icon star paper   HIV Articles  
Back grey_arrow_rt.gif
Leptin reverses nonalcoholic steatohepatitis in patients with severe lipodystrophy
  ---10 patients with lipodystrophy were studied. Liver biopsy and MRI were performed & it was found that 8 of 10 patients had fatty liver & NASH (a progressive metabolic liver disease). All 8 patients had liver injury (fibrosis). The study authors say--Lipodystrophy is characterized by leptin deficiency. After the study participants received treatment with recombinant Leptin, a second liver biopsy found significant reductions in NASH activity and significant reductions in triglycerides & fasting glucose, hemoglobin A1c, liver enzymes, and liver fat content. Perhaps we should consider further study to evaluate the treatment effect of this leptin on body changes, slowing liver disease, and if leptin therapy might improve response rates to peginterferon/ribavirin therapy.
April 2005
Edward D. Javor 1 *, Marc G. Ghany 2, Elaine K. Cochran 1, Elif Arioglu Oral 1, Alex M. DePaoli 3, Ahalya Premkumar 4, David E. Kleiner 5, Phillip Gorden 1 1Clinical Endocrinology Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 2Liver Diseases Section, Digestive Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD
3Amgen, Thousand Oaks, CA
4Department of Diagnostic Radiology, Warren G. Magnuson Clinical Center, National Institutes of Health, Bethesda, MD
5Laboratory of Pathology, National Cancer Institute, National Institutes of Health, Bethesda, MD
Patients & Methods
Author Discussion
Nonalcoholic steatohepatitis (NASH), a progressive metabolic liver disease, is one of the major consequences of the current obesity epidemic.[1-3] It lies on a spectrum of nonalcoholic fatty liver disease that ranges from ectopic lipid accumulation (steatosis) to cirrhosis.[4] Steatosis is believed to sensitize the liver to metabolic injury, leading to inflammation, necrosis, and fibrosis.[2][5-7] Thus, steatosis is a constant feature of NASH, but NASH is only distinguishable by liver biopsy. The assessment and severity of NASH is made histologically based on the patterns and degrees of hepatic steatosis, inflammation, and injury, in the absence of significant alcohol consumption.[8] Although steatosis is seen in both animal and human models, NASH is only fully appreciated in the human condition.[7] Thus, human models of NASH are critical for the development of therapeutic strategies for this condition.
Steatosis occurs with decreased leptin action, whether due to leptin deficiency or resistance.[9] Lipodystrophy represents a rare condition characterized by an absence of adipose tissue and resultant leptin deficiency. In the absence of adequate adipose tissue, combined with leptin deficiency, the liver becomes a major repository for triglycerides in lipodystrophic patients, often expanding far beyond the normal size.
We have previously reported decreased serum aminotransferases and liver volumes over a 4-month period of therapy with recombinant methionyl human leptin (r-metHuLeptin) in patients with severe lipodystrophy.[10] Decreased liver fat content has also been reported.[11] This study examined the prevalence of NASH in these patients with steatosis and assessed the histological changes in the context of biochemical and radiographic changes seen with r-metHuLeptin therapy.
Severe lipodystrophy is characterized by diminished adipose tissue and hypoleptinemia, leading to ectopic triglyceride accumulation. In the liver, this is associated with steatosis (fat accumulation in the liver), potentially leading to nonalcoholic steatohepatitis (NASH).
We investigated the prevalence of NASH and the effect of leptin replacement in these patients. Ten patients with either generalized lipodystrophy (8 patients) or Dunnigan's partial lipodystrophy (2 patients) were included in this analysis. Paired liver biopsy specimens were obtained at baseline and after treatment with recombinant methionyl human leptin (r-metHuLeptin), mean duration 6.6 months. The extents of portal and parenchymal inflammation, steatosis, ballooning, presence of Mallory bodies, and fibrosis in liver biopsy specimens were scored using a previously validated system developed to assess NASH activity. Histological disease activity was defined as the sum of ballooning, steatosis, and parenchymal inflammation scores. We concurrently tested serum triglycerides and aminotransferases and estimations of liver volume and fat content by magnetic resonance imaging.
Eight of 10 patients met histological criteria for NASH at baseline.
After treatment with r-metHuLeptin, repeat histological examinations showed significant improvements in steatosis (P = .006) and ballooning injury (P = .005), with a reduction of mean NASH activity by 60% (P = .002). Fibrosis was unchanged (ed note: second biopsy was done after 18 months leptin therapy. Perhaps longer followup would show improved fibrosis). Significant reductions were seen in mean serum triglycerides (1206 to 226 mg/dL, P = .002), glucose (220 to 144 mg/dL, P = .02), insulin (46.4 to 24.8 uIU/mL, P = .004), ALT (54 to 24 U/L, P = .02), AST (47 to 22 U/L, P = .046), liver volume (3209 to 2391 cm3, P = .007), and liver fat content (31 to 11%, P = .006). HbA 1c (%) improved from 8.5 to 6.7 (p=.005).
In conclusion, r-metHuLeptin therapy significantly reduced triglycerides, transaminases, hepatomegaly, and liver fat content. These reductions were associated with significant reductions in steatosis and the hepatocellular ballooning injury seen in NASH.
Patient Population and Study Design

Between July 2000 and December 2003, we evaluated 26 consecutive patients with severe forms of lipodystrophy, including congenital generalized lipodystrophy (CGL), acquired generalized lipodystrophy (AGL), and Dunnigan's partial lipodystrophy (FPLD). Inclusion criteria for leptin replacement included hypoleptinemia (males < 4 ng/mL, females < 5 ng/mL), metabolic abnormalities such as hypertriglyceridemia or type 2 diabetes, and ability to adhere to the leptin replacement protocol. Twenty patients met the inclusion criteria to initiate therapy with r-metHuLeptin in a protocol designed to study the effects of leptin replacement in patients with severe lipodystrophy. The protocol was approved by the institutional review board of the National Institute of Diabetes and Digestive and Kidney Diseases. Informed consents were obtained from each patient or his or her legal guardian.
To qualify for the current analysis, patients were required to have paired liver biopsies before and after initiating r-metHuLeptin. Two patients declined this procedure. Six patients younger than 18 years of age with normal liver chemistry did not undergo biopsies. Two patients with AGL were excluded because of initial biopsies showing changes consistent with autoimmune hepatitis, believed to be unrelated to steatosis. Thus, a total of 10 patients with severe lipodystrophy were included in this study. Four patients had CGL, 4 had AGL, and 2 had FPLD. Eight patients were female, and 2 were male. They ranged in age from 17 to 67 years. None had a history of significant alcohol consumption, defined as more than 140 g/week.
R-metHuLeptin therapy was given as a self-administered, twice-daily subcutaneous injection as previously described.[10] Patients were seen every 4 months for the first year and every 6 months thereafter. Laboratory data were collected on a metabolic unit during each visit. Diabetic and lipid medications were lowered or discontinued if indicated. Liver imaging studies were obtained and analyzed concurrently with liver biopsies.
Six of the 10 follow-up biopsies were done at 4 months because of the original length of the protocol. Three were done at 8 months. One patient (NIH-1) underwent the initial liver biopsy secondary to abnormal serum transaminases at age 15, 2 years before initiating r-metHuLeptin therapy. The follow-up biopsy was done after 18 months of r-metHuLeptin therapy, when she was 18 years old.
Liver Volume Measurements.

Axial T1-weighted magnetic resonance imaging (MRI) scans of the liver were obtained on a 1.5-tesla scanner (General Electric Medical Systems, Milwaukee, WI). Liver volumes were calculated using the MEDx image analysis software package (Medical Numerics, Sterling, VA), on a Sun workstation. By placing a seed point for an edge-following algorithm, tracings of the outer margins of the liver were made on individual contiguous slices. The liver volumes were then computed based on the pixel area and slice thickness.
Liver Fat Measurements.
Axial in-phase (IP) and out-of-phase (OP) breath-hold gradient echo MRI scans of the liver were obtained with the following parameters TR 9.3, TE 4.2 (IP); TR 7.3, TE 2 (OP), flip angle 30, 256 ~ 128 matrix, 2NEX, and a slice thickness of 10 mm. The analysis to quantify liver fat was done using the MEDx software analysis package (Medical Numerics) run on a LINUX platform. The modified Dixon method was used to assess the hepatic fat fraction. This is based on the fact that the signal intensity of the liver on OP MRI sequences would drop relative to IP sequences depending on the amount of liver fat present. Regions of interest were placed in the liver and the spleen to include maximum parenchyma without contamination with blood vessels or motion artifacts. Signal intensity measurements were obtained from the regions of interest in the liver and normalized to the spleen to offset differences in receive attenuation among the various sequences. The percent decrease in normalized liver signal intensity on the OP sequence relative to the IP sequence was calculated by the formula: Signal intensity decrease = 100 ~ [(L1/S1) - (L0/S0)] /L1/S1. (L1 and S1 = signal intensity of liver and spleen on IP images, and L0 and S0 = signal intensity of liver and spleen on OP images).[12][13]
Histological Analysis
Liver biopsies were performed through the percutaneous route under conscious sedation using a 16-gauge Klatskin needle. A liver specimen of at least 15 mm with a minimum of 10 portal tracts was considered adequate for evaluation. Liver biopsy specimens were fixed in 10% buffered formalin, embedded in paraffin, sectioned at 4-m intervals and stained with hematoxylin-eosin, Masson's trichrome, and Gomori iron stain. Anti-ubiquitin staining (Dako Z0458, 1:1000, heat-activated antigen retrieval in citrate buffer) was used to visualize Mallory bodies. Biopsy specimens were scored prospectively by a single hepatopathologist blinded to the clinical data and sequence of the biopsies using a previously described scoring method developed for NASH.[14] Briefly, 6 features of fatty liver disease are scored from 0 to 4: portal inflammation, parenchymal inflammation, ballooning hepatocellular injury, Mallory bodies, steatosis, and fibrosis. Three of the scores, parenchymal inflammation, ballooning injury, and steatosis, are combined to give a disease activity score from 0 to 12. The pattern of injury was characterized as steatohepatitis, borderline changes for steatohepatitis, or not diagnostic of steatohepatitis.
Baseline Clinical Characteristics

Ten patients with severe forms of lipodystrophy were studied. Eight patients had generalized lipodystrophy (4 CGL and 4 AGL), and 2 had FPLD. The mean age of the patients was 40 years (range, 17-67 years). All 10 patients were hypoleptinemic (males <3 ng/mL, females <4 ng/mL) and had uncontrolled diabetes with fasting glucose 126 mg/dL (range, 149-428 mg/dL) and hemoglobin A1c 7% (range, 7%-9.5%) despite aggressive anti-diabetic management. All patients had hypertriglyceridemia >150 mg/dL (range, 198-6,400 mg/dL) despite being on statins or fibrates. Anthropometric data at baseline and after r-metHuLeptin therapy were consistent with previously reported data.[15]
Baseline Liver Data
No patient had clinical evidence of decompensated liver disease. Four patients had elevated transaminases (aspartate aminotransaminase > 34 U/L, alanine aminotransaminase > 41 U/L), and 2 patients had elevated total bilirubin (>1 mg/dL). Nine patients had hepatomegaly (>2,000 mL), and 7 patients had fatty liver (19%) by MRI.
Eight patients had histological evidence of NASH. Of the 2 who did not have steatohepatitis, 1 had only a small amount of spotty lobular inflammation and the other had only mild steatosis and mild portal and lobular inflammation without the characteristic zone 3 injury of NASH. In addition to steatosis and inflammation, all 8 patients with steatohepatitis had ballooning hepatocellular injury varying from mild changes only in zone 3 to diffuse, panacinar ballooning changes and fibrosis that varied from mild perisinusoidal zone 3 fibrosis to cirrhosis. Four of the 8 had at least bridging fibrosis at baseline. Only 3 patients had identifiable Mallory bodies, despite staining all biopsy specimens with anti-ubiquitin.
r-metHuLeptin Therapy
Patients enrolled in an open-label pilot study of r-metHuLeptin for their severe insulin resistance, diabetes, dyslipidemia, fatty liver disease, and hypothalamic abnormalities. They were observed from 4 to 18 months (mean, 6.6 ± 1.4 months). During this period, there were significant reductions in liver volumes, liver fat percentage, transaminases, triglycerides, fasting glucose, fasting insulin, and hemoglobin A1c. These reductions were associated with significant reductions in NASH scores. The greatest changes were in the histological parameters that defined NASH activity: steatosis, ballooning injury, and parenchymal inflammation. The latter did not reach statistical significance. There was no detectable change in fibrosis. Of the 9 patients who had steatosis at baseline, only 3 had steatosis on follow-up biopsy. Similarly, of the 8 patients with ballooning hepatocellular injury, only 3 had ballooning identified in follow-up, and in all 3 the degree was mild and confined to zone 3. Six of the 8 patients (75%) with NASH at baseline no longer met the histological criteria for NASH. The reductions in NASH scores were strongly correlated with reductions in liver volumes, liver fat measured by MRI, serum transaminases, triglycerides, fasting insulin, and fasting glucose values.
R-metHuLeptin therapy in patients with severe forms of lipodystrophy led to significant reductions in NASH activity scores in conjunction with significant reductions in triglycerides, liver volumes and fat content, transaminases, fasting glucose values, and hemoglobin A1c. There were significant correlations between the NASH activity scores and MRI estimations of liver volume and fat content, serum transaminases, triglycerides, fasting insulin, and fasting glucose values. However, the causative relationship of these correlations to the clinical situation remains uncertain. For instance, note that elevated transaminases were not universally present in patients diagnosed with NASH.
Other limitations of this study include its relatively small sample size, the variable interval between liver biopsies, the study of a rare disease in terms of its general relationship to NASH, and the fact that this study was not placebo-controlled. Furthermore, we have no evidence that the results achieved in this population of patients with hypoleptinemia are relevant to the broader group of NASH patients with higher serum leptin concentrations. Nevertheless, the extreme nature of the metabolic defect in this group of patients makes this a valuable model system.
The diagnosis of NASH is usually considered in overweight/obese patients with elevated transaminases in the absence of alcohol use or viral hepatitis. NASH is a potential complication of steatosis, which is thought to represent the hepatic component of the metabolic syndrome,[16] characterized by obesity, insulin resistance, hypertriglyceridemia, and hypertension. Although a hallmark of the metabolic syndrome, insulin resistance alone does not appear sufficient for the development of steatosis or NASH. In our experience, patients with extreme insulin resistance as a result of receptor mutations[17][18] or autoantibodies[19] do not have NASH-associated findings such as hypertriglyceridemia, hepatomegaly, or abnormal liver function tests. Fat accumulation outside of adipose tissue appears to be the common link contributing to both insulin resistance and NASH in the general population and in these patients. In other words, insulin resistance and NASH appear to develop in parallel rather than sequentially. Severe lipodystrophy is an exaggerated model of ectopic fat accumulation leading to an exaggerated form of the metabolic syndrome.
Despite the remarkable success of r-metHuLeptin therapy in these patients, the precise mechanism of leptin action is still poorly understood. One action of r-metHuLeptin that is likely playing a role is the control of appetite in these patients. Leptin is known to act at the hypothalamus, inhibiting the synthesis and release of orexigenic peptides, including neuropeptide Y and agouti-related protein.[20] This decrease in energy through the system would potentially allow for mobilization of stored triglycerides from the liver.
How leptin might further mediate its metabolic actions, such as oxidizing fat, remains unclear. In rodent models, leptin increased expression of adenosine monophosphate-activated protein kinase[21] and peroxisome proliferator-activated receptor (PPAR)a,[22] activating enzymes of fatty acid oxidation, including carnitine palmitoyl transferase-1 and acyl-CoA oxidase, and inhibiting enzymes of lipogenesis, including acyl-CoA carboxylase and stearoyl-CoA desaturase-1.[21][23] Additionally, leptin increased hepatic and adipocyte expression of PPARĄ coactivator-1a,[22][24] which regulates mitochondria biogenesis.[25] Zucker diabetic fatty rats transfected with leptin had a dramatic augmentation in PPAR coactivator-1a expression in white adipose tissue with resultant delipidation.[24] Their shrunken adipocytes were filled with mitochondria, demonstrating conversion from fat-storing into fat-burning cells. Thus, leptin appears to enhance mitochondrial function. Conversely, conditions of decreased leptin action, whether due to deficiency or resistance, may lead to mitochondrial insufficiency or dysfunction, which has been implicated in ectopic fat accumulation and insulin resistance.[26]
Whether leptin primarily acts centrally or peripherally is still debated. For instance, the lipopenic action of leptin has been duplicated in rodents with small, centrally administered doses.[23] The effect is achieved even with knockout of peripheral leptin receptors.[27] This suggests that leptin signals one or more centrally mediated intermediates, including the sympathetic nervous system, which in turn lead to increased fatty acid oxidation and suppression of lipogenesis.
We interpret our data showing improvement in parameters of oxidative stress as improvement in metabolic liver disease, although we were unable to show a reduction in fibrosis. This interpretation may be controversial in the light of a longitudinal study of nonalcoholic fatty liver disease that suggests a progression of fibrosis with improvement of other histological parameters.[28] We believe that the short interval between biopsies was the primary factor limiting our observed change in fibrosis.
Rodent models have suggested that leptin may augment hepatic inflammation and fibrosis[29][30]; however, it must be remembered that our patients with hypoleptinemia had NASH with fibrosis at baseline. Thus, the relationship between rodent models and the human condition is unclear. Furthermore, recent studies have shown no relationship between leptin levels and severity of injury or fibrosis in humans.[31-33]
This model of removing ectopic fat from the liver is in contrast to other successful models using thiazolidinediones,[14][34-36] which are PPARĄ agonists that act primarily on adipose tissue to promote adipocyte development and differentiation.[37][38] These medications work in part by augmenting the fat-storing capacity of adipose tissue, lowering the burden of storing fat in ectopic locations such as the liver and skeletal muscles. The leptin and thiazolidinedione models both demonstrate the association of NASH with steatosis and the reversal of NASH by removing the fat.
A 6-month study using the first available PPARĄ agonist, troglitazone, in various forms of lipodystrophy also demonstrated an overall reduction of steatosis.[39] The patients with the most notable reductions had partial forms of lipodystrophy and could most likely repartition fat to their remaining adipose tissue. This class of drugs has a theoretical risk of actually worsening hepatic steatosis and NASH in patients with generalized lipodystrophy.[40] In the absence of adipose tissue, the liver may become the primary target of PPARĄ activation and fat accumulation.
This study highlights a successful application of leptin replacement in patients with leptin deficiency. The response to r-metHuLeptin therapy further supports an important role of leptin to maintain normal triglyceride homeostasis throughout the body. This may have broader implications for NASH in the general population, which is also characterized by ectopic triglyceride accumulation and secondary disorders such as insulin resistance. To date, using leptin in states of hyperleptinemia such as obesity has not led to significant fat loss, because these patients have leptin resistance. Recent rodent studies suggest that this can be potentially overcome with a high enough dose,[22][24] a theory similar to insulin resistance. Whether this is a feasible method of fat loss remains to be seen. Ultimately, research designed to elucidate the mechanisms of leptin action and leptin resistance is of greatest importance.
  icon paper stack View Older Articles   Back to Top   www.natap.org