2023 ANNUAL MEETING

A multicentre RCT has found no significant difference in the number of total, mature or fertilised oocytes by the pre-treatment administration of transdermal testosterone in likely poor responders. Nor did it improve cycle outcomes such as cancellation rate or CPR, and the rate of overall androgenic adverse events, such as hirsutism or acne, was higher in the testosterone-exposed group.

The findings from this T-TRANSPORT study were presented to a packed congress hall in a session on ‘new concepts’ in poor responders by Nikolaos Polyzos from Dexeus University Hospital in Barcelona, and seem likely to hit the final nail in the coffin of androgen pre-treatment in this challenging group.

Based on 290 women with poor ovarian reserve according to the Bologna criteria, this triple-blinded RCT run in ten centres from Spain, Belgium, Denmark and Switzerland, assessed the efficacy of transdermal testosterone gel applied prior to ovarian stimulation for ART.

The T-TRANSPORT study seemed justified by results from animal models which reported an increase in the number of ovarian follicles following exposure to testosterone and via a mechanism involving an increase in FSH receptors. In humans, dehydroepiandrosterone (DHEA) consequently became the androgen of choice, but the most recent meta-analysis appeared to support the use of testosterone and not DHEA.(1) In this context Polyzos somewhat ruefully commented on the ever-so-frequent paradox of publishing disproportionally more meta-analyses than primary studies, with tewn meta-analyses for DHEA based on 12 RCTs and seven meta-analyses for testosterone based on seven RCTs.

The power calculations for the T-TRANSPORT study targeted 400 participants; however, the pre-defined interim-analysis allowed for an early close of the trial if futility was confirmed when reaching 70% of the target. The study population had a mean age of 38 years, AMH level just under 0.8 ng/ml, and AFC under 5, which describe the typical poor responders routinely met today.

While outcomes related to the mean numbers of total oocytes (3.7 vs 3.4), mature oocytes (2.6 vs 2.6), and fertilised oocytes (2.2 vs 1.8) were similar between the testosterone and control groups, it was surprising to observe a higher treatment cancellation rate in the testosterone group (24.4%) than in the control group (13.5%). No significant differences were reported in terms of reaching the stage of oocyte collection (63% vs 70%) or having an embryo transfer (49.6% vs 52.9%).

Clinical pregnancy represented the most relevant outcome and was similar between the two groups (16.4% vs 17.5%), with no important differences in various age groups, such as under 35 (17.9% vs 24%), 36-39 (20.4% vs 21.1%) or 40+ (12.3% vs 10.5%).

One comment raised by the audience was the relatively low dose of testosterone (5.5 mg) compared with those used in previous studies, to which Polyzos answered that treatment aimed ‘to make women more fertile, not turn them into men’. Further, in the absence of a direct clinical benefit in terms of reproductive outcomes, Polyzos focused on the potential harm of testosterone treatment in terms of adverse events and reported a higher incidence of hirsutism, alopecia and acne in women exposed to testosterone.

The presentation concluded with the promise of further data analysis to answer relevant questions related to follicular fluid and cumulus cells on the day of oocyte collection, the effect of testosterone on ovarian reserve markers, and consequences for female libido and quality of life.



1. Neves AR, Montoya-Botero P, Polyzos NP. Androgens and diminished ovarian reserve: the long road from basic science to clinical implementation. A comprehensive and systematic review with meta-analysis. Am J Obstet Gynecol 2022; 227: 401-413.
doi: 10.1016/j.ajog.2022.03.051.