Diabetes Research and Clinical Practice - Volume 120 - Supplement 1 - October 2016

215 / 244

PJ-12
Comparisons of placebo effect of hypoglycemic drugs between
Asian and Caucasian type 2 diabetes patients
Xiaoling CAI1*, Lingli ZHOU1, Wenjia YANG1, Xueyao HAN1,
Linong JI 1. 1Peking University People’s Hospital, China
Aim: It was hypothesis that the placebo effect might be
different between different ethnicity. The aim of this study is
to compare the placebo effect of hypoglycemic treatment
between Asian and Caucasian type 2 diabetes patients.
Methods: The MEDLINE®, EMBASE® , CENTRAL were searched
and qualified studies were included. References were collected
until Dec. 2015. All the studies were double blind, placebo-
controlled randomized trials in type 2 diabetes patients; study
length of ≥12 weeks with the efficacy evaluated by changes in
HbA1c from baseline in groups.
Results: 250 studies compared a placebo with an active
hypoglycemic agent either in Asian (n = 40) or in Caucasian
(n = 210), placebo effect in HbA1c was comparable between
Asian (0.09%, 95%CI, 0 to 0.18%) and Caucasian (−0.02%, 95%CI,
− 0.15 to 0.10%), placebo effect in FPG was comparable between
Asian (0.21 mmol/L; 95% CI, 0.07 –0.34 mmol/L) and Caucasian
(0.1 mmol/L; 95% CI, −0.09 to 0.29 mmol/L), placebo effect in
weight was also comparable between Asian ( −0.22 kg; 95% CI,
− 0.48–0.03 kg) and Caucasian (−0.02 kg; 95% CI, −0.48–0.44 kg).
40 studies compared a placebo with an AGI either in Asian
(n = 25) or in Caucasian (n = 17), placebo effect in HbA1c was
comparable between Asian ( −0.13, 95%CI, −0.31 to 0.04) and
Caucasian (0.08, 95%CI, −0.11 to 0.27). 83 trials compared a
placebo with a TZD either in Asian (n = 38) or in Caucasian
(n = 46), placebo effect in HbA1c was superior in Asian ( −0.33,
95%CI, −0.64 to −0.03) to that in Caucasian (0.12, 95%CI, −0.03
to 0.27). 57 trials compared a placebo with a DPP-IV inhibitors
either in Asian (n = 20) or in Caucasian (n = 39), placebo effect in
HbA1c was comparable between Asian (0.02, 95%CI, −0.19 to
0.23) and Caucasian ( −0.13, 95%CI, −0.26 to 0.01). For SU
treatment, metformin treatment as well as SGLT2 inhibitors
treatment, no enough studies were found for the comparisons
between Asian and Caucasian population.
Conclusion: Results from this meta-analysis indicated that
the placebo effect in hypoglycemic drugs was comparable
between Asian and Caucasian type 2 diabetes patients.
PJ-13
Safety and efficacy comparison of different insulin regimens in
T2DM patients: A meta-analysis
Xueying GAO 1, Xiaoling CAI1, Wenjia YANG1, Linong JI1*.
1 Peking University People’s Hospital, China
To compare the safety and efficacy in T2DM patients treated
with different insulin regimens, we searched the following
databases: MEDLINE, EMBASE, CENTRAL. References were
collected until Dec. 2015. The main search concepts were
type 2 diabetes, NPH insulin, long acting insulin analogs,
human regular insulin, rapid insulin analogs, premixed
insulin, premixed insulin analogs, randomized controlled
trials (RCTs), and clinical trials. Inclusion criteria were: (1)
T2DM patients aged >18 years; (2) RCTs with at least 4 weeks of
follow-up; (3) different insulin regimens were evaluated and
compared.
A total of 74 articles were included in this review. (1) NPH
insulin versus long acting insulin analogs therapy (twenty-two
studies): Treatment with long acting insulin analogs was
associated with a significantly greater decrease in HbA1c and
FPG level, a significantly lower increase in body weight, and a
significantly lower risk of hypoglycemia. Treatment with NPH
insulin was associated with a significantly lower insulin
dosage. No statistically significant difference was found in
terms of all-cause mortality between two groups. (2) Human
regular insulin versus rapid insulin analogs therapy (nine
studies): Treatment with rapid insulin analogs was associated
with a significantly greater decrease in HbA1c. No statistically
significant difference was found in FPG change, body weight
change, insulin dosage, rate of hypoglycemia, and all-cause
mortality between two groups. (3) Premixed insulin versus
premixed insulin analogs therapy (six studies): Treatment
with premixed insulin was associated with a significantly
greater increase in body weight. No statistically significant
difference was found in HbA1c change, FPG change, insulin
dosage, rate of hypoglycemia, and all-cause mortality between
two groups. (4) Premixed insulin versus basal insulin therapy
(twenty studies): Treatment with premixed insulin was
associated with a significantly greater decrease in HbA1c
level, and a significantly greater increase in body weight.
Treatment with basal insulin was associated with a signifi-
cantly lower insulin dosage, and a significantly lower risk of
hypoglycemia. No statistically significant difference was
found in FPG change, and all-cause mortality between two
groups. (5) Premixed insulin versus basal-bolus/bolus insulin
therapy (twenty-one studies): Treatment with basal-bolus/
bolus insulin was associated with a significantly greater
decrease in FPG change. No statistically significant difference
was found HbA1c change, body weight change, insulin dosage,
rate of hypoglycemia, and all-cause mortality between two
groups.
Results from this meta-analysis comprehensively evaluated
the glucose control and the hypoglycemic rate as well as all-
cause mortality in different kinds of insulin treatment.
PJ-14
Metformin as an anticancer agent in breast cancer therapy by
regulating tumor associated macrophage polarization and
function
Chi-Fu CHIANG1*, Yi-Jen HUNG2, Yi-Shing SHIEH3,4 ,
Chien-Hsing LEE1,2. 1Graduate Institute of Medical Sciences,
National Defense Medical Center, 2Division of Endocrinology
and Metabolism, Department of Internal Medicine, Tri-Service
General Hospital, National Defense Medical Center, 3School of
Dentistry, National Defense Medical Center, 4Department of
Oral Diagnosis and Pathology, Tri-Service General Hospital,
Taipei, Taiwan
Background: Accumulated evidence suggests that diabetic
patients treated with metformin had a significantly lower risk
of developing cancer or a lower cancer mortality. Recent
evidence suggested that the phenotype of TAMs varies from
M1 to M2 with the stage of tumor progression. However, it is
not known if metformin is involved in TAMs phenotype switch
to affect tumor malignant behaviors.
Material and methods: We used the THP-1 macrophage
cultured with breast cancer conditioned medium as tumor
microenvironment model.
Results: We found that metformin significantly switched from
M2 to M1 phenotype in breast cancer conditioned medium, by
reduced expression of CD206, down-regulation of M2 marker
mRNA, and enhanced expression of CD16, up-regulation of M1
marker mRNA. But metformin not be direct influence THP-1
macrophage polarization. Moreover, we found that metformin
can affect cytokines secretion in the breast cancer, by reduced
expression of IL-4, IL-10, IL-13, and induced IFN-γ through
AMPK-NF-κB signaling to regulate, and then affect macro-
phage polarization. Administration of CC, another inhibitor of
AMPK, also blocked the switched fromM2 to M1 phenotype. In
tumor tissue, the percentage of M2-like macrophage was
decreased and M1-like macrophage was increased in the
metformin group.
Conclusion: These findings suggest that metformin
treatment can induce cytokines secretion and expression by
AMPK-NF-κB pathway in breast cancer and then affect TAM
polarization.
Poster Presentations / Diabetes Research and Clinical Practice 120S1 (2016) S65–S211 S197