Functional ovarian reserve in transgender men receiving testosterone therapy: evidence for preserved anti-Mullerian hormone and antral follicle count under prolonged treatment

STUDY QUESTION: Is the functional ovarian reserve in transgender men affected by testosterone therapy? SUMMARY ANSWER: Serum anti-Mullerian Hormone (AMH) levels slightly decrease during testosterone treatment but remain within the normal range, suggesting preserved follicular ovarian reserve. WHAT IS KNOWN ALREADY: Few small studies have investigated the impact of gender-affirming treatment on reproduction in transgender men. Conflicting results were reached concerning ovarian morphology and AMH levels in this context. STUDY DESIGN, SIZE, DURATION: The study consisted of two arms. The first arm was a prospective pilot study, which enrolled 56 transgender men (median age 22.5 [interquartile range (IQR)-19-27.7] years), 27 of whom had polycystic ovary syndrome (PCOS), prior to the initiation of gender-affirming testosterone therapy. A structured assessment was conducted prior to, and at 3 and 12months after treatment initiation. The second arm was a cross-sectional study that comprised 47 transgender men (median age 24 [IQR-20-31] years) who received testosterone for a median duration of 35 [IQR 13-62] months. The main outcome measures were serum AMH and antral follicle count (AFC) as indices of ovarian follicular reserve. PARTICIPANTS/MATERIALS, SETTING, METHODS: The study was conducted at a tertiary center for transgender health. Gender-affirming therapy was administered according to standard practice. AFC was determined by pelvic (abdominal or transvaginal) ultrasound and blood collection for measurements of AMH, testosterone, estradiol, LH and FSH was performed at the designated time-points. MAIN RESULTS AND THE ROLE OF CHANCE: Prospective arm for the entire group we observed a decrease of 0.71ng/ml in AMH levels between baseline and 12months (P=0.01). When expressed in age-specific percentiles, AMH went from the 47.37th to the 40.25th percentile at 12months (P<0.001). In a sub-group analysis, a decline of 9.52 points in age-specific percentile was seen in subjects with PCOS (P<0.001), while no changes were detected in the non-PCOS group. Testosterone treatment did not affect AFC over time in the entire cohort. In the sub-group analysis, a mean decrease of 5.0 follicles was detected between baseline and the 12months assessment (P=0.047) only in subjects with PCOS. In the cross-sectional study, AMH inversely correlated with age but not with treatment duration. Notably AMH did not deviate from the 50th age-specific percentile. Finally, four men fathered biological children after being under testosterone treatment for up to 12years. LIMITATIONS, REASONS FOR CAUTION: The limited sample size of the pilot study should be kept in mind. An additional limitation is the lack of a control group in the prospective study, as each participant served as his own control. Also, roughly 40% of the ultrasound examinations were performed transabdominally, potentially affecting the accuracy of the AFC measurements. As study participants were quite young, our reassuring data may not apply to older transgender men, either because of an age-related decline in ovarian reserve or to possible long-term effects of testosterone therapy. Furthermore, the chances for fertility preservation may be more limited in subjects with PCOS. WIDER IMPLICATIONS OF THE FINDINGS: This is an additional contribution to the emerging evidence that prolonged testosterone treatment may not be a major obstacle to later fertility potential in transgender men desirous of having children. Larger confirmatory studies, and particularly more with reproductive outcome data, are needed for evidence-based fertility counseling prior to treatment initiation in these subjects. STUDY FUNDING/COMPETING INTEREST(S): This study received no funding. The authors have no competing interests to declare.

Automated summary

The study found that testosterone therapy has a slight impact on the ovarian reserve of transgender men, but most participants maintained normal AMH levels. In subjects with PCOS, AMH levels decreased. AFC was not affected over time for the entire cohort.