original article

Exenatide improves type 2

diabetes concomitant with
non-alcoholic fatty liver disease
Exenatide melhora o diabetes tipo 2 de ocorrência concomitante
com doença hepática gordurosa não alcoólica

Hui Fan1, QingRong Pan1, Yuan Xu1, XinChun Yang2

ABSTRACT
1
Department of Endocrinology, Objective: To investigate the effects of exenatide on blood glucose, body weight and hepatic en-
Beijing Chaoyang Hospital, zymes in patients with type 2 diabetes mellitus (T2DM) and concomitant non-alcoholic fatty liver
Capital University of Medical disease (NAFLD). Subjects and methods: One hundred and seventeen patients with T2DM and
Science, Beijing, China
2
Heart Center, Beijing Chaoyang
NAFLD were randomly divided into exenatide group and metformin group. Patients were treated
Hospital, Capital University of with exenatide and metformin, respectively, for 12 weeks. Results: After 12 weeks of treatment, body
Medical Science, Beijing, China weight, body mass index (BMI), waist-to-hip ratio, HbA1c, FPG, 2-h PPG, ALT, AST, γ-GT, and hs-CRP
were significantly reduced, and the AST/ALT ratio and adiponectin were markedly increased in both
groups. BMI, waist-to-hip ratio, 2-h PPG, ALT, AST, γ-GT, and hs-CRP were markedly lower, and AST/
ALT ratio and adiponectin in the exenatide group were dramatically higher than in the metformin
group. Conclusion: Compared with metformin, exenatide is better to control blood glucose, reduces
body weight and improves hepatic enzymes, attenuating NAFLD in patients with T2DM concomitant
with NAFLD. Arq Bras Endocrinol Metab. 2013;57(9):702-8
Keywords
Exenatide; metformin; non-alcoholic fatty liver disease; type 2 diabetes mellitus

RESUMO
Correspondence to: Objetivo: Investigar os efeitos do exenatide sobre a glicose sérica, peso corporal e enzimas hepá-
Yuan Xu
Department of Endocrinology, ticas em pacientes com diabetes melito tipo 2 (T2DM) e doença hepática gordurosa não alcoólica
Beijing Chaoyang Hospital, (DHGNA). Sujeitos e métodos: Um total de 117 pacientes com T2DM e DHGNA foi aleatoriamente
Capital University of Medical Science,
Beijing 100020, China
separado em dois grupos, um tratado com exenatide e um tratado com metformina. Os pacientes fo-
xuyuan1234@126.com ram tratados por 12 semanas. Resultados: Após 12 semanas de tratamento, o peso corporal, índice
de massa corporal (IMC), relação cintura-quadril, HbA1c, FPG, glicose pós-prandial, ALT, AST, γ-GT e
Received on Mar/29/2013
Accepted on Oct/28/2013 proteína C-reativa foram significativamente reduzidos, e a relação AST/ALT e a adiponectina aumen-
taram marcadamente nos dois grupos. O IMC, relação cintura-quadril, glicose pós-prandial, ALT, AST,
γ-GT e proteína C-reativa foram marcadamente menores, e a relação AST/ALT e a adiponectina foram
dramaticamente mais altas no grupo tratado com exenatide do que no grupo tratado com metformi-
na. Conclusão: Comparado com a metformina, o exenatide controla melhor a glicose sérica, reduz
o peso corporal e melhora as enzimas hepáticas, atenuando a DHGNA em pacientes com T2DM de
ocorrência concomitante com a DHGNA. Arq Bras Endocrinol Metab. 2013;57(9):702-8
Descritores
Exenatide; metformina; doença hepática gordurosa não alcoólica; diabetes melito tipo 2

central obesity, hypertension, and insulin resistance (1).

iNTRODUCTION The prevalence of obesity and diabetes is increasing in
Copyright© ABE&M todos os direitos reservados.

T ype 2 diabetes mellitus (T2DM), obesity, and non-

-alcoholic fatty liver disease (NAFLD) are often
identified in patients simultaneously. In patients with
China (2) and, thus, one may expect an increase in the
incidence of NAFLD in T2DM in China. Although
NAFLD and non-alcoholic steatohepatitis (NASH) are
T2DM, the prevalence of NAFLD is as high as 75% generally asymptomatic, the increase in serum alanine
(1). NAFLD represents the hepatic manifestation of aminotransferase (ALT), a biochemical marker of liver
the metabolic syndrome, which includes dyslipidemia, injury, has been regarded as indicative of NAFLD (3).

702 Arq Bras Endocrinol Metab. 2013;57/9

 Exenatide improves type 2 diabetes

NAFLD can cause progressive fibrosis leading to sensitivity were improved over a 3.5-year follow-up
cirrhosis and corresponding complications, including period (11). Why can exenatide reverse hepatic bio-
portal hypertension and liver failure (1). In addition, markers and type 2 diabetes with fatty liver? Is it by
NAFLD is thought to be an independent determinant improving insulin sensitivity or by another way? When
of cardiovascular disease (CVD), besides affecting in- compared with metformin, exenatide is more effective
dividuals who already have multiple cardiovascular risk in reducing body weight. Thus, we wonder if exenati-
factors. This relationship may lead to an increased risk de may serve as a drug in the treatment of NAFLD to
of undesired outcomes in this population of patients, effectively improve the hepatic biomarkers.
and increased cardiovascular morbidity and mortality The present study aimed to evaluate the therapeutic
(4). Moreover, NAFLD is a complex syndrome that effect of exenatide in NAFLD. In order to achieve this
affects more than the liver. aim, plasma glucose, body weight, liver enzymology,
Although the pathogenesis of NAFLD is still poorly serum C reactive protein (CRP), and adiponectin were
understood, the relationship between NAFLD and in- evaluated after the treatment with exenatide or metfo-
sulin resistance make insulin resistance a therapeutic tar- min in T2DM patients with NAFLD.
get. Evidence has indicated that lifestyle modifications,
with weight loss and exercise, can improve, NAFLD (5-
7). This has stimulated the conduction of several clini- SUBJECTS AND METHODS
cal studies in which anti-diabetic drugs, such as insulin
sensitizers, including metformin, have been evaluated Patients
in NAFLD patients with insulin-resistance. From March 2009 to December 2010, a total of 134
Most of findings show no clear benefit from the patients with T2DM and NAFLD were enrolled in the
addition of metformin to diet or rosiglitazone, thou- present study. At the end of study, 117 patients were re-
gh the Alanine aminotransferase levels can be reduced cruited for analysis, including 66 males and 51 females.
by metformin treatment. It is indicated that metformin
The inclusion criteria were as follows: T2DM concomi-
might be of benefit in the treatment of NAFLD, while
tant with NAFLD, poor glucose control (FBG: 6.0-10.0
associated to hypocaloric diet and weight control. The
mmol/L or HbA1c: 7-9%), no acute complications or
lack of benefit was because patients in the control group
severe chronic complications of DM. The criteria for
improved on diet, and metformin had little effect on
DM developed by the World Health Organization in
liver histology (8). So, metformin as a hypoglycemic
1999 were employed for the diagnosis of DM. Exclu-
agent may be not reliable to alleviate NAFLD.
sion criteria were as follows: abnormal kidney function,
A new drug, glucagon-like peptide-1 analogue,
serum Cr ≥ 133 μmol/L, presence of diseases affec-
such as exenatide, may be an effective anti-diabetic
ting blood glucose levels (such as hyperthyroidism and
agent. Exenatide is a synthetic analog of exendin-4,
hypercortisolism), treatments affecting blood glucose
a 39-amino-acid agonist of glucagon-like peptide 1
(GLP-1) receptor. Exendin-4 is found in the saliva of levels (such as steroid hormones), pregnancy, breast-
the Gila monster, Heloderma suspectum, and shares 53% -feeding, and patients with history of chronic drinking.
sequence homology with GLP-1 (9). GLP-1 is a gas- All patients were diagnosed with T2DM concomitant
trointestinal hormone secreted by the L cells of the in- with NAFLD; 52% of patients had abnormal liver func-
testine, which can regulate blood glucose primarily via tion (ALT > upper limit of normal; ALT > 2.5 times of
stimulation of glucose-dependent insulin release. The- upper limit of normal in 46 patients). Ultrasonography
refore, exenatide may not function as a direct insulin was carried out for the diagnosis of fatty liver (Philips
sensitizer, but may induce clinically significant weight iu22; C5-2 probe) by two medical sonographers, based
loss, leading to an insulin-sensitizing effect. on the following criteria: 1) enlarged liver, increase and
Copyright© ABE&M todos os direitos reservados.

However, in a murine model of fatty liver, exen- intensification in echoes and bright spots (bright liver);
din-4 significantly reduced glucose levels, improved 2) unclear or absent intrahepatic blood vessels; 3) re-
insulin sensitivity, and reduced hepatic steatosis (10). duced or attenuated deep echoes. Patients who met the
In an open-label, uncontrolled clinical trial using exe- diagnostic criteria and had no history of drinking, acute
natide to assess the safety of the drug in patients with fatty liver, drug-induced fatty liver, hepatitis B, and he-
diabetes, aspartate aminotransferase (AST) and insulin patitis C were recruited for the present study.

Arq Bras Endocrinol Metab. 2013;57/9 703

 Exenatide improves type 2 diabetes

Therapeutic regimen body weight was 78.59 ± 11.03 kg, and mean body
Patients were divided into two age- and gender-ma- mass index (BMI) was 27.02 ± 4.43 kg/m2. These pa-
tched groups. All the patients underwent lifestyle in- tients were characterized by mild to moderate increase
terventions. In Group A, patients (n = 49) were treated in ALT, AST and γ-GT, as well as reduction in AST/
with exenatide, and in Group B (n = 68), with metfor- ALT.
min. The therapeutic regimen was as follows: Group At baseline, there were no marked differences in
A: Exenatide injection (Eli Lilly and Company, USA) age, body weight, WHR, HbA1c, FPG, 2-h PPG, TG,
was administrated from week 1 to week 4 at 5 μg (bid), HDL-C, LDL-C, Tch-C, ALT, AST, γ-GT, adiponec-
and from week 5 to week 12 at 10 μg (bid). Group B: tin, and hs-CRP (Table 1).
metformin was initially administered at 0.5 g (bid). The Table 1. Demographics of patients with T2DM and NAFLD at baseline
dose of metformin (Sino-American Shanghai Squibb Group A Group B
t P
Pharmaceutical Ltd.) was adjusted to a maximum of (n = 49) (n = 68)
2.0 g/d on the basis of FPG and 2-h PPG. Age (yr) 51.02 ± 10.10 54.68 ± 12.14 -1.68 0.10
Gender (M/F) 28/21 38/30 0.018 0.89
Observations Body weight (kg) 80.61 ± 11.23 78.58 ± 10.76 2.122 0.168

The measurement of height, body weight, waist-to-hip BMI 28.18 ± 1.86 27.61 ± 1.77 1.58 0.12

ratio and blood pressure, ultrasonography of the liver, WHR 0.98 ± 0.08 0.97 ± 0.06 -0.06 0.95

gallbladder and spleen, and detection of HbA1c, FPG, ALT (U/L) 65.74 ± 18.13 65.81 ± 17.58 -0.02 0.99
2-h PPG, FIN, TG, HDL-C, LDL-C, Tch-C, ALT, AST (U/L) 35.92 ± 12.27 34.28 ± 13.69 0.62 0.54
AST, γ-GT, hs-CRP and adiponectin were done at ba- AST/ALT 0.60 ± 0.06 0.60 ± 0.06 0.05 0.96
seline and after 12 weeks of treatment. The hs-CRP γ-GT (U/L) 70.21 ± 21.54 62.65 ± 19.02 1.79 0.08
was detected with Latex-enhanced immune turbidime- HbA1c (%) 8.14 ± 0.51 8.09 ± 0.59 -0.86 0.59
tric method (Orion, Fineland), and serum adiponectin FPG (mmol/L) 8.45 ± 0.64 8.41±0.69 0.34 0.73
with radioimmunoassay (Linco, USA). Patients were 2-h PG (mmol/L) 15.09 ± 5.42 14.64 ± 5.12 0.81 0.42
re-examined after 4 weeks of the treatment, when FBG TG (mmol/L) 2.47 ± 1.49 2.51 ± 1.73 -1.01 0.32
and 2-h PPG at each meal were measured. TCH-C (mmol/L) 5.03 ± 0.97 5.15 ± 0.95 -0.62 0.54
LDL-C (mmol/L) 2.56 ± 0.83 2.62 ± 0.82 -1.57 0.12
Statistical analysis HDL-C (mmol/L) 1.05 ± 0.24 0.99 ± 0.20 1.34 0.18
Quantitative data were expressed as mean ± standard hs-CRP (mg/L) 3.14 ± 0.58 3.16 ± 0.68 -0.18 0.86
deviation ( ± SD). Statistical analysis was done with Adiponectin (mg/L) 8.43 ± 1.01 8.13 ± 1.23 1.27 0.21
SPSS version 17.0. Means before and after treatment HOMA-IR 3.05 ± 0.55 2.94 ± 0.59 0.95 0.34
were compared between the two groups with indepen-
dent t test or rank sum test. Means before and after
Variables after the 12-week treatment
treatment in each group were compared with paired
t test or rank sum test. Qualitative data were compa- In the treatment, the major side effects included nau-
red with chi-square test. Two-sided P value of < 0.05 sea, upper abdominal discomfort and decreased appe-
was considered statistically significant. HOMA-IR was tite. Three patients treated with metformin developed
calculated as follow: fasting blood glucose vs. fasting diarrhea. Patients were tolerant to these side effects
blood insulin/22.5. within 2-3 weeks. At 12 weeks of treatment, body wei-
ght, BMI, WHR, HbA1c, FPG, 2-h PPG, ALT, AST
and γ-GT reduced significantly when compared with
RESULTS the values observed before treatment. There were no
Copyright© ABE&M todos os direitos reservados.

marked differences in HbA1c, FPG, HOMA-IR, TG,

Characteristics of patients at baseline HDL-C, LDL-C, and Tch-C between two groups af-
A total of 144 patients were recruited for the present ter treatment. However, body weight (76.09 ± 9.85 vs.
study, and 27 of them withdrew from the study or were 77.22 ± 10.15, P = 0.02), BMI (25.87 ± 1.48 vs. 26.91
lost to follow-up. In the end, 117 patients completed ± 1.79, P = 0.02), WHR (0.95 ± 0.05 vs. 0.96 ± 0.04,
this study. Mean age was 52.35 ± 11.83 years, mean P = 0.01), ALT (39.82 ± 14.05 vs. 51.48 ± 18.89, P =

704 Arq Bras Endocrinol Metab. 2013;57/9

 Exenatide improves type 2 diabetes

0.00), AST (25.61 ± 7.87 vs. 31.54 ± 10.75, P = 0.01), more obvious than those in the metformin group. Ho-
AST/ALT (0.81 ± 0.17 vs. 0.69 ± 0.15, P = 0.00), wever, reductions in FPG, HbA1c and HOMA-IR were
γ-GT (47.53 ± 15.80 vs. 53.44 ± 15.00, P = 0.05), hs- comparable between two groups (Table 3).
-CRP (2.18 ± 0.34 vs. 2.69 ± 0.53, P = 0.01), and 2-h
PPG (10.31 ± 3.17 vs. 12.05 ± 4.03, P = 0.036) were Table 3. Variables after 12 weeks of treatment in the two groups
markedly reduced, and adiponectin increased dramati-
Changes in variables
cally in the exenatide group when compared with the Factors
Exenatide Metformin
metformin group (Table 2). group (n = 49) group (n = 68)
t P

Body weight (kg) 4.16 ± 5.32 1.98 ± 3.28 2.01 0.000**

Table 2. Variables in two groups after 12 weeks of treatment BMI 1.31 ± 0.98 0.69 ± 0.94 3.22 0.000**
Exenatide Metformin WHR 0.04 ± 0.02 0.02 ± 0.02 3.08 0.000**
Variables t P
group (n = 49) group (n = 68)
ALT (U/L) 27.32 ± 15.96 12.85 ± 11.38 5.73 0.002**
Body weight (kg) 76.09 ± 9.85 77.22 ± 10.15 -2.764 0.02*
AST (U/L) 7.89 ± 7.87 5.11 ± 6.98 2.03 0.048*
BMI 25.87 ± 1.48 26.91 ± 1.79 -2.28 0.02*
AST/ALT -0.23 ± 0.08 -0.11 ± 0.09 -1.02 0.043*
WHR 0.95 ± 0.05 0.96 ± 0.04 -2.59 0.009*
γ-GT (U/L) 26.48 ± 17.34 10.26 ± 14.11 5.98 0.000**
ALT (U/L) 39.82 ± 14.05 51.48 ± 18.89 -3.36 0.000**
HbA1c (%) 0.91 ± 0.65 0.89 ± 0.57 1.30 0.203
AST (U/L) 25.61 ± 7.87 31.54 ± 10.75 2.62 0.012*
FPG (mmol/L) 1.09 ± 0.83 1.12 ± 0.80 -0.08 0.581
AST/ALT 0.81 ± 0.17 0.69 ± 0.15 4.34 0.00*
2hPG (mmol/L) 4.16 ± 3.01 2.63 ± 2.28 2.03 0.022*
γ-GT (U/L) 47.53 ± 15.80 53.44 ± 15.00 -1.92 0.047*
TG (mmol/L) 0.31 ± 1.44 0.33 ± 1.06 -0.11 0.904
HbA1c (%) 7.15 ± 0.36 7.16 ± 0.58 -1.73 0.082
TCH-C (mmol/L) 0.29 ± 0.71 0.26 ± 0.80 0.17 0.939
FPG (mmol/L) 7.11 ± 0.68 7.32 ± 0.76 -1.52 0.131
LDL-C (mmol/L) 0.19 ± 0.63 0.16 ± 0.71 0.20 0.646
2-h PG (mmol/L) 10.31 ± 3.17 12.05 ± 4.03 -2.43 0.024*
HDL-C (mmol/L) 0.02 ± 0.50 0.02 ± 0.61 0.10 0.827
TG (mmol/L) 1.84 ± 0.51 1.78 ± 0.60 0.95 0.345
hs-CRP (mg/L) 0.89 ± 0.59 0.61 ± 0.54 2.59 0.018**
TCH-C (mmol/L) 4.88 ± 1.51 4.82 ± 1.50 -0.09 0.934
Adiponectin (mg/L) -1.86 ± 2.22 -0.76 ± 1.30 -7.37 0.001**
LDL-C (mmol/L) 2.42 ± 0.52 2.45 ± 0.55 -0.029 0.687
HOMA-IR 0.57 ± 0.36 0.56 ± 0.49 0.89 0.367
HDL-C (mmol/L) 1.06 ± 0.22 1.03 ± 0.23 1.08 0.084
* P ≤ 0.05; ** P ≤ 0.01.
hs-CRP (mg/L) 2.18 ± 0.34 2.69 ± 0.53 -3. 21 0.023*
Adiponectin (mg/L) 10.44 ± 3.29 8.48 ± 2.67 3.45 0.000**
HOMA-IR 2.47 ± 0.44 2.48 ± 0.51 -0.10 0.923 HbA1c and ALT after treatment
* P ≤ 0.05; ** P ≤ 0.01.
After the 12-week treatment, the proportion of patients
with HbA1c < 7% was comparable between two groups,
but the proportion of patients with normal ALT level
Body weight and biochemical variables before and (< 40 U/L) in the exenatide group was significantly
after treatment lower than in the metformin group. Re-examination re-
In both groups, body weight, BMI, WHR, HbA1c, vealed that the percentage of patients with concomitant
FPG, 2-h PPG, HOMA-IR, ALT, AST, γ-GT, TG, and NAFLD in the exenatide group was slightly lower than
hs-CRP reduced significantly; HDL-C, LDL-C, and in the metformin group, which might be related to the
Tch-C remained unchanged; and AST/ALT and adi- short-term treatment or small sample size (Table 4).
ponectin increased dramatically when compared with
values observed before treatment. In the exenatide
group, reductions in body weight (4.16 ± 5.32 vs. 1.98 DISCUSSION
± 3.28), BMI (1.31 ± 0.98 vs. 0.69 ± 0.94), WHR (0.04 With the elevation in the standard of living, prevalence
Copyright© ABE&M todos os direitos reservados.

± 0.02 vs. 0.02 ± 0.02), ALT (27.32 ± 15.96 vs. 12.85 ± of obesity and T2DM increases. Studies have shown
11.38), AST (7.89 ± 7.8 vs. 75.11 ± 6.98), γ-GT (26.48 that obesity may increase the risk of DM, as well as
± 17.34 vs. 10.26 ± 14.11), and hs-CRP (0.89 ± 0.59 cardiovascular diseases and NAFLD. In clinical practi-
vs. 0.61 ± 0.54), and increases in AST/ALT (0.23 ± ce, we usually encounter patients that develop T2DM,
0.08 vs. 0.11 ± 0.09), adiponectin (1.86 ± 2.22 vs. 0.76 obesity and NAFLD simultaneously (12,13). Insulin
± 1.30) and 2-h PPG (4.16 ± 3.01 vs. 2.63 ± 2.28) were resistance may compromise the ability of insulin to

Arq Bras Endocrinol Metab. 2013;57/9 705

 Exenatide improves type 2 diabetes

Table 4. Proportion of patients with HbA1c > 7% and ALT < 40 U/L
Concomitant
HbA1c < 7% HbA1c > 7% Absence of NAFLD ALT < 40 U/L ALT > 40 U/L
NAFLD
Exenatide (n = 49) 18 (36.7%) 31 (63.3%) 39 10 27 (55.1%) 22 (44.9%)
Metformin (n = 68) 28 (41.2%)* 40 (58.8%)* 61 7 18 (26.5%) 50 (73.5%)
Value 0.24 2.35 9.86
P 0.62 0.13 0.02*
* P < 0.05 between two groups after treatment.

inhibit lipase, which leads to the increased degradation T2DM, and metabolic syndrome (16). Currently, some
of peripheral adipose tissue and elevated free fatty acid. studies reveal that NAFLD is closely related to the in-
Then, oxidation and consumption of free fatty acid are creased risk of cardiovascular disease (17). In NAFLD
insufficient, resulting in lipoylation of fatty acid into patients, severe insulin resistance, sustained hyperinsu-
triglycerides, and increased triglyceride levels. Triglyce- linism, disordered lipid metabolism (increase in TG,
ride is then used to synthesize fat, which deposits in the reduction in HDL-C, and elevation in LDL-C) and
liver, resulting in fatty liver. The deposition of fat in the subclinical inflammatory state may serve as risk factors
liver may reduce the clearance of insulin in the liver, to promote the occurrence and development of athe-
which aggravates the peripheral and hepatic insulin re- rosclerosis (17,18). On the other hand, NAFLD may
sistance, resulting in a positive feedback amplification. result in progressive hepatitis, hepatic fibrosis or cirrho-
Thus, hepatic insulin resistance has been regarded as sis, or even liver failure (19). Thus, for T2DM patients
an important characteristic of NAFLD. In addition, the with concomitant NAFLD, therapy targeting insulin
imbalanced activities of adipocytokines might be a cau- resistance may improve NAFLD. For example, body
se of NAFLD progression after body weight loss and weight loss may not only reduce blood glucose, but
reduction in body fat. may also be beneficial for blood pressure, blood lipids
The adipose tissue is not only an important organ and the liver, leading to reduction in risk of cardiovas-
for energy storage, but also a key endocrine organ. It cular diseases.
can produce some adipocytokines via autocrine and pa- In previous therapies for T2DM (such as application
racrine mechamisms. Of these adipocytokines, leptin, with insulin and insulin secretagogues), hypoglycemia
adiponectin, resistin, tumor necrosis factor, and inter- and body weight increase were usually encountered
leukin have been found to be closely related to NAFLD. in clinical practice. In a large prospective study (AC-
In addition, hs-CRP is another important inflammatory CORD study), the investigators have proposed caution
factor and has been regarded as a pivotal factor that can on the goal and safety of lowering glucose levels (20).
predict chronic inflammatory reaction, metabolic syn- In addition, previous therapies often fail to improve
drome, and cardiovascular diseases (14). Adiponectin insulin resistance and have limitations in improving
is one of proteins with high expression in the adipose cardiovascular complications. Besides lifestyle interven-
tissues, which can inhibit inflammation. Arvaniti and tions, such as exercise and diet control, metformin and
cols. (15) found that hypoadiponectinemia could indu- thiazolidinedione have been applied in the treatment
ce fat deposition in the liver and the progression of fatty of DM, but their therapeutic efficacy is not ideal. A
hepatitis, and it is the most important adipocytokine in recent meta-analysis reveals that lifestyle intervention
the initiation and progression of NAFLD. Thus, insu- is comparable to metformin in the improvement of li-
lin resistance, increase in inflammatory cytokines, such ver histology and hepatic enzymes in NAFLD patients
as hs-CRP, and reduction in adiponectin act synergis- (21). Thus, we pose the question if GLP-1 may be a
tically to promote the occurrence and development of promising anti-DM drug.
Copyright© ABE&M todos os direitos reservados.

T2DM and NAFLD. Exenatide is an incretin analogue and can mimic

The majority of patients with NAFLD have no evi- the actions of GLP-1. For example, it can promote the
dent clinical symptoms and only present increase in glucose-dependent insulin secretion, inhibit the inap-
ALT and/or AST. The mild to moderate increase in propriate secretion of glicentin, delay gastric emptying,
ALT and γ-GT and reduction in AST/ALT have been and suppress appetite to reduce blood glucose and
considered as characteristics closely related to NAFLD, body weight (22). In a randomized, controlled study

706 Arq Bras Endocrinol Metab. 2013;57/9

 Exenatide improves type 2 diabetes

on T2DM, exenatide was superior to insulin glargine in ALT, which influences liver function, increases the
in improving the risk factors of cardiovascular diseases risk for hepatic cirrhosis and liver failure, and limits the
(such as adiponectin and hs-CRP) (23). It is a question application of oral anti-diabetic drugs. In addition, ap-
to be answered whether exenatide is superior to me- plication of insulin may increase body weight. Thus,
tformin in improving NAFLD and related biochemical short term application of exenatide may interrupt this
variables. vicious cycle to control blood glucose, reduce body
In the present study, T2DM patients with NAFLD weight, and improve the liver function.
were recruited, and the effects of exenatide and me- Therefore, for T2DM patients with NAFLD, exena-
tformin on the blood glucose, NAFLD-related varia- tide as a GLP-1 receptor agonist may not only reduce
bles (such as ALT, AST, and γ-GT) were compared. In blood glucose and body weight, but also improve liver
addition, ultrasonography was done to evaluate the fatty enzymes, attenuating NAFLD. For NAFLD, the im-
liver before and after treatment. Our results demons- provement of liver histology is a gold standard in the
trated that exenatide was comparable to metformin in evaluation of therapeutic efficacy. However, liver histo-
the reduction of blood glucose, but it was superior to logy was not conducted in the present study. In future
metformin in reducing body weight, decreasing WHR, studies, long-term treatment with exenatide and a large
and improving hepatic enzymes. Although ultrasono- sample size are required to confirm the effects of exena-
graphy showed that the proportion of patients with im- tide on NAFLD, insulin resistance, and atherosclerosis
provement of fatty liver was comparable between two and to determine whether exenatide can be of benefit
groups, possibly due to short term treatment and/or to NAFLD patients.
small sample size, exenatide had a slightly better ability
Disclosure: no potential conflict of interest relevant to this article
to improve the fatty liver. HOMA-IR was employed to was reported.
evaluate insulin resistance, and results did not indicate
differences between the two groups. Thus, exenatide
has special effect on the liver, which is different from REFERENCES
metformin. Moreover, exenatide is better to control 2-h 1. Sass DA, Chang P, Chopra KB. Nonalcoholic fatty liver disease: a
PPG when compared with metformin, and our results clinical review. Dig Dis Sci. 2005;50 (1):171-80.
showed that serum adiponectin in patients treated with 2. Yang W, Lu J, Weng J, Jia W, Ji L, Xiao J, et al. Prevalence of
diabetes among men and women in China. N Engl J Med.
exenatide was higher than that in metformin-treated pa- 2010;362(12):1090-101.
tients, but hs-CRP was reduced in exenatide-treated pa- 3. Leavens KF, Birnbaum MJ. Insulin signaling to hepatic lipid
tients when compared with metformin-treated patients. metabolism in health and disease. Crit Rev Biochem Mol Biol.
2011;46:200-15.
Whether exenatide has a sustained effect in improving
4. Targher G, Day CP, Bonora E. Risk of cardiovascular disease in
abdominal obesity or whether exenatide interrupts the patients with nonalcoholic fatty liver disease. N Engl J Med.
vicious circle between insulin resistance and NAFLD 2010;363:1341-50.
is still unclear. The failure of metformin to improve 5. Huang MA, Greenson JK, Chao C, Anderson L, Peterman D,
Jacobson J, et al. One-year intense nutritional counseling
NAFLD might be related to the absence of improve- results in histological improvement in patients with
ment of adiponectin and hs-CRP, which suggests that nonalcoholic steatohepatitis: a pilot study. Am J Gastroenterol.
2005;100(5):1072-81.
metformin has no ability to improve oxidative stress in-
6. Wang RT, Koretz RL, Yee HF Jr. Is weight reduction an effective
duced by fatty acid deposition and inflammation in the therapy for nonalcoholic fatty liver? A systematic review. Am J
liver (24). The GLP-1 analogue may improve fatty liver Med. 2003;115(7):554-9.
and insulin resistance. This may be explained by the fact 7. Kadayifci A, Merriman RB, Bass NM. Medical treatment of non-
alcoholic steatohepatitis. Clin Liver Dis. 2007;11(1):119-40.
that GLP-1 may directly act on hepatocytes and activate
8. Musso G, Gambino R, Cassader M. A meta-analysis of
the expression of genes related to the β-oxidation of randomized trials for the treatment of nonalcoholic fatty liver
fatty acids and insulin sensitivity (such as PPAR-γ and disease. Hepatology. 2010;52(1):79-104.
Copyright© ABE&M todos os direitos reservados.

PPAR-α), which lead to the improvement of insulin re- 9. Eng J, Kleinman WA, Singh L, Singh G, Raufman JP. Isolation
and characterization of exendin-4, an exendin-3 analogue, from
sistance in the NAFLD patients (25). Heloderma suspectum venom. Further evidence for an exendin
Currently, exenatide treatment is costly, which sig- receptor on dispersed acini from guinea pig pancreas. J Biol
nificantly limits its wide application in clinical practice. Chem. 1992;267(11):7402-5.
10. Ding X, Saxena NK, Lin S, Gupta NA, Anania FA. Exendin-4, a
However, most of T2DM patients with abdominal obe- glucagon-like protein-1 (GLP-1) receptor agonist, reverses hepatic
sity develop NAFLD characterized by evident increase steatosis in ob/ob mice. Hepatology. 2006;43(1):173-81.

Arq Bras Endocrinol Metab. 2013;57/9 707

 Exenatide improves type 2 diabetes

11. Klonoff DC, Buse JB, Nielsen LL, Guan X, Bowlus CL, Holcombe 18. Pacana T, Fuchs M. The cardiovascular link to nonalcoholic fatty
JH, et al. Exenatide effects on diabetes, obesity, cardiovascular liver disease: a critical analysis. Clin Liver Dis. 2012;16(3):599-613.
risk factors and hepatic biomarkers in patients with type 2 diabetes 19. Ali R, Cusi K. New diagnostic and treatment approaches in non-
treated for at least 3 years. Curr Med Res Opin. 2008;24(1):275-86. alcoholic fatty liver disease (NAFLD). Ann Med. 2009;41(4):265-78.
12. Bray GA. Medical consequences of obesity. J Clin Endocrinol 20. American Gastroenterological Association. American Gastroente-
Metab. 2004;89(6):2583-9. rological Association medical position statement: nonalcoholic fa­
13. Mokdad AH, Ford ES, Bowman BA, Dietz WH, Vinicor F, Bales VS, tty liver disease. Gastroenterology. 2002;123(5):1702-4.
et al. Prevalence of obesity, diabetes, and obesity-related health 21. Buse JB, Ginsberg HN, Bakris GL, Clark NG, Costa F, Eckel R, et
risk factors, 2001. JAMA. 2003;289(1):76-9. al.; American Heart Association; American Diabetes Association.
Primary prevention of cardiovascular diseases in people with
14. Zimmermann E, Anty R, Tordjman J, Verrijken A, Gual P, Tran A, et
diabetes mellitus: a scientific statement from the American Heart
al. C-reactive protein levels in relation to various features of non-
Association and the American Diabetes Association. Diabetes
alcoholic fatty liver disease among obese patients. J Hepatol.
Care. 2007;30:162-72.
2011;55(3):660-5.
22. Gallwitz B. Glucagon-like peptide-1 analogues for Type 2 diabetes
15. Arvaniti VA, Thomopoulos KC, Tsamandas A, Makri M,
mellitus: current and emerging agents. Drugs. 2011;71(13):1675-88.
Psyrogiannis A, Vafiadis G, et al. Serum adiponectin levels in
23. Bunck MC, Diamant M, Eliasson B, Cornér A, Shaginian RM,
different types of non alcoholic liver disease. Correlation with
Heine RJ, et al. Exenatide affects circulating cardiovascular risk
steatosis, necroinflammation and fibrosis. Acta Gastroenterol
biomarkers independently of changes in body composition.
Belg. 2008;71(4):355-60. Diabetes Care. 2010;33(8):1734-7.
16. Schindhelm RK, Diamant M, Dekker JM, Tushuizen ME, Teerlink T, 24. Van Wagner LB, Rinella ME. The role of insulin-sensitizing agents
Heine RJ. Alanine aminotransferase as a marker of non-alcoholic in the treatment of nonalcoholic steatohepatitis. Therap Adv
fatty liver disease in relation to type 2 diabetes mellitus and Gastroenterol. 2011;4(4):249-63.
cardiovascular disease. Diabetes Metab Res Rev. 2006;22(6):437-43. 25. Svegliati-Baroni G, Saccomanno S, Rychlicki C, Agostinelli L, De
17. Hanley AJ, Williams K, Festa A, Wagenknecht LE, D’Agostino RB Minicis S, Candelaresi C, et al. Glucagon-like peptide-1 receptor
Jr, Haffner SM. Liver markers and development of the metabolic activation stimulates hepatic lipid oxidation and restores hepatic
syndrome: the insulin resistance atherosclerosis study. Diabetes. signalling alteration induced by a high-fat diet in nonalcoholic
2005;54(11):3140-7. steatohepatitis. Liver Int. 2011;31(9):1285-97.
Copyright© ABE&M todos os direitos reservados.

708 Arq Bras Endocrinol Metab. 2013;57/9