The effect of Metformin on endometrial tumor-regulatory genes and systemic metabolic parameters in polycystic ovarian syndrome – a proof-of-concept study

1. http://informahealthcare.com/gye
ISSN: 0951-3590 (print), 1473-0766 (electronic)
Gynecol Endocrinol, Early Online: 1–5
! 2014 Informa UK Ltd. DOI: 10.3109/09513590.2014.989982
ORIGINAL ARTICLE
The effect of Metformin on endometrial tumor-regulatory genes and
systemic metabolic parameters in polycystic ovarian syndrome –
a proof-of-concept study
Mohamad Nasir Shafiee1,2
, Dahlia Abd Malik1
, Ryia Illani Mohd Yunos3
, William Atiomo2
, Mohd Hashim Omar1
,
Nur Azurah Abdul Ghani1
, Ahmad Zailani Hatta1
, Claire Seedhouse4
, Caroline Chapman5
, and
Norfilza Mohd Mokhtar3,6
1
Department of Obstetrics and Gynaecology, Faculty of Medicine, Universiti Kebangsaan Malaysia, Cheras, Kuala Lumpur, Malaysia, 2
Division of
Obstetrics and Gynaecology and Child Health, Faculty of Medicine and Health Sciences, School of Medicine, Queen’s Medical Centre, Nottingham
University Hospital, Nottingham, UK, 3
UKM Medical Molecular Biology Institute (UMBI), Universiti Kebangsaan Malaysia Medical Centre, Kuala
Lumpur, Malaysia, 4
Department of Haematology, Clinical Sciences Building, 5
Division of Medical Sciences and Graduate Entry Medicine, Faculty of
Medicine and Health Sciences, University of Nottingham, Nottingham, UK, and 6
Department of Physiology, Faculty of Medicine, Universiti
Kebangsaan Malaysia, Jalan Raja Muda Abdul Aziz, Kuala Lumpur, Malaysia
Abstract
The aim of this proof-of-concept study was to determine the effects of three-month Metformin
therapy on the expression of tumor-regulatory genes (p53, cyclin D2 and BCL-2) in the
endometrium of women with polycystic ovary syndrome (PCOS). A total of 40 women, aged
between 21 and 45 years with PCOS (Rotterdam criteria) were recruited. The participants were
assessed at pre- and 3-month-post-Metformin therapy for the menstrual regularities, weight
reduction, Ferriman Galway scores, fasting blood glucose (FBG), total cholesterol, LDL, HDL and
p53, BCL-2 and cyclin D2 gene expression. Five participants conceived spontaneously after the
initial recruitment. Majority (68%) resumed regular menstrual cycles after Metformin. There
were significant reduction in BMI (p ¼ 0.001), weight (p ¼ 0.001) and Ferriman Galway scores
(p ¼ 0.001). A significant improvement was seen in mean FBG (p ¼ 0.002), total cholesterol
(p ¼ 0.001), LDL (p ¼ 0.003) and HDL cholesterol levels (p ¼ 0.015). Tumor suppressor gene (p53)
was significantly up-regulated after Metformin (10 out of 14 women), with p value 0.016. BCL-2
and cyclin D2 (oncogenes) were slightly up-regulated without significant difference (p ¼ 0.119
and 0.155, respectively). In conclusion, Metformin therapy improved clinical and metabolic
parameters in women with PCOS and up-regulated p53 tumor suppressor gene significantly.
Further studies are however required to independently validate our findings.
Keywords
Endometrial cancer, genes, metformin, PCOS
History
Received 8 October 2014
Accepted 16 November 2014
Published online 11 December 2014
Introduction
Polycystic ovarian syndrome (PCOS) has a wide disease spectrum
that affects 5–10% of women in the reproductive age [1,2]. Even
though many committees have proposed different criteria to
diagnose this condition, the Rotterdam criteria, 2004 is now
widely accepted in many parts of world. Symptoms of chronic
anovulation (oligo or amenorrhoea), clinical and/or biochemical
evidence of excess androgens and sonographic polycystic appear-
ance of the ovaries are now the integral features used to diagnose
PCOS using the Rotterdam criteria, 2004 [3].
The disease (PCOS) implications include menstrual irregula-
rities, subfertility, cardiovascular diseases, sleep apnoea, endo-
metrial hyperplasia and endometrial cancer (EC). Women with
PCOS are approximately three times more likely to develop EC
compared with women without the condition which translates into
a 9% lifetime risk of EC in Caucasian women with PCOS
compared with 3% in women without the condition [4]. The
precise molecular mechanisms which increase EC risk in women
with PCOS are, however unclear but they may include raised
estrogen levels, hyperinsulinemia and reduced apoptosis in
women with PCOS [5].
Current strategies to reduce EC risk in PCOS are aimed at
ameliorating risk factors for EC in women with PCOS such as
obesity, raised estrogen levels, anovulation and endometrial
hyperplasia (EH). Weight loss is the induction of regular menstrual
withdrawal bleeding with progestogens in obese amenorrhoeic
women with PCOS. The treatment of EH with oral progestogens,
the levonorgestrel releasing IUCD or a hysterectomy in PCOS
women with atypical hyperplasia are currently the mainstay of
Address for correspondence: Dr Mohamad Nasir Shafiee, MD, MMed
(O&G), MRCOG, FICS, Division of Obstetrics and Gynaecology and
Child Health, The School of Medicine, Faculty of Medicine and Health
Sciences, Queen’s Medical Centre, Nottingham University Hospital,
Derby Road, Nottingham NG7 2UH, UK. Tel: +44 7565686679. E-mail:
mgxmnsh@nottingham.ac.uk
GynecolEndocrinolDownloadedfrominformahealthcare.comby86.30.66.46on12/11/14
Forpersonaluseonly.

2. reducing EC risk in PCOS [6]. Progesterone hormones currently
used to treat EH in women with PCOS are associated with many
side effects which make compliance a problem, weight loss is not
always effective and its efficacy is relatively untested. Even though
levonorgestrel releasing IUCD is the available option, but it is
relatively more expensive and not acceptable for certain women.
Besides, hysterectomy is invasive, risky and not the best option in
young women.
Alternative treatments are therefore required. It has, for
example, been shown in both animal and human studies that
Metformin, a drug which improves insulin resistance may be of
value in preventing EC in PCOS [6–8]. A case report found
that Metformin reversed atypical endometrial hyperplasia in two
PCOS women when progestogen failed to respond [9]. In
another study of endometrial cancer cell lines, Metformin
inhibited carcinogenesis through mTOR pathways [7]. There
are, however, no in vivo studies so far specifically investigating
the molecular effects of Metformin on endometrial carcinogen-
esis in high-risk individuals such as women with PCOS. In vivo
studies to investigate the molecular mechanisms which
underpin EC risk prevention in PCOS by Metformin are
required to justify its wider use for this indication and optimize
treatment.
The aim of this study therefore was, as a proof-of-concept, to
investigate the effects of a short course of three months Metformin
therapy on the expression of a panel of key tumor-regulatory
genes (p53, cyclin D2 and BCL-2) in the endometrium of women
with PCOS, as well as the impact of Metformin on their clinical
and systemic biochemical parameters.
Materials and methods
Study protocol
This prospective study was conducted in the Department
of Obstetrics and Gynecology, Universiti Kebangsaan
Malaysia (UKM) Medical Centre in collaboration with UKM
Medical Molecular Biology Institute (UMBI) from June 2012 to
December 2013. It was approved by the UKMMC Research
Ethics Committee (FF-392-2012) and funded by the Young
Researcher’s Grant, Universiti Kebangsaan Malaysia (GPM-
082-2012).
Women with PCOS were diagnosed using the Rotterdam
European Society for Human Reproduction and Embryology
(ESHRE) and the American Society of Reproductive Medicine
(ASRM) criteria when two out of three criteria were met
(oligoovulation and/or anovulation, clinical and/or biochemical
evidence of androgen excess and polycystic appearance of
ovaries). As PCOS is a diagnosis by exclusion, we have excluded
other diseases prior to recruitment. The participants were between
21 and 45 years of age and not on any hormonal treatment.
Pregnancy was excluded prior to the recruitment. The appoint-
ment list in the Gynecology Outpatient Clinic was screened and
any potential women were approached and assessed for their
eligibility to participate in this study. Prior to the recruitment, the
Patient Information Sheet (PIS) about the study was circulated
and discussed before a written consent was signed up.
The study proforma comprised of two parts. The first part
was the participant’s background information, obtained from
history taking by the researcher. This included age, race,
menstrual cyclicity, age of menarche, medical history of diabetes
mellitus and hypertension and family history of endometrial
malignancy. The weight and BMI were calculated and assess-
ment of hirsutism was carried out using Ferriman Galway score.
The second part was a set of information to assess the clinical
outcomes following three-month therapy of Metformin 850 mg
twice daily. The participants were informed regarding the
possible side effects and were asked to contact the research
team members directly for any advice including potential
withdrawal from this study, without jeopardizing their subse-
quent clinical care. Any side effects or reasons for Metformin
withdrawal were recorded. There was no specific advice
on weight reduction programme given, but the participants
were allowed to perform their routine exercises. Blood inves-
tigations for fasting blood sugar, low density lipoprotein (LDL),
high density lipoprotein (HDL) and triglycerides levels were
analyzed using the Cobas8000Õ
System by Roche (Basel,
Switzerland). For follicular stimulating hormone (FSH), leuti-
nizing hormone (LH) and testosterone levels, we utilized an
immunoassay method using Architect 12000SRÕ
equipment by
Abbott (Illinois, USA). The result was validated after a serial of
verification methods namely precision, accuracy, linearity and
method comparison studies.
Endometrial tissue samples were taken using a Pipelle sampler
device pre- and 3-month post-Metformin proliferative phase
endometrium. The specimen was snap frozen in the liquid
nitrogen until further analysis.
Quantitative real-time polymerase chain reaction (qPCR)
RNA extraction was performed using the QIAshredder kit (Qiagen,
Valencia, CA) and further purified by the RNeasy Mini-kit
(Qiagen, Valencia, CA). The reverse transcription was performed
using the RevertAid First Strand cDNA Synthesis Kit (Thermo
Scientific,Waltham, MA). Reverse transcription was carried out at
25
C for 5 min, 42
C for 60 min and 70
C for 5 min. PCRs were
performed using Solaris qPCR Gene Expression Master Mix
(Thermo Scientific) in Rotor Gene Q (Qiagen, Valencia, CA).
The PCR conditions consist of 15 min step at 95
C, 40 cycles at
95
C for 15 s each and 1 min at 60
C. A housekeeping gene
glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as
an internal control to correct differences in the amount of RNA in
each sample. The primer and probe used for gene of interest and the
housekeeping gene are listed in Table 1. Each experiment was
performed in triplicate and repeated twice to assess for consistency
of the results. Relative gene expression analysis calculation was
carried out by using the 2ÀDDCT
method.
Statistics
We were unable to do a baseline sample size calculation prior to
this proof-of-concept study, as we could not identify any previous
studies on the effect of Metformin on the expression of the genes
in the endometrium of women with PCOS to enable us to do a
baseline calculation. We however felt a sample size of 40 would
produce sufficient numbers to be able to measure the variables of
interest and prove the concept.
SPSS version 21 (Chicago, IL) statistical program was used to
capture all the data. Prevalence was presented as percentage. The
normality test was performed using Shapiro normality test.
Depending on the number of independent variables, one-way-
ANOVA test or independent t-test was used for the baseline
characteristics data. For normally distributed continuous variable,
means ± standard deviations was advocated whilst medians for
not normally distributed variables. The non-parametric test and
Wilcoxon-signed rank test were used for the comparison between
pre- and post-Metformin, and Kruskal–Wallis test was used for
the correlations between variables.
Using GAPDH as the housekeeping gene, the relative thresh-
old cycle (Ct) value of the genes of interest (p53, BCL2 and cyclin
D2) was measured at pre- and post-Metformin, and the fold
changes of genes expression was calculated using the
Comparative Ct Method (2ÀDDCt
).
2 M. N. Shafiee et al. Gynecol Endocrinol, Early Online: 1–5
GynecolEndocrinolDownloadedfrominformahealthcare.comby86.30.66.46on12/11/14
Forpersonaluseonly.

3. Results
Demographics
A total of 40 women with PCOS were recruited in this study and
started on Metformin 850 mg twice daily. However, five of them
conceived spontaneously following the initial sampling and were
excluded. Out of 35 participants, only 14-pair samples were
suitable for the gene analysis. The other samples from 21
participants were not analyzed due to either inadequate sample,
denatured or no pre/post-therapy match.
The mean age of participants was 32.33 years (SD 5.68).
A total of 17 participants were nulliparous with subfertility. The
other 23 participants were multiparous, with irregular menses or
secondary subfertility. Majority (85%) of them had irregular
menses. Twenty-two percent of the participants had impaired
glucose tolerance, 9% had hypertension and 5% had both. A total
of 70% of the participants had a family history of diabetes
mellitus. In addition, 30% and 40% of them had a family history
of hypertension and ischaemic heart disease respectively. Twenty-
three percent had a first and/or second degree family history of
gynaecological cancer.
Side-effects of Metformin
Majority of the participants (33/40) were ideal users (fully
compliance) and only 17.5% did not take Metformin regularly as
instructed (but managed to complete more than 85% of the
medication). Five participants (subfertile) became pregnant after
three months compliance to Metformin. The commonest side
effects were nausea and vomiting, accounted for 55% of all
participants, followed by gastric discomfort, indigestion, diarrhea
and headache in 18%, 12%, 9% and 7%, respectively. All the side
effects were reported as transient and occurred at the beginning of
Metformin use. These resolved within four weeks of Metformin
commencement in all participants. During the study period, none
of the participants reported episodes of hypoglycemia.
Clinical effects of Metformin therapy
The menstrual cycle regularity was improved. About 67.5% of the
participants had regular menses (in previously irregular menses)
after 3 months Metformin. The five pregnant participants had
irregular menses prior to their conception. There were statistically
significant improvements in their weights (3 kg mean reduction,
p ¼ 0.001), BMI (3.3 kg/m2
mean reduction, p ¼ 0.001) and FGS
(p ¼ 0.001) after Metformin.
Biochemical effects of Metformin therapy
Four participants (10%) had elevated fasting blood glucose (FBG)
and 26 of them had abnormal total cholesterol (65%) at initial of
the study. Table 1 shows the statistical difference of FBG and lipid
profiles at baseline and after 3 months of Metformin therapy.
After Metformin use, the glycemic control improved significantly
(p ¼ 0.002) with a reduction of mean FBG of 0.30 mmol/L. There
was also a decline in means of total cholesterol, LDL and
triglycerides levels, and an increase in HDL levels. The differ-
ences in total cholesterol and LDL levels were statistically
significant (p50.05). The mean LH level dropped from 7.6 to
7.15 IU/L (p ¼ 0.015) (Table 1).
Association between post-treatment weight reduction
and menstrual regularity and biochemical profiles
The mean weight was reduced from 75.00 to 72.00 kg whilst the
BMI reduced from 31.30 to 28.00 (p50.05). A statistically
significance association was seen between the menstrual change
post-Metformin and weight lost (p ¼ 0.007). Participants who
attained regular menses (from previously irregular) had higher
percentage of weight reduction (Table 2).
Tumor-regulatory genes expression following 3-month
Metformin
A total of 14 paired endometrial tissue samples were successfully
obtained and analyzed from PCOS women in our study. The p53
tumor suppressor gene was significantly up-regulated in 10
participants after Metformin. The mean fold-change was 2.08.
Statistically with n ¼ 14 (adequate samples for gene analysis), the
p53 gene up-regulation was significant with p ¼ 0.016 (Figure 1).
The BCL-2 and cyclin D2 gene expression were down-regulated
in 6 and 7 participants out of 14 samples, respectively (p50.05).
However, the non-parametric t-test revealed both genes (BCL-2
and cyclin D2) were also up-regulated, but not statistically
significant (p ¼ 0.119 and p ¼ 0.155 respectively) (Figure 1).
Subgroup analysis found that eight participants (out of 10) whom
had up-regulated p53 gene regained their regular menses after
Table 1. Comparison of fasting blood glucose and fasting lipid profile and luteinizing hormone at baseline
and after 3 months of Metformin therapy.
Metformin
Pre-Metformin (n ¼ 40) Post-Metformin (n ¼ 35)
Mean Mean p Value
Fasting blood glucose (mmol/L) 5.12 (4.36–8.61) 4.55 (4.25–7.98) 0.002*
Total cholesterol (mmol/L) 5.50 (3.76–7.23) 5.20 (3.51–6.71) 0.000*
HDL (mmol/L) 1.37 (1.02–2.15) 1.57 (1.32–2.54) 0.074
LDL (mmol/L) 3.39 (2.23–4.53) 3.03 0.003*
Triglycerides (mmol/L) 1.12 (0.98–1.64) 1.10 (0.87–1.56) 0.200
LH (IU/L) 7.60 (6.23–8.75) 7.15 (6.67–8.34) 0.015*
HDL, high-density lipoprotein; LDL, low-density lipoprotein; LH, luteinizing hormone.
*Significant at p50.05.
Table 2. Relationship between weight loss and menstrual cycle following
Metformin therapy.
No change in
menstrual cycle
post-
Metformin
Positive change
in menstrual
cycle post-
Metformin p Value
Weight loss 0–4.9%, N (%) 5 (12.5) 3 (7.5) 0.007*
Weight loss 5–9.9%, n (%) 8 (20) 14 (35)
Weight loss 410%, n (%) 0 (0) 10 (25)
*Significant at p50.05.
DOI: 10.3109/09513590.2014.989982 Metformin effects on PCOS 3
GynecolEndocrinolDownloadedfrominformahealthcare.comby86.30.66.46on12/11/14
Forpersonaluseonly.

4. Metformin. Spearman correlation test found a significant correl-
ation between age (r ¼ 0.567; CI ¼ 0.034–0.849) and p53 expres-
sion (p ¼ 0.037). In addition, there was an inverse correlation
between p53 expression with cholesterol (r ¼ À0.597; CI ¼
À0.861 to À0.079) and triglyseride (r ¼ À0.689; CI ¼ À0.897
to À0.233) with p value 0.026 and 0.008, respectively. However,
the weight loss was not significantly correlated with the gene
expression (p ¼ 0.361).
Discussion
This is the first study as far as we know to have investigated the
effect of Metformin on genes associated with cancer in the
endometrium of women with PCOS. P53, the tumor suppressor
gene was significantly up-regulated while BCL-2 and cyclin D2
(both oncogenic genes) were down regulated in only half of the
participants. However, statistically they (BCL-2 and cyclin D2)
were insignificantly up-regulated (p40.05). Other effects of
Metformin included an improvement in menstrual regularity,
weight loss and biochemical parameters.
P53 is a tumor suppressor gene involved in many cancers
including those involving the endometrium [10], cervix [11] and
breast [12]. Its key role is in conserving stability by preventing
genome mutation. Mutation or damage of this gene results in
failure to produce a 53-kDa protein, which functions as an
essential element in the cell cycle. Inability to repair the faulty
DNA or cellular stresses leads to cell death [13]. The major
regulator of p53 is MDM2, which can trigger the degradation of
p53 by the ubiquitin system. A previous study reported that p53
and MDM2 levels were associated in endometrial tumor tissue,
signifying that p53 is inactivated by MDM2 in endometrial
cancer [14]. The finding of up-regulation of p53 in the
endometrium of women with PCOS following Metformin
treatment in our study, therefore supports the hypothesis that
Metformin could prevent endometrial cancer in high risk
women, namely PCOS by up-regulating p53 expression in the
endometrium.
With respect to BCL-2 and cyclin D2, a study on endometrial
cancer cell lines found that increased phosphorylated BCL2
antagonist of cell death (BAD) was associated with cisplatin
resistance [15]. Besides, an activation of estrogen receptors was
also found to indirectly increase BCL2/BAX expression in the
type 1 endometrial cancer tissue and in endometrial hyperplasia
[16]. The down regulation of BCL-2 by Metformin as found in our
study again supports the hypothesis that Metformin could prevent
endometrial cancer namely PCOS by modulating BCL-2 and
cyclin D2 expression in the endometrium. This hypothesis is also
supported by findings with respect to cyclin D2 which has an anti-
apoptotic effect in the cell cycle. In a study on endometrial cancer
cell lines, histone deacetylase inhibitors were found to have a
profound antigrowth activity associated with a down regulation of
both BCL2 and cyclin D1 and D2 [17].
Our study also found that Metformin modified other clinical
and biochemical risk factors for endometrial cancer, particularly
reducing the body weight, regulating menstrual cycles and
maintaining normoglycaemia. It was therefore not unreasonable
to expect similar changes molecularly. For example, we found a
significant drop in mean LH following three-month Metformin
and a significant correlation between menstrual regularities and
LH value (p ¼ 0.044). Hyperinsulinaemia which is the core
problem in PCOS, indirectly synergizes LH production leading to
hormonal imbalance and its consequences [18]. Undoubtedly
improving insulin insensitivity with Metformin could normalize
the pituitary–ovarian axis and improve menstrual cycles reducing
endometrial cancer risk in women with PCOS. How this pathway
interacts with endometrial gene expression requires further
research.
Majority (85%) of the participants had irregular menses, and
strikingly Metformin reverted their menstrual irregularities in
67.5% of them. Attaining regular menses may reflect ovulatory
cycles and a sound pituitary–ovarian axis, reducing endometrial
cancer risk which is associated with anovulatory menstrual cycles
[19]. The fact that 29.4% (5 in 17 participants whom resumed
their regular menstrual cycles) conceived spontaneously after
Metformin, was also consistent with the restoration of ovulatory
cycles.
Obesity is a known risk factor for endometrial cancer [19]. In
our study, the use of Metformin for three months resulted in a
statistically significant weight reduction and the average weight
loss was 3 kg. This was consistent with a study by Haas et al. in
2003 [20], which showed that Metformin produced statistically
significant reductions in weight and BMI. Barbieri [21] also
mentioned that in obese PCOS woman, the use of Metformin with
a low caloric diet was associated with more weight loss than a low
caloric diet alone.
These findings are consistent with other studies of the impact
of Metformin on cancer risk outside the context of PCOS [6]. For
example, in an animal study [22], Metformin reversed estrogen-
induced endometrial hyperplasia in mice. Studies have also shown
that insulin and insulin-like growth factors (IGFs) are thought to
be the precursors in carcinogenesis, including reports of an
exaggeration of insulin receptors in endometrial cancer [9,23].
Metformin may therefore also modulate endometrial cancer risk
in PCOS through its effect on the insulin pathway. The fact that
p53 is also indirectly involved in PI3K/Akt signalling pathways
[14,24], which is responsible for glucose metabolism, further
strengthens our hypothesis.
The strengths of our study were in its novelty, the rigorous
selection of women with PCOS using the most up to date
criteria, and the relatively reasonable sample size in the context
of a proof-of-concept study. It may be argued that the use of
placebo or a control group would have been desirable, we
however felt it would be unethical to administer Metformin to
women without PCOS and subject them to endometrial biopsies
or subject women with PCOS to endometrial biopsies but then
offering them a placebo drug, without data from a proof of
concept study justifying a comparative study, which we have
now achieved in this study.
BC
L-2
post-M
etform
in
p53
pre-M
etform
in
p53
post-M
etform
in
cyclin
D
2
post-M
etform
in
cyclin
D
2
pre-M
etform
inBC
L-2
pre-M
etform
in
0
1
2
3
Pre and Post Metformin
p53, cyclin D2 and BCL-2 expression
Relativegeneexpression
Figure 1. Relative p53, cyclin D2 and BCL-2 gene expression in pre- and
3-month post-Metformin therapy among PCOS women.
4 M. N. Shafiee et al. Gynecol Endocrinol, Early Online: 1–5
GynecolEndocrinolDownloadedfrominformahealthcare.comby86.30.66.46on12/11/14
Forpersonaluseonly.

5. Conclusion
In this proof-of-concept study, the use of Metformin for 3 months
in this cohort of women with PCOS showed that p53 the tumour
suppressor gene was significantly up-regulated while BCL-2 and
cyclin D2 both oncogenic genes were down regulated in 50% of
participants. However, generally both BCL-2 and cyclin D2 were
not significantly up-regulated. Other effects of Metformin found
included an improvement in menstrual regularity, weight loss and
biochemical parameters. These findings are consistent with the
hypothesis that the use of Metformin in women with PCOS may
reduce endometrial cancer risk by modulating cancer related
genes in the endometrium. Our findings now justify larger
placebo controlled studies to test this hypothesis. In addition,
discovery-driven research approaches involving genomic, prote-
omic and metabolomics experiments may enable a more com-
prehensive molecular analysis to identify new pathways that first
explain the molecular link between endometrial cancer and PCOS
and secondly identify other molecular pathways through which
Metformin may reduce endometrial cancer risk in PCOS, before
its regular use is recommended.
Acknowledgements
Our deepest gratitude to all participants and medical staff involved in this
study.
Declaration of interest
All authors declared no conflict of interest.
References
1. Chang RJ, Heindel JJ, Dunaif A, eds. Polycystic ovarian syndrome.
New York: Marcel Dekker, Inc; 2002:361–5.
2. Dahlgren E, Friberg LG, Johansson S. Endometrial carcinoma;
ovarian dysfunction – a risk factor in young women. Ere J Obstet
Gynecol Reprod Biol 1991;41:143–50.
3. The Rotterdam ESHRE/ASRM-sponsored PCOS Consensus
Workshop Group. Revised 2003 concensus on diagnostic criteria
and long-term health risks related to PCOS. Hum Reprod 2004;19:
41–7.
4. Haoula Z, Salman M, Atiomo W. Evaluating the association between
endometrial cancer and polycystic ovary syndrome. Hum Reprod
2012;27:1327–31.
5. Shafiee MN, Chapman C, Barrett D, et al. Reviewing the molecular
mechanisms which increase endometrial cancer (EC) risk in women
with polycystic ovarian syndrome (PCOS): time for paradigm shift?
Gynecol Oncol 2013;131:489–92.
6. Shafiee MN, Khan G, Ariffin R, et al. Preventing endometrial cancer
risk in polycystic ovarian syndrome (PCOS) women: could
metformin help? Gynecol Oncol 2014;132:248–53.
7. Cantrell LA, Zhou C, Mendivil A, et al. Metformin is a potent
inhibitor of endometrial cancer cell proliferation – implications for a
novel treatment strategy. Gynecol Oncol 2010;116:92–8.
8. Tan BK, Adya R, Chen J, et al. Metformin treatment exerts
antiinvasive and antimetastatic effects in human endometrial
carcinoma cells. J Clin Endocrinol Metab 2011;96:808–16.
9. Shen ZQ, Zhu HT, Lin JF. Reverse of progestin-resistant atypical
endometrial hyperplasia by metformin and oral contraceptives.
Obstet Gynecol 2008;112:465–7.
10. Dumont P, Leu JI, Della Pietra AC, et al. The codon 72 polymorphic
variants of p53 have markedly different apoptotic potential.
Nat Genet 2003;33:357–65.
11. Klug SJ, Ressing M, Koening J, et al. TP53 codon 72 polymorphism
and cervical cancer: a pooled analysis of individual data from 49
studies. Lancet Oncol 2009;10:772–84.
12. Zhang Z, Wang M, Wu D, et al. P53 codon 72 polymorphism
contributes to breast cancer risk: a meta-analysis based on 39 case-
control studies. Breast Cancer Res Treat 2009;120:509–17.
13. Kubbutat MH, Jones SN, Vousden KH. Regulation of p53 stability
by Mdm2. Nature 1997;387:299–303.
14. Soslow RA, Shen PU, Chung MH, Isacson C. Distinctive p53 and
MDM2 immunohistochemical expression profiles suggest different
pathogenetic pathways in poorly differentiated endometrial carcin-
oma. Int J Gynecol Pathol 1998;17:129–34.
15. Chon HS, Marchion DC, Xiong Y, et al. The BCL2 antagonist of cell
death pathway influences endometrial cancer cell sensitivity to
cisplatin. Gynecol Oncol 2012;124:119–24.
16. Zhang R, He Y, Zhang X, et al. Estrogen receptor-regulated
microRNAs contribute to the BCL2/BAX imbalance in endometrial
adenocarcinoma and precancerous lesions. Cancer Lett 2012;314:
155–65.
17. Takai N, Desmond JC, Kumagai T, et al. Histone deacetylase
inhibitors have a profound antigrowth activity in endometrial cancer
cells. Clin Cancer Res 2004;10:1141–9.
18. Konishi I, Koshiyama M, Mandai M, et al. Increased expression of
LH/hCG receptors in endometrial hyperplasia carcinoma in anovula-
tory women. Gynecol Oncol 1997;65:273–80.
19. Chittenden BG, Fullerton G, Maheshwari A, Bhattacharya S.
Polycystic ovary syndrome and the risk of gynaecological cancer:
a systematic review. Reprod BioMed Online 2009;19:398–405.
20. Haas DA, Carr BR, Attia GR. Effect of Metformin on BMI,
menstrual cyclicity and ovulation induction in women with
polycystic ovary syndrome. Fertil Steril 2003;79:469–81.
21. Barbieri RL. Metformin for the treatment of polycystic ovary
syndrome. Obstet Gynecol 2003;101:785–93.
22. Tas M, Kutuk MS, Serin IS, et al. Comparison of
antiproliferative effects of metformin and progesterone on estro-
gen-induced endometrial hyperplasia in rats. Gynecol Endocrinol
2013;29:311–14.
23. Gunter MJ, Hoover DR, Yu H, et al. Prospective evaluation of
insulin and insulin like growth factor-I as risk factors of endometrial
cancer. Cancer Epidemiol Biomarkers Prev 2008;17:921–9.
24. Levav-Cohen Y, Haupt S, Haupt Y. MDM2 in growth signaling and
cancer. Growth Factors 2005;23:183–92.
DOI: 10.3109/09513590.2014.989982 Metformin effects on PCOS 5
GynecolEndocrinolDownloadedfrominformahealthcare.comby86.30.66.46on12/11/14
Forpersonaluseonly.