Introduction

Polycystic ovary syndrome (PCOS) is a heterogeneous endocrine disorder among women of reproductive age [1]. The worldwide prevalence of PCOS varies depending on the inclusion criteria or ethnicity. Nevertheless, the incidence ranges from approx. 6% to 20% [2].

Polycystic ovary syndrome is characterized by a spectrum of symptoms and clinical features that includes ovarian dysfunction (chronic anovulation with or without menstrual cycle disturbances), clinical and/or biochemical hyperandrogenism, and evidence of morphologically polycystic ovaries seen in ultrasound examination [3]. Although there are several diagnostic criteria for PCOS, the Rotterdam criteria are the most widely used. According to them, PCOS can be diagnosed after exclusion of diseases presenting similar symptoms, when 2 of the 3 above-mentioned features are found [4]. It should be highlighted that the clinical course of this syndrome varies depending on the type of disorder that predominates in the patient. Polycystic ovary syndrome is one of the most common causes of anovulatory infertility [5]. Concerning reproduction, subfertility/infertility and ovarian dysfunction are related to complex factors, e.g. hyperandrogenism, malfunctioning hypothalamic-pituitary-gonadal (HPG) axis, and hormonal abnormalities [6]. Furthermore, metabolic profile disturbances are associated with PCOS. Metabolic disorders such as obesity and insulin resistance (IR) are commonly seen among this group. Obesity is present in 40–70% of patients with this endocrinopathy [3]. Insulin resistance is closely linked to PCOS. Interestingly, amongst individuals with PCOS, insulin resistance is partly independent of obesity [3]. The clinical consequences of disturbed metabolic functions with a high prevalence of obesity and IR in this population are the increased risk of prediabetes, type 2 diabetes mellitus (T2DM), and other metabolic diseases [3, 6].

This review summarizes and discusses findings regarding obesity and its relation to hyperandrogenism and insulin resistance in PCOS, with special attention paid to the pharmacological treatment of adiposity with glucagon-like peptide 1 (GLP-1) analogues.

PCOS aetiology and pathophysiology

It has been widely accepted that the aetiology of PCOS is complex. However, to date, there still is a need to establish all the factors underlying this syndrome. According to the literature, the aetiology of PCOS is a combination that involves an interaction between developmental, environmental, genetic, and epigenetic mechanisms [3]. Non-genetic conditions that are supposed to be involved in PCOS are prenatal androgen exposure related to maternal, foetal, or placental sources, lifestyle factors including poor-quality diet and physical inactivity, endocrine-disrupting chemicals (EDC), altered light exposure, sleep disturbance, heightened levels of stress, lifestyle-induced changes in the gastrointestinal tract microbiome, and others [7]. Extensive evidence indicates that foetal genomic programming of metabolic and endocrine pathways can increase the further susceptibility to develop PCOS following exposure to specific nutritional and environmental factors. Interestingly, there is a theory that PCOS is a result of an evolutionary adaptation to lifestyle and the environment. It seems highly reasonable that, in the past, specific features (IR, hyperandrogenism, adiposity, and subfertility) were metabolic survival responses to food shortages [7].

The exact pathophysiology of PCOS has not yet been fully explained [3]. Although there is a need to conduct further studies for a better understanding of all molecular mechanisms, it is accepted that both hyperandrogenism and insulin resistance are key factors in the development and sustaining metabolic and hormonal disturbances linked to PCOS [8]. It should be mentioned that the above-mentioned malfunctions are interconnected, and their effects are overlapping. Another important pathogenic factor underlying PCOS is disrupted activity of the HPG axis. This axis is largely affected by hyperandrogenism and IR [8].

In general, hyperandrogenism is related to dysfunction of reproductive functions as well as impaired metabolic homeostasis. Androgens in women are derived from the ovaries and adrenals. In the course of PCOS androgen excess from the ovaries is associated with overproduction in the thecal cells caused by enhanced luteinizing hormone (LH) levels leading to the promotion of steroidogenic enzyme expression and, consequently, intensified androgen biosynthesis [6]. Furthermore, a decrease in androgen to oestrogen conversion in the granulosa cells due to attenuated enzyme activity results in hyperandrogenaemia [6]. The enhanced adrenal production of androgens in PCOS cannot also be omitted. The most probable mechanism of adrenal hyperandrogenaemia is an abnormal response to adrenocorticotropic hormone (ACTH) stimulation and disturbed metabolism of adrenal products [9].

Androgen overactivity results in direct and indirect effects on different organs. Firstly, androgens have been shown to change the fat distribution in PCOS from a female to a male pattern with dominating abdominal obesity [6]. It is widely reported that abdominal adiposity is related to increased visceral adipose tissue (VAT). In turn, the enhanced VAT amount is closely linked with disturbed adipokine secretion profile, pro-inflammatory activity, hyperinsulinaemia, and insulin resistance. All those abnormalities may result in metabolic diseases (e.g. T2DM2, dyslipidaemia, and others) as well as an increased risk of cardiovascular disease [10].

Moreover, conversion from testosterone into oestrogens takes place in the adipose tissue [6]. Consequently, high levels of oestrogens disturb the hypothalamic gonadotropin-releasing hormone (GnRH) generator activity and pulsatility, leading to increased LH secretion from the pituitary [11]. Furthermore, androgens indirectly affect LH secretion by influencing the negative feedback loop of oestradiol and progesterone — GnRH/LH secretion [6].

Multiple lines of evidence confirm the negative role of hyperandrogenaemia on ovarian function by affecting folliculogenesis. High levels of androgens induce the growth of preantral follicles and increase the recruitment of follicles, which promotes the secretion of the anti-Müllerian hormone (AMH) from granulosa cells [12]. In addition, androgens at the beginning of folliculogenesis generate increased oestrogen synthesis, which negatively influences the hypothalamus [13]. Elevated AMH concentration inhibits follicular development and lowers follicle-stimulating hormone (FSH) and aromatase activity, resulting in the decreased ovarian conversion of androgens to oestrogens, which additionally intensifies hyperandrogenaemia [12]. This mechanism explains at least partially the phenomenon of increased AMH levels and the high number of early antral follicles observed in subjects with PCOS. Similarly to AMH, enhanced androgens concentrations are responsible for folliculogenesis inhibition and intensified aromatase activity. The arrest of the dominant follicle development and lack of corpus luteum-derived progesterone intensifies the GnRH pulse generator/LH secretion abnormalities [9].

It is also worth noting that hyperandrogenaemia leads to decreased production of sex hormone-binding globulin (SHBG) in the liver. As a consequence of lower SHBG concentration, a higher amount of bioactive androgens is found in the blood of PCOS individuals [12]. Therefore, it could be stated that hyperandrogenaemia and its consequences are a classic model of a vicious circle.

Finally, it should be mentioned that abnormally high secretion of androgens is related not only to metabolic and endocrine complications but also could independently aggravate the development of hypertension, atherosclerosis, and cardiac hypertrophy and enhance the risk of cardiovascular disease and coronary heart disease [14].

Another factor that plays a key role in the pathogenesis of PCOS is insulin resistance coexisting with hyperinsulinaemia. The mechanisms in which insulin disturbances exert their negative role in PCOS are very similar to those of androgens. Peripheral hyperinsulinaemia related to insulin resistance has multidirectional unfavourable central effects on the hypothalamus and pituitary. In detail, insulin stimulates GnRH as well as LH pulse secretion by changing their amplitude and frequency [11]. The indirect effect of hyperinsulinaemia is that it increases GnRH neuron activity [15]. Moreover, elevated insulin concentration stimulates insulin receptors in the pituitary gland, resulting in the release of LH. It has also been suggested that insulin could change pituitary gonadotropin sensitivity to GnRH [15]. The above-mentioned disturbances result in preferential LH secretion from the pituitary gland that leads to ovarian hyperandrogenism and ovulatory dysfunction.

Insulin resistance is defined as insufficient tissue receptor response to insulin [16]. However, not all organs remain insensitive. In obesity as well as in PCOS, adrenals and ovaries have preserved responses to insulin while other organs and tissues including skeletal muscles, adipose tissue, and liver are insulin resistant [16]. This phenomenon has negative consequences in the form of increased androgen production. In detail, in the ovaries insulin influences enzymes that promote steroidogenesis. Furthermore, hyperinsulinaemia enhances LH-binding sites in thecal cells and increases the androgen-producing response to LH [13]. Not surprisingly, high insulin levels have a direct negative impact on ovary function. Similarly to androgens, insulin in excess possesses the ability to stop follicle growth and maturation. In addition, hyperinsulinaemia by affecting insulin-like growth factor 1 (IGF-1) binding protein synthesis in the liver is responsible for enhanced IGF-1 bio-availability. In turn, increased IGF-1 concentration stimulates androgen synthesis in thecal cells, accelerates granulosa cell apoptosis, and inhibits folliculogenesis [6, 9, 17]. It should not be forgotten that insulin also affects SHBG synthesis, and the correlation between these 2 compounds is negative. Therefore, in the course of PCOS hyperinsulinaemia, lowering the SHBG concentration leads to an increase in free testosterone levels with all known consequences [6]. Finally, it is widely accepted that insulin influences adipose tissue. It can stimulate adipogenesis and lipogenesis as well as inhibit lipolysis [18]. Thus, insulin anomalies are linked to increased adiposity, especially in visceral locations. As a factor affecting adipose tissue activity, insulin also plays a role in the deregulation of adipokine secretion [6]. Likewise, hyperinsulinaemia can intensify low-grade inflammation connected with the presence of adipose tissue [6]. Of note, although the majority of women suffering from PCOS are overweight or obese, which could explain insulin resistance, it is likely that all women with PCOS, whether obese or lean, have reduced insulin sensitivity [7]. This hypothesis has been confirmed in a systematic review indicating that PCOS individuals have a 27% reduction in insulin sensitivity compared to body mass index (BMI)- and age-matched controls [19]. Not surprisingly, hyperinsulinaemia and insulin resistance are closely related to a higher risk of metabolic syndrome and cardiovascular complications in women with PCOS.

Adiposity and PCOS

Obesity is a common feature of PCOS. However, it has been reported that also lean women with this endocrinopathy, whose BMI is within the normal range, may present alteration in regional fat distribution due to hyperandrogenaemia [20, 21].

The relationship between the development of PCOS and obesity is bilateral. On the one hand, weight gain and adiposity may contribute to PCOS, while on the other, disturbances found in this syndrome drive further weight gain [22]. It is believed that PCOS individuals have a higher amount of visceral adipose tissue. However, findings based on the magnetic resonance imaging (MRI) examination suggest that women with PCOS manifest global adiposity, rather than a dominance of visceral fat [23].

Unsurprisingly, adiposity in women with PCOS promotes metabolic dysfunction, including insulin resistance. The role of IR and hyperinsulinaemia as important pathological factors in PCOS has been comprehensively discussed above. Hence, obesity is another mediator of reproductive impairment and hyperandrogenaemia. The role of obesity in the pathomechanism of PCOS is confirmed by the findings of weight loss in obese and overweight women with PCOS that resulted in improvement of insulin sensitivity and serum insulin levels, which could serve as an equivalent of the favourable impact on metabolic health. Additionally, amelioration of reproductive functions (including restoration of ovulation, menstrual cyclicity, and fertility) and hyperandrogenic features were also achieved [24].

Furthermore, a wide spectrum of obesity-related mechanisms leads to the development and intensification of disorders observed in the course of PCOS. First of all, adipose tissue can transform steroid hormones. Indeed, due to adipose-derived enzyme up-regulation, the conversion from testosterone into oestrogens, androstenedione into testosterone, testosterone into dihydrotestosterone, and estrone into oestradiol occurs [6]. These enzyme activities result in aggravated hyperandrogenism and hyperoestrogenism, which in turn harm the pathologies seen in POCS. Moreover, conversion of cortisone to active cortisol takes also place in adipose tissue, especially visceral tissue. Relative hypercortisolaemia may worsen the metabolic status and intensify insulin-related disorders [6].

The next important issue concerns inappropriate adipokine secretion in overweight or obese patients. Interestingly, hyperandrogenism in PCOS drives dysfunctional adipocyte secretion of potentially harmful adipokines [25]. Adiponectin, one of the adipose tissue-derived compounds, plays a positive role in carbohydrate metabolism and many endocrine mechanisms. Its concentration is inversely proportional to the amount of adipose tissue. Notably, results from 2 meta-analyses of studies on PCOS subjects revealed that serum adiponectin levels are lower in women with PCOS compared with BMI-adjusted control women, possibly contributing to insulin resistance and further enhancement of ovarian androgen production [26, 27]. Under physiological conditions, adiponectin is found to inhibit androgen synthesis in theca cells, trigger oestradiol secretion in granulosa cells, and play a role in ovulation. It has also been reported that this adipokine decreases GnRH activity and reduces LH secretion from the pituitary [28]. Therefore, decreased adiponectin concentration has a negative impact on PCOS. Obesity is also connected with enhanced secretion of leptin. High leptin levels influence aromatase activity in granulosa cells and that way interrupt androgens to oestrogens conversion [9]. Possibly, hyperleptinaemia is also related to the absence of folliculogenesis [28].

Underestimated complications of obesity, which may also aggravate the disorders present in PCOS, include oxidative stress and chronic low-grade inflammation. Oxidative stress is defined as an imbalance between pro-oxidants and antioxidants. The oxidative molecules may interact with ovary functions, including steroidogenesis. They also influence different mechanisms like signalling pathways, cell growth, and differentiation [6, 29]. A growing body of data from the literature shows that PCOS individuals have increased oxidative stress expression. Consequently, the overproduction of oxidative compounds results in the presence of damaged lipids, proteins, and deoxyribonucleic acid (DNA). In addition, oxidative stress affects the production of pro-inflammatory cytokines and is involved in an intensification of the inflammatory pathways. Finally, an imbalance between pro-oxidants and antioxidants causes inhibition of the insulin signalling pathway that intensifies insulin resistance. It should also be emphasized that oxidative stress possesses a role in increasing mature adipocyte size and stimulating pre-adipocyte proliferation [6, 29]. Consequently, this process contributes to an increase in adiposity.

It is widely accepted that low-grade chronic inflammation accompanies obesity. Inflammatory markers like pro-inflammatory cytokines, higher levels of white blood cells, or c-reactive protein are also commonly found in the peripheral blood of PCOS patients [6]. Chronic inflammation contributes to the deregulation of ovary function. In detail, pro-inflammatory compounds may interfere with the insulin signalling pathways worsening insulin resistance. Moreover, inflammation may stimulate androgen production [29]. Pro-inflammatory cytokines are believed to prompt theca cell proliferation, induce ovarian fibrosis, and disturb follicular formation. Furthermore, they might also reduce the activity of FSH and LH receptors in ovaries [6]. Thus, inhibition of follicular development and ovulation may occur in the course of obesity-related inflammation.

In addition to the issues described above, women with PCOS may present different obesity-related comorbidities including diabetes, cardiovascular diseases (e.g. hypertension, dyslipidaemia, atherosclerosis, venous thromboembolism), non-alcoholic fatty liver disease, obstructive sleep apnoea, susceptibility to mental health problems, and others [30–32]. Hence, it is of high importance to maintain normal weight with adequate amounts of visceral adipose tissue in patients with PCOS.

Therapeutic approaches to obesity in the course of PCOS

Very recently a systematic review and meta-analysis on the effectiveness of lifestyle modification in PCOS patients with obesity has been published. The results indicate a beneficial role of healthy lifestyle modifications in the course of this particular endocrinopathy. The authors indicated a significant improvement in reproductive function in the studied group compared to the controls. Moreover, a combination of diet and exercise had better effects on metabolic and androgenic parameters than monotherapy. In addition, moderate weight loss by a minimum of 5% of primary body weight was effective in improving fasting insulin levels that served as the metabolic index. The BMI and body weight of the intervention group were markedly lower than those of the controls [33]. Thus, it is important to recommend dietary interventions to reduce caloric intake as well as regular physical activity as the primary treatment for weight management in obese women with PCOS.

Besides necessary lifestyle interventions, pharmacological therapies should also be considered in this group of patients, to interrupt the vicious circle of obesity-insulin resistance-hyperandrogenaemia. Therefore, possible treatment methods will be discussed below.

Metformin is a biguanide that is an insulin-sensitizing agent. It is commonly used in the treatment of PCOS, although in PCOS metformin is prescribed off-label. Nevertheless, metformin is effective in restoring insulin sensitivity in the peripheral tissues ameliorating glucose homeostasis, thus preventing T2DM, and improving lipid profiles in women with PCOS. Studies also indicated a significant decrease in BMI independent of lifestyle modification, reduction in androgen levels, and effectiveness in anovulatory infertility [34–36].

Another insulin-sensitizing agent that could potentially be used to treat PCOS is a group of thiazolidinediones — peroxisome proliferator-activated receptor-gamma (PPAR-g) agonists. Amongst patients with this syndrome thiazolidinediones positively affect insulin resistance, hyperandrogenaemia, and ovulatory dysfunction. In addition, it has been reported that treatment with pioglitazone, one of the thiazolidinediones, resulted in a marked decrease in fasting serum insulin and free androgen index. Moreover, SHBG levels were increased [36]. Even though a meta-analysis comparing the effect of metformin and pioglitazone revealed a significant improvement in ovulation and menstrual cycle in the pioglitazone group, there was an increase in BMI in this group, probably due to fluid retention [37]. These results are in concordance with findings of a recently published meta-analysis by Abdall et al. These authors confirmed a weight gain in the PCOS patients treated with thiazolidinediones in comparison to the placebo or metformin group [38]. Therefore, thiazolidinediones should not be used as a treatment method in obese individuals with PCOS.

Dipeptidyl peptidase-4 (DPP-4) inhibitors, named gliptins, are oral antihyperglycaemic compounds registered for the treatment of T2DM. However, animal studies indicate potential benefits of gliptins in PCOS. In the rat model of PCOS, sitagliptin reduced fasting blood glucose, lowered androgen levels, and improved ovarian fibrosis [39]. However, further studies are needed to evaluate the possibilities of treatment with the use of gliptins in PCOS women.

Sodium-glucose co-transporter-2 inhibitors (SGLT-2) are also used in the management of T2DM. Based on data from studies in patients with T2DM indicating a reduction in fat mass, preliminary studies involving PCOS patients were performed [36]. In a 12-week study of empagliflozin versus metformin in obese women with PCOS, it was found that the empagliflozin group showed significant amelioration of anthropometric parameters and body composition, but unfortunately without changes in metabolic parameters [40].

Orlistat is a lipase inhibitor used for obesity treatment. It decreases the hydrolysis of triglyceride and reduces the absorption of dietary fat. Several studies indicated that, in comparison to metformin, treatment with orlistat in PCOS women resulted in a significant weight reduction, BMI, and waist circumference as well as in amelioration of lipid profiles, insulin resistance parameters, and androgen levels [36]. However, orlistat has significant side effects and is often poorly tolerated.

The combination of bupropion and naltrexone is an effective weight-loss medication. However, there are no data concerning the effects of this treatment on PCOS. Interestingly, in PCOS, monotherapy with naltrexone for 6 months resulted in a marked decrease in BMI, fasting serum insulin, LH levels, LH/FSH ratio, and testosterone concentration [41]. Therefore, it seems reasonable to introduce research on the efficacy of bupropion/naltrexone treatment of obesity in the PCOS cohort.

Glucagon-like peptide-1 analogues were initially introduced as a treatment for T2DM. To date, this group includes liraglutide, semaglutide, dulaglutide, lixisenatide, and exenatide. It has been observed that GLP-1 analogues significantly reduce body weight, which was also related to the improvement of insulin sensitivity [36]. The mechanisms in which these agents contribute to enhanced insulin sensitivity and amelioration of glucose homeostasis are complex. It has been suggested that GLP-1 agonists may modulate the molecular pathways, including those related to inflammation, oxidative stress, lipid metabolism, beta-cells function, endoplasmic reticulum, and insulin activity [42]. As has already been mentioned, GLP-1 agonists are potent weight-reducing agents. They elicit multidirectional actions leading to a decrease in body weight and include, among others, reduced appetite and delayed gastric emptying [43]. However, only two GLP-1 analogues — liraglutide 3.0 mg daily and semaglutide 2.4 mg daily — fulfilled the requirements of the Food and Drug Administration (FDA) and the European Medicines Agency (EMA) of a minimum 5% of bodyweight reduction during one year of therapy and are approved for obesity treatment. In Poland, these 2 medications also have such registration, but at the moment of writing this article liraglutide 3.0 is the only available drug. Nevertheless, growing data from the literature indicate a positive role of GLP-1, mainly liraglutide, in the treatment of obesity and obesity-related complication in patients with PCOS [36].

Bariatric surgery is an invasive method of treatment for morbidly obese patients. It may be an option for women with PCOS and severe obesity (BMI ≥ 40 kg/m2 or ≥ 35 kg/m2 with a high-risk obesity complication) if standard weight-loss strategies have failed [44]. As a result of the surgery, the following achievements are expected: weight loss, improvement in metabolic abnormalities, and amelioration of insulin resistance. In the literature, there is a small group of studies referring to obese PCOS subjects who underwent bariatric procedures. However, very recently published data indicate improvement in the parameters of insulin resistance, hirsutism, and androgen profile as well recovery of menstrual cycles [44]. Nonetheless, bariatric surgery is associated with a high risk of complications both perioperative and postoperative, but it still could be a longer-term therapeutic option in selected cases.

GLP-1 analogues in the treatment of obese women with PCOS

Glucagon-like peptide 1 is mostly produced and secreted by intestinal enteroendocrine cells. It was first isolated in 1986 from the gut [45]. The first studies indicated that GLP-1 stimulated insulin production and inhibited glucagon secretion. Subsequent data showed an inhibitory effect on appetite and food intake. The GLP-1 receptors were found on pancreatic beta cells, in the regions of the hypothalamus involved in the regulation of appetite, satiety, food/energy intake, and energy expenditure, as well as in adipocytes and the reproductive system [46, 47].

Interestingly, GLP-1 seems to modulate the activity of hypothalamic GnRH neurons. Moreover, the influence of GLP-1 at the level of the pituitary is indirect as it is predominantly mediates through hypothalamic stimulation of GnRH release [48].

The primary indication for GLP-1 analogues has been type 2 diabetes. Currently, selected GLP-1 analogues find registered indications for use in the treatment of obesity. There is also an increasing amount of research on the potential benefits of GLP-1 analogues in overweight/obese patients with PCOS. However, there may be a problem with the research methodology due to the heterogeneity of the study groups, especially when the results are compared. Nonetheless, there have been enough studies and meta-analyses published over the last decade to indicate the possible pros and cons of GLP-1 analogue use in overweight/obese individuals suffering from PCOS.

One of the first meta-analyses of the role of GLP-1 analogues in PCOS was published in 2016 by Niafar et al. The authors included 7 studies in which obese PCOS subjects were treated with liraglutide for a limited period of 3 months. This short time intervention resulted only in a drop in BMI and serum testosterone. Other variables including waist circumference, fasting insulin levels, insulin resistance index (HOMA-IR), and SHBG did not change significantly [49]. Of note, the first studies included in this meta-analysis had limitations in terms of small group sizes, a relatively low dose of GLP-1 (mostly 1.2 mg of liraglutide once daily), and a short intervention time.

A comparison between GLP-1 (liraglutide or exenatide) treatment and the use of metformin in a cohort of PCOS patients was presented in the study by Han et al. This meta-analysis indicated that GLP-1 analogues had the same therapeutic effect as metformin in decreasing testosterone or free androgen index. Moreover, there were no marked differences between the studied group when the following parameters were compared: improvement of menstrual frequency, SHBG, androstenedione, LH, lipid profile, fasting blood glucose, fasting insulin, blood pressure, and Ferriman-Gallwey scores. By contrast, the use of liraglutide or exenatide had an advantage over metformin in terms of improving insulin resistance, reducing BMI, and decreasing abdominal circumferences [50].

The question had been raised whether the combination of GLP-1 analogue and metformin would result in intensifying beneficial results in overweight/obese women with PCOS. The answer came from a recently published meta-analysis by Lyu et al. and separately from the Ma et al.

Both groups compared the effects of GLP-1 (exenatide or liraglutide) alone versus metformin, and the combination therapy of GLP-1 with metformin versus metformin. Data from the study by Lyu et al. confirmed that the anti-obesity effect of GLP-1 analogue alone or combined with metformin was superior to metformin alone in terms of reduction of weight, waist circumference, and BMI [51]. In turn, Ma et al. presented extended results. In concordance with the findings of Liu et al. GLP-1 analogue exhibited a greater efficiency in reduction of body mass and decrease of abdominal girth than metformin. Moreover, treatment with GLP-1 analogue resulted in improvement of the HOMA-IR parameter. Furthermore, the comparison between combination therapy and GLP-1 alone failed to find any significant differences in menstrual frequency, total testosterone, BMI, and HOMA-IR [52].

Noticeably, infertility amongst women with PCOS is an essential problem. Results from the animal PCOS model suggest that treatment with GLP-1 analogues may have a positive impact on infertility treatment. Indeed, in the rat model, administration of exenatide resulted in an increased number of granulosa cell layers, a thinner theca cell layer, more corpora lutea, and decreased dilated follicles, suggesting that GLP-1 was able to reverse the polycystic ovary morphology [53]. In addition, it has been speculated that PCOS is linked to endometrial abnormalities related to steroid hormone action, insulin resistance, and inflammation that may result in implantation failure and defective implantation. Most markers of endometrial dysfunction in PCOS are linked to steroid hormone action, glucose metabolism, insulin resistance, endometrial receptivity/decidualization, and inflammation [54]. Data obtained from the diabetes animal model can be at least partially transferred to the studies on PCOS because those 2 diseases share common features. In the study by Artunc-Ulkumen et al., the use of exenatide in diabetic rats resulted in an improvement of histological degeneration and stromal fibrosis in the ovaries, a decrease in inflammatory and oxidative stress markers, and an increase in serum AMH level, as well as a decrease in histological degeneration and fibrosis in the endometrium as an effect of reduced inflammation and oxidative stress [55]. In addition, a clinical study with the use of liraglutide for 26 weeks in obese human PCOS subjects revealed ovarian volume reduction versus placebo [56].

The impact of GLP-1 treatment on pregnancy rates has also been evaluated. Salamun et al., in their pilot study involving 28 women with obesity and PCOS, assessed whether the addition of low-dose liraglutide (1.2 mg daily) to metformin would increase the pregnancy rate. They found that a combination of liraglutide and metformin in the preconception period was superior to metformin alone, despite comparable weight reduction in both groups, in increasing in vitro fertilization pregnancy rates (85.7% vs. 28.6% per embryo transfer, respectively), as well as cumulative pregnancy rate for 12 months including spontaneous pregnancies after treatment in women who had previously been resistant to lifestyle modification and first-line reproductive treatment (69% vs. 36%, respectively) [48]. The results concerning the natural pregnancy rate are in concordance with the observation from a study by Liu et al, who investigated 176 overweight or obese women with PCOS who were randomly assigned to the exenatide or metformin group. The treatment was performed for either exenatide 10 μg twice daily or metformin 1000 mg twice daily for the first 12 weeks. Then all patients were given metformin alone for another 12 weeks. The authors observed that during the second 12 weeks the rate of natural pregnancy of exenatide-treated patients was markedly higher than that of metformin-treated patients (43.60% vs. 18.70%) [57]. Thus, therapy with GLP-1 analogues may be useful in assisted reproductive techniques, especially in patients who are poor responders to the first-line procedures. In addition, women with PCOS who are overweight/obese have higher risks during controlled ovarian stimulation and throughout pregnancy, which can be minimized by a pre-treatment resulting in reduced body weight [24].

The above data unequivocally prove the validity of the use of GLP-1 analogues in the treatment of overweight/obese patients with PCOS. However, it should be emphasized that GLP-1 analogues, despite their effects on insulin resistance and hyperandrogenism, should not be used as therapeutic agents for PCOS not coexisting with adiposity. Furthermore, based on the current registrations and evidence-based data, the only recommended GLP-1 analogue for obese PCOS subjects is liraglutide. Although semaglutide is registered for obesity treatment, no clinical studies confirm its benefits in the therapy of PCOS. Finally, exenatide exerting similar properties to liraglutide in obesity concomitant with PCOS is registered to treat diabetes mellitus only.

Conclusions

Undoubtedly, obesity is involved in exacerbating the negative effects of insulin resistance, hyperinsulinaemia, and hyperandrogenaemia in the course of PCOS. Therefore, it is essential to maintain normal weight or effectively treat overweight/obesity in patients suffering from the consequences of this syndrome. As well as diet and lifestyle interventions, an appropriate pharmacological or surgical therapy should be individually selected for patients. GLP-1 analogue treatment is a new therapeutic opportunity for weight loss and the achievement of a wide range of benefits for obese women with PCOS.

Acknowledgments

ABB is the only author of this review.

Financial support and conflict of interest

None declared.