Contents
pdf Download PDF
pdf Download XML
41 Views
15 Downloads
Share this article
Original Article | Volume 12 Issue 11 (November, 2026) | Pages 327 - 337
Study on Effects of Ethinyl Estradiol–Cyproterone Acetate Versus Metformin in Overweight Women with Polycystic Ovarian Syndrome: A Randomized Comparative Study
 ,
 ,
 ,
 ,
 ,
1
Associate Professor, Department of Gynaecology and obstetrics, Sarat Chandra chattopadhyay govt medical college Uluberia. Howrah west bengal India
2
Assistant professor, Department of Gynaecology and obstetrics, Sarat Chandra chattopadhyay govt medical college Uluberia. Howrah west bengal India
3
Assistant professor Department of obstetrics and gynaecology , Uluberia sarat Chandra chattopadhyay govt medical college, Howrah westbengal India
4
Specialist medical officer, Department of Gynaecology and obstetrics, Contai subdivisional hospital Contai purbamedinipur westbengal India
5
Professor and HOD, Department of Gynaecology and obstetrics, Sarat Chandra chattopadhyay govt medical college Uluberia. Howrah west bengal India
6
Ex professor and HOD, Department of Gynaecology and obstetrics, Gouri Devi institute of medical science Durgapur west Bengal India,
Under a Creative Commons license
Open Access
Received
Oct. 15, 2026
Revised
Oct. 21, 2026
Accepted
Nov. 6, 2026
Published
Nov. 17, 2026
Abstract
Background: Polycystic ovary syndrome (PCOS) is the commonest endocrine disorder of reproductive-age women and carries a clustering of reproductive, cutaneous and cardiometabolic risk. Combined oral contraceptives (COCs) and insulin sensitizers are the two principal pharmacological options for women not seeking pregnancy, but direct comparisons in young overweight Indian women are limited.Objective: To compare the effects of ethinyl estradiol–cyproterone acetate (EE-CPA) and metformin on clinical, endocrine and metabolic parameters, and on adverse effects, in overweight women with PCOS.Methods: This randomized, prospective, open-label, comparative interventional study was conducted over one year (1 May 2013 to 30 April 2014) in the Departments of Obstetrics and Gynaecology and Endocrinology, Nilratan Sircar Medical College and Hospital, Kolkata. One hundred overweight women (BMI >25 kg/m²) aged 15–25 years with PCOS (Rotterdam 2003 criteria) were randomized equally to EE-CPA (35 µg ethinyl estradiol + 2 mg cyproterone acetate, 21 days per 28-day cycle) or metformin (500 mg twice daily) for 6 months. Menstrual pattern, BMI, waist–hip ratio (WHR), modified Ferriman–Gallwey (FG) score, acne, acanthosis nigricans, LH/FSH ratio, fasting insulin–glucose ratio, testosterone, lipid profile and adverse effects were assessed at baseline, 3 and 6 months.Results: All 100 women completed the study and were analysed (50 per group). Groups were comparable at baseline. At 6 months, EE-CPA produced significantly more regular cycles (92% vs 70%, p = 0.011), normal cycle length (82% vs 60%, p = 0.028), menstrual duration ≥3 days (90% vs 66%, p = 0.008), acne improvement (92% vs 78%, p = 0.033), lower FG score (7.94 vs 8.54, p = 0.038), lower LH/FSH ratio (1.456 vs 2.050, p < 0.0001) and fewer women with elevated testosterone (10% vs 32%, p = 0.014). Metformin produced significantly greater reductions in BMI (26.14 vs 27.71 kg/m², p = 0.0001), WHR (0.771 vs 0.795, p = 0.003), acanthosis nigricans (8% vs 38%, p = 0.0009) and fasting insulin–glucose ratio (4.010 vs 4.802, p < 0.0001). Lipid profile did not differ significantly. Gastrointestinal effects were commoner with metformin; spotting and weight gain occurred only with EE-CPA.Conclusion: EE-CPA was superior for cycle control and hyperandrogenic features, whereas metformin was superior for anthropometric and insulin-related outcomes. Treatment should be individualized to the dominant clinical problem, and combination or sequential strategies merit further study
Keywords
INTRODUCTION
Polycystic ovary syndrome (PCOS) is among the most common endocrine disorders of women of reproductive age, with global prevalence estimated at roughly 6–13% depending on the diagnostic criteria and population studied [1,2,36]. It is the leading cause of hyperandrogenism and anovulatory subfertility and is characterized by oligo-/anovulation, clinical or biochemical hyperandrogenism and polycystic ovarian morphology [1–3]. The condition was first described in 1935 by Stein and Leventhal, who reported amenorrhoea, hirsutism and enlarged polycystic ovaries in a small series of women [4]. Diagnosis today rests on the Rotterdam consensus, which requires two of three features after exclusion of mimicking disorders such as hyperprolactinaemia, thyroid dysfunction, non-classical congenital adrenal hyperplasia, androgen-secreting tumours and Cushing’s syndrome [5]. Reported prevalence among Indian women varies widely, from under 5% to over 20% according to the population and criteria used [6,7]. The pathogenesis is multifactorial and incompletely understood. Genetic susceptibility interacts with environmental and lifestyle factors to produce dysregulated gonadotrophin secretion, intrinsic theca-cell steroidogenic abnormalities, and insulin resistance with compensatory hyperinsulinaemia [8–11,33]. Hyperinsulinaemia amplifies LH-driven ovarian androgen production, suppresses hepatic sex hormone-binding globulin and thereby raises free testosterone, and interferes with follicular maturation [8,10]. Obesity, particularly central adiposity, aggravates this cycle, although insulin resistance is also present in lean women with PCOS [8,10]. The resulting phenotype includes menstrual disturbance, hirsutism, acne, acanthosis nigricans, impaired glucose tolerance, dyslipidaemia, and an increased long-term risk of type 2 diabetes, cardiovascular disease and endometrial hyperplasia or carcinoma [2,3,12,13]. With increasingly sedentary lifestyles and energy-dense diets, the co-occurrence of obesity and PCOS in adolescents and young women is rising, and early, effective treatment is needed to prevent long-term sequelae [3,14,15]. For women who do not currently desire fertility, the pharmacological mainstays are COCs, which regularize withdrawal bleeding, suppress ovarian androgen production and raise SHBG, and insulin sensitizers such as metformin, which reduce hyperinsulinaemia and secondarily lower androgens [1,14,16,17]. Cyproterone acetate (CPA) is a progestogen with potent antiandrogenic activity and is widely used in combination with ethinyl estradiol for hirsutism and acne. COCs may adversely affect glucose tolerance and lipids, depending on the estrogen dose and the type and dose of progestin, whereas metformin may improve metabolic variables but is less reliable for cycle regularization [16–19]. Systematic reviews and guidelines therefore recommend individualized therapy, but head-to-head randomized evidence remains limited, heterogeneous and often derived from small trials of mixed body weight [14,16–19]. Data on young, overweight women from the Indian subcontinent are particularly scarce. We therefore conducted a randomized comparative study of EE-CPA versus metformin over six months in overweight women aged 15–25 years with PCOS, evaluating clinical, endocrine and metabolic parameters together with adverse effects, to inform the choice of first-line therapy in this population. Aims and Objectives 1. To evaluate and compare the effects of EE-CPA and metformin on clinical (menstrual pattern, BMI, WHR, hirsutism, acne, acanthosis nigricans), endocrine (LH/FSH ratio, testosterone) and metabolic (fasting insulin–glucose ratio, lipid profile) parameters in overweight women with PCOS. 2. To compare the adverse effects of the two drugs.
MATERIALS AND METHODS
Study design and setting This was a randomized, prospective, open-label, comparative interventional study conducted in the Department of Obstetrics and Gynaecology and the Department of Endocrinology, Nilratan Sircar Medical College and Hospital (NRSMCH), Kolkata, India, over a period of one year, from 1 May 2013 to 30 April 2014. Participants Overweight women (BMI >25 kg/m²) aged 15–25 years attending the gynaecology or endocrinology outpatient departments with oligomenorrhoea/amenorrhoea and/or hirsutism were screened. Ultrasonography (Philips HD-7 with a 7.5 MHz transvaginal probe) was performed, and PCOS was diagnosed by the Rotterdam (ESHRE/ASRM, 2003) criteria, i.e., at least two of: (i) oligo-/anovulation; (ii) clinical hyperandrogenism (hirsutism, acne) or biochemical hyperandrogenism (raised LH, testosterone, androstenedione or insulin); (iii) polycystic ovaries on ultrasonography (≥12 follicles of 2–9 mm and/or ovarian volume >10 cm³) [5]. Only women who did not desire pregnancy at the time of the study were included. Exclusion criteria were hyperprolactinaemia, thyroid dysfunction, Cushing’s syndrome, non-classical congenital adrenal hyperplasia, virilizing ovarian tumours, cardiovascular disease, a history of hypertension or diabetes, use of oral contraceptives, antiandrogens, insulin sensitizers or lipid-lowering agents in the preceding six months, and pregnancy. Thyroid function and serum prolactin were measured in all candidates to exclude these disorders. Sample size, randomization and interventions A total of 100 women meeting all criteria were enrolled after written informed consent. They were allocated by a computer-generated randomization sequence to one of two equal groups (n = 50 each): The study was open-label. Participants in both groups received identical clinical follow-up at 3 and 6 months. • EE-CPA group: a combined oral contraceptive containing 35 µg ethinyl estradiol and 2 mg cyproterone acetate, one tablet daily for 21 days followed by a 7-day pill-free interval, for 6 months. • Metformin group: metformin 500 mg twice daily with meals for 6 months. Married or sexually active women in this group used barrier contraception throughout the study. • Clinical, endocrine and metabolic assessment • A detailed history and examination were completed at baseline. Laboratory investigations were performed on day 3 of the cycle (or at a random time in amenorrhoeic women). All parameters were evaluated at baseline, 3 months and 6 months. • Menstrual history: regularity, cycle length (21–35, 36–50, >50 days), duration (<3 or ≥3 days) and amount of flow. • Anthropometry: height by stadiometer and weight by calibrated scale; BMI = weight (kg)/height (m)²; overweight was defined as 25–30 and obesity as >30 kg/m². Waist (narrowest point of the torso) and hip (widest gluteal point) circumferences were measured with a tape, and WHR >0.85 was taken to suggest insulin resistance. • Hirsutism was scored using the modified Ferriman–Gallwey (FG) method across nine body areas (score 0–36; >8 considered hirsutism) [26]. Acne and acanthosis nigricans (AN) were recorded as present or absent; acne outcome was classified as unchanged or improved (reduced or absent). • Hormones (serum LH, FSH, free and total testosterone, insulin, free T3, free T4, TSH and prolactin) were assayed by ELISA. Elevated free/total testosterone was categorized above the assay reference range. • Metabolic parameters: fasting glucose by the glucose oxidase–peroxidase method; fasting insulin–glucose ratio as a surrogate index of insulin resistance [21]; total cholesterol, HDL, LDL, VLDL and triglycerides by end-point method on a Transasia XL-600 autoanalyser. Lipid profile was categorized as normal or abnormal. • Adverse effects: nausea, diarrhoea, abdominal pain, weight gain, spotting/breakthrough bleeding, chloasma, hypertension, leg-vein thrombosis, gallstones and others were recorded at each visit. Statistical analysis Data were tabulated and analysed using standard statistical methods. Continuous variables are expressed as mean ± standard deviation (SD) and compared between groups using the unpaired Student’s t-test. Categorical variables are expressed as number (percentage) and compared using the chi-square test, with the contingency coefficient as a measure of association. A two-sided p value <0.05 (95% confidence level) was considered statistically significant. Ethical considerations The study protocol was approved by the Institutional Ethics Committee of NRSMCH, Kolkata. The purpose of the study was explained to each participant in her own language, and written informed consent was obtained (with guardian consent for participants who were minors).
RESULTS
Participant flow and baseline characteristics Of 123 women assessed for eligibility, 100 were randomized, 50 to each arm (Figure 1). All allocated women received the assigned treatment, completed 6 months of follow-up and were included in the analysis; there were no exclusions after randomization. The mean age of the cohort was 19.57 ± 2.31 years, and 56% were aged 15–19 years. Three-quarters were rural residents and from low socio-economic backgrounds, and 63% were unmarried. Mean BMI was 28.05 ± 2.18 kg/m², mean WHR 0.799 ± 0.042 and mean FG score 9.67 ± 1.91. At enrolment, 65% had irregular cycles, 47% had acne, 44% had acanthosis nigricans, 51% had elevated testosterone and 52% had an abnormal lipid profile. The two groups were statistically comparable for all demographic, clinical, endocrine and metabolic variables at baseline (all p > 0.05; Table 1). Menstrual outcomes Menstrual outcomes improved in both groups but more in the EE-CPA group (Table 2). Irregular cycles fell from 72% to 20% at 3 months and 8% at 6 months with EE-CPA, compared with 58%, 42% and 30% with metformin. Differences in regularity (p = 0.031 at 3 months; p = 0.011 at 6 months), cycle length (p = 0.036; p = 0.028) and menstrual duration (p = 0.014; p = 0.008) were statistically significant at both time points. The proportion with normal blood loss increased in both groups, but the between-group difference was not significant at 3 months (p = 0.066) or 6 months (p = 0.098). 4.3 Clinical parameters Mean BMI changed little with EE-CPA (27.89 → 27.77 → 27.71 kg/m²) but fell progressively with metformin (28.21 → 26.92 → 26.14 kg/m²); the between-group difference was significant at 3 months (p = 0.047) and highly significant at 6 months (p = 0.0001). WHR was unchanged with EE-CPA (0.795 throughout) and decreased with metformin (0.802 → 0.787 → 0.771), the difference reaching significance at 6 months (p = 0.003) but not at 3 months (p = 0.321). Hyperandrogenic features improved with both drugs, but more with EE-CPA. Mean FG score fell from 9.50 to 7.94 with EE-CPA and from 9.84 to 8.54 with metformin (p = 0.013 at 3 months; p = 0.038 at 6 months). Acne was improved (reduced or absent) in 78% versus 52% of women at 3 months (p = 0.012) and in 92% versus 78% at 6 months (p = 0.033). In contrast, acanthosis nigricans regressed substantially only with metformin: 42% → 38% with EE-CPA versus 46% → 22% → 8% with metformin (p = 0.084 at 3 months; p = 0.0009 at 6 months) (Table 3). Endocrine and metabolic parameters The LH/FSH ratio declined in both groups, from 2.68 to 1.456 with EE-CPA and from 2.64 to 2.050 with metformin; the between-group difference was significant at 3 months (p = 0.022) and 6 months (p < 0.0001). Elevated testosterone persisted in only 10% of the EE-CPA group at 6 months versus 32% of the metformin group (p = 0.014). The fasting insulin–glucose ratio showed opposite trends: it was unchanged with EE-CPA (4.792 → 4.794 → 4.802) but decreased with metformin (4.788 → 4.466 → 4.010), a significant difference at 3 months (p = 0.0002) and 6 months (p < 0.0001). The proportion with an abnormal lipid profile fell numerically in both groups (EE-CPA 56% → 46%; metformin 48% → 28%) but the between-group difference was not significant (p = 0.098 at 6 months) (Table 4). 4.5 Adverse effects Both regimens were well tolerated and no participant withdrew. At 6 months, 92% of the EE-CPA group and 82% of the metformin group reported no adverse effect (Table 5). Gastrointestinal complaints (nausea, diarrhoea, abdominal pain) were confined mainly to the metformin group (18% at 6 months), whereas spotting and weight gain occurred only with EE-CPA (6% each at 3 months, 2% each at 6 months). No thromboembolic event, hypertension, chloasma or gallstone disease was recorded in either group. Table 1. Baseline characteristics of the study groups Characteristic EE-CPA (n = 50) Metformin (n = 50) p value Age, years (mean ± SD) 19.62 ± 2.42 19.52 ± 2.22 0.830 Age 15–19 years, n (%) 32 (64) 24 (48) 0.159 Literate, n (%) 30 (60) 36 (72) 0.291 Rural residence, n (%) 40 (80) 35 (70) 0.356 Low socio-economic status, n (%) 40 (80) 35 (70) 0.513 Unmarried, n (%) 30 (60) 33 (66) 0.679 BMI, kg/m² (mean ± SD) 27.89 ± 2.03 28.21 ± 2.33 0.466 WHR (mean ± SD) 0.795 ± 0.042 0.802 ± 0.043 0.407 FG score (mean ± SD) 9.50 ± 1.97 9.84 ± 1.85 0.376 Acne present, n (%) 19 (38) 28 (56) 0.109 Acanthosis nigricans, n (%) 21 (42) 23 (46) 0.840 Irregular cycles, n (%) 36 (72) 29 (58) 0.208 Cycle length 21–35 days, n (%) 23 (46) 18 (36) 0.373 Menstrual duration <3 days, n (%) 24 (48) 24 (48) 0.841 Scanty flow, n (%) 39 (78) 34 (68) 0.368 LH/FSH ratio (mean ± SD) 2.68 ± 0.53 2.64 ± 0.53 0.706 Fasting insulin–glucose ratio (mean ± SD) 4.792 ± 0.454 4.788 ± 0.436 0.964 Elevated testosterone, n (%) 29 (58) 22 (44) 0.230 Abnormal lipid profile, n (%) 28 (56) 24 (48) 0.548 p values: chi-square test for categorical and unpaired t-test for continuous variables. Table 2. Menstrual outcomes at baseline, 3 months and 6 months, n (%) Outcome Group Baseline 3 months 6 months p (3 mo) p (6 mo) Regular cycles EE-CPA 14 (28) 40 (80) 46 (92) 0.031 0.011 Metformin 21 (42) 29 (58) 35 (70) Cycle length 21–35 days EE-CPA 23 (46) 38 (76) 41 (82) 0.036 0.028 Metformin 18 (36) 27 (54) 30 (60) Menstrual duration ≥3 days EE-CPA 26 (52) 42 (84) 45 (90) 0.014 0.008 Metformin 26 (52) 30 (60) 33 (66) Normal blood loss EE-CPA 11 (22) 35 (70) 36 (72) 0.066 0.098 Metformin 16 (32) 25 (50) 27 (54) Table 3. Clinical parameters during treatment Parameter Group Baseline 3 months 6 months p (3 mo) p (6 mo) BMI, kg/m² EE-CPA 27.89 ± 2.03 27.77 ± 2.06 27.71 ± 2.00 0.047 0.0001 Metformin 28.21 ± 2.33 26.92 ± 2.16 26.14 ± 1.97 WHR EE-CPA 0.795 ± 0.042 0.795 ± 0.043 0.795 ± 0.044 0.321 0.003 Metformin 0.802 ± 0.043 0.787 ± 0.037 0.771 ± 0.035 FG score EE-CPA 9.50 ± 1.97 8.56 ± 1.47 7.94 ± 1.28 0.013 0.038 Metformin 9.84 ± 1.85 9.40 ± 1.83 8.54 ± 1.56 Acne improved*, n (%) EE-CPA – 39 (78) 46 (92) 0.012 0.033 Metformin – 26 (52) 39 (78) Acanthosis nigricans present, n (%) EE-CPA 21 (42) 20 (40) 19 (38) 0.084 0.0009 Metformin 23 (46) 11 (22) 4 (8) Acne improved = reduced or absent at follow-up, as a proportion of all 50 women in each group. Values are mean ± SD unless stated. Table 4. Endocrine and metabolic parameters during treatment Parameter Group Baseline 3 months 6 months p (3 mo) p (6 mo) LH/FSH ratio EE-CPA 2.68 ± 0.53 2.064 ± 0.550 1.456 ± 0.385 0.022 <0.0001 Metformin 2.64 ± 0.53 2.316 ± 0.535 2.050 ± 0.530 Fasting insulin–glucose ratio EE-CPA 4.792 ± 0.454 4.794 ± 0.425 4.802 ± 0.423 0.0002 <0.0001 Metformin 4.788 ± 0.436 4.466 ± 0.430 4.010 ± 0.432 Elevated testosterone, n (%) EE-CPA 29 (58) 8 (16) 5 (10) 0.040 0.014 Metformin 22 (44) 18 (36) 16 (32) Abnormal lipid profile, n (%) EE-CPA 28 (56) 28 (56) 23 (46) NS 0.098 Metformin 24 (48) 19 (38) 14 (28) Table 5. Adverse effects, n (%) Adverse effect EE-CPA, 3 mo Metformin, 3 mo EE-CPA, 6 mo Metformin, 6 mo None 41 (82) 44 (88) 46 (92) 41 (82) Nausea 3 (6) 4 (8) 2 (4) 4 (8) Diarrhoea 0 1 (2) 0 4 (8) Abdominal pain 0 1 (2) 0 1 (2) Spotting 3 (6) 0 1 (2) 0 Weight gain 3 (6) 0 1 (2) 0
DISCUSSION
In this randomized comparison of 100 young overweight women with PCOS, the two drugs showed complementary rather than uniformly superior profiles. EE-CPA was significantly better for restoring cyclicity and for controlling hyperandrogenic manifestations (hirsutism, acne, testosterone and LH/FSH ratio), whereas metformin was significantly better for reducing BMI, WHR, acanthosis nigricans and the fasting insulin–glucose ratio. Neither drug significantly altered the lipid profile over six months, and both were well tolerated, with gastrointestinal effects predominating with metformin and spotting and weight gain confined to EE-CPA. Because the randomized groups were closely matched at baseline for demographic, clinical and biochemical variables, the differences observed are unlikely to reflect confounding by baseline imbalance. Menstrual pattern Cycle regularity at six months was achieved by 92% of women on EE-CPA versus 70% on metformin, with parallel advantages in normal cycle length and menstrual duration. The advantage of the COC appeared early (3 months) and widened by 6 months. This agrees with the Cochrane reviews comparing insulin sensitizers with COCs, which found metformin less effective than the pill for establishing a regular menstrual pattern [18,19], and with the randomized study of Morin-Papunen et al. in obese women with PCOS, in which EE-CPA regularized cycles more consistently than metformin [20]. The biological explanation is straightforward: a cyclical COC supplies exogenous estrogen and progestogen, suppresses the hypothalamic–pituitary–ovarian axis, and produces scheduled withdrawal bleeding irrespective of the underlying ovulatory status, so that regularity is partly pharmacological rather than physiological. Metformin, by contrast, acts upstream by lowering hyperinsulinaemia and thereby ovarian androgen output, which can restore spontaneous ovulation in a proportion of women but more slowly and less predictably [22,23,27]. The fact that 70% of metformin-treated women had regular cycles by six months, compared with 42% at baseline, indicates that this effect is nonetheless clinically meaningful, particularly for women who prefer to avoid hormonal contraception. Blood loss improved in both groups, with no significant difference between them, which is consistent with the endometrial normalization expected once anovulation and unopposed estrogen exposure are interrupted. A point of clinical importance is that regular withdrawal bleeding on a COC provides endometrial protection but does not equate to restored ovulation, and cycles may revert after discontinuation. Our six-month design does not address post-treatment recurrence, which is a known limitation of all short-term PCOS trials [1,14]. Hirsutism and acne Both drugs reduced the FG score, but EE-CPA did so to a greater extent (7.94 vs 8.54 at 6 months) and also produced more acne improvement (92% vs 78%). These findings are mechanistically coherent. Ethinyl estradiol suppresses LH and therefore ovarian androgen synthesis and raises SHBG, lowering free testosterone, while cyproterone acetate adds competitive androgen-receptor blockade and inhibition of 5α-reductase activity in the pilosebaceous unit [14,16]. Metformin reduces androgens only indirectly through a fall in insulin, which modifies ovarian cytochrome P450c17α activity and increases SHBG [22,23]. Earlier studies of metformin have reported reductions in hirsutism and androgens [23,24], and the overall evidence suggests that its effect on hirsutism and acne is real but smaller than that of COCs, which is the pattern we observed [18,19]. The Cochrane analysis likewise suggests an advantage of COCs over metformin for hirsutism [18]. Because terminal hair follicles cycle over many months, a six-month assessment will underestimate the eventual cosmetic benefit of any systemic therapy, and guidelines accordingly advise at least 6–12 months of treatment before judging response and the addition of cosmetic measures or antiandrogens when response is inadequate [1,30]. The absolute FG reductions in our study (about 1.5 points with EE-CPA and 1.3 with metformin) were modest, underscoring this point. Anthropometric parameters Metformin produced a progressive fall in BMI (28.21 to 26.14 kg/m²) and WHR (0.802 to 0.771), whereas BMI changed by only 0.18 kg/m² and WHR not at all with EE-CPA. This replicates the findings of Morin-Papunen et al., who found a reduction in BMI and WHR with metformin but not with EE-CPA [20], and is consistent with meta-analytic data showing modest but significant reductions in BMI with metformin, particularly when combined with lifestyle measures [17,28]. Metformin is thought to reduce weight through lowered insulin, reduced hepatic gluconeogenesis, decreased appetite and increased peripheral glucose utilization. The reduction in WHR is clinically relevant because central adiposity is more closely tied than BMI to insulin resistance and cardiovascular risk in South Asian women [10,13]. It should be noted that the average weight loss achieved (about 2 kg/m² of BMI over six months) is modest, and that lifestyle modification remains the foundation of therapy in overweight PCOS irrespective of the pharmacological agent chosen [1,32]. The small weight gain reported by a few women taking EE-CPA (6% at 3 months, 2% at 6 months) is in keeping with the generally neutral effect of low-dose COCs on body weight; importantly, BMI did not rise significantly in this group, which is reassuring for women with existing overweight [16]. Insulin resistance and acanthosis nigricans The fasting insulin–glucose ratio decreased only in the metformin group, and acanthosis nigricans, a cutaneous marker of hyperinsulinaemia, regressed markedly with metformin (46% to 8%) but hardly at all with EE-CPA (42% to 38%). These concordant biochemical and clinical observations support a genuine insulin-sensitizing effect of metformin in this population. The influence of COCs on insulin sensitivity is more controversial: Dahlgren et al. demonstrated reduced insulin sensitivity by euglycaemic clamp in women with PCOS treated with EE-CPA [25], while other investigators have reported neutral or mildly adverse effects, depending on the estrogen dose, the progestogen and the baseline degree of insulin resistance [16,18]. The more recent meta-analysis of different COC types in PCOS similarly concluded that metabolic effects are heterogeneous and that no single formulation is clearly superior [16]. In our study, EE-CPA did not improve the insulin index and the ratio drifted marginally upward, which is a reason for periodic metabolic surveillance in overweight women on long-term COCs, even though the change was small. A limitation is that the fasting insulin–glucose ratio is a surrogate index, validated against clamp-derived measures in PCOS in its glucose-to-insulin form [21], and was not supplemented by HOMA-IR or an oral glucose tolerance test. LH/FSH ratio and testosterone The LH/FSH ratio fell more with EE-CPA (2.68 to 1.456) than with metformin (2.64 to 2.050), and the prevalence of elevated testosterone fell from 58% to 10% with EE-CPA versus 44% to 32% with metformin. Suppression of pituitary gonadotrophins by ethinyl estradiol with CPA, and the SHBG-raising effect of the estrogen, explain the pronounced fall in free and total testosterone and LH [14,16]. Metformin also lowered androgens in some studies, although the effect has been inconsistent and often limited to smaller series [22–24]; in our cohort the decline in the proportion with elevated testosterone was numerically present but small. Although a raised LH/FSH ratio is no longer a diagnostic criterion, it remains a useful pharmacodynamic marker of the degree of gonadotrophin dysregulation in PCOS [1,5]. Lipid profile The proportion of women with an abnormal lipid profile fell numerically in both groups, but without a significant between-group difference. Our categorical assessment and six-month duration may have been insufficient to detect modest changes, and our data should not be interpreted as proving the lipid neutrality of either drug. Previous literature has reported both favourable and unfavourable lipid effects of COCs in PCOS, depending on estrogen dose and progestogen, and generally small effects of metformin on lipids [16,17,35]. The results mirror this heterogeneity and are in line with the overall conclusion of the 2023 guideline that evidence of a differential effect on lipids is limited [1]. Safety and tolerability Adverse effects were mild and infrequent. In the metformin group, gastrointestinal symptoms affected 18% of women at six months, a figure consistent with the known dose-related intolerance of metformin, which can often be mitigated by slow titration, administration with meals and use of extended-release formulations [34]. In the EE-CPA group, spotting and weight gain were the only reported effects and were uncommon. No thromboembolic event occurred; however, with 50 women per arm the study is far too small to assess rare serious events. Combined pills containing cyproterone acetate are generally considered to carry a venous thromboembolic risk that is not lower than that of levonorgestrel-containing pills, and obesity itself is an additional risk factor, so a careful personal and family history of thrombosis and blood-pressure monitoring are essential in this population [1,16]. High cumulative doses of cyproterone acetate (25 mg/day and above, as used in high-dose antiandrogen regimens) have been associated in a large French cohort with a dose-dependent increase in intracranial meningioma [29]; the 2 mg dose in COCs is much lower, but this signal supports use of the lowest effective dose for the shortest necessary time and review of the need for continued treatment. Clinical implications The findings support an individualized, phenotype-driven approach. For an overweight young woman whose predominant complaints are irregular cycles, hirsutism or acne and who does not wish to conceive, EE-CPA offers the most rapid and complete control, consistent with guideline recommendations that COCs are first-line therapy for menstrual irregularity and hyperandrogenism in adolescents and adults [1,14,15]. For women whose principal problems are central adiposity, acanthosis nigricans and hyperinsulinaemia, or in whom estrogen is contraindicated or declined, metformin is a rational alternative with distinct metabolic advantages, consistent with guidance that metformin may be considered for anthropometric and metabolic outcomes in adults with BMI ≥25 kg/m², and in adolescents with a clear diagnosis, in addition to lifestyle therapy [1,17]. Combination therapy appears a logical option for women with both hyperandrogenic and metabolic phenotypes, and a number of systematic reviews support exploring COC plus metformin or adding anti-obesity or other agents in selected patients, although data remain limited [17,31]. Counselling must also address contraception for sexually active women on metformin, since metformin is not contraceptive and, as in our study, barrier methods or another method must be used. Strengths and limitations Strengths of this study include its randomized design, balanced baseline characteristics, complete follow-up with no attrition, a standardized dose of both agents, assessment at two time points and evaluation of clinical, endocrine, metabolic and safety outcomes together in a young, overweight, predominantly rural and low-income Indian population that is under-represented in the literature. Several limitations should be acknowledged. First, the study was open-label and single-centre, so performance and assessment bias cannot be excluded, particularly for subjective outcomes such as acne and FG scoring. Second, the follow-up of six months is too short to judge hair response, post-treatment recurrence, long-term metabolic safety or cardiovascular outcomes. Third, ovulation was not confirmed biochemically, so cycle regularity on EE-CPA reflects withdrawal bleeding. Fourth, insulin resistance was assessed by a surrogate ratio without HOMA-IR or oral glucose tolerance testing, and lipid and testosterone outcomes were analysed categorically; between-group comparisons at each time point were reported and formal within-group tests of change were not performed. Fifth, the Rotterdam criteria were applied to adolescents, whereas the 2023 guideline advises against using ultrasound to diagnose PCOS within about eight years of menarche because of the physiological multifollicular appearance of young ovaries [1,15]; moreover, the broad inclusion of raised insulin or LH as “biochemical hyperandrogenism” differs from current definitions, which rely on total/free testosterone or the free androgen index [1]. Sixth, diet and exercise were not standardized or quantified, and a BMI threshold of 25 kg/m² was used, whereas lower Asian cut-offs would classify more women as overweight. Finally, the sample size was not powered for rare adverse events or for subgroup analysis by age or phenotype. Larger, longer, ideally blinded multicentre trials, including combination arms and lifestyle-matched controls, are warranted.
CONCLUSION
In overweight women with PCOS, EE-CPA is more effective than metformin for achieving regular, normal-length cycles and for reducing hirsutism, acne, LH/FSH ratio and testosterone, whereas metformin is more effective for lowering BMI, waist-to-hip ratio, acanthosis nigricans and the fasting insulin–glucose ratio; neither drug significantly changed the lipid profile over six months. Both were well tolerated. These complementary benefits suggest that therapy should be individualized according to the predominant clinical problem, the woman’s contraceptive needs and her cardiovascular and thrombotic risk profile, always alongside lifestyle modification, and that combination regimens deserve further evaluation in adequately powered, longer trials.
REFERENCES
1. Teede HJ, Tay CT, Laven JJE, et al. Recommendations from the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome. Fertil Steril. 2023;120(4):767-793. 2. Joham AE, Norman RJ, Stener-Victorin E, et al. Polycystic ovary syndrome. Lancet Diabetes Endocrinol. 2022;10(9):668-680. 3. Stener-Victorin E, Teede H, Norman RJ, et al. Polycystic ovary syndrome. Nat Rev Dis Primers. 2024;10(1):27. 4. Stein IF, Leventhal ML. Amenorrhea associated with bilateral polycystic ovaries. Am J Obstet Gynecol. 1935;29:181-191. 5. Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril. 2004;81(1):19-25. 6. Ganie MA, Vasudevan V, Wani IA, Baba MS, Arif T, Rashid A. Epidemiology, pathogenesis, genetics and management of polycystic ovary syndrome in India. Indian J Med Res. 2019;150(4):333-344. 7. Sachdeva G, Gainder S, Suri V, Sachdeva N, Chopra S. Comparison of the different PCOS phenotypes based on clinical metabolic and hormonal profile, and their response to clomiphene. Indian J Endocrinol Metab. 2019;23(3):326-331. 8. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev. 1997;18(6):774-800. 9. Dapas M, Dunaif A. Deconstructing a syndrome: genomic insights into PCOS causal mechanisms and classification. Endocr Rev. 2022;43(6):927-965. 10. Zhao H, Zhang J, Cheng X, Nie X, He B. Insulin resistance in polycystic ovary syndrome across various tissues: an updated review of pathogenesis, evaluation, and treatment. J Ovarian Res. 2023;16(1):9. 11. Siddiqui S, Mateen S, Ahmad R, Moin S. A brief insight into the etiology, genetics, and immunology of polycystic ovarian syndrome (PCOS). J Assist Reprod Genet. 2022;39(11):2439-2473. 12. Azziz R, Carmina E, Chen Z, et al. Polycystic ovary syndrome. Nat Rev Dis Primers. 2016;2:16057. 13. Escobar-Morreale HF. Polycystic ovary syndrome: definition, aetiology, diagnosis and treatment. Nat Rev Endocrinol. 2018;14(5):270-284. 14. Legro RS, Arslanian SA, Ehrmann DA, et al. Diagnosis and treatment of polycystic ovary syndrome: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2013;98(12):4565-4592. 15. Ibáñez L, Oberfield SE, Witchel S, et al. An international consortium update: pathophysiology, diagnosis, and treatment of polycystic ovarian syndrome in adolescence. Horm Res Paediatr. 2017;88(6):371-395. 16. Forslund M, Melin J, Alesi S, et al. Different kinds of oral contraceptive pills in polycystic ovary syndrome: a systematic review and meta-analysis. Eur J Endocrinol. 2023;189(1):S1-S16. 17. Melin JM, Forslund M, Alesi S, et al. The impact of metformin with or without lifestyle modification versus placebo on polycystic ovary syndrome: a systematic review and meta-analysis of randomized controlled trials. Eur J Endocrinol. 2023;189(2):S37-S63. 18. Fraison E, Kostova E, Moran LJ, et al. Metformin versus the combined oral contraceptive pill for hirsutism, acne, and menstrual pattern in polycystic ovary syndrome. Cochrane Database Syst Rev. 2020;8:CD005552. 19. Costello M, Shrestha B, Eden J, Sjoblom P, Johnson N. Insulin-sensitising drugs versus the combined oral contraceptive pill for hirsutism, acne and risk of diabetes, cardiovascular disease, and endometrial cancer in polycystic ovary syndrome. Cochrane Database Syst Rev. 2007;(1):CD005552. 20. Morin-Papunen LC, Vauhkonen I, Koivunen RM, Ruokonen A, Martikainen HK, Tapanainen JS. Endocrine and metabolic effects of metformin versus ethinyl estradiol-cyproterone acetate in obese women with polycystic ovary syndrome: a randomized study. J Clin Endocrinol Metab. 2000;85(9):3161-3168. 21. Legro RS, Finegood D, Dunaif A. A fasting glucose to insulin ratio is a useful measure of insulin sensitivity in women with polycystic ovary syndrome. J Clin Endocrinol Metab. 1998;83(8):2694-2698. 22. Nestler JE, Jakubowicz DJ. Decreases in ovarian cytochrome P450c17α activity and serum free testosterone after reduction of insulin secretion in polycystic ovary syndrome. N Engl J Med. 1996;335(9):617-623. 23. Kolodziejczyk B, Duleba AJ, Spaczynski RZ, Pawelczyk L. Metformin therapy decreases hyperandrogenism and hyperinsulinemia in women with polycystic ovary syndrome. Fertil Steril. 2000;73(6):1149-1154. 24. Kazerooni T, Dehghan-Kooshkghazi M. Effects of metformin therapy on hyperandrogenism in women with polycystic ovarian syndrome. Gynecol Endocrinol. 2003;17(1):51-56. 25. Dahlgren E, Landin K, Krotkiewski M, Holm G, Janson PO. Effects of two antiandrogen treatments on hirsutism and insulin sensitivity in women with polycystic ovary syndrome. Hum Reprod. 1998;13(10):2706-2711. 26. Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab. 1961;21:1440-1447. 27. Harborne L, Fleming R, Lyall H, Norman J, Sattar N. Descriptive review of the evidence for the use of metformin in polycystic ovary syndrome. Lancet. 2003;361(9372):1894-1901. 28. Naderpoor N, Shorakae S, de Courten B, Misso ML, Moran LJ, Teede HJ. Metformin and lifestyle modification in polycystic ovary syndrome: systematic review and meta-analysis. Hum Reprod Update. 2015;21(5):560-574. 29. Weill A, Nguyen P, Labidi M, et al. Use of high dose cyproterone acetate and risk of intracranial meningioma in women: cohort study. BMJ. 2021;372:n37. 30. Spritzer PM, Marchesan LB, Santos BR, Fighera TM. Hirsutism, normal androgens and diagnosis of PCOS. Diagnostics (Basel). 2022;12(8):1922. 31. Alesi S, Forslund M, Melin J, et al. Efficacy and safety of anti-obesity medications for improving clinical and biochemical outcomes in adults with polycystic ovary syndrome: a systematic review and meta-analysis. EClinicalMedicine. 2023;63:102162. 32. Lim SS, Hutchison SK, Van Ryswyk E, Norman RJ, Teede HJ, Moran LJ. Lifestyle changes in women with polycystic ovary syndrome. Cochrane Database Syst Rev. 2019;3:CD007506. 33. Sanchez-Garrido MA, Tena-Sempere M. Metabolic dysfunction in polycystic ovary syndrome: pathogenic role of androgen excess and potential therapeutic strategies. Mol Metab. 2020;35:100937. 34. Bonnet F, Scheen A. Understanding and overcoming metformin gastrointestinal intolerance. Diabetes Obes Metab. 2017;19(Suppl 1):473-481. 35. Mastorakos G, Koliopoulos C, Creatsas G. Androgen and lipid profiles in adolescents with polycystic ovary syndrome who were treated with two forms of combined oral contraceptives and flutamide. Fertil Steril. 2002;77(5):919-927. 36. Sadeghi HM, Adeli I, Calina D, et al. Polycystic ovary syndrome: a comprehensive review of pathogenesis, management, and drug repurposing. Int J Mol Sci. 2022;23(2):583.
Recommended Articles
Review Article
Accuracy of Artificial Intelligence vs. Human Clinicians in Predicting Dental Implant Success Rates in Compromised Bone Morphologies: A Systematic Review
...
Published: 17/11/2026
Research Article
Causes of recurrence after inguinal hernia repair, surgical factors, patients factors and strategies for prevention
...
Published: 30/09/2026
Research Article
Diabetic Ketoacidosis (DKA): Precipitating Factors, Management Gaps, and Outcomes at Ayub Teaching Hospital, Abbottabad: A Retrospective Observational Cohort Study
...
Published: 28/08/2026
Research Article
Clinical and Hematological Spectrum of Hemoglobinopathies and Thalassemias Detected by High-Performance Liquid Chromatography: A Three-Year Retrospective Study at a Tertiary Care Centre in Southern Assam
...
Published: 16/11/2026
Chat on WhatsApp
© Copyright Journal of Contemporary Clinical Practice