A Systematic Review of Intravaginal Testosterone for the Treatment of Vulvovaginal Atrophy

Robin J. Bell, MBBS, PhD, MPH, FAFPHM; Farwa Rizvi, MBBS, MPH; Rakabul M. Islam, MPH, PhD; Susan R. Davis, MBBS, PhD, FRACP


Menopause. 2018;25(6):704-709. 

In This Article


Of the 1,922 nonduplicate citations reviewed, 8 were considered appropriate for inclusion and 1,914 were excluded after review of titles or abstracts. Of the eight manuscripts reviewed in detail, one was subsequently excluded, as the aim of treatment was to achieve a systemic outcome as opposed to local vaginal effects.[15] This resulted in a total of seven papers included and 1,915 excluded (Figure 1). The seven publications included two from the same study;[16,17] therefore six separate clinical trials were identified. The included studies ranged in size from 10 to 80 participants, with treatment durations of 1 day to 12 weeks. Three studies included women without breast cancer,[16,18,19] and three included women with breast cancer taking an AI (Table 1).[20–22]

Figure 1.

Flowchart for the selection of included papers.

Findings From Studies of Intravaginal Testosterone in Women Without Breast Cancer

Raghunandan et al[18] compared intravaginal conjugated equine estrogens (CEEs) or CEE plus testosterone with a water-based vaginal lubricant gel. Fernandes et al[16] randomized women to intravaginal lubricant gel, CEE, polyacrylic acid, or testosterone cream. Both studies had a 12-week treatment period. Raghunandan et al asked women to insert the study drug twice a week, whereas women in the study by Fernandes et al inserted their allocated therapy three times each week.

Across both studies, improvement in the vaginal maturation index was only seen for women allocated to CEE[17,18] (Table 2). Estrogen or testosterone alone, and combined therapy were associated with significant improvements in vaginal pH,[17] the proportion of vaginal lactobacilli,[17] and the vaginal health score,[17,18] compared with the lubricant gel.

Raghunandan et al reported a significantly greater improvement in the total sexuality score for each treatment arm from baseline. Although the greatest improvement was seen in the CEE plus testosterone treatment group, no between-group differences were reported for the change in the sexuality score.[18] Fernandes et al reported the impact of intravaginal testosterone on sexual function, assessed by the Female Sexual Function Index (FSFI), in a separate paper.[16] Intravaginal testosterone, polyacrylic acid, and lubricant, but not estrogen, were associated with an improvement in the total FSFI score from baseline. However, it is not possible to discern from the analyses presented whether any one therapy was more effective than the others. This is because of the substantial differences in the baselines total FSFI scores of the treatment groups, with no adjustment for baseline differences.[16] The same issue applied to each of the FSFI subdomain scores. The lubricant group was reported as having a significant improvement in the total FSFI score from baseline 13.1 ± 9.0 to 12 weeks 15.8 ± 10, whereas the CEE group were reported as having no significant improvement from baseline (12.7 ± 10.1) to 12 weeks (18.2 ± 13.0).

An increase in serum estradiol, measured by radioimmunoassay, was seen in the treatment arms that included CEE in both studies.[17,18] Raghunandan et al[18] reported a significant increase in free testosterone in the group treated with CEE plus testosterone. A double-blind, cross-over pharmacokinetic study of high-dose intravaginal testosterone compared with placebo, was undertaken by Apperloo et al[19] in premenopausal women. The study showed absorption of testosterone with corresponding increases in total and free testosterone into the supraphysiological female range, but no measurable change in serum estradiol.

Intravaginal Testosterone Therapy in Women With Breast Cancer Taking an AI

Three open-label studies reported on the effects of intravaginal testosterone therapy in women with breast cancer concurrently taking an AI (Table 3). Witherby et al[20] compared the effects of two doses of intravaginal testosterone cream, 300 mcg and 150 mcg applied daily for 28 days. They reported a statistically significant improvement in the vaginal maturation index with the higher dose of testosterone.[20] Significant improvements were also seen for the total symptom score and dyspareunia, from baseline, with no between-dose differences (Table 3). All participants in the study by Witherby et al[20] had low baseline estradiol levels, measured by organic solvent extraction followed by radioimmunoassay. Across the 28 days of the study, serum estradiol levels remained below 5 pg/mL in 18 of the 20 women, with two women having levels reported as 7 pg/mL.

Witherby et al[20] reported pretreatment testosterone levels to be less than 21 ng/dL in all participants, and serum levels remained at or below 42 ng/dL, in all except one woman.

Dahir and Travers-Gustafson[22] reported on the effects of daily 300 mcg of testosterone for 28 days compared with baseline. They reported that daily use of intravaginal testosterone was associated with significant improvements from baseline in the total FSFI and the subdomains of desire, arousal, lubrication, orgasm, satisfaction, and pain over 28 days.[22]

Melisko et al[21] investigated the intravaginal application of 5,000 mcg of testosterone compared with a estradiol-releasing vaginal ring, releasing 7.5 mcg estradiol/day, over 12 weeks. Using a testosterone dose that was more than 10-fold the dose used in the Witherby et al study, Melisko et al[21] reported significant improvements in vaginal rugae, pallor, petechiae, elasticity, and dryness from baseline that were similar to those seen in women treated with the estradiol-releasing ring. They also reported that the improvement on the single item Cancer Rehabilitation Evaluation System sexual satisfaction subscale was greater for women treated with the estradiol ring than testosterone, without adjusting for baseline differences.[21] Baseline serum estradiol levels, measured by tandem liquid chromatography-mass spectrometry (LCMS), were reported as elevated in 37% of the participants.[21] Some women were described as having transient elevations of serum estradiol and others had persistently elevated levels. One woman in the testosterone arm and 2 in the estradiol-ring group had elevated estradiol levels at 12 weeks.[21] Supraphysiological total testosterone levels were recorded in the Melisko et al[21] study, by LCMS in the majority of the women randomized to testosterone (mean 171 ng/dL and median 67 ng/dL at week 12).