The analysis, led by Xuanling Li and colleagues at Longhua Hospital, Shanghai University of Traditional Chinese Medicine, pooled data from 14 clinical controlled studies involving 2,132 patients with diminished ovarian reserve. The team searched seven databases, including Scopus, Web of Science, PubMed, Springer, VIP Information, Wanfang and the China National Knowledge Infrastructure, for studies published as of July 5, 2026. They included both randomized controlled trials and non-randomized controlled trials in which patients received dehydroepiandrosterone, commonly known as DHEA, or transdermal testosterone gel before IVF. The protocol was registered on PROSPERO under identifier CRD420261438943, and the authors followed PRISMA guidelines throughout.

Diminished ovarian reserve is a major driver of female infertility, accounting for an estimated 10 to 30 percent of cases. It is typically diagnosed when follicle-stimulating hormone levels fall between 10 and 25 international units per liter, or when antral follicle counts drop below seven, or when anti-Müllerian hormone falls under 1.1 nanograms per milliliter. A nationwide survey in China from 2019 to 2020 found the detection rate of ovarian reserve dysfunction among women of childbearing age approaching 30 percent, and the condition has become one of the most common obstacles in assisted reproduction. In the United States, diminished ovarian reserve was the indication for 27 percent of IVF cycles in 2021, according to Centers for Disease Control and Prevention data.

The rationale for androgen priming rests on a growing body of reproductive biology. Androgen receptors are expressed at every level of the hypothalamic-pituitary-gonadal axis and in reproductive tissues including the uterus and placenta. Animal studies have shown that female mice lacking androgen receptors are subfertile with underdeveloped follicles, while moderate androgen supplementation promotes the development of preantral follicles. Androgens appear to act at multiple stages of follicular maturation: they help recruit follicles from the resting pool into the growing pool, induce expression of the follicle-stimulating hormone receptor on granulosa cells, enhance follicle sensitivity to FSH, and support oocyte maturation and ovulation. Because circulating androgen levels decline with reproductive age, some IVF centers have begun offering androgen pretreatment to patients with poor ovarian response, and a previous survey found that more than 40 percent of doctors in Europe and Australia prescribe off-label androgens to these patients.

The new meta-analysis confirmed the strongest evidence for one outcome in particular. In randomized controlled trials, androgen priming increased the number of retrieved oocytes by a mean difference of 0.79 eggs, with a 95 percent confidence interval of 0.50 to 1.09 and a p-value below 0.00001. In non-randomized studies, the effect on retrieved oocytes was not statistically significant, with a mean difference of 0.11. However, the non-randomized trials did show significant gains in secondary outcomes: the number of top-quality day-3 embryos rose by a mean difference of 0.47, the number of fertilized embryos increased by 0.78, and the clinical pregnancy rate more than doubled, with an odds ratio of 2.23 and a 95 percent confidence interval of 1.49 to 3.36. The authors caution that these non-randomized results carry a high risk of bias and should be treated as exploratory.

For the hormonal markers of ovarian reserve themselves, the picture was less encouraging. Across randomized trials, androgen priming produced no statistically significant changes in anti-Müllerian hormone, follicle-stimulating hormone, or antral follicle count, although the authors note that only two or three studies contributed data to each of these analyses, limiting statistical power. Similarly, randomized trials showed non-significant favorable trends for metaphase II oocytes, live birth rates, and miscarriage rates. The live birth odds ratio of 1.24 and miscarriage odds ratio of 0.39 both pointed in a potentially beneficial direction but did not reach significance, which the authors attribute to insufficient testing power from limited sample sizes.

The subgroup analyses revealed several patterns that may guide future clinical practice. DHEA pretreatment in randomized trials significantly increased retrieved oocytes and metaphase II oocytes and reduced miscarriage rates, while testosterone gel did not reach significance in these subgroups, though the authors emphasize that only three randomized trials tested the gel and no head-to-head comparison of the two supplements exists. Timing also mattered: two months of androgen priming before the IVF cycle yielded significantly more retrieved oocytes, whereas longer interventions did not show the same benefit, and one study using four months of treatment actually retrieved fewer eggs than its control group. Notably, patients with an average age over 40 appeared to benefit more, showing significant increases in both retrieved oocytes and metaphase II oocytes, a finding consistent with evidence that older women convert androgen precursors less efficiently and may respond better to direct androgen administration.

The choice of ovarian stimulation protocol also influenced outcomes. Subgroup analysis indicated that pairing androgen priming with gonadotropin-releasing hormone antagonist protocols produced significantly more retrieved oocytes than agonist protocols. The authors point out that antagonist regimens are already widely preferred for poor responders because they offer shorter stimulation times, lower gonadotropin doses, no risk of cyst formation, and reduced incidence of ovarian hyperstimulation symptoms. Sensitivity analyses confirmed that the pooled conclusions were robust, and funnel plots showed no obvious asymmetry, though formal tests for publication bias could not be run because fewer than ten studies contributed to each outcome.

The study has important limitations that temper its conclusions. Five of the included trials did not use random allocation, and eleven did not implement blinding, resulting in risk-of-bias ratings ranging from some concerns to serious. The included studies varied in diagnostic criteria, intervention doses, durations, and stimulation protocols, and several lacked detailed baseline data, preventing meta-regression to explore sources of heterogeneity. Safety data were also sparse: seven studies reported no adverse reactions and four did not state whether any occurred, while the remaining trials documented mild effects such as acne in 3.1 percent of patients, voice deepening in 1.9 percent, and hair loss in 1.3 percent after DHEA use. Long-term DHEA use can also cause stomach discomfort and dizziness, and testosterone gel carries a risk of secondary exposure to others despite its otherwise favorable safety profile.

The authors conclude that androgen priming with DHEA or transdermal testosterone gel may be a useful adjuvant for improving ovarian response in diminished ovarian reserve patients undergoing IVF, with increased oocyte retrieval as the main demonstrated benefit. Whether the therapy genuinely improves clinical pregnancy rates and live births remains unresolved, since only the lower-certainty non-randomized evidence showed significant pregnancy benefits. The research team calls for larger, higher-quality randomized trials with unified diagnostic criteria and standardized outcome definitions to determine the optimal supplement, dose, and duration, and to establish whether androgen priming can deliver on its promise of better fertility outcomes for this growing patient population.

Subject of Research: Effects of androgen priming with DHEA and transdermal testosterone gel on IVF outcomes in women with diminished ovarian reserve

Article Title: The improvement effect of androgen priming on fertility in patients with diminished ovarian reserve and its impact on in vitro fertilization outcomes: a systematic review and meta-analysis

Article References: Li, X., Sui, C., Zhang, B., Bai, Y., Zhao, Q., & Xu, L. (2026). The improvement effect of androgen priming on fertility in patients with diminished ovarian reserve and its impact on in vitro fertilization outcomes: a systematic review and meta-analysis. Frontiers in Endocrinology, 17, Article 1939223. https://doi.org/10.3389/fendo.2026.1939223

Image Credits: AI Generated

DOI: 10.3389/fendo.2026.1939223

Keywords: diminished ovarian reserve, androgen priming, DHEA, transdermal testosterone gel, in vitro fertilization, meta-analysis, oocyte retrieval, clinical pregnancy rate, ovarian reserve, reproductive endocrinology, GnRH antagonist protocol, follicle development