslipidemia was uncontrollable. At the age of 17 years, she visited our department. On admission, her height was 149.4 cm; body weight 41 kg; and body mass index (BMI) was 18.4. She had a moon face-like appearance, characterized by rich buccal fat and a double chin; however, Cushing’s signs were undetected. Conversely, her arms and legs were slim, with low subcutaneous fat. Additionally, her gluteal fat was very thin; she experienced buttock pain when sitting. Acanthosis nigricans was not evident, and no heart murmur was detected.

Blood tests revealed elevated levels of alanine aminotransferase (ALT), uric acid, HbA1c, and triglycerides. Lipoprotein fractionation showed an increase in the pre-beta fraction (Table 1). Furthermore, hepatorenal echo contrast was increased on ultrasonography, even though hepatomegaly was not apparent (Fig. 1a). An oral glucose tolerance test indicated hyperinsulinemia (Fig. 1b, dashed line) and severe insulin resistance (HOMA-R was 4.07). Her plasma corticotropin, serum cortisol, dehydroepiandrosterone sulfate, growth hormone, and insulin-like growth factor 1 levels were within normal limits. Her serum cortisol level decreased to 2.2 µg/dL at midnight, indicating no autonomous secretion. She had no hyper- or hypothyroidism. She was receiving estradiol and dydrogesterone for Kaufmann therapy.

Table 1.

Laboratory findings

Findings	Before treatment	After treatment	Reference range
Physical findings
Body mass index (kg/m2)	18.4	19.6	NA
Systolic blood pressure (mmHg)	105	127	NA
Diastolic blood pressure (mmHg)	77	81	NA
Biochemical test
Total bilirubin (mg/dL)	0.9	0.3	0.3–1.2
AST (IU/L)	27	18	13–33
ALT (IU/L)	37	22	6–27
γ-GTP (IU/L)	39	29	10–47
Uric acid (mg/dL)	7.4	6.6	2.6–7.0
FPG (mg/dL)	103	100	80–110
HbA1c (%)	6.3	6.2	4.6–6.2
LDL cholesterol (mg/dL)	96	148	59–139
HDL cholesterol (mg/dL)	46	50	45–67
Triglycerides (mg/dL)	413	403	30–149
M2BPGi	0.48	0.48	< 1.00
Lipoprotein (agarose gel electrophoresis)
α fraction (%)	27.8	26.9	32.6–52.5
Pre-β fraction (%)	28.9	27.7	6.6–20.8
β fraction (%)	38.7	44.3	33.6–52.0
Origin (%)	1.8	0.5	NA
Endocrinology
ACTH (pg/mL)	26	NA	7.2–63.3
Cortisol (µg/dL)	10.9	NA	3.0–19.6
DHEA-S (ng/mL)	2413	NA	510–3210
IGF-1 (ng/mL)	246	NA	175–488
TSH (µU/mL)	1.01	NA	0.35–4.94
Free T4 (ng/dL)	0.89	NA	0.70–1.48
Urine c-peptide (µg/day)	32.67	26.78	NA
Adiponectin (µg/mL)	2.0	2.3	> 4.0
Leptin (ng/mL)	14.8	84.6	NA

All blood samples were collected following an overnight fast

AST aspartate transaminase, ALT alanine aminotransferase, γ-GTP γ-glutamyl transpeptidase, FPG fasting plasma glucose, LDL low-density lipoprotein, HDL high-density lipoprotein, M2BPGi mac-2 binding protein glycosylation isomer, ACTH adrenocorticotropic hormone, DHEA-S dehydroepiandrosterone sulfate, IGF insulin-like growth factor, TSH thyroid-stimulating hormone, T4 thyroxine, NA not available

Fig. 1.

Impaired fat accumulation in the body, sequential changes in administered drugs, and metabolic measurements. a Hepatorenal echo contrast in ultrasonography. b Oral glucose tolerance test before and after the treatment with pemafibrate and metreleptin. Dashed line; before the treatment (-pre). Solid line; after the treatment (-post). Black line; plasma glucose (PG), Gray line; serum insulin (IRI). c Magnetic resonance imaging of the body. The MRI imaging was conducted under T1-weighted conditions. Left panel; abdominal and gluteal fat. Middle panel; subcutaneous fat in the arms. Right panel; subcutaneous fat in the legs. d Upper panel; drug administration history. Lower panel; sequential changes in serum triglycerides (TG, black, dashed line, square markers), total cholesterol (T-cho, gray, dotted line, diamond markers), and LDL cholesterol (LDL-cho, gray, dashed line, round markers) levels under ad libitum conditions. e Upper panel; drug administration history. Lower panel; sequential changes in plasma glucose (PG, black, dotted line, square markers), serum insulin (IRI, black, dashed line, round markers), HbA1c (gray, dotted line, diamond markers) and body weight (BW, gray, solid line, square markers) levels under ad libitum conditions

We hypothesized that her metabolic complications without obesity were due to unbalanced distribution of body fat, possibly influencing insulin resistance. Therefore, we conducted magnetic resonance imaging (MRI) and found that the gluteal subcutaneous fat was markedly thinner than was the abdominal subcutaneous fat (Fig. 1c). The area of visceral fat measured by the impedance method was 72 cm². Subcutaneous fat in the legs was markedly thinner than that in the arms. Based on these results, we concluded that her metabolic complications were attributed to partial lipodystrophy after the HSCT treatment. Notably, her serum leptin level was within normal range (14.8 ng/mL), while her adiponectin level was almost 50% of the minimum range of normal individuals (2.0 µg/mL). Furthermore, MRI revealed a smaller muscle mass compared to those in normal individuals of the same age. Her grasping power was 5 kg per hand, whereas the average grasping power of 17-year-old Japanese female is 26.81 ± 4.8 kg [16].

The dietitian revealed that she was consuming 20.4 kcal per ideal body weight (IBW) daily, with a higher proportion of calories coming from fat rather than from protein and carbohydrates. Additionally, she was having snacks between lunch and supper due to hunger. The ideal body weight was determined based on a BMI of 22. She was advised to increase her daily calorie intake to 30 kcal per IBW and to achieve a more balanced distribution of major nutrients. In addition to dietary intervention, we administered 400 mg of bezafibrate per day to normalize her serum triglyceride level; however, she immediately experienced severe muscle weakness. Consequently, we prescribed 600 mg/day of tocopherol and 0.2 mg/day of pemafibrate, which decreased her serum triglyceride level to 200 mg/dL. However, her casual serum low-density lipoprotein (LDL) cholesterol level markedly increased from 131 to 194 mg/dL. The clinical course of serum lipid and the drugs administered are shown in Fig. 1d, while the changes in body weight and diabetic features are displayed in Fig. 1e. At 19 years of age, we initiated leptin administration (metreleptin), as described previously [14], at a daily dose of 0.08 mg/kg body weight/day, following a priming period.

After 1 year of leptin treatment, her metabolic complications were re-examined. As shown in Table 1, serum ALT and uric acid levels were restored to the normal ranges; markers of hepatic fibrosis did not increase, indicating no progression of fibrosis due to fatty liver. In contrast, dyslipidemia remained unchanged. The oral glucose tolerance test revealed improved insulin resistance (HOMA-R, 2.02) (Fig. 1b, solid line). Her serum leptin level was elevated (84.6 ng/mL), while her adiponectin level remained low (2.3 µg/mL). According to the dietitian's assessment, she was consuming 24 kcal per IBW with the previous dietary recommendations, which resulted in weight gain. We concluded that leptin administration in combination with the selective peroxisome proliferator-activated receptor α (PPARα) modulator, attenuated insulin resistance and fatty liver, but only partially affected other metabolic complications. Notably, despite being under the satiety signal of leptin, she gained body weight without any changes in height. At least, the administration of leptin and PPARα modulators effectively mitigated the rise in hyperglycemia and hypertriglyceridemia attributable to increased food intake and body weight.

After 3 years of metreleptin administration, she suffered from an adjustment disorder unrelated to lipodystrophy, leading to a loss of appetite and reduced food intake. As a result, her body weight also decreased (BMI 15.7). Consequently, metreleptin was discontinued. More importantly, her hyperglycemia, hyperinsulinemia, and hypertriglyceridemia were attenuated without metreleptin, along with reduced food intake and body weight.

Discussion

We herein report a 17-year-old woman who developed insulin resistance, hypertriglyceridemia, and fatty liver after HSCT. Since her gluteal fat was thin and her blood adiponectin level was low, her metabolic abnormalities were attributed to partial lipodystrophy. In the present case, leptin administration partially attenuated severe insulin resistance when administered in combination with a selective PPARα modulator. This is the fifth case reporting the administration of metreleptin to treat lipodystrophy after HSCT [14, 15, 17, 18], and to the best of our knowledge, the first case describing the efficacy of pemafibrate on lipodystrophy. More importantly, this is the first case involving a patient with an original low BMI (BMI 18.9), who also suffered from an adjustment disorder unrelated to lipodystrophy, resulting in a loss of appetite and weight loss (BMI 15.7). This highlights the influence of diet and body weight in treatment-resistant acquired partial lipodystrophy after HSCT and its potential for metabolic improvement.

Lipodystrophy is a very rare cause of dyslipidemia, but it needs to be considered when dyslipidemia occurs in young patients with no obesity or no previous history of hyperlipidemia [19]. Lipodystrophy is a genetic disease, but is sometimes acquired. The present case had no apparent family history of genetic dyslipidemia, obesity, or severe proteinuria. Although the patient had previously been administered corticosteroids, the treatment had finished more than 3 years before her serum triglyceride level increased again. Partial lipodystrophy appeared to be the most appropriate diagnosis to explain her prominent insulin resistance and hyperlipidemia. Her clinical features, such as rich buccal and neck fat, reduced gluteal fat, and narrow limbs, were consistent with the features previously reported [11], which described cases of partial lipodystrophy after HSCT. Furthermore, the serum adipokine level in the present case was similar to those in previous cases; relatively unchanged leptin and decreased adiponectin levels [11, 20]. In lipodystrophy, insulin resistance and attenuated adipokine signals due to white adipose tissue loss lead to increased synthesis of triglycerides in the liver, resulting in non-alcoholic fatty liver [19]. To control the elevated serum triglycerides in patients with lipodystrophy, fibrate, a PPARα agonist, is considered [1]. Activated PPARα reduces serum triglycerides via increased production of lipoprotein lipase and enhanced catabolism of triglyceride-rich lipoproteins [22]. We prescribed pemafibrate, a selective PPARα modulator, designed to reduce undesirable off-target effects [23]. Pemafibrate reduced the serum triglyceride level without any side effects in the present case, but increased the serum LDL cholesterol level. This increase may be attributed to the fact that triglycerides were mostly included in very low-density lipoprotein (VLDL), as speculated from the lipoprotein fractionation results (Table 1). In general, pemafibrate decreases VLDL and increases LDL [24], but it was more pronounced in our case. Elevated LDL cholesterol level was sustained during pemafibrate administration for about 5 years, regardless of administering metreleptin and the amount of food intake and body weight, which is contrary to previous report [24].

Administration of metreleptin, a recombinant leptin analog, effectively reduced hypertriglyceridemia and increased insulin sensitivity in patients with lipodystrophy, particularly in cases of generalized lipodystrophy with more severe leptin deficiency [21, 25, 26]. In the current case, additional metreleptin treatment partially relieved insulin resistance; however, her serum triglyceride level remained elevated. Possible explanations are as follows: First, her serum triglyceride level was < 500 mg/dL and leptin level was > 4 ng/mL, which are negative predictors of a treatment response to metreleptin. Second, unlike in the previous case [15, 27], our patient experienced weight gain after metreleptin treatment. While the specific changes in body composition could not be determined in this case, it is hypothesized that the increase in food intake led to weight gain. Generally, leptin treatment improves postprandial satiety in patients with lipodystrophy and supports dietary therapy [28]. Given that the patient increased her calorie intake in line with the dietitian's recommendations, it remains uncertain how metreleptin treatment affected her appetite in our case. Additionally, it is suspected that the triglyceride level measured at the initial examination before treatment may have been underestimated due to her low-calorie intake. Nevertheless, metreleptin and pemafibrate treatments did appear to suppress the rise in triglycerides associated with increased food intake. The third explanation regarding the unaltered triglyceride levels after the leptin treatment is as follows: the reduction in muscle mass accounted for metabolic complications in the present case, rather than a decreased fat mass [20]. In partial lipodystrophy after HSCT, exercise may effectively attenuate insulin resistance and improve the quality of life. In this regard, adiponectin replacement therapy may also be an ideal treatment because serum adiponectin levels were very low in this case, and it has been reported to ameliorate insulin resistance in the muscle [29]. In rodents, adiponectin deficiency leads to the elevated plasma triglyceride concentration accompanied by the downregulation of PPARα in the liver and muscles [30]. It is speculated that low adiponectin levels may attenuate the triglyceride-lowering effects of pemafibrate. Similar to our case, in a report by Adachi et al. [15], triglyceride improvement with metreleptin treatment was not consistently maintained when food intake increased and physical activity decreased, despite steady improvements in blood glucose levels and liver enzyme levels. In our case, unfortunately, we did not measure changes in physical activity, but it is presumed that both diet and exercise therapy are essential for the success of metreleptin treatment, particularly in terms of triglyceride improvement. In our current case, the impact of calorie restriction and subsequent weight loss on metabolic improvement was indistinguishable since the weight loss was always a result of decreased food intake. The preceding case reports have shown that successful cases of metreleptin treatment often resulted in weight loss [15, 18], and in one case, the patient experienced weight loss without altering calorie intake during metreleptin treatment [15]. This suggests that weight loss during metreleptin treatment may have more significance than just reduced food intake.

Based on expert consensus, lifestyle changes that limit the energetic load on residual fat stores are considered the cornerstone of lipodystrophy management [1]. In adults, caloric restriction lowers triglycerides and glucose levels [31]. However, it's important to note that BMI is not an accurate tool to stratify cardiometabolic disease risk in patients with lipodystrophy, and caloric restriction should be considered even at a normal or low-normal BMI [32]. In our case, metabolic disorders were evident when her BMI was 18.9 without metreleptin, and partially improved with metreleptin. Importantly, the improvements were much more significant when her BMI was 15.7, with a 5% body weight loss even without metreleptin. Among the previous four cases, two case reports indicated the BMI before leptin treatment as 16.9 [17] and 15.6 [18], respectively. In both instances, metreleptin alleviated metabolic complications while further reducing weight. The question arises whether a BMI of 22 should be considered as the IBW for determining calorie requirements in lipodystrophy patients, although diet therapy and exercise therapy form the foundation for addressing metabolic disorders associated with lipodystrophy [33].

Notably, our case had sustained leptin levels, and leptin administration did not affect the hyperphagia believed to be induced by leptin deficiency [28]. In rodents, mice overexpressing leptin exhibit increased calorie intake when fed a high-fat diet [34]. However, in this case, the patient increased her food intake in accordance with the dietitian's recommendations, and the impact of the elevated leptin signal on satiety was uncertain. In this regards, the effect of adiponectin on food intake is still controversial, as it is believed to regulate food intake under glucose levels [35]. Indeed, metreleptin administration, especially when combined with a proper diet, may be particularly effective in partial lipodystrophy cases, especially those with relatively unchanged leptin levels and decreased adiponectin levels.

In summary, the current case suggests the modest effect of pemafibrate in controlling hypertriglyceridemia in partial lipodystrophy after HSCT, and highlights the importance of diet and weight loss in metreleptin treatment. Additionally, it underscores the need for endocrinologists to be aware of metabolic complications in young cancer survivors, as they may have a greater risk of late-onset metabolic complications [36].

Data availability

All the data supporting our findings is available from corresponding author upon request.

Declarations

Conflict of interest

None of the authors have any potential conflicts of interest associated with this report.

Human rights statement and informed consent

All procedures followed were in accordance with the Helsinki Declaration of 1964 and later versions. Informed consent for it was obtained from the patient and her parent.