Impact of androgen excess on metabolism

Hyperandrogenism is the main feature of PCOS and is also partly responsible for the development of metabolic disturbances. Androgens exert their effects at both central and peripheral levels on a variety of tissues . They have diverse effects on white adipose tissue. First of all, high androgen concentrations shift the pattern of fat distribution. Women with PCOS exhibit male pattern fat accumulation with increased visceral adiposity . This altered fat deposition pattern has detrimental metabolic consequences and is a major risk factor for the development of metabolic syndrome . The exact mechanism of the fat redistribution caused by hyperandrogenism remains largely unknown. Rodent studies have shown that exposure to excess androgens was associated with increased adipocyte size in both visceral and subcutaneous fat depots . When adipocytes enlarge, the microenvironment of adipose tissue changes, tissue hypoxia occurs, proinflammatory cytokines and free fatty acids are released, macrophage recruitment begins, and eventually IR emerges. In addition, IR and exceeding adipose tissue capacity trigger the release of free fatty acids and triglycerides into the circulation, causing ectopic fat accumulation in skeletal muscle, liver and pancreas ( Fig. 1 ). As a result, ectopic fat accumulation further aggravates IR in these peripheral tissues . Hyperandrogenism also affects adipocyte differentiation by suppressing adipogenesis via the Wnt signaling pathway . Another potential mechanism of adipogenesis suppression is related to a ligand-enhanced coactivator, activation of androgen receptor-associated protein 70 (ARA70). Peroxisome proliferator-activated receptor-γ (PPAR-γ) activates ARA70 and induces adipogenesis. When activated by androgens, the androgen receptor competes with PPAR-γ to bind ARA70, which leads to blunted PPAR-γ effects on ARA70 and decreased adipogenesis . This impairment further complicates the lipid storage capacity of adipose tissue and contributes to IR. Androgens also alter adipokine dynamics. Androgen excess leads to a decrease in circulating levels of insulin sensitizer adipokines, such as adiponectin . Apart from white adipose tissue, androgens can modify brown adipose tissue as well. Uncoupling protein-1 levels are decreased by the effect of androgen excess, resulting in lower brown adipose tissue activity in women with PCOS .

Pathogenesis of metabolic dysfunction in PCOS. Hyperandrogenism causes alterations in adipose tissue microenvironment. Exceeding adipose tissue storage capacity and tissue hypoxia trigger the release of FFAs, triglycerides and proinflammatory cytokines to the circulation, causing ectopic fat accumulation in skeletal muscle, liver and pancreas. Compensatory hyperinsulinemia stimulates LH secretion through upregulation of hypothalamic GnRH expression. Both LH and insulin stimulate ovarian theca cells to produce more androgen. Adrenal androgen production is also increased. Hyperinsulinemia inhibits sex hormone-binding globulin synthesis in liver, causing increased free androgen levels. FFA , free fatty acid; GnRH , gonadotropin releasing hormone; IR , insulin resistance; LH , luteinizing hormone.

Androgen excess also decreases insulin sensitivity in the skeletal muscle in women with PCOS . Androgens reduce insulin-mediated glucose transport in skeletal muscle by affecting the insulin-signaling pathway at different levels. Androgen excess decreases insulin receptor phosphorylation in skeletal muscle cells of women with PCOS, which leads to IR . Hyperandrogenemia may also lead to impaired insulin secretion. Pancreatic β cell failure has been demonstrated in female mice following exposure to androgens .

Androgen receptors are expressed in various parts of the central nervous system . Through their actions on the central nervous system, androgens can modify metabolic homeostasis . In an experimental mouse model, key metabolic and reproductive traits of PCOS including irregular menses, obesity, and dyslipidemia, did not develop after dihydrotestosterone treatment in neuron-specific androgen receptor knockout mice . These data suggest that the direct action of androgens on the brain is associated with metabolic and reproductive disturbances in PCOS.

Impact of altered metabolic status on androgen action

Although IR is more severe in overweight and obese individuals with PCOS, it is also frequently seen in lean PCOS subjects . Due to decreased sensitivity to insulin in specific tissues such as adipose tissue, muscle and liver, a higher amount of insulin is required, and compensatory hyperinsulinemia emerges. However, women with PCOS exhibit a specific form of IR called selective IR. Despite the existence of IR in adipose tissue, muscle and liver, hypothalamo-pituitary-ovarian (HPO) axis remains sensitive to insulin actions .

Insulin upregulates gonadotropin-releasing hormone (GnRH) expression in hypothalamus and enhances gonadotropin response to GnRH at pituitary level. Increased gonadotropin stimulates ovarian androgen secretion and contributes to ovarian dysfunction. Ovary is also affected by hyperinsulinemia. Insulin acts like a co-gonadotropin and induces ovarian steroidogenesis. Moreover, not only ovarian steroid production but also adrenal steroid production is increased due to excess insulin activity. Insulin indirectly provokes corticotropin-releasing hormone (CRH) secretion from the hypothalamus and potentiates the hypothalamo-pituitary-adrenal axis. Finally, insulin inhibits sex hormone-binding globulin (SHBG) synthesis and secretion in liver, and leads to increased free (bioavailable) androgen levels ( Fig. 1 ) .

Metabolic dysfunction and PCOS phenotype

The clinical presentation of PCOS is not uniform. Four different phenotypes of PCOS have been defined so far; phenotype A: Hyperandrogenism (HA) + ovulatory dysfunction (OD) + polycystic ovarian morphology (PCOM), phenotype B: HA + OD, phenotype C: HA + PCOM and phenotype D: OD + PCOM . Although HA seems to be a key factor in the pathophysiology of PCOS and the main predictor of the associated metabolic dysfunction, different pathophysiological processes may be involved in different PCOS phenotypes, especially in nonhyperandrogenic phenotype (phenotype D).

Identification of phenotype in each PCOS patient is important. Cardiometabolic risks appear to be particularly higher and more severe in PCOS phenotypes characterized with hyperandrogenism and anovulation. Metabolic disturbances, such as glucose intolerance may develop earlier than expected in these phenotypes, especially in the presence of accompanying obesity . Therefore, phenotypic characteristics of the patients should be considered while deciding which intervention to choose and when, in order to treat metabolic disturbances in PCOS.

Lifestyle modifications

Excess body fat worsens the underlying hormonal disturbances in PCOS. Therefore, implementing a healthy lifestyle with regular physical activity and optimal weight control may improve disease-related metabolic, cardiovascular and reproductive outcomes. There is only one evidence-based recommendation regarding lifestyle intervention in recent international guidelines: “Lifestyle intervention (preferably multicomponent including diet, exercise and behavioral strategies) should be recommended in all those with PCOS and excess weight, for reductions in weight, central obesity and insulin resistance” . High-quality evidence about lifestyle modifications in the management of PCOS is currently lacking.

In a Cochrane systematic review, including 15 randomized controlled trials and 498 participants, the effectiveness of lifestyle interventions on metabolic and reproductive traits of PCOS were analyzed . The results suggested that lifestyle modifications (diet, exercise, behavioral or combined) may improve biochemical and/or clinical hyperandrogenism (total testosterone levels, SHBG levels, free androgen index, hirsutism scores), anthropometric measures (weight, waist circumference, and body mass index [BMI]) and metabolic parameters (total cholesterol, low-density lipoprotein [LDL] cholesterol, fasting insulin, and oral glucose tolerance test [OGTT] insulin) when compared to minimal or no treatment. In this analysis, there was no clear evidence regarding the impact of lifestyle modifications on glucose tolerance. None of the studies included in the analysis reported the effect of lifestyle interventions on live birth, miscarriage, pregnancy or menstrual regularity rates. Although the studies were heterogeneous, emotions and infertility domains of the quality of life scores improved with lifestyle modifications . To date, no study has compared the effectiveness of lifestyle modifications according to different PCOS phenotypes. Moreover, almost all participants included in the analyses were overweight or obese. Therefore, the data regarding the effect of lifestyle modifications in lean subjects with PCOS is lacking. In addition, studies are heterogeneous regarding lifestyle interventions in PCOS. Even though there is no standard lifestyle recommendation, it is logical to recommend a healthy lifestyle to all patients with PCOS with more emphasis on special subgroups considering metabolic and cardiovascular consequences of the syndrome ( Fig. 2 ) .

Proposed management algorithm for metabolic risk in PCOS. All women with PCOS should be encouraged for having a healthy lifestyle. COCs are used to improve hyperandrogenism if pregnancy is not desired. Metformin is recommended in overweight/obese patients with high metabolic risk. GLP-1 RAs could be used in obese patients as add-on to metformin to provide greater weight loss, or as a single agent in case of metformin-intolerance. Bariatric surgery is for selected patients who were unable to lose weight with prior interventions. Ovulation could return with insulin-sensitizing interventions; patients should be informed about pregnancy categories of medications. Women undergoing bariatric surgery need to postpone conception for at least a year postoperatively. BMI , body mass index; COC , combined oral contraceptive; GLP-1 RA , glucagon like peptid-1 receptor agonist.

Exercise should be offered to women with PCOS in addition to dietary recommendations. International PCOS Network recommends age-appropriate exercise interventions for all women with PCOS ( Fig. 2 ) . However, the evidence regarding the effect of exercise on health-related outcomes in women with PCOS is limited as well. A recent metaanalysis including 10 randomized controlled studies including 533 subjects evaluated the data on the effect of different types of physical activity on reproductive and metabolic traits of PCOS . Studies have tested three different types of physical activity interventions (aerobic, resistance or combined). The result of this analysis indicated that exercise had little or no effect on reproductive outcomes whereas aerobic exercise reduced BMI . However, most of the included studies had small sample sizes (average n : 32, ranging 15–124) and relatively short follow-up periods (8–32 weeks). The outcome measures were also heterogeneous. Like other lifestyle modifications, the effect of various types of exercise on PCOS subjects with different phenotypes or on lean subjects has not been studied yet.

Finally, smoking cessation should be recommended to all women with PCOS. Apart from other known adverse effects for health, smoking increases the amount of bioavailable testosterone and exacerbates IR in women with PCOS .

Metformin

Metformin is a biguanide derivative drug of plant origin which has been in use since 1957 in Europe and since 1995 in the United States. It has been the first-line medication in treatment guidelines of T2D for a long time. Metformin acts mainly by reducing endogenous hepatic glucose production, mostly through the inhibition of hepatic gluconeogenesis. It also inhibits lipogenesis in liver, skeletal muscle and visceral adipose tissue. Metformin enhances insulin-mediated glucose disposal in skeletal muscle, and also increases peripheral glucose utilization by liver, adipose tissue, the intestine, and ovaries. Moreover, metformin suppresses obesity-related low-grade chronic inflammation associated with endothelial dysfunction, which in the long term is linked to poor cardiovascular outcomes. The ability of metformin to improve insulin sensitivity and inflammation has made it a considerable treatment option in women with PCOS. Metformin improves hyperandrogenemia and menstrual cyclicity in PCOS by reducing IR, increasing the synthesis of SHBG in the liver, and also by directly inhibiting androgen production in the ovarian theca cells .

Following lifestyle modifications, current guidelines recommend metformin treatment in overweight/obese women with PCOS in order to achieve desired goals of weight reduction and metabolic improvement, either alone or in combination with combined oral contraceptives (COCs), since the favorable metabolic effects of metformin are rather evident in patients with a BMI ≥ 25 kg/m ² ( Fig. 2 ) . Adding metformin to lifestyle modifications could specially be considered in patients with high metabolic risk (carrying the conventional risk factors for T2D or having impaired glucose tolerance, and belonging to certain ethnic groups such as Caucasians of North America and Australia, Africans, South East Asians, and Indigenous Australians), . When compared to placebo, metformin combined with lifestyle modifications is associated with a significant decrease in BMI, homeostasis model assessment of IR (HOMA-IR) score, total cholesterol, LDL-cholesterol, and triglycerides, as well as serum testosterone levels in overweight women with PCOS . The quality of evidence regarding the decrease in blood pressure with metformin treatment is low, and the data on alterations of serum lipids is inconsistent.

Both lifestyle modifications and metformin significantly reduce BMI in patients with PCOS. Metaanalyses have been controversial regarding whether the addition of metformin to lifestyle modifications leads to greater decrease in BMI or not . This may probably be due to varying compliance of study participants with lifestyle modifications and/or metformin therapy.

Metformin is the recommended first-line medical treatment for the management of IR in PCOS. Recent data indicates that the combination of lifestyle modifications and metformin treatment is the optimal intervention to improve IR . Metformin improves IR and BMI when used as either monotherapy or add-on therapy to COCs and/or antiandrogens. Additionally, reductions in serum total cholesterol, LDL cholesterol, but not triglyceride levels have been reported in patients treated with the combination of metformin and COCs. Metformin is the only insulin-sensitizing agent that has been evaluated in adolescents with PCOS by randomized controlled trials which revealed favorable metabolic effects similar to adult PCOS. Guidelines recommend metformin treatment in insulin-resistant adolescent PCOS patients with a BMI ≥ 25 kg/m ² , both as monotherapy and as add-on therapy to COC .

Another indication of metformin in PCOS is to induce ovulation. In overweight/obese women with PCOS, combination of metformin with clomiphene citrate has been superior to clomiphene citrate monotherapy for ovulation induction. The use of metformin for this indication has caused a growing number of women to be exposed to metformin during the early weeks of pregnancy. To date, no increased risk of miscarriage due to metformin has been defined if stopped just after the detection of pregnancy, but there is not enough evidence to justify the use of metformin throughout pregnancy .

The most common side effects of metformin are related to the gastrointestinal system, such as nausea, diarrhea, flatulence, abdominal pain and swelling. Gastrointestinal intolerance could occur in up to 1 in 5 patients. In the general population, discontinuation of metformin due to gastrointestinal side effects is reported to be 1%–5%. Gradual dose uptitration or extended/delayed-release formulations could be used to diminish the gastrointestinal problems . Headache has also been reported as a side effect. Despite being an antihyperglycemic agent, metformin does not cause clinical hypoglycemia. Lactic acidosis is a concern in a group of patients with poor general condition in whom this agent is already contraindicated .

Metformin has been the most commonly used medicine in overweight/obese, insulin-resistant, glucose-intolerant and/or dyslipidemic women with PCOS, and despite being included in the PCOS treatment guidelines, metformin use in PCOS is still off-label. Good quality of evidence is needed to identify the optimal dose and duration of treatment. Notably, studies investigating the modification of cardiometabolic risk factors in the long term are lacking. Nevertheless, with its good safety profile and multipathway beneficial effects, metformin merits to be the first drug of choice in PCOS if the metabolic risk factors are present .

Thiazolidinediones

Thiazolidinediones (TZDs) are PPAR-γ agonists. By altering the postreceptor actions of insulin, TZDs make the liver, fat and muscle cells become more receptive to insulin actions. While improving glucose utilization and clearance, TZDs also inhibit ectopic fat accumulation . Troglitazone was the first TZD in use but was withdrawn in 2000 due to inducing irreversible hepatotoxicity. Rosiglitazone and pioglitazone remained to be in use, although rosiglitazone has been restricted by Food and Drug Administration in 2010 due to cardiovascular safety issues. Both have shown beneficial effects on cardiovascular risk factors such as dyslipidemia, hypertension and endothelial dysfunction .

A 2017 Cochrane Review revealed that both rosiglitazone and pioglitazone slightly increased BMI, but decreased fasting insulin levels. In addition, both drugs increased menstrual frequency and ovulation while only pioglitazone improved hyperandrogenism in PCOS patients compared to placebo or no treatment . In a metaanalysis including overweight women with PCOS, while TZDs have led to gain weight, they have improved IR, hyperandrogenemia and menstrual regularity, in a noninferior manner compared to either monotherapies with metformin or glucagon-like peptide-1 receptor agonists (GLP-1 RAs) . Although the results of the metaanalyses comparing the efficacy of metformin and TZDs regarding hyperandrogenism have been inconsistent , combined treatment of TZD with metformin might be effective in order to restore menstrual frequency in PCOS patients with severe IR . Pioglitazone use in patients with nonalcoholic fatty liver disease (NAFLD) reduces inflammation and steatosis, and possibly fibrosis , yet it is not considered as first-line pharmacotherapy of NAFLD in PCOS .

Few studies including adolescent nonobese PCOS patients suggested that the combination of low dose spironolactone, metformin and pioglitazone has improved IR and ovulation rates, and reduced visceral fat accumulation . However, limited evidence from these small studies do not yet justify routine use of TZD in adolescent girls with PCOS.

Besides weight gain, peripheral edema, mastopathy, muscle cramps and insomnia are possible adverse effects of TZD treatment. Notably, since TZDs are drugs of pregnancy category C, contraception should be assured when using these medications. The increase in BMI and lack of data upon safety in the long-term being the main concerns, current treatment guidelines do not recommend the use of TZDs in PCOS .

GLP-1 receptor agonists

Glucagon-like peptide-1 (GLP-1) is an incretin hormone secreted from enteroendocrine cells. Following food intake, GLP-1 is secreted from the small gut in order to stimulate pancreatic beta cells to secrete insulin while suppressing alpha cells to inhibit glucagon secretion. GLP-1 RAs are used to maintain both glycemic control and weight loss. They display an incretin-like action in order to lower blood glucose by inducing glucose-dependent insulin secretion while increasing glucose uptake in target tissue and organs, leading to an improvement in IR. By stimulating multiple GLP-1 Receptor (GLP-1R) expressing regions in the central nervous system, GLP-1 RAs create a feeling of satiety, reduce food intake and slow gastric emptying, and thus provide weight loss .

Given that disordered eating is common in PCOS, the appetite suppressing effect of GLP-1 RAs could be of significant benefit. Endogenous GLP-1 response to OGTT has been demonstrated to be lower in lean hyperandrogenic PCOS women with normal glucose tolerance when compared to healthy controls . In addition, the findings that demonstrate the expression of GLP-1Rs in ovaries and that GLP-1 stimulates hypothalamo-pituitary-ovarian (HPO) axis have led the researchers to investigate whether GLP-1 could be a treatment target in PCOS .

Studies in overweight and obese women with PCOS have reported that GLP-1 RAs exenatide and liraglutide develop favorable metabolic effects including weight loss, either as monotherapy or in combination with metformin. Therefore, overweight/obese women with PCOS who have not achieved adequate weight loss with lifestyle modifications plus metformin may be candidates for GLP-1 RAs ( Fig. 2 ). Exenatide has been found to be more effective if given as add-on therapy to metformin in PCOS regarding weight loss, improving hyperandrogenism and menstrual frequency compared to exenatide monotherapy in a duration of 24 weeks . An open-label prospective randomized trial revealed that women with PCOS treated with exenatide for 12 weeks followed by metformin for another 12 weeks achieved higher reductions in body weight, waist circumference, and total fat mass than the metformin-only group. Almost half of the exenatide-treated patients reached the goal of losing 5% of total body weight while none in metformin group did. Fasting plasma insulin and HOMA-IR scores were lower in patients in exenatide group, whereas fasting glucose levels were similar. The improvement in markers of hyperandrogenism was also similar in both groups, yet the rate of women being able to have regular menses and become pregnant was higher in the exenatide group . Studies with short treatment durations of 12–26 weeks in women with PCOS revealed that liraglutide in daily doses of 1.2 or 1.8 mg was superior to metformin regarding reductions in body weight and waist circumference, some of which reported even better outcomes with the combination of metformin and liraglutide . However, greater reduction in body weight occurred with liraglutide monotherapy 3 mg daily in obese PCOS women than combination of liraglutide 1.2 mg plus metformin in a duration of 12 weeks . Liraglutide monotherapy provided 5% loss in body weight in 16%, 42%, and 57% of obese women with PCOS, with respective daily doses of 1.2, 1.8, and 3 mg . In addition to a significant reduction in weight, some studies reported further favorable effects on other metabolic variables such as a decrease in visceral adipose tissue area and liver fat content, as well as serum markers of hyperandrogenism . Despite comparable reductions in body weight, visceral adipose tissue area and serum androgen levels, a study reported higher IVF success rates with the combination of metformin and liraglutide versus metformin monotherapy after 12 weeks of treatment in obese PCOS patients . It was suggested that this result might be related to the effect of GLP-1 on gonadotropin-releasing hormone secretion by activating the HPO axis .

A metaanalysis of eight randomized studies comparing the effectiveness of GLP-1 RAs and metformin in obese women with PCOS revealed greater reductions in body weight and waist circumference with the use of GLP-1 RAs, but similar declines in serum androgens and lipids . On the other hand, a metaanalysis of 14 studies including overweight adult women with PCOS showed no superiority of GLP-1 RAs to metformin regarding the reduction in BMI, neither as monotherapy or in combination with metformin. However, women treated with the combination of GLP-1 RAs and metformin achieved greater reductions in waist circumference, fasting plasma glucose and androstenedione levels, and higher levels of SHBG than either with metformin or GLP-1 RA alone . There is no sufficient data to determine if the improvement in hyperandrogenism is due to the direct effect of GLP-1 RAs on GLP-1 receptors expressed in the ovaries or due to weight loss and/or improved IR .

The most common side effects of GLP-1 RAs are gastrointestinal disturbances such as nausea, dyspepsia, vomiting, and diarrhea. The frequency of gastrointestinal adverse events with the use of metformin and GLP-1 RAs was reported to be similar. The severity of gastrointestinal adverse events may increase with combination, although drop-out rates in studies was not higher with combination therapy. Other possible side effects are rash at the injection site, constipation, and insomnia .

Long-term high-quality studies are needed to evaluate the efficacy and safety of GLP-1 RAs in overweight/obese PCOS women with and without glycemic abnormalities. While our knowledge about the use of GLP-1 RAs in the general diabetic and/or obese population is widening and the findings of the short-term studies in adult women with PCOS so far are promising, the data on adolescent PCOS is inadequate. Also, it should be kept in mind that there is no sufficient data about pregnancy outcomes in women exposed to GLP-1 RAs before gestation. Contraception should be assured while using these pregnancy category C medications. Nevertheless, GLP-1 RAs may be an alternative or add-on treatment option to metformin in obese women with PCOS who were not able to lose weight despite lifestyle changes. By creating a feeling of satiety and slowing gastric emptying, these medications could be especially beneficial in PCOS patients with eating disorders. Current guidelines suggest the use of these agents for the management of obesity in PCOS as per general population recommendations.

Dipeptidyl peptidase-4 inhibitors

Dipeptidyl peptidase-4 (DPP-4), formerly known as CD26, is the enzyme involved in the cleavage of GLP-1. DPP-4 inhibitors are a class of antidiabetic drugs which controls postprandial hyperglycemia in a glucose-dependent fashion by decreasing the degradation of GLP-1, and favored in a group of diabetic patients due to low hypoglycemia risk and for being weight-neutral .

Studies with a small sample size reported some beneficial metabolic effects with DPP-4 inhibitors in women with PCOS. The use of sitagliptin along with lifestyle modifications for 12 weeks prevented metformin-intolerant obese women with PCOS from becoming glucose intolerant or diabetic . Also, sitagliptin prevented weight regain after 12 weeks of 3 mg daily liraglutide treatment, whereas metformin monotherapy failed to do so . Alogliptin monotherapy or its combination with pioglitazone improved IR better than metformin alone in PCOS patients with high metabolic risk . Saxagliptin as monotherapy or as add-on therapy to metformin for 16 weeks provided normal glucose tolerance in a higher percent of formerly glucose-intolerant patients compared to metformin monotherapy. Moreover, patients treated with saxagliptin plus metformin had lesser BMI and waist circumference measurements, better insulin sensitivity, lower serum triglyceride and mean blood glucose levels after treatment . Another study on newly diagnosed diabetic subjects with PCOS revealed comparable control of hyperglycemia with metformin, saxagliptin and combination of the two . To date, no study reported a decrease in serum androgen levels with DPP-4 inhibitor monotherapy, yet the combination of alogliptin with metformin and/or pioglitazone and saxagliptin with metformin improved hyperandrogenemia .

Besides GLP-1, DPP-4 is involved in the degradation of various other endogenous peptides, such as growth hormone-releasing hormone (GHRH). Since patients with PCOS and abdominal obesity have relatively low GH levels, it was hypothesized that DPP-4 inhibitors might restore GH levels by inhibiting the degradation of endogenous GHRH. In overweight nondiabetic women with PCOS, daily use of sitagliptin 100 mg decreased the maximal glucose response to OGTT and visceral adipose tissue mass but not overall body fat percentage, while increased GH half-life. GH and free IGF-1 levels were not affected by sitagliptin treatment .

DPP-4 inhibitors have a relatively good safety profile, yet a recent metaanalysis revealed a slight but insignificant increased risk of acute pancreatitis . Notably, DPP-4 is expressed on inflammatory cells and involved in the degradation of several cytokines and chemokines. Inhibition of DPP-4 might be altering inflammatory pathways, and that needs to be clarified by further studies. Studies to date investigating DPP-4 inhibitor use in PCOS are short term with small sample sizes, and these drugs are not yet recommended in current international guidelines. It remains to be determined whether DPP-4 inhibitors might be an alternative in overweight/obese PCOS who are intolerant to metformin and/or GLP-1 RAs due to gastrointestinal side effects.

Combined oral contraceptives/antiandrogens

Hyperandrogenism related symptoms, such as hirsutism, acne, hair loss, and menstrual irregularity, are usually the main concerns in most women with PCOS. Consequently, COCs as monotherapy or in combination with antiandrogens are the first drugs of choice in patients who do not desire pregnancy . The antiandrogenic effects of COCs occur through suppression of luteinizing hormone (LH) secretion by progestin component, and inhibition of ovarian androgen production. Also, estradiol component reduces serum-free androgen concentrations by increasing SHBG levels. Some progestins prevent androgens to bind their receptors or inhibit 5-alpha reductase enzyme activity. COCs also cause a slight decrease in adrenal androgen secretion .

Hyperandrogenism contributes to IR by previously mentioned mechanisms, and improving hyperandrogenism could provide beneficial metabolic effects. Along with lifestyle modifications, treatment of PCOS patients with COCs in general do not have an adverse effect on body weight or IR while resulting in temporary increases in high-density lipoprotein (HDL)-cholesterol and triglycerides . Currently, there is not enough evidence to favor one combination over another, or to conclude if antiandrogens have additional advantageous effects on metabolic profile . Antiandrogens could be used alone in adolescent patients with PCOS, but given the teratogenicity, they should always be combined with contraceptives in childbearing aged women .

Statins

Dyslipidemia is very common in women with PCOS, mostly characterized with hypertriglyceridemia and an atherogenic lipid profile with high LDL and low HDL cholesterol levels . Statins selectively and reversibly inhibit HMG CoA reductase, the rate-limiting enzyme in cholesterol synthesis, and reduce serum total cholesterol and LDL cholesterol levels, and decrease oxidative stress. Statins are widely used in order to modify cardiovascular risk factors. In women with PCOS, a metaanalysis reported that statins improve serum total and LDL cholesterol and triglyceride levels, as well as total testosterone, but not HDL cholesterol. Also, the combination of statins and a COC has shown a greater reduction in total testosterone than COC monotherapy . For the fact that statins are in pregnancy category X, contraception should be assured in fertile women . Nevertheless, based on limited available evidence, treatment guidelines for PCOS do not advocate routine statin use at this time .

Sodium-glucose cotransporter-2 inhibitors

Sodium-glucose cotransporter-2 (SGLT-2) inhibitors improve hyperglycemia by increasing urinary excretion of glucose in an insulin-independent manner. They have been included in diabetes treatment guidelines in 2019. Helping to lose weight, not causing hypoglycemia, and the cardiovascular benefits of SGLT-2 inhibitors have brought them to the upper steps of T2D treatment algorithms, although they carry the risk of certain adverse effects which have never been a major concern before, such as euglycemic diabetic ketoacidosis .

To date, there have been two small short-term studies in overweight/obese women with PCOS comparing the cardiometabolic outcomes of empagliflozin with metformin. First study reported an insignificant reduction in weight (− 1.4 kg) while no impact on hyperinsulinemia, dyslipidemia, blood pressure, and hyperandrogenism with 25 mg daily empagliflozin . The second study hypothesized that empagliflozin may decrease the levels of endothelial microparticles in the plasma samples of overweight/obese PCOS patients, such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 that are biomarkers of endothelial dysfunction, but found the opposite . Given the lack of data, for now, SGLT-2 inhibitors could only be used in PCOS for investigational purposes.

Bariatric surgery

In general population, bariatric surgery (BS) is considered in patients with a BMI above 40 kg/m ² or a BMI above 35 kg/m ² in the presence of obesity-related comorbidities such as severe cardiopulmonary problems or diabetes, who were unable to achieve sufficient weight loss despite prior interventions . In addition to being an effective method for weight loss, BS has become a treatment alternative for obesity-related comorbidities, especially T2D. According to recent evidence, BS can also help obese adolescents lose weight effectively . Although it is controversial if PCOS could be counted as one of the comorbidities to be included among the indications for BS in obese adolescents and adult women, current clinical practice guidelines for BS recommend to test serum androgens preoperatively if there is a suspicion for PCOS .

In the United States, retrospective analysis of female patients undergoing BS revealed that 13% of these women already had PCOS , which is similar to the background prevalence of PCOS in obese American women . The reported prevalence of PCOS among women undergoing BS in the United Kingdom was 15%, and subfertility was the main reason for surgery in 7.4% . Particularly, PCOS and infertility were more common in women who were obese at the age of 18 compared to those became obese at a later age .

Metaanalyses agree that BS is superior to medical therapy in general population in terms of providing weight loss, remission of T2D and metabolic syndrome, and improvement of dyslipidemia . Rates of weight loss 20 years after surgery appear to be stable, although the number of patients remaining on follow-up decreases significantly over time . High remission rates for T2D and hypertension were reported in a metaanalysis, including 234 young obese women with PCOS undergoing BS, mainly Roux-Y gastric bypass (RYBG) . Notably, BS has also improved PCOS phenotype .

The amount and durability of the beneficial effects of BS differ between surgical techniques. Despite the fact that they can cause relatively more serious adverse events, malabsorptive procedures like RYGB and biliopancreatic diversion (BPD) seem to have better metabolic outcomes than restrictive procedures such as sleeve gastrectomy and gastric banding . This is mainly because that the metabolic effects of BS are not solely derived from weight loss, but related to the impact on intestinal neuroendocrine hormones such as GLP-1, ghrelin and glucose-dependent insulin-trophic peptide (GIP).

Eating disorders such as binge eating, which are common in women with PCOS as well as in the general population undergoing BS may not resolve after surgery . Preoperative psychiatric evaluation of depression or anxiety that frequently accompany PCOS has grave importance, and patients should be assessed postoperatively in order to detect the possible “transfer of food reward to addictive substances” .

BS also has efficacy in improving NAFLD and obstructive sleep apnea (OSA) , which may be associated with PCOS . BS provides histological improvement in patients with NAFLD and nonalcoholic steatohepatitis (NASH), which is only partly sustainable for long durations . Data is growing but yet insufficient to conclude whether one surgical procedure is superior to another in this regard . In obese patients with OSA, current evidence suggests that BS improves apnea-hypopnea index especially in patients with higher preoperative apnea-hypopnea indexes; although further studies are needed to obtain long-term high-quality data permitting comparison between surgical procedures .

Obese women with PCOS who meet the suggested clinical criteria may be candidates for BS ( Fig. 2 ). These patients should be well informed about the benefits and risks of the procedures, the importance of multidisciplinary management and follow up during pre- and postoperative periods. In women with PCOS at childbearing age, special counselling is needed to inform that fertility status may improve after BS and that early postoperative pregnancy should be avoided for at least 6–12 months due to increased risk of perinatal morbidity and mortality.