Testosterone gel before IVF did not lift pregnancy rates in women with low ovarian reserve

Women with diminished ovarian reserve, meaning few eggs remaining, are among the hardest to help with IVF, and fertility clinics have reached for anything that might improve the odds. One popular add-on is testosterone, often prescribed off-label as a skin gel in the weeks before treatment, on the theory that priming the ovary with androgens makes its follicles respond better to stimulation. It is widely used despite thin evidence. A randomised trial has now tested it head-to-head against placebo, and found no benefit [s1].

What the trial tested

The study was a triple-blind, placebo-controlled, randomised trial, meaning participants, clinicians and analysts were all kept unaware of who received the drug [s1]. It recruited women aged 18 to 43 with infertility and diminished ovarian reserve by the standard Bologna criteria, at 10 fertility clinics across Europe between April 2015 and November 2022 [s1]. Of 316 women assessed, 290 were enrolled and randomised; after two exclusions tied to the onset of the COVID-19 pandemic, 288 were analysed, 134 assigned to testosterone and 154 to placebo [s1].

Participants applied 5.5 mg of transdermal testosterone gel, or a matching placebo, once daily for about nine weeks before ovarian stimulation [s1]. Everyone then had the same stimulation protocol and a fresh embryo transfer if an embryo was available [s1]. The primary outcome was the clinical pregnancy rate, defined rigorously as an intrauterine gestational sac with a heartbeat at seven weeks or more [s1]. The trial was registered as NCT02418572 [s1].

What it found

Testosterone did not help. A clinical pregnancy occurred in 21 of 134 women (15.7%) in the testosterone group and 23 of 154 (14.9%) in the placebo group, a risk ratio of 1.05 (95% confidence interval 0.61 to 1.81; P=0.86) [s1]. The two rates are, for practical purposes, identical.

The result was so clearly heading nowhere that the trial was stopped early for futility at a prespecified interim analysis, once 70% of the planned participants had been randomised [s1]. That is a notable decision: an interim look calculated that continuing to full enrolment had little chance of changing the answer, so the researchers halted it rather than expose more women to a treatment showing no signal [s1]. Stopping a trial for futility is a mark of discipline, not failure; it ends the study on the strength of its own data rather than on a hunch that more patients might rescue the result [s1].

Reading past the marketing

Two features make this trial worth taking seriously. The first is its rigour. Triple-blinding and a placebo control remove the main routes by which an ineffective add-on can appear to work, from patients' expectations to clinicians' hopes. The second is its funding. The trial was funded by Ferring, a pharmaceutical company with commercial interests in fertility treatment [s1]. Industry-funded studies more often report favourable results; here an industry sponsor funded a trial that returned a clearly negative one, which strengthens rather than weakens confidence in the finding [s1].

The result also fits the wider evidence. A 2024 Cochrane review examined DHEA and testosterone as add-ons in assisted reproduction, treatments proposed to boost the ovary's response to stimulation and, in turn, egg yield and pregnancy [s2]. The existence of that review, and the continued off-label use it describes, reflects how far these androgens have spread on hope ahead of proof, exactly the gap a trial like this is built to close [s1][s2].

How to read it

For a woman with low ovarian reserve being offered a testosterone gel before IVF, the practical takeaway is direct: in the best test so far, it did not raise the chance of a pregnancy with a heartbeat, and it was stopped early because it was going nowhere [s1]. That does not make the disappointment of diminished reserve any easier, but it spares women the cost, the daily application and the false reassurance of an ineffective add-on.

The caveats are narrow. This trial studied one androgen, one dose and one roughly nine-week schedule before a specific stimulation protocol, and a different regimen has not been excluded, though nothing here suggests one would fare better [s1]. Enrolment ran over several years and was cut short, which limits some secondary comparisons [s1]. And the finding is about diminished ovarian reserve specifically, not every difficult IVF case [s1].

The liftable summary: transdermal testosterone before IVF did not improve clinical pregnancy rates in women with diminished ovarian reserve, and the trial, funded by the drug's commercial field, was halted early for futility [s1]. The burden of proof now sits firmly with anyone still recommending it.

This article is informational and is not medical advice. IVF add-on treatments should be discussed with a fertility specialist.

Sources

• [s1] Transdermal testosterone gel vs placebo in women with diminished ovarian reserve prior to in vitro fertilization: a randomized, clinical trial. Nature Communications, 12 February 2026; 17(1). Funded by Ferring. ClinicalTrials.gov NCT02418572. https://doi.org/10.1038/s41467-026-69557-z
• [s2] Androgens (dehydroepiandrosterone or testosterone) for women undergoing assisted reproduction. Cochrane Database of Systematic Reviews, 5 June 2024. https://doi.org/10.1002/14651858.CD009749.pub3