Abstract
BACKGROUND: Hirsutism (excessive and unwanted hair growth) is a distressing and relatively common endocrine problem in women that may prove difficult to manage. Cyproterone acetate (CPA), an antiandrogen, is frequently used to treat hirsutism, usually in combination with ethinylestradiol. This is an update of a Cochrane review first published in 2003. OBJECTIVES: To assess the benefits and harms of cyproterone acetate (CPA) alone, or in combination with ethinylestradiol, and other medication in reducing hair growth and improving the endocrine profile in women with hirsutism secondary to ovarian hyperandrogenism as well as idiopathic hirsutism. SEARCH METHODS: We searched the Cochrane Gynaecology and Fertility specialised register, CENTRAL, MEDLINE, Embase, PsycINFO and CINAHL to 20 December 2022. We also searched the reference lists of relevant papers and clinical trial registries. SELECTION CRITERIA: We included randomised controlled trials (RCT) examining women of reproductive age with either idiopathic hirsutism or hirsutism secondary to ovarian hyperandrogenism. Hirsutism was defined as a Ferriman Gallwey (FG) score greater than 7. DATA COLLECTION AND ANALYSIS: We included 23 studies. Only one study had more than 100 women included in the analysis. The primary outcomes were the objective and subjective assessment of clinical parameters (FG scores, linear hair growth and hair shaft diameter, women's feedback and frequency of hair removal). Secondary outcomes included endocrine parameters, side effects and withdrawals during therapy. MAIN RESULTS: Cyproterone acetate greater than 2 mg plus ethinylestradiol compared to cyproterone acetate 2 mg or less plus ethinylestradiol (dose comparison) After six months of treatment, we are uncertain of any difference in effect on FG scores (mean difference (MD) 0.63, 95% confidence interval (CI) -1.02 to 2.28; I(2) = 0%; 2 RCTs, 145 women; very low-certainty evidence). There was likely little to no difference in free testosterone (MD 0.35 pmol/L, 95% CI -0.61 to 1.31; 1 RCT, 113 women; moderate-certainty evidence). Cyproterone acetate alone (no ethinylestradiol) compared to other interventions alone (no ethinylestradiol) There was little or no difference for CPA compared with gonadotropin-releasing hormone (GnRH) analogues on total testosterone at three months (MD 0.17 nmol/L, 95% CI -0.15 to 0.49). There was little or no evidence of an effect for CPA compared with GnRH agonist on androstenedione at three months (MD 0.66 nmol/L, 95% CI -0.44 to 1.76). Both results from one RCT (20 women; very low-certainty evidence). Cyproterone acetate plus ethinylestradiol compared to other interventions alone (no ethinylestradiol) There was little to no difference in effect for CPA plus ethinylestradiol on FG scores at six months compared with finasteride (MD 4.70, 95% CI -1.86 to 11.26; 1 RCT, 27 women; low-certainty evidence), spironolactone (MD 0.90, 95% CI -2.86 to 4.66; 1 RCT, 77 women; moderate-certainty evidence), ketoconazole (MD 0.70, 95% CI -0.84 to 2.24; 1 RCT, 81 women; moderate-certainty evidence), or pioglitazone plus flutamide plus metformin (Pio-Flu-Met) (MD 0.90, 95% CI -0.79 to 2.59; 1 RCT, 34 women; low-certainty evidence). CPA plus ethinylestradiol may improve hirsutism slightly compared with flutamide (MD 4.00, 95% CI 0.10 to 7.90; 1 RCT, 28 women). CPA plus ethinylestradiol likely results in little to no difference in total testosterone at six months compared with spironolactone (MD -0.06 nmol/L, 95% CI -1.25 to 1.13; 1 RCT, 77 women; moderate-certainty evidence), ketoconazole (MD -0.02 nmol/L, 95% CI -0.37 to 0.33; 1 RCT, 81 women; moderate-certainty evidence), or Pio-Flu-Met (MD -0.39 nmol/L, 95% CI -0.82 to 0.04; 1 RCT, 81 women; low-certainty evidence). CPA plus ethinylestradiol may be more effective than finasteride at six months (MD -1.60 nmol/L, 95% CI -2.39 to -0.81; 1 RCT, 27 women; low-certainty evidence). CPA plus ethinylestradiol may lower free testosterone slightly at six months compared to finasteride (MD -9.02 nmol/L, 95% CI -12.44 to -5.60; 1 RCT, 27 women; low-certainty evidence) or flutamide (MD -4.16 nmol/L, 95% CI -6.62 to -1.70; 1 RCT, 28 women; low-certainty evidence). There was little to no difference between CPA plus ethinylestradiol and spironolactone (MD 0.35 nmol/L, 95% CI -0.62 to 1.32; 1 RCT, 77 women; low-certainty evidence). CPA plus ethinylestradiol may be less effective than ketoconazole (MD 1.39 nmol/L, 95% CI 0.43 to 2.35; 1 RCT, 81 women; low-certainty evidence). Cyproterone acetate plus ethinylestradiol compared to other interventions plus ethinylestradiol CPA plus ethinylestradiol likely results in little to no difference in effect on FG scores at six months compared with finasteride plus ethinylestradiol (MD -0.91, 95% CI -1.82 to 0; 1 RCT, 26 women; moderate-certainty evidence) or spironolactone plus drospirenone plus ethinylestradiol (MD 0.69, 95% CI -0.80 to 2.18; 1 RCT, 89 women; high-certainty evidence). However, CPA plus ethinylestradiol may lower FG scores compared to spironolactone plus ethinylestradiol, but the evidence is very uncertain (MD -0.93, 95% CI -1.68 to -0.19; 3 RCTs, 103 women; very low-certainty evidence). There was probably little to no difference in effect for CPA plus ethinylestradiol on total testosterone at six months compared with spironolactone plus ethinylestradiol (MD -0.20 nmol/L, 95% CI -0.64 to 0.24; 1 RCT, 45 women; moderate-certainty evidence), and likely results in little to no difference compared with drospirenone plus ethinylestradiol (MD 0.15 nmol/L, 95% CI 0.08 to 0.22; 1 RCT, 91 women; moderate-certainty evidence). Six months of CPA plus ethinylestradiol treatment likely lowered free testosterone levels compared to drospirenone plus ethinylestradiol (MD -0.31 pmol/L, 95% CI -0.53 to -0.09; 1 RCT, 91 women; moderate-certainty evidence). Data were lacking for all other outcomes in our main review comparisons. AUTHORS' CONCLUSIONS: There were some differences in clinical outcome between CPA and spironolactone, CPA and flutamide, and CPA and finasteride. There were no clinical differences between CPA and the other medical therapies, possibly because of small study size, lack of standardised assessment and objective determinants of improvement in many studies. There are insufficient data presented to compare the adverse effects of all the treatment options. Larger, carefully designed studies are needed to compare efficacy and safety profiles between the therapeutic options available.