PCOS in Special Populations: Adolescents, Perimenopause, Transgender Men, and Beyond

Why PCOS Requires a Population-Specific Approach

Standard Rotterdam criteria define PCOS by requiring two of three features: oligo-anovulation, clinical or biochemical hyperandrogenism, and polycystic ovarian morphology on ultrasound [1]. These criteria were validated primarily in reproductive-age women of European descent. They perform less reliably at the extremes of reproductive life and across diverse populations.

The 2023 International Evidence-Based Guideline for the Assessment and Management of PCOS, endorsed by over 40 societies including the Endocrine Society and the American Association of Clinical Endocrinology (AACE), explicitly addresses diagnostic modifications for adolescents and perimenopausal women [2]. The guideline states: "Diagnostic criteria must be adapted for life stage, as physiological changes in adolescence and the menopausal transition overlap with PCOS features." This recognition marks a shift from the one-size-fits-all diagnostic model that dominated clinical practice for decades.

Metabolic consequences compound across populations. A 2022 meta-analysis in the Journal of Clinical Endocrinology & Metabolism (N=19,044) found that women with PCOS had a 2.87-fold increased risk of type 2 diabetes (95% CI 1.44 to 5.72) compared to age-matched controls [3]. That risk varies by ethnicity, body composition, and hormonal milieu.

Adolescents: Diagnostic Uncertainty and Conservative Management

Diagnosing PCOS in adolescents is difficult because normal puberty mimics core PCOS features. Irregular cycles are physiologic for the first 1 to 2 years after menarche, and polycystic ovarian morphology appears in up to 40% of healthy adolescent girls on ultrasound [4]. The 2023 international guideline recommends against using ultrasound alone for diagnosis in adolescents within 8 years of menarche [2].

Confirmed diagnosis requires both hyperandrogenism (biochemical or clinical) and persistent oligomenorrhea, defined as cycles longer than 90 days at any point or cycles longer than 45 days more than 2 years post-menarche [2]. Anti-Mullerian hormone (AMH) levels may aid diagnosis in this group, as values above age-specific thresholds correlate with polycystic morphology, but the guideline considers AMH an emerging rather than established criterion.

Treatment in adolescents centers on combined oral contraceptives (COCs) for menstrual regulation and anti-androgen effects. Metformin is recommended as second-line for metabolic features, particularly when BMI exceeds the 85th percentile [5]. A randomized trial by Ibáñez et al. (N=38) demonstrated that low-dose metformin-flutamide-pioglitazone triple therapy reduced androgen levels by 60% over 12 months compared to COCs alone in obese adolescents [6]. Lifestyle intervention remains first-line. Even a 5 to 10% reduction in body weight improves ovulatory function in 30 to 40% of overweight adolescents with PCOS [2].

Clinicians should use a "diagnosis at risk" designation for adolescents who meet only partial criteria, with reassessment at 3-year intervals through early adulthood [2].

Perimenopausal and Postmenopausal Women: The Myth of Resolution

A common misconception holds that PCOS resolves after menopause. It does not. Hyperandrogenism may partially improve with age, but cardiometabolic risk accelerates. A longitudinal study published in the Journal of Clinical Endocrinology & Metabolism followed 2,566 women with PCOS for 20 years and found persistent elevation in testosterone levels, fasting insulin, and triglycerides through the menopausal transition [7].

Diagnosis in perimenopause is complicated by the natural decline in ovulatory frequency. The 2023 guideline notes that oligo-anovulation loses diagnostic specificity after age 40 [2]. Clinicians should rely more heavily on a documented history of PCOS from reproductive years, persistent hyperandrogenism, and metabolic phenotyping.

Dr. Richard Legro, a lead author on the Endocrine Society's PCOS clinical practice guideline, has noted: "Women with PCOS carry a cardiovascular risk profile that does not diminish with menopause. Screening for type 2 diabetes, dyslipidemia, and obstructive sleep apnea should continue indefinitely" [8].

Screening recommendations for this group include fasting glucose or HbA1c every 1 to 3 years, lipid panels every 1 to 2 years, and blood pressure monitoring at each visit [2]. Statin therapy should follow standard cardiovascular risk guidelines, though women with PCOS may reach treatment thresholds earlier than age-matched peers. The AACE recommends a 10-year ASCVD risk calculation starting at age 40 for all women with a PCOS history [9].

Transgender Men on Testosterone: Overlapping Phenotypes

Transgender men receiving exogenous testosterone develop biochemical and sometimes clinical features that overlap substantially with PCOS: elevated androgens, oligo-amenorrhea, and polycystic ovarian morphology. A cross-sectional study by Baba et al. (N=44 transgender men) found that 58% met Rotterdam criteria after 12 months of testosterone therapy [10].

This overlap creates two distinct clinical scenarios. First, a transgender man may have had undiagnosed PCOS before initiating testosterone. Second, testosterone therapy itself may induce PCOS-like ovarian changes without true underlying pathology. Distinguishing these requires a careful pre-testosterone medical history focusing on menstrual patterns, acne, and hirsutism before hormone initiation.

Metabolic screening matters regardless of etiology. Testosterone therapy increases erythrocytosis risk and may worsen insulin resistance in genetically predisposed individuals [11]. The Endocrine Society's 2017 guideline on gender-affirming hormone therapy recommends monitoring fasting glucose, lipids, and hematocrit every 6 to 12 months in transgender men on testosterone [12]. For transgender men with confirmed pre-existing PCOS, metformin can be added for insulin sensitization without interfering with masculinizing hormone therapy.

Ethnic and Racial Variation in PCOS Presentation

PCOS prevalence and phenotype vary markedly across ethnic groups. South Asian women show the highest reported prevalence at 20 to 25% depending on diagnostic criteria used, compared to 4 to 8% in Caucasian populations [13]. This disparity is driven partly by higher rates of insulin resistance and central adiposity at lower BMI thresholds. The WHO has recognized that standard BMI cutoffs underestimate metabolic risk in South Asian populations, recommending overweight classification at BMI 23 rather than 25 [14].

Black women with PCOS present differently than white women. A study by Engmann et al. (N=1,399) in the Reproductive Medicine Network found that Black women had lower AMH levels, fewer polycystic-appearing ovaries on ultrasound, but higher rates of metabolic syndrome (39% vs. 27%, P<0.01) and insulin resistance compared to white women with PCOS [15].

Hispanic women with PCOS show intermediate metabolic profiles but carry disproportionately high rates of gestational diabetes. East Asian women tend toward the lean PCOS phenotype with lower BMI but equivalent rates of hyperandrogenism [16].

These differences have direct clinical implications. Using a uniform BMI threshold of 30 to trigger metabolic screening would miss a substantial proportion of South Asian and East Asian women with significant insulin resistance. The 2023 international guideline recommends ethnicity-adjusted BMI cutoffs and universal metabolic screening for all women with PCOS regardless of weight [2].

PCOS and Pregnancy: Heightened Surveillance Required

Women with PCOS face compounded obstetric risk. A systematic review and meta-analysis by Boomsma et al. (N=7,151) found that PCOS conferred a 3.47-fold increase in gestational diabetes (95% CI 2.44 to 4.94), a 3.07-fold increase in pregnancy-induced hypertension, and a 1.75-fold increase in preterm birth compared to controls [17].

Pre-conception counseling should address weight optimization, folic acid supplementation (at least 400 mcg daily), and metformin continuation or initiation. The ADA's 2024 Standards of Care recommend early glucose tolerance testing (first trimester) rather than the standard 24 to 28 week screen for women with PCOS due to elevated baseline risk [18].

Ovulation induction protocols require careful monitoring. Letrozole 2.5 to 7.5 mg daily for 5 days is now first-line for ovulation induction in PCOS, replacing clomiphene citrate, based on the NICHD Reproductive Medicine Network trial (N=750) that showed higher live birth rates with letrozole (27.5% vs. 19.1%, P=0.007) [19].

Metformin during pregnancy reduces gestational diabetes incidence by approximately 40% in women with PCOS according to the PregMet2 trial (N=487), though it crosses the placenta and long-term offspring data remain limited to 10-year follow-up showing no adverse developmental effects [20]. GLP-1 receptor agonists such as semaglutide and liraglutide are contraindicated in pregnancy and must be stopped at least 2 months before conception per FDA labeling [21].

Lean PCOS: Normal BMI Does Not Mean Normal Metabolism

Approximately 20 to 30% of women with PCOS have a BMI <25. This subgroup is frequently underdiagnosed because clinicians associate PCOS primarily with obesity [22]. Lean women with PCOS still demonstrate insulin resistance at rates of 50 to 75%, elevated free testosterone, and anovulation.

A study by Moran et al. published in Human Reproduction (N=263) found that lean women with PCOS had fasting insulin levels 35% higher than BMI-matched controls without PCOS and comparable rates of dyslipidemia [23]. The metabolic phenotype exists independent of adiposity, driven by intrinsic defects in insulin receptor signaling and steroidogenesis.

Treatment differs from that of overweight PCOS. Weight loss is not the primary intervention. COCs remain first-line for menstrual regulation and hyperandrogenism. Metformin may still improve ovulation rates even at normal BMI. Inositol supplementation (myo-inositol 4 g plus D-chiro-inositol 400 mg daily) has shown modest benefit in lean PCOS, reducing testosterone and improving ovulation in a meta-analysis of 10 RCTs (N=601) [24].

PCOS and Disordered Eating: A Bidirectional Relationship

Eating disorders co-occur with PCOS at 2 to 3 times the rate seen in the general female population [25]. Binge eating disorder is the most common diagnosis, present in 12 to 20% of women with PCOS compared to 3 to 5% of women without PCOS. The relationship is bidirectional: hyperandrogenism and insulin resistance may drive appetite dysregulation, while restrictive eating and weight cycling worsen metabolic parameters.

Prescribing weight loss as first-line therapy requires clinical judgment in this population. The 2023 PCOS guideline recommends screening for disordered eating before initiating any weight-focused intervention and involving a multidisciplinary team including psychology or psychiatry when eating pathology is identified [2].

GLP-1 receptor agonists present a nuanced consideration here. While semaglutide 2.4 mg produced 14.9% mean weight loss at 68 weeks in the STEP-1 trial (N=1,961) vs. 2.4% with placebo [26], and off-label use in PCOS is growing, prescribing appetite-suppressing agents to patients with active eating disorders requires careful psychiatric co-management. There are no published RCTs of GLP-1 agonists specifically in PCOS populations with comorbid eating disorders.

Mental Health in PCOS Across Populations

Depression and anxiety affect approximately 40% of women with PCOS, with odds ratios of 3.78 for depression and 5.62 for anxiety compared to controls in a meta-analysis by Cooney et al. (N=3,050) [27]. Adolescents with PCOS report particularly high rates of body image dissatisfaction and social anxiety linked to hirsutism and acne.

The 2023 guideline recommends routine mental health screening using validated tools (PHQ-9 for depression, GAD-7 for anxiety) at diagnosis and annually thereafter [2]. Cognitive behavioral therapy (CBT) has the strongest evidence base for PCOS-associated psychological distress, with one RCT (N=126) showing significant improvement in depression scores, quality of life, and even menstrual regularity over 8 weeks [28].

SSRIs remain appropriate pharmacotherapy when indicated, though clinicians should be aware that some SSRIs (particularly paroxetine) can cause weight gain that may worsen metabolic features.

GLP-1 Receptor Agonists in PCOS: Emerging but Off-Label

Liraglutide and semaglutide are increasingly used off-label for PCOS-associated obesity and insulin resistance. A randomized trial by Frøssing et al. (N=72) compared liraglutide 1.8 mg daily to placebo in overweight women with PCOS over 26 weeks and found 5.2 kg greater weight loss, a 1.1 kg/m² BMI reduction, and improved menstrual cyclicity with liraglutide [29]. Free testosterone decreased significantly in the treatment group.

No phase III trials of semaglutide specifically in PCOS populations have been completed as of May 2026, though several are underway (NCT05246787). Off-label prescribing should follow shared decision-making, with explicit discussion of cost, gastrointestinal side effects (nausea in 40 to 44% of patients), and the 2-month pre-conception washout requirement [21].

Metformin remains the better-studied insulin sensitizer in PCOS, with over 30 years of safety data and pregnancy compatibility that GLP-1 agents lack. Combination therapy (metformin plus a GLP-1 agonist) is used in clinical practice but lacks randomized evidence in PCOS-specific cohorts.