If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Obstetrics and Gynecology, Herlev Gentofte Hospital, Herlev Ringvej 75, DK-2730 Herlev, DenmarkFaculty of Health and Medical Sciences, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen, Denmark
Faculty of Health and Medical Sciences, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen, DenmarkDepartment of Internal Medicine, Endocrine Unit, Herlev Gentofte Hospital, Herlev Ringvej 75, DK-2730 Herlev, Denmark
Department of Obstetrics and Gynecology, Herlev Gentofte Hospital, Herlev Ringvej 75, DK-2730 Herlev, DenmarkFaculty of Health and Medical Sciences, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen, Denmark
Faculty of Health and Medical Sciences, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen, DenmarkDepartment of Internal Medicine, Endocrine Unit, Herlev Gentofte Hospital, Herlev Ringvej 75, DK-2730 Herlev, Denmark
Faculty of Health and Medical Sciences, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen, DenmarkDepartment of Internal Medicine, Endocrine Unit, Herlev Gentofte Hospital, Herlev Ringvej 75, DK-2730 Herlev, Denmark
Department of Obstetrics and Gynecology, Herlev Gentofte Hospital, Herlev Ringvej 75, DK-2730 Herlev, DenmarkFaculty of Health and Medical Sciences, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen, Denmark
Liraglutide improved the ovarian dysfunction in an overweight PCOS population.
•
Normalization of bleeding pattern was driven by weight loss rather than reduced insulin resistance.
•
Liraglutide therapy caused increased SHBG and decreased free testosterone levels.
•
Nausea and constipation were more prevalent with liraglutide than with placebo.
Abstract
Polycystic ovary syndrome (PCOS) encompasses an ovarian and a metabolic dysfunction. Glucagon-like peptide-1 (GLP-1) analogues facilitate weight loss and ameliorate metabolic dysfunction in overweight women with PCOS, but their effect on ovarian dysfunction is scarcely reported. In a double-blind, randomized trial, 72 women with PCOS were allocated to intervention with the GLP-1 analogue liraglutide or placebo (1.8 mg/day), in a 2:1 ratio. At baseline and 26-week follow-up, bleeding pattern, levels of AMH, sex hormones and gonadotrophins were assessed and ovarian morphology evaluated. Liraglutide caused 5.2 kg (95% CI 3.0 to 7.5, P < 0.0001) weight loss compared with placebo. Bleeding ratio improved with liraglutide: 0.28 (95% CI 0.20 to 0.36, P < 0.001); placebo: 0.14 (95% CI 0.02 to 0.26, P < 0.05); between-group difference: 0.14 (95% CI 0.03 to 0.24, P < 0.05). In the liraglutide group, SHBG increased by 7.4 nmol/L (95% CI 4.1 to 10.7) and free testosterone decreased by 0.005 nmol/L (95% CI −0.009 to −0.001). Ovarian volume decreased by −1.6 ml (95% CI −3.3 to 0.1) with liraglutide versus placebo. Nausea and constipation were more prevalent in the liraglutide group. Liraglutide improved markers of ovarian function in overweight women with PCOS, and might be a possible intervention.
). Central for PCOS is the ovarian dysfunction, clinically seen as oligo-/anovulation, androgen excess and polycystic ovarian morphology. These three represent the Rotterdam criteria upon which the PCOS diagnosis is based (
Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS).
Folliculogenesis is a highly complex process directed by autocrine, paracrine and endocrine factors, e.g. gonadotrophins, sex steroids and anti-Müllerian hormone (AMH) (
). By inhibiting follicular recruitment and FSH-dependent follicular growth, AMH is thought to have a protective effect on the ovarian reserve in the healthy individual (
). As a result of a large pool of early developing follicles and greater AMH production per follicle, women with PCOS have higher AMH levels than controls (
Anti-Mullerian hormone levels reflect severity of PCOS but are negatively influenced by obesity: relationship with increased luteinizing hormone levels.
Am. J. Physiol. Endocrinol. Metab.2009; 296: E238-E243
Serum antimüllerian hormone in response to dietary management and/or physical exercise in overweight/obese women with polycystic ovary syndrome: secondary analysis of a randomized controlled trial.
Metformin effects on ovarian ultrasound appearance and steroidogenic function in normal-weight normoinsulinemic women with polycystic ovary syndrome: a randomized double-blind placebo-controlled clinical trial.
The metabolic status modulates the effect of metformin on the antimullerian hormone-androgens-insulin interplay in obese women with polycystic ovary syndrome.
). Indeed, insulin is a central factor in PCOS, in which intrinsic and obesity-related insulin resistance causes compensatory hyperinsulinaemia. Insulin amplifies the excessive, LH-stimulated, ovarian androgen production (
The ideal PCOS treatment addresses both the metabolic and the ovarian dysfunction of the syndrome. Over the past decade, glucagon-like peptide 1 (GLP-1) analogues are increasingly used in treating type 2 diabetes and obesity, where they mimic the effect of the endogenous incretin hormone, promoting weight loss, improved glycaemic control and reduced dyslipidaemia (
). Because of the resemblance between type 2 diabetes and PCOS, and the fact that metformin affects ovarian dysfunction, GLP-1 analogues might have a role in treating the syndrome. The GLP-1 analogue liraglutide has been found to cause weight loss in several small trials on women with PCOS (
Short term monotherapy with GLP-1 receptor agonist liraglutide or PDE 4 inhibitor roflumilast is superior to metformin in weight loss in obese PCOS women: a pilot randomized study.
A 12-week treatment with the long-acting glucagon-like peptide 1 receptor agonist liraglutide leads to significant weight loss in a subset of obese women with newly diagnosed polycystic ovary syndrome.
), but results on bleeding frequency are sparse. Losing weight and maintaining a stable body weight is difficult, and interventions effectively treating both the body weight and the ovarian dysfunction are warranted.
We hypothesized that intervention with a GLP-1 analogue would improve ovarian dysfunction in overweight women with PCOS, possibly through weight loss and altered glucose metabolism. In this randomized, clinical trial, we aimed to investigate the effect of liraglutide on markers of ovarian dysfunction: bleeding ratio, ovarian morphology, levels of AMH and androgens, in women with PCOS.
Materials and methods
Protocol and ethics
This was an investigator-initiated, double blind, placebo-controlled, randomized trial, conducted between March 2014 and December 2015 at Herlev Gentofte Hospital, University of Copenhagen, Denmark. The study was approved by the regional Ethics Committee of the Capital Region of Denmark (ID: H-2-2013-142), the Danish Health Authority (EudraCT: 2013-003862-15) and the Danish Data Protection Agency. The study was conducted in accordance with Good Clinical Practice guidelines and the declaration of Helsinki and registered at www.clinicaltrials.gov (NCT02073929). Oral and written consent were obtained for each participant before inclusion. The protocol has been published elsewhere (
The LIPT-Study: on risk markers of vascular thrombosis in polycystic ovary syndrome. A randomized, double-blind, placebo-controlled study of the effect of Liraglutide.
). At baseline and 26-week follow-up, anthropometrics and Ferriman-Gallway score were assessed, fasting blood tests and a transvaginal ultrasound were conducted. Blood was drawn between 8.00 and 10.00 am after overnight fasting. Blood for determination of AMH and insulin levels was centrifuged and serum was stored at −80°C until analysis.
Population
In Denmark, women with PCOS are usually managed by a general practitioner or a private-practising gynaecologist, rather than in a hospital setting. Because of this, women were mainly recruited from social media (www.facebook.com/PCOSkliniskforsoeg), from advertising in the local area and from general practitioners or private gynaecologists, but also from our outpatient clinic. Briefly, inclusion criteria were as follows: age 18 years or above, PCOS according to Rotterdam criteria (
Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS).
), body mass index 25 kg/m2 or more and/or insulin resistance (defined as fasting plasma C-peptide >600 pmol/L). Irregular bleeding was defined as cycle length more than 35 days. Hyperandrogenism was defined as total testosterone above reference levels, free testosterone above reference level (1.80 nmol/l and 0.0034 nmol/l, respectively), Ferriman-Gallway score >8. In brief, exclusion criteria were pregnancy, diabetes, use of hormonal contraceptives (within 6 weeks before inclusion), anti-diabetic or anti-androgenic agents (within three months before inclusion). Other causes of irregular menstruation and androgen excess, e.g. hyperprolactinaemia, thyroid and adrenal diseases, were excluded, using biochemical work-up performed at screening. As pregnancy was an exclusion criterion, participants had a copper intrauterine device inserted at the baseline visit. In case of contraindications for IUD, the participant was instructed in concomitant use of diaphragm and condom, in accordance with guidelines from the Danish Health Authority. A pregnancy test was carried out at every visit.
Randomization and intervention
Women were randomized, in blocks of six, to 26 weeks of intervention with liraglutide or placebo in a 2:1 ratio. Novo Nordisk A/S, Bagsværd, Denmark, provided identically packed and labelled study drug (liraglutide and placebo) as well as a randomization list. All participants and investigators were blinded as an independent secretary handled the randomization list and instructed the investigators in which serial numbers, i.e. drug packages, to be handed out to each participant. The study drug was administered subcutaneously 1.8 mg/day, starting at 0.6 mg/day and 1.2 mg/day for the first and second week, respectively.
Outcomes
Pre-specified outcomes included the between-group difference in bleeding ratio at follow-up, as well as the between-group differences in change (from baseline to follow-up) in bleeding ratio, ovarian volume, stromal volume, antral follicle count, Ferriman-Gallway score, serum levels of AMH, SHBG and androgens.
Anthroprometrics, bleeding and ultrasound
Body weight was assessed in light clothing, after overnight fasting. Waist circumference was measured by a single observer in a standardized way: half way between the 12th rib and the anterior superior iliac spine. At screening and baseline, participants were questioned about number of menstrual bleedings for the last six months. The women were asked to keep a bleeding diary during the study. Bleeding ratio was defined as number of menstrual bleedings divided by study period (months). A bleeding ratio of 1.0 corresponds to six menses in six months, i.e. a cycle length of 30.4 days. A cycle length of 35 days corresponds to a bleeding ratio of 0.87 (365 days/35 days/month)/12 months = 0.87). Transvaginal three-dimensional ultrasound was carried out by a single investigator (MN) using a Voluson E6, GE Healthcare, Chicago, IL., USA. Data were analysed with the VOCAL and SonoAVC-tools, 4DView, GE Healthcare, Chicago, IL, USA. Follicles measuring 2–9 mm were counted, and stromal volume was calculated as ovarian volume – follicular volume. Intra-observer, intra-class correlations for volume and antral follicle count were 0.957 and 0.987, respectively.
Assays
All serum AMH samples were measured in the same run, using an AL-105-i ultra-sensitive AMH/MIS ELISA (Ansh Labs, Webster, TX, USA) with intra-assay coefficient of variability less than 2.0%. Serum levels of LH, FSH and oestradiol were determined using an ADVIA Centaur Immunoassay (Siemens Healthcare GmbH, Erlangen, Germany) with inter-assay coefficients of variability less than 3.8%, 3.9% and 10.2%, respectively. Testosterone, androstenedione, dehydroepiandrosterone sulfate and 17-OH progesterone levels were determined using liquid chromatography and double mass spectrometry (UPLC-MSMS TQ-S System, Waters Corporation, Milford, MA, USA), with inter-assay coefficients of variability 12% or less for all. Levels of SHBG were measured using a sandwich chemiluminescence immunometric method (Immulite 2000, Siemens Healthcare GmbH, Erlangen, Germany) with inter-assay coefficients of variability less than 7%. Free testosterone (
) and free androgen index (total testosterone x 100/SHBG) were calculated from total testosterone and SHBG. Fasting plasma glucose and HbA1c levels were assessed using in-house routine analyses. Plasma insulin levels were determined with an electro-chemiluminescense immunoassay and Cobas e422 reader (Roche Diagnostics GmbH, Mannheim, Germany) with an intra-assay coefficient of variability of 2.1%. An online calculator (www.dtu.ox.ac.uk/homacalculator) was used for calculating the homeostasis model assessment-estimated insulin resistance (HOMA2-IR) from fasting insulin and fasting glucose. Levels of glycosylated hemoglobin (HbA1c) were assessed using routine analyses at the Department of Clinical Biochemistry, Herlev Gentofte Hospital, Denmark.
Statistics
SAS Enterprise Guide 7.1, SAS 9.4, SAS Inc., Cary, NC, USA was used for analyses. Distribution of data was evaluated using histograms and qq-plots. Non-normally distributed data were subject to logarithmic transformation. Data are presented as mean (SD), median (p25–p75) and mean differences (95% confidence interval). Difference from baseline to follow-up in each group was estimated with a paired t-test, and between-group difference in effect size was calculated using a linear mixed model with repeated measurements and maximum likelihood. As some of the data were logarithmically transformed for the mixed model some effect sizes are presented as ratios. Every other woman contributed with her right ovary and every second with her left for ovarian morphology outcomes. Associations between change in bleeding ratio and baseline bleeding ratio, as well as change in anthropometric, endocrine, metabolic and ovarian morphology variables were assessed using Pearson's or Spearman's correlation coefficients, as appropriate, and further analysed using multiple linear regression analyses. This study was a part of a trial investigating the effect of liraglutide on thrombin generation, where a sample size calculation stated that 63 women, randomized in a 2:1 ratio, were needed. For the outcome: between-group difference in change in bleeding ratio, with 63 women included, a standard deviation of 0.33 and a significance level of 0.05 the study had 95% power to find an effect size of 0.33. This effect size corresponds to two menstruations every six months and is supported by results from a previous study, using exenatide in oligoovulatory and anovulatory PCOS (
found a SD of 0.1. By using the Rotterdam criteria, however, we will, per definition, include women with regular menstruations and our SD will be higher; therefore, SD 0.33 was chosen. Seventy-two women were included to allow for drop-outs. Post-hoc power calculation using 63 participants, significance level of 0.05 and a common SD of 3.6 ml, revealed an 80% power to detect the observed between-group difference in ovarian volume. The corresponding numbers for total testosterone (n = 65, common SD = 0.57 nmol/l) and AMH (n = 65, common SD = 32.1 pmol/ml) were 59% and 53%, respectively.
Results
Of 138 women assessed for eligibility, 72 were randomized and 65 completed the trial (Figure 1). In the placebo group, one woman dropped out on the day of randomization, owing to personal reasons. Apart from this, reasons for dropout were similar in the two groups: lost to follow-up (4.2%) and abdominal pain (4.2%). At baseline, no between-group differences were observed in bleeding ratio, ovarian morphology, levels of AMH, sex hormones or SHBG (Table 1 and Table 2). Even though not statistically significant, the liraglutide group seemed older, with lower proportion of smokers and, therefore, between-group differences were adjusted for age, body mass index (BMI) and smoking status. This did not change the results.
Figure 1CONSORT flow chart presenting the inclusion criteria, randomization and follow-up of participants.
Results presented as mean(SD), median(p25–p75) or percentage(n). Differences at baseline assessed with unpaired t-test, Mann–Whitney U-test, Chi-square-test or Fisher's exact test, as appropriate.
BMI, body mass index; GFR, glomerular filtration rate; hyperandrogenism; HbA1c, glycosylated hemoglobin; HOMA2-IR, Homeostasis model assessment-estimated insulin resistance; OA, oligomenorrhoea/amenorrhoea; PCO, polycystic ovaries.
After six months, the liraglutide group had a mean weight loss of 5.2 kg (95% CI 3.0 to 7.5, P < 0.0001) and mean reductions in fasting glucose and HbA1c of 0.24 mM (95% CI 0.05 to 0.43, P < 0.05) and 1.4 mmol/mol (95% CI 0.3 to 2.5, P < 0.05), respectively, compared with the placebo group. We observed no effect on fasting insulin or HOMA2-IR (data not shown).
Bleeding
At follow-up, 62% (26 of 42) of women in the liraglutide group had a bleeding ratio over 0.87 and corresponding number in the placebo group was 28% (5 of 18) (P < 0.05). At baseline, the number of women with amenorrhoea, oligomenorrhoea and regular menstrual bleedings were 8, 29 and 11 (n = 48) in the liraglutide group and 5, 14 and 5 (n = 24) in the placebo group. At follow-up, corresponding numbers were 2, 14 and 26 (n = 42), and 1, 12 and 5 (n = 18). Within- and between-group differences in bleeding ratio are presented in Table 2.
Change in bleeding ratio correlated with bleeding at baseline (r = −0.66, P < 0.001) and change in BMI (r = −0.28, P < 0.05). Nosignificant correlation was found between the change in bleeding ratio after intervention and the change in waist circumference, levels of androgens, gonadotrophins, AMH, fasting glucose, fasting insulin or HOMA2-IR (data not shown). Excluding the 16 women with regular menstruation at baseline did not alter the results. In a multiple regression analysis with change in bleeding ratio as dependent variable, both study drug and baseline bleeding ratio had significant impact (model: R2 = 0.49, P < 0.0001, study drug: β = 0.13, P < 0.05, baseline bleeding ratio: β = −0.53, P < 0.0001). When adding change in BMI to the model, the study drug-effect became non-significant, whereas adding change in HOMA2-IR did not alter the results.
Ovarian morphology and endocrine markers
Results of ovarian morphology and endocrine markers are presented in Table 2. We observed no effect on Ferriman-Gallway score in either group.
Adverse effects
The most prevalent adverse effects were nausea (liraglutide 79% versus placebo 13%, P < 0.01), primarily seen in the start-up phase, and constipation (26% versus 0%, P < 0.01). Gallstone-related pain was experienced by 6% of women in the liraglutide group and 4% in the placebo group (NS). Two of the women in the liraglutide group had a cholecystectomy and, owing to discontinuation of the study drug during diagnosis and surgery, their overall compliance was 64% and 87%, respectively, and both were included in final analyses. In the placebo group, one woman had a severe gallstone event, and chose not to have a cholecystectomy, but to drop out of the study.
Discussion
In this randomized, placebo-controlled trial, investigating the effect of the GLP-1 analogue liraglutide on the ovarian dysfunction in PCOS, we found improved bleeding ratio, decreased free testosterone and a trend towards lower ovarian and stromal volume.
found improved bleeding pattern with the GLP-1 analog exenatide. They treated 42 overweight women with oligo-ovulatory PCOS with exenatide, metformin or a combination for 24 weeks and found improved ovulation rate in all groups, with the combined treatment being superior (
). Additionally, they found reduced body weight and insulin resistance, as measured by HOMA-IR. Contrary to our results, several studies on liraglutide in PCOS find unaltered bleeding frequency despite an effect on body weight (
Short term monotherapy with GLP-1 receptor agonist liraglutide or PDE 4 inhibitor roflumilast is superior to metformin in weight loss in obese PCOS women: a pilot randomized study.
A 12-week treatment with the long-acting glucagon-like peptide 1 receptor agonist liraglutide leads to significant weight loss in a subset of obese women with newly diagnosed polycystic ovary syndrome.
Short term monotherapy with GLP-1 receptor agonist liraglutide or PDE 4 inhibitor roflumilast is superior to metformin in weight loss in obese PCOS women: a pilot randomized study.
A 12-week treatment with the long-acting glucagon-like peptide 1 receptor agonist liraglutide leads to significant weight loss in a subset of obese women with newly diagnosed polycystic ovary syndrome.
Short term monotherapy with GLP-1 receptor agonist liraglutide or PDE 4 inhibitor roflumilast is superior to metformin in weight loss in obese PCOS women: a pilot randomized study.
A 12-week treatment with the long-acting glucagon-like peptide 1 receptor agonist liraglutide leads to significant weight loss in a subset of obese women with newly diagnosed polycystic ovary syndrome.
). No human data are available on the use of GLP-1-analogues in pregnancy. Therefore, GLP-1 analogues are contraindicated in pregnancy and pre-pregnancy counselling is a must in this population.
Weight loss is known to cause improved bleeding pattern in overweight women with PCOS, found both in lifestyle intervention studies (
Randomized comparison of the influence of dietary management and/or physical exercise on ovarian function and metabolic parameters in overweight women with polycystic ovary syndrome.
proposed reduced insulin resistance to be the mechanism for this improved bleeding pattern, as women, experiencing improved bleeding with weight loss had reduced fasting insulin levels, compared with non-responders (
Metformin effects on ovarian ultrasound appearance and steroidogenic function in normal-weight normoinsulinemic women with polycystic ovary syndrome: a randomized double-blind placebo-controlled clinical trial.
The metabolic status modulates the effect of metformin on the antimullerian hormone-androgens-insulin interplay in obese women with polycystic ovary syndrome.
Randomized comparison of the influence of dietary management and/or physical exercise on ovarian function and metabolic parameters in overweight women with polycystic ovary syndrome.
Randomized comparison of the influence of dietary management and/or physical exercise on ovarian function and metabolic parameters in overweight women with polycystic ovary syndrome.
). We found reduced fasting glucose and HbA1c, indicating improved insulin sensitivity, in the liraglutide group compared with the placebo group. We could not, however, demonstrate this by a decrease in HOMA2-IR. Additionally, we found the improved bleeding ratio to be driven by weight loss rather than by changes in HOMA2-IR or by a direct effect of liraglutide. Most likely, both weight loss and subtle alterations in insulin levels play a role in restoring the bleeding pattern.
We found increased SHBG levels, decreased free testosterone levels and unaltered total testosterone levels with 26 weeks of liraglutide treatment. Corroborating this, liraglutide caused increased SHBG levels and unaltered total testosterone levels in a 12-week study comparing liraglutide with the combination of liraglutide and metformin in obese women with PCOS (
Metformin as an initial adjunct to low-dose liraglutide enhances the weight-decreasing potential of liraglutide in obese polycystic ovary syndrome: Randomized control study.
). Inconsistent with our results, 24 weeks of exenatide was found to cause reduced total testosterone levels and unaltered SHBG levels, in a population with higher baseline testosterone levels than ours (
Short term monotherapy with GLP-1 receptor agonist liraglutide or PDE 4 inhibitor roflumilast is superior to metformin in weight loss in obese PCOS women: a pilot randomized study.
A 12-week treatment with the long-acting glucagon-like peptide 1 receptor agonist liraglutide leads to significant weight loss in a subset of obese women with newly diagnosed polycystic ovary syndrome.
) liraglutide was found to have no effect on either total testosterone or SHBG levels. Ovarian stroma volume is highly correlated with androgen levels (
Ultrasound in polycystic ovary syndrome – the measuring of ovarian stroma and relationship with circulating androgens: results of a multicentric study.
) and the trend observed in our study, towards decreased stroma volume, might indicate a reduced ovarian androgen production with liraglutide.
We did not observe any effect on Ferriman-Gallway score, most likely owing to too short follow-up for this parameter. As combined oral contraceptives effectively raise SHBG (
), thereby decreasing the levels of free testosterone the effect of discontinuation of combined oral contraceptives could counteract the effect of liraglutide, even though we applied a well-accepted wash-out period of at least six weeks. The percentage of women, who within three months before randomization discontinued these drugs, was similar between the two groups (15% in liraglutide group and 21% in placebo group). Moreover, in sub-analyses, excluding women who discontinued combined oral contraceptives, metformin or spironolactone therapy less than five months before enrolment, results regarding SHBG, androgens, hirsutism and ovarian morphology were unaltered.
The effect of GLP-1 analogs on AMH has, to our knowledge, never been investigated before. We observed a trend towards decreased AMH over time in the liraglutide group, but no between-group difference. Decreased AMH levels have been observed together with improved bleeding frequency in women with PCOS after bariatric surgery (
Serum antimüllerian hormone in response to dietary management and/or physical exercise in overweight/obese women with polycystic ovary syndrome: secondary analysis of a randomized controlled trial.
Serum antimüllerian hormone in response to dietary management and/or physical exercise in overweight/obese women with polycystic ovary syndrome: secondary analysis of a randomized controlled trial.
Serum antimüllerian hormone in response to dietary management and/or physical exercise in overweight/obese women with polycystic ovary syndrome: secondary analysis of a randomized controlled trial.
). Conversely, a 20-week dietary programme in 52 overweight women with PCOS had no significant effect on AMH despite improved bleeding frequency, reduced body weight and total testosterone levels (
). As the bleeding ratio improved in this study, we assume that some of the AMH-producing follicles bypassed the follicular arrest and were eliminated through ovulation, thereby making the pool of AMH-producing follicles smaller, which theoretically would cause lower AMH levels and antral follicle count. Unfortunately, this was not the case in our study, most likely because of lack of power as indicated by the post-hoc power calculation.
This study has several strengths: the study design, the high compliance and the use of the double mass spectroscopy for determination of androgen levels. Moreover, ovarian morphology was assessed with three-dimensional ultrasound, both total and stromal volumes were analysed and all examinations were carried out by a single, blinded observer (MN). The study, however, has some limitations: recruiting participants from social media might have caused a selection bias, i.e. including only resourceful and highly motivated women. Moreover, participants did not keep a bleeding diary before inclusion, and the reported bleeding ratio at baseline could be associated with recall bias. This, as well as bleeding due to the IUD, could be the reason for us finding improved bleeding pattern in the placebo group. Nevertheless, a significant difference was found between the two groups, demonstrating an effect of liraglutide. Additionally, bleeding pattern was assessed using menstrual bleedings rather than ovulations (biochemical), which would have been ideal. Since most of our participants had irregular cycles we were not able to evaluate gonadotrophins in early follicular phase as appropriate. Evaluating changes in the bleeding pattern in a population, including women with regular menstruation (owing to the use of Rotterdam criteria) might seem inappropriate. Nevertheless, this reflects the true clinical situation, and excluding the women with regular menstruation from analyses did not change the results. Finally, several of the outcome measures showed trends rather than being statistically significant, indicating that the study might be limited by a type 2 error.
In conclusion, 26 weeks of liraglutide intervention altered the ovarian dysfunction in an overweight PCOS population. It improved the bleeding ratio, reduced free testosterone levels and caused substantial weight loss. Side-effects of liraglutide included nausea, especially in the up-titrating phase, and constipation. We believe that liraglutide has a potential role in treating overweight and obese women with PCOS without current pregnancy wish, especially when weight loss is desired. Larger, randomized studies with longer follow-up are warranted, as are studies investigating the combination of liraglutide, metformin and lifestyle intervention in obese women with PCOS.
Acknowledgements
The authors would like to thank the participants, the personnel at the Endocrine Unit research laboratory and the nurses and secretaries at the Fertility Clinic, Herlev Gentofte Hospital. Additionally, the authors want to thank statisticians Tobias Wirenfeldt and Mathias Ejdrup Bredkjær for statistical support.
References
Bartsch W.
Interrelationships between sex hormone-binding globulin and testosterone, 5 alpha-dihydrotestosterone and oestradiol-17 beta in blood of normal men.
The LIPT-Study: on risk markers of vascular thrombosis in polycystic ovary syndrome. A randomized, double-blind, placebo-controlled study of the effect of Liraglutide.
Ultrasound in polycystic ovary syndrome – the measuring of ovarian stroma and relationship with circulating androgens: results of a multicentric study.
Short term monotherapy with GLP-1 receptor agonist liraglutide or PDE 4 inhibitor roflumilast is superior to metformin in weight loss in obese PCOS women: a pilot randomized study.
A 12-week treatment with the long-acting glucagon-like peptide 1 receptor agonist liraglutide leads to significant weight loss in a subset of obese women with newly diagnosed polycystic ovary syndrome.
Metformin as an initial adjunct to low-dose liraglutide enhances the weight-decreasing potential of liraglutide in obese polycystic ovary syndrome: Randomized control study.
Randomized comparison of the influence of dietary management and/or physical exercise on ovarian function and metabolic parameters in overweight women with polycystic ovary syndrome.
Serum antimüllerian hormone in response to dietary management and/or physical exercise in overweight/obese women with polycystic ovary syndrome: secondary analysis of a randomized controlled trial.
Anti-Mullerian hormone levels reflect severity of PCOS but are negatively influenced by obesity: relationship with increased luteinizing hormone levels.
Am. J. Physiol. Endocrinol. Metab.2009; 296: E238-E243
Metformin effects on ovarian ultrasound appearance and steroidogenic function in normal-weight normoinsulinemic women with polycystic ovary syndrome: a randomized double-blind placebo-controlled clinical trial.
The metabolic status modulates the effect of metformin on the antimullerian hormone-androgens-insulin interplay in obese women with polycystic ovary syndrome.
Malin Nylander, MD, received her medical degree from the University of Copenhagen in 2012 and her PhD from the Faculty of Health and Medical Sciences, University of Copenhagen and Department of Obstetrics and Gynecology, Herlev Gentofte Hospital in 2017. She is currently a junior doctor at the Department of Obstetrics and Gynecology, Herlev Gentofte Hospital. Her areas of interest are gynaecological endocrinology in general, and polycystic ovary syndrome in particular.
Key message
In this double-blind, placebo-controlled, randomized trial, the glucagonlike peptide-1 analogue liraglutide was found to ameliorate ovarian dysfunction in overweight women with PCOS. Improved bleeding regularity, reduced levels of free testosterone and substantial weight loss were observed. Liraglutide could serve as a treatment in overweight women with PCOS.
Article info
Publication history
Published online: April 24, 2017
Accepted:
March 28,
2017
Received in revised form:
March 20,
2017
Received:
December 2,
2016
Declaration: The main author (MN) received a full-salary grant from the University of Copenhagen throughout the study period. The study was partly funded by a grant from Novo Nordisk A/S, Denmark and a grant from the Danish Toyota Foundation. Funders had no influence on study design, on collection, analysis or interpretation of data or on writing of the manuscript, but Novo Nordisk A/S had access to read the manuscript prior to submission. CK and JF are members of NovoNordisk A/S advisory board on the use of liraglutide in diabetes. No other disclosures are reported for any of the authors. JF, SOS and CK initiated the study, all authors contributed to the protocol and to the intellectual interpretation of the results. SF and MN conducted the study and collected data. MN performed the statistical analyses and produced the manuscript, which all co-authors have read and approved.
30194-3&id=rbmo1727-fig-0001.jpg){kind=link}