GLP-1 Receptor Analogs: Evidence Linking to Effect on Metabolic and Reproductive Functions in Patients with PCOS and Obesity

Abstract

Background

Polycystic ovary syndrome (PCOS) is a multifaceted hormonal disorder that impacts both metabolic functions and reproductive health in women. PCOS pathophysiology is characterized by insulin resistance and chronic inflammation, which contribute to disrupted gonadotropin secretion, elevated androgen levels, and irregular ovulation. There is a bidirectional relationship between obesity and PCOS. Obesity contributes to the pathophysiology of PCOS, particularly by exacerbating insulin resistance – a core feature of the condition. Insulin resistance can lead to further weight gain and hinder successful weight loss.

Summary

According to the latest international guidelines, improving insulin sensitivity and promoting weight loss are key strategies in the effective management of PCOS. GLP-1 receptor analogs (GLP-1 RAs) are a group of agents that are used in type 2 diabetes mellitus to regulate blood glucose levels, in addition to being indicated for weight loss in obesity. Thus, GLP-1 RAs may play a role in the management of PCOS, both due to weight loss effect and through pleiotropic effects like ameliorating insulin resistance, decreasing inflammation, and modulation of adiposity.

Key Messages

In light of these effects, this review aimed to perform a systematic search and narrative synthesis on the impact of GLP-1 receptor agonists (GLP-1 RAs) on the metabolic and reproductive outcomes in PCOS, with the goal of elucidating the mechanistic link between the action of GLP-1 RAs and the underlying pathophysiology of the condition.

Introduction

Polycystic ovary syndrome (PCOS) is a complex endocrine disorder that affects many aspects of a woman’s health, particularly her reproductive and metabolic systems. As there have been numerous historical definitions for PCOS, the latest international guideline, updated in 2023, has revised both the diagnosis and the management of PCOS. As the guideline stresses, the global prevalence of PCOS is 12% [1]. PCOS manifests with reproductive abnormalities, ovarian dysfunction, hirsutism, acne, androgenic alopecia, in addition to metabolic abnormalities. The disease has increased risk of infertility, obstetrical complications, type 2 diabetes mellitus (DM), obesity, dyslipidemia, metabolic dysfunction-associated steatotic liver disease (MASLD), cardiovascular disease (CVD), obstructive sleep apnea, mood disorders, endometrial and ovarian cancer [1–5]. Obstetrical complications that may be observed in patients with PCOS are higher gestational weight gain, gestational DM, hypertension in pregnancy, preeclampsia, miscarriage, preterm delivery, caesarean section, intrauterine growth restriction and low birth weight in the fetus [1].

The pathophysiology of PCOS is not entirely understood but implicates a combination of hormonal, genetic, and environmental factors. Some key aspects of PCOS pathophysiology include genetic factors and insulin resistance [6]. PCOS is characterized by disturbances in gonadotropin secretion, resulting in impaired luteinizing hormone-to-follicle stimulating hormone (LH-to-FSH) ratio, anovulation, and hyperandrogenism. An increased LH/FSH ratio enhances hypersecretion of androgens in the ovaries, whereas insufficient FSH secretion and its inhibition result in irregular ovulation or anovulation. LH hypersecretion may result in premature luteinization in the granulosa cells. There is uncertainty about the etiology of insulin resistance in PCOS. It has been elaborated to be due to a receptor defect or a post-receptor defect [7]. Compensatory hyperinsulinemia may enhance steroidogenesis, possibly acting through the mitogen activated protein kinase pathway [8]. This contributes to enhanced androgen production and ovulatory dysfunction. Hyperinsulinemia may also have gonadotropic effects on the theca cells, leading to enhancement of LH pulse amplitude [9, 10] and decreased sex hormone-binding globulin (SHBG), thus increasing bioavailable testosterone [11, 12]. Another factor that contributes to insulin resistance is adipocyte dysfunction [4, 13]. There is evidence suggesting that there are abnormal adipocytes and disturbed adipocyte function in patients with PCOS, resulting in subclinical inflammation that downregulates insulin-mediated glucose transport in adipocytes [4]. Androgens stimulate lipolysis and impair differentiation of adipocytes [14]. Larger adipocytes have been shown in women with PCOS compared with body mass index (BMI)-matched controls and the size of the adipocytes seems to be correlated with androgen levels [15]. Moreover, there is dysregulated lipolysis in PCOS leading to lipotoxicity and ectopic fat accumulation [16]. Abnormal adipokine release is another contributor to adipocyte dysfunction in PCOS [17].

Considering the complications, insulin resistance and hyperestrogenic anovulation have an increased risk for endometrial cancer [1]. As for the cardiovascular complications, factors like insulin resistance, low grade inflammation, hyperandrogenism, oxidative stress, and changes in adipokine levels have been implicated [18].

The relationship between PCOS and obesity is bidirectional [19, 20]. Additionally, obesity enhances insulin resistance and chronic inflammation, which further contribute to hormonal imbalance and metabolic alterations in PCOS such as type 2 DM [21]. Abdominal obesity contributes to hyperandrogenemia and anovulation [22]. Visceral fat accumulation is increased in women with PCOS, with possible contribution to insulin resistance through adipokine and fatty acid release [23–25]. Besides, coexistence of PCOS and NAFLD has been demonstrated [26, 27]. However, it has to be mentioned that visceral adipose tissue (VAT) may be increased regardless of the BMI, indicating that even women with PCOS who have normal weight may have increased VAT [28]. Thus, visceral abdominal obesity is more common in phenotypes A and B of PCOS whereas phenotype B is least affected [13].

GLP-1 RAs are indicated as a first-line treatment for patients with type 2 DM, with known CVD or risk factors for CVD [29, 30]. They are also indicated in patients with obesity and overweight, who have comorbidities due to increased body weight [31, 32]. As the latest international guidelines on PCOS suggest, weight loss is an important component in the management of PCOS [1]. Because of the presence of insulin resistance, its effect on hyperandrogenemia and reproductive complications, and possible prediabetes and diabetes, treatment of insulin resistance and weight loss are essential in the management of PCOS. Guidelines mention lifestyle changes and metformin for providing weight loss and treatment of metabolic components. Moreover, they suggest the use of GLP-RAs in higher weight adults with PCOS for the management of their weight [1]. This class of medications besides being efficient in enabling weight loss have additional pleiotropic effects, like increasing insulin sensitivity, decreasing inflammation, and modulation of adiposity. The interplay between GLP-1 signaling, energy balance, insulin sensitivity, and reproduction is highly complex and not fully understood. The exact mechanism of action of GLP-1 RAs on the ovaries is not fully eloborated, and research in this area is ongoing. However, some possible mechanisms of GLP-1 RAs in PCOS have been proposed. In this regard, the aim of this review was to elaborate on the hypothetical and mechanistic potential benefits of GLP-RAs in these patients, in relation to hormonal, reproductive, and metabolic outcomes.

Methods

This study was designed as a systematic review conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [33]. PubMed, Web of Science, Embase (Elsevier), and the Cochrane Central Register of Controlled Trials (Wiley) were used as databases to identify articles with the combinations of keywords “polycystic ovarian syndrome (PCOS)” AND “obesity” AND “liraglutide” OR “exenatide” OR “dulaglutide” OR “semaglutide” OR “Glucagon like peptide 1 (GLP1) Receptor analogues” with “weight loss” OR “metabolic outcomes” OR “Reproductive outcomes” in the “title or abstract.” Articles that were published in English between January 2000 and December 2024 were included. Guidelines and systematic reviews were also assessed and the relevant articles within them were included. Furthermore, the reference lists of the reviews and original articles were screened to search for additional studies.

Studies included in this review focused on premenopausal women older than 18, diagnosed with obesity and PCOS, based on the criteria established by the National Institutes of Health (NIH), the National Institute of Child Health and Human Development (NICHD), and the ASRM-ESHRE Rotterdam criteria (2003).Two independent reviewers (D.Y. and O.C.) screened and extracted data, while all authors independently reviewed the articles. Any discrepancies were resolved through consensus-based discussions. Original studies considered for inclusion provided details such as article title, year of publication, participant characteristics, PCOS diagnostic criteria, study groups, interventions, sample size, follow-up duration, and metabolic, inflammatory, and reproductive outcomes (Table 1). One reviewer (O.C.) conducted the data extraction and the other authors subsequently verified their results. In cases where overlapping samples were identified, preference was given to studies that assessed a broader range of relevant outcomes. Only studies with a minimum duration of 12 weeks were included.

Table 1.

Summary of the original studies included in the review

Author, year, country, study design	Population/setting	Intervention details		Study duration	Primary outcome		Secondary outcome	Metabolic outcome	Reproductive outcome	BW loss, kg
Elkind Hirsch et al., 2008, USA, RCT [34]	2003 Rotterdam: 18–40 years; overweight, oligo-ovulatory, IR women (BMI >27)	(1) MET: 1,000 mg BID		24 wks	Menstrual frequency		Ovulation rate, anthropometric measures, androgen levels, inflammatory markers	COM therapy was superior to EXE or MET monotherapy in improving insulin sensitivity measures and reducing weight and abdominal fat. TT significantly decreased with all treatments. FAI was significantly more reduced with COM treatment	Menses more regular with COM therapy versus single-agent therapy with MET	(1) MET: −1.6±0.2
		(2) EXE: 10 μg BID							Ovulation rate: COM group 86%, 50% in the EX group, 29% in the MET group (p < 0.01)	(2) EXE: −3.2±0.1
		(3) COM: MET 1,000 mg BID + EX 10 μg BID								(3) COM (MET+ EXE): −6.0±0.5
Rasmussen et al., 2017, Denmark, observational study [35]	2003 Rotterdam: 84 overweight or obese women with PCOS who had failed to lose any weight after 6 months with MET and life style intervention	Add-on of LIRA 1.2–1.8 mg (n = 84)		6 months	Weight loss		NA	NA	NA	Mean: −9.0 (95% CI: 7.8–10.1)
Kahal et al., 2014, UK, case-control study [36]	2003 Rotterdam: 18–45 years; obese women (BMI = 30–45); 19 PCOS, 17 controls	LIRA: 1.8 mg		6 months	Markers of liver fibrosis		Weight loss	Insulin, FPG, HOMA-IR, hsCRP, triglycerides, and urinary isoprostane significantly reduced in both groups. PIIINP significantly reduced in the PCOS group (p < 0 01) but not in controls	NA	PCOS: −3.0±4.2
										Controls: −3.8±3.0 (p = 0.56)
Ma RL et al., 2021, China, open-label RCT [37]	2003 Rotterdam: 18–40 years; overweight or obese (BMI ≥25)	(1) MET: 1.5 mg/day		12 wks	Anthropometric changes		Changes in reproductive hormone levels, glucose and lipid metabolism, and CRP	COM therapy was more effective than MET alone in reducing BW, BMI, and WC and improving insulin sensitivity in women with PCOS	NA	(1) MET: −2.1±3.0
		(2) COM: EXE 2 mg/wk + MET 1.5 mg/day								(2) COM (MET+ EXE): −3.8±2.4
		COCP (each group)
Tao T et al. 2021, China, randomized, open-label, parallel-group controlled trial [38]	2003 Rotterdam: BMI ≥25 kg/m2 and diagnosed with prediabetes	(1) MET: 1,000 mg BID		12 wks	Remission rate of prediabetes		Anthropometric, hormonal, metabolic, and pancreatic β-cell function parameters	Sustained prediabetes remission rate was 50.7%. Remission rate of COM group (64%) or EXE group (56%) was significantly higher than that of the MET group (32%). There were no between-group differences for weight, BMI, or any of the androgen, metabolic, and hormonal levels	NA	(1) MET: −4
		(2) EXE: 10 μg BID								(2) EXE: −5.45
		3) COM: MET 1,000 mg BID + EX 10 μg BID								(3) COM (MET+ EXE): −6.5 (NS)
Jensterle Sever et al., 2014, Slovenia, open-label, prospective study [39]	NICHD criteria: 40 PCOS, obese (BMI ≥30), nondiabetic women who had lost ≥5% body weight during pretreatment with MET for at least 6 months	(1) MET: 1,000 mg BID (n = 14)		12 wks	Weight loss		Body composition	No major changes in FG, insulin, and insulin during OGTT. Glucose value after 120 min during OGTT significantly reduced in the COM arm versus MET (p = 0.006). Only androstenedione level decreased significantly in COM group. Other hormones did not change	NA	(1) MET: −1.2±1.4
										(2) LIRA: −3.8±3.7
		(2) LIRA: 1.2 mg (n = 11)					IR, hormones			(3) COM: −6.5±2.8
		(3) COM: MET 1,000 mg BID + LIRA 1.2 mg (n = 11)								COM therapy was superior to LIRA and MET monotherapy in reducing weight
Frossing et al., 2018, Denmark, RCT [40]	2003 Rotterdam, PCOS women with BMI >25 and/or insulin resistance	(1) LIRA: 1.8 mg (n = 48)		26 wks	Liver fat VAT		Weight change	LIRA reduced liver fat content by 44%, VAT by 18%, the prevalence of NAFLD by two-thirds (all p < 0.01). With LIRA, SHBG levels increased by 19% (p = 0.03) and FT decreased by 19% (p = 0.054). HbA1c, fasting glucose, and leptin were reduced (all p < 0.05), whereas measures of TT, IR, adiponectin, and glucagon did not change	NA	(1) LIRA: −5.2±0.7
		(2) Placebo (n = 24)			Prevalence of NAFLD		OGTT, SHBG, testosterone			(2) Placebo: +0.2±0.9 (p < 0.001)
Elkind-Hirsch et al., 2022, USA, RCT [41]	PCOS, modified NIH 1990 criteria Women (BMI ≥30) with newly diagnosed PCOS, 18–45 years	(1) LIRA: 3 mg (n = 55)		32 wks	Weight loss, body composition by DXA, FAI		OGTT, Matsuda’s insulin sensitivity index, HOMA-IR, sex steroids, lipids	Total fat mass% decreased (p = 0.028) with LIRA. There were no significant changes in lean body mass. FAI decreased with LIRA (p = 0.006). FBG, HOMA-IR, Matsuda (ISOGTT), GI/HOMA improved significantly in LIRA group. CHOL, HDL-C, and LDL-C levels were not consistently altered with any treatment. In contrast, blood pressure, TG, and TG/HDL ratio were consistently reduced with LIRA	Menses significantly increased with LIRA (p: 0.0001)	(1) LIRA: −5.7±0.75
		(2) Placebo (n = 27) with life style								(2) Placebo: −1.4±1.09 (p < 0.001)
Zhang Y et al., 2023, China, single-center, randomized, controlled, open-label clinical trial [42]	2003 Rotterdam, 68 overweight or obese PCOS (BMI ≥24), 18–40 years	(1) CRD		6 months	Difference in the change in VAT area reduction		Menstrual frequency, metabolic profiles, hormonal parameters, liver fat, and body composition	There was no significant between-group difference in area change of VAT reduction. DULA + CRD had significant advantages in reducing HA1c and postprandial plasma glucose levels	No changes in menstruation frequency	(1) CRD: −5.44
		(2) CRD + DULA: 1.5 mg/wk						Metabolic profiles, hormonal markers, liver fat, and body composition between the two groups did not differ significantly		(2) CRD+ DULA: −5.42 (NS)
Jensterle Sever et al, 2015, Slovenia, open-label, prospective study [43]	2003 Rotterdam, women aged 18 years to menopause and with obesity (BMI ≥30)	(1) MET:1,000 mg BID (n = 14)		12 wks	Weight loss		Hormonal and metabolic changes	HOMA-IR decreased in all treatment arms (difference was not statistically significant in groups). LIRA was superior to MET in reducing glucose at 30–120 min of OGTT. Significant TT and FAI reduction was noted in ROF arm when compared to baseline. No statistically significant differences were found in FT, SHBG, androstenedione, DHEAS, LH, and FSH, neither over time nor when analyzing it separately by therapeutic arm	Menstrual frequency increased with all treatments. The increase was shown as being slightly greater in patients treated with ROF compared with MET (p = 0.090) and LIRA (p = 0.165). However, the between-treatment differences were not statistically significant	(1) MET: −0.8±1.0
		(2) LIRA: 1.2 mg (n = 14)								(2) LIRA: −3.1±3.5
		(3) ROF: 500 mg qd (n = 14)								(3) ROF: −2.1±2.0. LIRA was superior to MET in reducing weight (p = 0.022)
Jensterle et al., 2017, Slovenia, prospective randomized open-label design [44]	2003 Rotterdam, 18 years to menopause and obese (BMI ≥30)	(1) COM: MET 1,000 mg BID + LIRA 1.2 mg (n = 15)		12 wks	Weight loss		Metabolic and hormonal changes	Significant decreases in glucose at 0 and 120 min, insulin at 0 min of OGTT, HOMA-IR, and LDL cholesterol were observed in the COM arm. LIRA 3 mg resulted in significant improvement of glucose and insulin levels at 120 min of OGTT. COM was superior in the decrease of LDL-C. TT decreased in COM arm (p = 0.023). SHBG (p = 0.018) is increased in LIRA 3 mg arm	NA	(1) COM: −3.6±2.5
		(2) LIRA: 3 mg (n = 15)								(2) LIRA 3: −6.3±3.7 (<0.01)
Nylander et al., 2017, Denmark, double-blind, placebo-controlled, randomized trial [45]	2003 Rotterdam, Overweight (BMI ≥25), PCOS women and/or IR, ≥18 years	(1) LIRA: 1.8 mg (n = 44)		26 wks		Bleeding pattern	BMI, AMH, sex hormones, gonadotropins, ovarian morphology: ovarian volume, stromal volume, antral follicle count, F-G score	Mean reductions in FBG and HbA1c of 0.24 mm (p < 0.05) and 1.4 mmol/mol (p < 0.05), respectively, compared with the placebo group. No effect on fasting insulin or HOMA-IR	Bleeding ratio significantly improved with LIRA versus placebo. LIRA: SHBG increased and FT decreased. Ovarian volume decreased with LIRA versus placebo. Other: NS	LIRA versus placebo: −5.2 kg (95% CI: 3.0–7.5)
		(2) Placebo (n = 21)
Liu X et al. 2017, China, open-label, prospective, randomized, clinical study [46]	2003 Rotterdam, 176 PCOS with overweight or obesity	(1) MET: 1,000 mg BID (n = 88)		12 wks; after 12 wks, all patients treated with MET alone for additional 12 wks		Weight loss	Fat mass, hormonal levels, IR, lipid profile inflammatory marker, menstrual frequency, rate of pregnancy	Decrease in android fat mass% and total fat mass% with EXE, but not with MET (p = 0.023, 0.011 respectively)	Menstrual frequency increased significantly in both groups (p < 0.001)	(1) MET: −2.28±0.55
		(2) EXE: 10 μg BID (n = 88)						Greater decrease in HOMA-IR and insulin levels with EXE than with MET. FBG, TC, TG, HDL-C, and LDL-C levels did not change. hsCRP levels decreased significantly in the EXE group only. TT decreased, but not significantly, in both treatment groups. The FAI (p < 0.001) reduced significantly and SHBG levels raised (p < 0.001) in both groups with no significant difference between types of treatment	Rate of natural pregnancy significantly higher following EXE treatment than following MET treatment (p < 0.05)	2) EXE: −4.29±1.29 (p < 0.001)
	18–40 years
Salamun et al. 2018, Slovenia, prospective randomized open-label study [47]	2003 Rotterdam, 28 infertile women (BMI ≥30), first or second IVF attempt, age ≤38 years	(1) MET 1,000 mg BID (n = 14)		12 wks		IVF PR	Weight loss, DXA, VAT	All treatment interventions resulted in significant reduction of VAT	The PR per ET was significantly higher in the COM (85.7%) compared with the MET (28.6%) group (p = 0.03)	(1) MET: −7.0±6.0
		(2) COM: MET 1,000 mg BID + LIRA 1.2 mg (n = 14)						Mass, volume, and area as assessed by DXA. HOMA-IR score was reduced after both interventions, the reduction being significant only in the COM group (p = 0.04). Fasting and post-OGTT decreased significantly also in the COM group only (p = 0.004 and p = 0.05, respectively)	The cumulative PR in the time frame of 12 months was 69.2% in the COM group compared to 35.7% in the MET group	(2) COM: −7.5±3.9 (p = 0.246)
								There was a significant increase in SHBG noted in the MET and COM treatment arm compared to the baseline (MET: p = 0.019, COM: p = 0.03)
Jensterle Sever et al., 2024, Slovenia, open-label, prospective study [48]	2003 Rotterdam, 25 PCOS with obesity, mean age: 33.7±5.3 years	COM: SEMA 1 mg/wk +MET 1,000 mg BID + lifestyle intervention. At wk 16, SEMA was discontinued. Treatment with metformin 2,000 mg/day and promotion of lifestyle intervention were continued during the 2-year follow-up period	16 wks			Weight change, cardio-metabolic, and endocrine parameters were assessed 2 years after SEMA discontinuation	NA	There was significant weight loss of about 9 kg in the patients in the short term and this was maintained in the 2-year follow-up period. Improvements in cardio-metabolic parameters including decrease in total and LDL cholesterol, FBG, and glucose after OGTT that had been seen during SEMA treatment phase reverted toward baseline 2 years after SEMA cessation. FT levels significantly decreased during SEMA treatment (p = 0.012) and did not significantly deteriorate after SEMA discontinuation	NA	Baseline: 101 (90–106.8)
										COM: 92 (83.3–100.8)
										Off-treatment phase: 95 (77–104) <0.001
Carmina and Longo, 2023, Italy [49]	2003 Rotterdam, 27 PCOS women with BMI >30 who were unresponsive to a lifestyle modification program, 65 control	SEMA: 0.5 mg/wk	3 months. In responsive patients (weight loss > 5%), the SEMA treatment was prolonged for an additional 3 months			Weight loss and metabolic changes	Menstrual cycles	First period BMI, body weight, FBG, insulin, and HOMA-IR values significantly decreased	Of the patients who lost at least 5% of their BW, about 80% normalized their menstrual cycles	There was a mean decrease in BW of 7.6 kg. In patients who had at least > 5% weight loss with SEMA, the treatment was continued for additional 3 months, reaching a total weight loss of 11.5 kg
								No further changes in FBG, insulin or HOMA-IR were found in the second period
Zheng et al., 2017, China, open-label, randomized, prospective trial [50]	2003 Rotterdam, adult women with PCOS aged 18–40 years	(1) MET: 1,000 mg BID (n = 32)	12 wks			Weight loss	Hormonal, metabolic, inflammatory parameters, menstrual frequency	There was a significantly greater effect of EXE compared to MET with respect to 2hINS, HOMA-IR, INSAUC, and ISI (p < 0.001, p = 0.004, p < 0.001, p < 0.021, respectively) also in reducing hsCRP. But both treatments leading to significant decreases in 2hINS, HOMA-IR, and INSAUC and a significant increase in ISI. FAI decreased while SHBG increased significantly with both treatments with no significant difference between two groups	Menstrual periods, FG score, and Rosenfield score were not significantly different before and after treatment for either group	(1) MET: −1.2±0.3
		(2) EXE: 10 μg BID (n = 31)								(2) EXE: −3.1±0.2 (p < 0.001)
Dawson AJ et al., 2019, UK, open-label study [51]	2003 Rotterdam, 20 adult women with PCOS (mean age: 27±4 years)	EXE: 5 μg BID for 4 wks, then 10 μg BID	12 wks			Weight loss	FAI, endothelial function, serum endothelial markers, change in inflammation (hsCRP), and alteration in clot structure and function	There was no change in FAI or SHBG. There was no effect on lipid parameters, but there was an improvement in TG: 1.3±0.1 to 1.2±0.1 mmol/L (p = 0.04). There was no difference in glucose, insulin, and HOMA-IR compared before and after exenatide treatment. Serum markers of endothelial function, inflammation, and clot function were improved	NA	Reduction in weight with EXE: 111.8±4.8 to 108.6±4.6 kg (p = 0.003)
Chuan Xing et al., 2022, China, prospective, randomized, open-label, parallel group controlled trial [52]	2003 Rotterdam, 60 adult women with PCOS aged 18–40 years	(1) MET: 1,000 mg BID (n = 25)	12 wks			Anthropometric measurements, menstrual cycle changes, gonadal profiles, and OGTT	NA	Both MET and COM improved anthropometric parameters and glucose metabolism after the 12-wk treatment; however, there was no statistical difference between the two groups. COM treatment improved TT, SHBG, FAI, LH, and progesterone more effectively than MET monotherapy; however, there were no significant differences on E2, FSH, and LH/FSH between the two groups	Both MET and COM improved menstrual cycles (p = 0.9)	There was no significant difference between the two groups in improving body weight, BMI, and WC
		(2) COM: MET 1,000 mg BID + LIRA 1.2 mg (n = 27)

RCT, randomized controlled trial; IR, insulin resistance; BMI, body mass index; wk, week; MET, metformin; EXE, exenatide; COM, combination; TT, total testosterone; FAI, free androgen index; LIRA, liraglutide; NA, not applicable; FPG, fasting plasma glucose; HOMA-IR, Homeostatic Model Assessment for Insulin Resistance; PIIINP, procollagen type 3 amino-terminal peptide; BW, body weight; CRP, C-reactive protein; PCOS, polycystic ovary syndrome; WC, waist circumference; NS, nonsignificant; NICHD, National Institute of child health and human development; FG, fasting glucose; OGTT, oral glucose tolerance test; VAT, visceral adipose tissue; NAFLD, nonalcoholic fatty liver disease; SHBG, sex hormone binding globulin; FT, free testosterone; DXA, dual energy X-ray absorptiometry; ISOGTT, insulin sensitivity index; CHOL, cholesterol; LDL-C, LDL cholesterol; HDL-C, HDL cholesterol; CRD, calorie-restricted diet; DULA, dulaglutide; ROF, roflumilast; AMH, anti-mullerian hormone; F-G score, Ferriman gallwey score; IVF, in vitro fertilization; PR, pregnancy rate; SEMA, semaglutide. INS-AUC, insulin under the curve; ISI, insulin sensitivity index.

Clinical Studies and Evidence on the Use of GLP-1 RAs in Patients with PCOS

First-line treatment of PCOS includes lifestyle modifications; however, it does not seem to be very effective in reducing weight or treating symptoms of PCOS. Pharmacotherapy, added on to lifestyle modifications, can be used to augment the efficacy of weight loss. GLP-1 RAs have gained attention as potential therapeutic agents owing to their metabolic effects, including glycemic control, weight reduction, and improvement in insulin sensitivity. This section aims to evaluate clinical studies and evidence on the effect of GLP-1 RAs in management of PCOS.

Effects of GLP1-RA on Weight Control and Metabolic Endpoints

Several studies have investigated the role of GLP-1 RAs in improvement of metabolic components and enable significant weight loss in women with PCOS, who are overweight or who have obesity. Clinical studies of GLP-1 RAs in patients with PCOS who are overweight or who have obesity are summarized in Table 1. In 2008, Hirsch et al. investigated the impact of exenatide (EXE), metformin (MET), and their combination (COM) on metabolic profiles and weight management in women with PCOS, who have insulin resistance and who are overweight. Both EXE treatment groups demonstrated greater efficacy in promoting weight loss compared to MET alone. All treatment groups showed significant improvements in insulin resistance, as measured by Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), and in insulin sensitivity, assessed using the Matsuda index (ISOGTT). Notably, the COM therapy outperformed either monotherapy in enhancing insulin sensitivity and reducing body weight and abdominal fat. While HDL-C and LDL-C levels remained largely unchanged, total cholesterol (TC) and triglyceride (TG) levels significantly decreased with COM therapy, unlike MET alone, which did not consistently produce such improvements [34]. Tao et al. [38] compared the combined use of EXE (10–20 μg/day) and MET (1.5–2 g/day) with MET alone in patients with prediabetes. The overall remission rate of prediabetes was 50.7%. EXE or the COM therapy led to a higher rate of prediabetes remission in PCOS patients by enhancing postprandial insulin secretion. However, no significant differences were observed between the groups in terms of weight, BMI, or androgen levels (including free androgen index [FAI], total testosterone [TT], sex hormone binding globulin [SHBG], dehyroepiendrosterone sulfate [DHEAS], and androstenedione [A]). Metabolic markers such as fasting blood glucose (FBG), fasting insulin, HOMA-IR, and lipid profiles, HDL-C, TG also showed no significant differences, except MET was more effective than EXE + MET in lowering 2-h insulin and LDL-C.

Ma et al. [37] evaluated EXE (2 mg weekly) with MET (1.5 g/day) versus MET alone, with both groups receiving co-treatment with cyproterone acetate 2 mg + ethinyl estradiol 35 μg (Diane-35). The COM therapy showed greater reductions in body weight, BMI, and waist circumference (WC) and led to improved insulin sensitivity in individuals with overweight and obesity and PCOS, with tolerable short-term side effects compared to MET alone. COM therapy with EXE and MET outperformed LIRA and MET alone in reducing weight, BMI, and WC in PCOS patients with obesity unresponsive to MET. Although HOMA-IR scores tended to decline across all treatments, none reached statistical significance. Fasting glucose and insulin levels during OGTT were inconsistently affected, but the combination group showed the most significant reduction in glucose at 120 min during OGTT (p = 0.006 vs. MET) [39].

PCOS is associated with increased liver fat and a fourfold higher prevalence of metabolic dysfunction associated steatotic liver disease (MASLD) compared to BMI-matched controls, despite conflicting findings from imaging studies. Frossing et al. [40] conducted a randomized, placebo-controlled trial in 72 PCOS women (BMI >25 and/or IR), treated with LIRA (1.8 mg/day) or placebo for 26 weeks. LIRA significantly reduced body weight (−5.2 kg), liver fat (−44%), visceral adipose tissue (VAT) (−18%), and MASLD prevalence (by two-thirds), alongside improvements in SHBG (+19%) and free testosterone (FT) (−19%). Reductions were also seen in HbA1c, FBG, and leptin, although measures like insulin resistance, adiponectin, and glucagon remained unchanged. Other data also confirm reductions in weight, insulin, FBG, HOMA-IR, and TG with LIRA in PCOS and metabolic dysfunction-associated steatohepatitis patients [36].

LIRA, approved up to 1.8 mg for T2DM, requires higher doses for optimal weight loss. LIRA 3 mg leading to weight loss about 5–15% is indicated for obesity treatment in many countries. Jensterle et al. [53] found LIRA 3 mg to be more effective than low-dose LIRA (1.2 mg) plus MET in decreasing BMI and WC. Additionally, combination therapy further improved androgen profiles and was better tolerated. Recently, Hirsch et al. [41] assessed LIRA 3 mg versus placebo over 32 weeks in obese PCOS patients. LIRA 3 mg led to greater weight loss (5.7% vs. 1.4%, p = 0.002), reductions in WC and waist-hip ratio (WHR). There was improved fat mass percentage with LIRA 3 mg, with no change in lean body mass. Free androgen index (FAI) levels dropped significantly with LIRA 3 mg, while the placebo group saw a slight increase. Glucose metrics during OGTT, HOMA-IR, SIOGTT, and early insulin response (IGI/HOMA) all improved significantly. Although TC, HDL-C, and LDL-C remained stable, TG and TG/HDL ratios dropped with LIRA 3 mg, consistent with earlier findings in T2DM patients. This finding was consistent with previous data that found LIRA dramatically decreased TG in patients with T2DM [36]. Niafar et al. [54] conducted a meta-analysis of seven RCTs, showing LIRA reduced BMI by 1.65 kg/m² after 3 months and lowered serum TT, though insulin sensitivity and SHBG levels remained unchanged. Despite the effectiveness of GLP-1 RAs in weight reduction for women with PCOS having overweight or obesity, study heterogeneity was high. Another meta-analysis of eight RCTs, including studies from Asian populations, demonstrated GLP-1 RAs to be more effective than MET in decreasing BMI, WC, and insulin resistance in patients with PCOS [55].

Recent evidence highlights the potential role of semaglutide (SEMA) in managing PCOS and obesity. A study evaluated the short-term effects of SEMA (1.0 mg) along with MET (2,000 mg/day) and lifestyle modifications in 25 women with PCOS and obesity over a 16-week period. This was followed by a 2-year follow-up phase with continued MET and lifestyle interventions alone. Participants experienced a notable short-term weight loss of around 9 kg, which was sustained throughout the 2-year follow-up. Additionally, FBG and OGTT glucose levels significantly improved at 16 weeks but returned to baseline by the end of the follow-up. Regarding androgenic markers, the FAI decreased during treatment and remained stable after discontinuation of SEMA [48]. In an Italian population, patients with PCOS and obesity were on SEMA (0.5 mg) for 3 months and the BW decreased by 7.6 kg. FBG levels normalized in 80% of patients. In patients with at least >5% weight loss with SEMA, the treatment was continued for an additional 3 months, where weight loss continued at a slower rate, reaching a weight loss of 11.5 kg in total. Among patients who achieved at least a 5% reduction in body weight, approximately 80% experienced normalization of their menstrual cycles [49].

Moreover, Zhang et al. [42] found that patients with PCOS who received dulaglutide (DULA) combined with a calorie-restricted diet (CRD) reached a 7% weight loss target in a significantly shorter median time compared to those on CRD alone, indicating DULA’s potential for more efficient weight reduction. However, no significant difference was observed between the groups in terms of reduction in VAT area. While the combination therapy showed greater improvements in HbA1c and postprandial plasma glucose levels, there were no notable differences between the groups regarding menstrual frequency, other metabolic parameters, hormonal profiles, liver fat content, or body composition.

In patients with PCOS, visceral fat deposition is increased and even in patients who do not have overweight or obesity; thus, it is reasonable to determine the compartmental body composition in these patients [28, 56]. There are few studies within the scope of this review that have determined the body fat distribution in women with PCOS. LIRA added on to MET resulted in decreases in VAT [47]. Elkind-Hirsch et al. [41] demonstrated that LIRA decreased total fat mass, whereas the lean mass was not changed. There was a decrease in liver fat content with LIRA in patients with PCOS and NAFLD [40]. Android fat mass and total fat mass were reduced with EXE [46]. No significant changes in lean body mass by DULA added on to CRD was evident [42].

There were two studies included in the review where there was coexistence of PCOS with NAFLD. In the study by Frossling et al. [40], no biochemical markers about NAFLD were detected. In the other study, LIRA combined with weight loss led to a decrease in procollagen type 3 amino-terminal peptide (PIIINP) levels, a marker associated with liver cirrhosis risk in women with PCOS and obesity [36].

Overall, GLP-1 RAs demonstrate varying effectiveness in promoting weight loss over different time frames, which aligns with metabolic improvements such as reductions in visceral fat mass in women with PCOS. However, further research, including larger randomized controlled trials and long-term safety studies, are necessary to establish optimal treatment protocols and to fully comprehend the risks and benefits associated with GLP-1 RAs with/without metformin in this population.

Effects of GLP1-RA on Reproductive End Points

Beyond their role in metabolic regulation and weight loss, GLP-1 receptor agonists have also demonstrated potential reproductive and hormonal benefits in women with PCOS. A 24-week combination therapy of EXE and MET proved more effective than either agent alone in improving menstrual cycle regularity and lowering hyperandrogenism in individuals with overweight and PCOS. Although progesterone levels were measured in only one cycle, the ovulation rate exceeded 85% in the combination group, compared to less than 30% with MET monotherapy, underscoring the advantage of dual therapy. TT and FAI levels significantly declined across all treatment groups, while SHBG levels increased, though not significantly. Notably, FAI reduction was significantly greater with combination therapy compared to MET alone but not when compared to EXE alone [34]. This finding could be due to increased SHBG levels observed with EXE, but not MET, reducing the bioavailability of circulating androgens.

Jensterle et al. [43] randomized PCOS patients to MET 1,000 mg BID or LIRA 1.2 mg QD or roflumilast (ROF) 500 μg QD. After 12 weeks of treatment, ROF group showed significant reductions in TT and FAI compared to baseline. However, no significant changes were observed in FT, SHBG, A, DHEAS, LH, or FSH levels, either over time or between treatment groups. All treatment groups experienced an increase in menstrual frequency, with a slightly greater, though not statistically significant, increase in the ROF group compared to those treated with MET and LIRA. In a separate study, Nylander et al. [54] assessed 72 women with PCOS receiving either LIRA (1.8 mg/day) or placebo over 26 weeks. LIRA led to a significant weight loss of 5.2 kg (95% CI: 3.0–7.5; p < 0.0001) compared to placebo. Additionally, LIRA improved bleeding patterns, increased SHBG levels, and reduced free testosterone (FT). Ovarian volume also declined by 1.6 mL (95% CI: −3.3 to 0.1) in the LIRA group versus placebo. Hirsch et al. [41] demonstrated that in the LIRA 3 mg group, the mean frequency of menstrual cycles increased from 4.5 ± 0.3 to 8.65 ± 0.4 cycles per year. In contrast, the placebo group showed no change, with cycles remaining at 4.8 ± 0.5 and 4.8 ± 0.65 per year. Although TT levels did not decrease significantly, treatment with LIRA 3 mg led to an increase in SHBG, which in turn lowered FAI, resulting in an improvement in hyperandrogenism that was not observed with placebo. In contrast, a few studies on LIRA in PCOS showed no effect on menstrual frequency despite a decrease in body weight [39]. Another group demonstrated that EXE treatment manifested distinct advantages in losing weight but without any changes in menstrual periods, testosterone, DHEAS levels, mean FBG, and Rosenfield scores [46].

The impact of LIRA treatment on pregnancy outcomes has been evaluated in two studies. In one study, Liu et al. [46] assessed 176 overweight or obese women with PCOS who were randomized to receive either EXE 10 μg twice daily or MET 1,000 mg twice daily for 12 weeks. This was followed by an additional 12-week period during which all participants received MET alone. After the initial 12-week intervention, those in the EXE group experienced significantly greater weight loss, improved insulin resistance, and increased menstrual frequency compared to the MET group. Notably, during the second phase, the natural pregnancy rate (PR) in the EXE group was significantly higher than in the MET group (43.6% vs. 18.7%, p < 0.05). Salamun et al. [47] evaluated the impact of low-dose LIRA combined with MET (COM) versus MET alone on in vitro fertilization PRs and cumulative PRs (including in vitro fertilization and spontaneous pregnancies) in infertile women with PCOS and obesity. These participants had previously shown poor response to lifestyle interventions and were resistant to first-line fertility treatments such as clomiphene citrate or aromatase inhibitors. At baseline, there were no significant differences between the groups in terms of anthropometric, metabolic, or hormonal profiles, except for 120-min insulin levels during OGTT. All participants received lifestyle counseling from a multidisciplinary team. Both MET and COM groups showed significant increases in SHBG levels from baseline (MET: p = 0.019; COM: p = 0.03). Notably, the PR per embryo transfer was significantly higher in the COM group (85.7%) compared to the MET group (28.6%, p = 0.03). Additionally, the cumulative 12-month PR was 69.2% in the COM group versus 35.7% in the MET group. Larger studies should be conducted using these preliminary results.

Despite the data indicating that GLP-1 RAs would increase rate of pregnancy, there are no data on the effect of these agents on pregnancy complications. This will be an interesting area of research as the use of these medications will probably be increased in this patient group, along with increased rate of fertilization.

There has been an extensive meta-analysis published before the 2023 guideline and it has analyzed the effect of all anti-obesity medications in 10 years time up to July 2022. This meta-analysis included eight trials that evaluated the efficacy of GLP-1RAs in women with PCOS. They have observed that treatment with LIRA and SEMA led to prominent weight loss compared to placebo and that led to relevant metabolic and reproductive outcomes [57]. Our systematic review included randomized controlled trials conducted up to December 2024, expanding on available data in the literature. Besides, the current review included only GLP-based therapies, in order to focus more on their role in patients with PCOS regardless of the presence of obesity. Moreover, we have elaborated on the postulated specific mechanisms of GLP-RAs in PCOS besides the effect on weight.

Weight Regain

Weight regain is mostly observed when GLP-1RAs are discontinued, based on trials conducted in patients with obesity. These data come from the STEP trials, where in the STEP 4 trial, after a run-in treatment period with SEMA for 20 weeks, there was a re-randomization to SEMA continuation and placebo for additional 48 weeks. In the group that continued SEMA, there was a 17.4% reduction in body weight from baseline, whereas the group that switched to placebo experienced a 6.9% weight gain. [58]. Similar findings have been observed in the STEP 1 extension trial, where 327 participants were observed after treatment withdrawal, where SEMA and placebo group participants gained back 11.6% and 1.9% of lost weight, respectively, at the end of 120 weeks, along with the improvements in cardiometabolic parameters reverting to baseline [59]. The weight regain is possibly due to alterations in the levels of hypothalamic and gut hormones regulating satiety and decreased metabolic rate [60–62].

Considering the weight regain in patients with PCOS, the recent study that we had included in this manuscript by Jensterle et al. [48] mentioned above looked into weight regain, changes in cardiometabolic and endocrine parameters after discontinuation of SEMA following 16 weeks of treatment. Two years after SEMA withdrawal, regain in weight was 2 kg and the net weight loss after discontinuation of semaglutide was still significant when compared to baseline. Furthermore, the improvements in cardiometabolic variables observed during the SEMA treatment phase reverted to baseline levels, except for changes in FT levels that persisted afterward. Importantly, patients were continuing with MET in the following SEMA-free period. Based on these findings, it may be beneficial to continue MET treatment after discontinuing GLP-1 RA plus MET in order to prevent weight regain. However, weight regain with GLP-1 RAs has to be determined in larger scale placebo controlled trials in patients with PCOS.

Mechanisms

Below and highlighted in Figure 1 are the postulated mechanisms of action of GLP-1 RAs in patients with PCOS.

Fig. 1.

Postulated benefits of GLP-1 RAs on the different pathophysiological pathways in PCOS. SHBG, sex hormone-binding globulin; GLP-1 RAs, glucagon-like peptide 1 receptor agonists; LH, luteinizing hormone; FSH, follicle-stimulating hormone; GnRH, gonadotropin-releasing hormone; ACTH, adrenocorticotropin hormone.

Improvement of Insulin Sensitivity

Insulin resistance is a central pathophysiological feature of PCOS, occurring independently of obesity; however, obesity can exacerbate insulin resistance, contribute to adipocyte dysfunction, and amplify androgen excess in patients with PCOS [63–65]. Besides, it has been demonstrated by Yildiz et al. [66] that there is increased risk of metabolic syndrome, type 2 DM, hypertension, and dyslipidemia in the fathers, and higher risk of hypertension in the brothers of women with PCOS, so this would stress the effect of insulin resistance in the pathophysiology of PCOS per se. Literature emphasizes the improvement of insulin sensitivity by GLP-RAs in people with PCOS [34, 37, 38]. Whether the effect of these agents is regardless of weight loss and decreased adipose tissue and is specific to insulin resistance of PCOS has to be elucidated in future trials.

Reduction in Androgen Levels

Improving insulin sensitivity reduces the disruption of the normal menstrual cycle, leading to irregular ovulation by lowering excess androgens [34]. Besides, increased SHBG levels by GLP-RAs have caused decreases in TT [34] and FAI [34, 41].

Regulation of Ovarian Function

GLP-1 RAs may also exert direct effects on the ovaries, as GLP-1 receptors are expressed in ovarian tissue, and their activation may enhance ovarian function [67]. Research in animal models suggests that GLP-1 RAs could help regulate ovarian steroidogenesis and improve follicular development, potentially restoring normal ovulation [67, 68]. GLP-1 receptors have also been demonstrated in the pituitary gland and their activation could also potentially influence the production and secretion of hormones as GnRH, LH, and FSH and restore the LH-to-FSH ratio, essential for normal ovarian function [68–71].

Improvement of Menstrual Irregularities

By improving insulin sensitivity and reducing hyperinsulinemia, GLP-1 RAs may help to restore regular menstrual cycles and improve ovulatory function, both of which are often disrupted in women with PCOS [45, 72].

Anti-Inflammatory Effects

Chronic low-grade inflammation is believed to play a role in insulin resistance and other metabolic abnormalities associated with PCOS. GLP-1 receptor agonists exhibit anti-inflammatory effects, which may help mitigate inflammation and enhance metabolic outcomes in women with PCOS [72].

Cardiovascular Protection

GLP-1 RAs have shown cardiovascular benefits, including improved lipid profiles and reduced blood pressure, which may be particularly advantageous for women with PCOS, who are at elevated risk for CVD [34, 73]. Additionally, reductions in cardiovascular risk markers have been observed in women with PCOS treated with these agents [51, 74, 75].

Effect on Adipose Tissue

Previous data indicate that GLP-RAs induce thermogenesis in the adipose tissue and browning of the adipocytes [76]. Clinical studies revealed no difference of brown adipose tissue with LIRA but an increase with EXE [76]. GLP-1 RAs can influence the distribution of body fat, shifting it away from visceral fat to subcutaneous fat. GLP-RAs have reduced VAT, subcutaneous, liver, and epicardial adipose tissue in patients with obesity [77, 78]. This redistribution of fat can ameliorate the metabolic profile and decrease insulin resistance in women with PCOS, as demonstrated above in studies in women with PCOS [40, 41, 47, 46].

Gut-Brain Axis Modulation

GLP-1 RAs are involved in regulating appetite and satiety through the gut-brain axis [79, 80]. By modulating this pathway, GLP-1 RAs may help control hunger and reduce overeating, which could contribute to better weight management in women with PCOS. Moreover, there is the possibility that disturbed patterns of GLP secretion may be evident in PCOS in the first place. According to Yildiz et al. [81], lean women with PCOS exhibit a diminished GLP response both at fasting and during a mixed meal challenge when compared to healthy controls. Notably, short-term oral contraceptive therapy does not appear to affect GLP levels in these individuals. Current evidence suggests that women with PCOS, regardless of the presence of obesity, do not show consistent differences in basal or stimulated GLP-1 levels. In lean women with PCOS, studies have reported varying results – ranging from unchanged to decreased or even increased fasting GLP-1 levels and either unchanged or reduced post-stimulation GLP-1 levels [82]. This inconsistency highlights the need to clarify the regulation of basal and stimulated GLP-1 secretion in PCOS.

In summary, the proposed mechanisms through which GLP-1 RAs exert their effects in PCOS include improvements in insulin resistance and chronic inflammation, modulation of adipose tissue function, potential direct ovarian actions, and normalization of GLP secretion patterns. Further research is necessary to determine whether GLP-1 RAs act differently in women with PCOS based on obesity status, which may help uncover underlying pathophysiological mechanisms. These hypotheses warrant exploration in future clinical trials.

Limitations

First, the longest duration of all the studies included was up to 32 weeks. The data available for people with obesity is up to about 4 years [83] and discontinuation of the medications is commonly associated with weight regain [58, 59]. There is only one study available looking into weight regain in patients with PCOS [48]. Moreover, the doses of GLP-RAs used are variable among the studies. In most of the studies, the doses were doses indicated for DM, only two studies included LIRA that was used in doses indicated for obesity [41, 44].

Conclusion

GLP-1 RAs show potential in the management of PCOS due to their ability to improve glycemic control, promote weight loss, and enhance both hormonal and metabolic profiles. As previous reviews also have mentioned [84], more preliminary data are becoming available on the possible efficacy of these agents in patients with PCOS, particularly in those who are overweight or have obesity. Studies exploring the use of GLP-1 RAs in patients with PCOS have yielded promising findings, demonstrating improvements in metabolic outcomes such as enhanced insulin sensitivity, reduced inflammatory markers, and increased prediabetes remission rates as well as reproductive benefits, including increased menstrual regularity, higher ovulation rates, and reduced androgen levels. These promising effects are thought to stem from improved insulin resistance, potential anti-inflammatory actions, modulation of adipocyte differentiation, and the weight loss induced by GLP-1 RAs. Additionally, direct effects on the pituitary and ovarian function may play a role. Future trials will be essential to determine whether these benefits are solely attributable to weight loss and improvements in obesity-related insulin resistance, or if they also exert independent effects on insulin resistance and directly influence the pituitary and ovaries in women with PCOS. Furthermore, medical weight management for PCOS patients will likely remain an area of interest because new dual acting GLP1-RAs have recently become available. However, more work is needed to exhibit the long-term effects, optimal dosing, and potential effects on metabolic, reproductive, and fertility outcomes in this population. In particular, placebo-controlled studies are urgently needed.

Conflict of Interest Statement

Dragan Micic and Volkan Yumuk were members of the journal’s Editorial Board at the time of submission. The other authors declare that they have no conflicts of interest.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

Ö.C.: summarizing the clinical data on the effect of GLP-1 RAs on metabolic and reproductive functions in PCOS patients with obesity. D.Y.: bringing together the parts and preparing the manuscript for submission. A.C.: summarizing the pathophysiology of PCOS and the related effect of GLP-1 RAs in PCOS and obesity. D.M., D.M., V.D.Y., and B.O.Y.: reviewing and editing the manuscript.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

Research data supporting this publication are available from the NN repository located at www.NNN.org/download/.