The naturally occurring polyphenol resveratrol may have beneficial hormonal effects in women with polycystic ovary syndrome (PCOS), according to the results of a new study.

Previous in vitro research by the lead author, Beata Banaszewska, MD, from the division of infertility and reproductive endocrinology, in the department of gynecology, obstetrics, and gynecological oncology, Poznan University of Medical Sciences, Poznan, Poland, and her colleagues, suggested that resveratrol may inhibit cell growth and reduce androgen production in rat theca-interstitial cells, which are implicated in excessive androgen production in PCOS in humans.

“Importantly, it seems that resveratrol actions on ovarian steroidogenesis are selective, given that it has no effect on progesterone production by theca cells,” the authors wrote in the October 18 online edition of the Journal of Clinical Endocrinology & Metabolism.

They conducted a placebo-controlled trial , randomizing 34 women with PCOS to 1,500 mg oral resveratrol or placebo daily, with clinical, endocrine and metabolic assessments performed at baseline and three months after initiating treatment. The mean age of the women was 27 years; their mean BMI was 27.1 to 27.6 kg.m2.

At three months, serum total testosterone, the primary outcome, had declined significantly by 23.1% in the resveratrol group (P = .01), but increased by a non-significant 2.9% in the placebo group. The reduction in serum testosterone was even greater among individuals with a lower body mass index ( J Clin Endocrinol Metab. 101: 3575–81, 2016. October 18. doi: 10.1210/jc.2016-1858 ).

Similarly, levels of dehydroepiandrosterone sulfate declined by a significant 22.2% in the resveratrol group (P = .01), but increased by a non-significant 10.5% in the placebo group, suggesting an effect on ovarian as well as adrenal androgen production.

“The magnitude of improvement of hyperandrogenemia observed in response to resveratrol is comparable to or greater than that found in response to OC [oral contraceptive] pills or metformin, with the exception of preparations containing cypretorone acetate, which are not available in the United States,” the authors wrote. They cited another study showing a 19% reduction in testosterone either with 12 months of treatment with the oral contraceptive pill or with metformin.

They also noted that while reductions in testosterone with metformin occur gradually over 3 to 6 months, their study showed a marked reduction in just three months.

Researchers also saw a significant 31.8% decline in fasting insulin (P = .007) and a 66.3% increase (P =.04) in the Insulin Sensitivity Index among patients treated with resveratrol.

Resveratrol was not associated with significant effects on BMI, ovarian volume, gonadotropins, lipid profile, or markers of inflammation and endothelial function. The women on placebo did show a significant reduction in ovarian volume, and increases in total and high-density lipoprotein cholesterol levels, prompting the authors to suggest that resveratrol may have prevented an increase in cholesterol that would otherwise have occurred.

Other than two patients on resveratrol reporting transient numbness, no other adverse events were noted.

While resveratrol, which is found in grapes, nuts and berries, is known to have anti-inflammatory, antioxidant, and cardioprotective properties, the authors said this was the first clinical study examining its effects in PCOS.

“Although identification of the mechanisms of action of resveratrol is not possible in this clinical trial, several possible mechanisms may be considered,” including a reduction of growth of theca cells, “and the improvement of insulin sensitivity with consequent reduction of insulin levels,” they wrote.

“Furthermore, given that insulin is known to stimulate androgen production in both ovarian and adrenal tissues, it is likely that the resveratrol-induced reduction of insulin observed in the present study may have contributed to a decrease of androgen levels,” they added.

RevGenetics provided the resveratrol for the study. The study was supported by the authors’ own institutions, and no conflicts of interest were declared.