Abstract
Intravaginal dehydroepiandrosterone (DHEA) is a locally metabolised estrogen and androgen precursor, licensed in 2018 in the EU for moderate to severe vulvovaginal atrophy in postmenopausal women. A literature search revealed four original trials suitable for appraisal, three evaluating change in dyspareunia or dryness as a primary outcome, one evaluated safety as a primary outcome. In two trials of 255 and 558 women without cancer, the benefit of placebo (nightly vaginal suppositories with a lipophilic base) was a 0.9 and 1 point reduction in dyspareunia as measured on a 3 point scale, an unvalidated outcome measure. With nightly DHEA, dyspareunia was reduced by an additional 0.4 points compared to placebo. When 464 women with gynaecological cancer were randomised, those using nightly plain moisturiser gel reported a reduction of ‘most bothersome symptom’ (either dyspareunia or dryness) of 1.5 points on a 3 point scale. Those using nightly DHEA reported an additional symptom reduction of 0.3 points. This is also an unvalidated outcome measure. Data evaluating the efficacy of DHEA over placebo is unconvincing and based on unvalidated primary outcome measures that also do not reflect the complex psycho-sexual and socio-cultural components of genitourinary menopausal symptoms. The efficacy and safety data excluded women taking systemic HRT, applies to postmenopausal, not perimenopausal, women and had relatively short follow up. It is important further independent trials use sophisticated and validated assessment tools to better establish the efficacy, safety and cost effectiveness of intravaginal DHEA in clinically representative groups of women before being routinely prescribed.
Introduction
Vulvovaginal atrophy (VVA) is common in postmenopausal women and can present with vaginal symptoms such as dryness, dyspareunia, itching and irritation, affecting the sexual health and quality of life in up to two thirds of postmenopausal women.1–3 Women can also report urinary symptoms, termed ‘genitourinary syndrome of menopause’ (GSM) 4 and may present to both primary and secondary care. Treatment of VVA includes pharmacological and non-pharmacological options, with topical estrogen therapy commonly used. However, not all women respond to vaginal estrogen and some may be advised against using it or are intolerant. Intravaginal dehydroepiandrosterone (DHEA, prasterone), is a nightly pessary, converted intracellularly to estrogens and androgens exerting its effect locally. The following review evaluates available efficacy and safety data for intravaginal DHEA, approved in Europe in 2018. 5
Methods
Search strategy
A search strategy (supplementary Figure 1) used The Healthcare Databases Advanced Search tool 6 to search nine databases (Medline, PubMed, BNI, EMBASE, CINAHL, EMCARE, Psychinfo, AMED, HNIC) finding papers evaluating intravaginal DHEA.
Results
Once duplications were removed, there were 72 results and 34 relevant papers. Case reports, editorials and non-English language papers were excluded. Original studies in human subjects and reviews were considered and trial data was reviewed through the ClinicalTrials.gov website, 7 a database of international privately and publicly funded clinical studies.
Critical appraisal of trials
Summary of RCT trials, adequate quality trials highlighted.
A phase III placebo-controlled double blind prospective RCT (included in Table 1) conducted by Alliance for Clinical Trials in Oncology, funded by the National Cancer Institute, was described by Barton et al. 14 Women reporting moderate to severe dyspareunia or vaginal dryness who had received treatment for breast or gynaecological cancer were randomised to three arms; vaginal plain moisturiser (PM) as placebo, DHEA 0.25% and DHEA 0.5% nightly gel for 12 weeks. 16% of women in each group were taking Tamoxifen and 55% in each group were taking aromatase inhibitors (AI).
Significant numbers of participants were removed from the analysis after randomisation in ERC-210 and ERC-234 jeopardising the validity of the study; hence, their results are not discussed in this review. Follow up rates in ERC-234 were also very different between placebo and intervention groups (66% vs 90%). The open label study efficacy data was not included due to the lack of placebo group. 12 Participants were from American and Canadian study locations, a mean age of 57 to 59 and at least 90% white Caucasian in each study. Baseline demographics of intervention and placebo groups were similar in all trials.
Inclusion and exclusion criteria for ERC-231 and ERC-238 included findings on vaginal smear. This may result in an atypical trial population as microscopy is not routinely performed in clinical practice. Women with complex medical histories or previous cancer were excluded from the ERC studies limiting generalisation of findings to these populations.
Barton et al. did not exclude women with systemic illness. 14
Primary outcomes in ERC trials included % parabasal and superficial cells (maturation value, indicating levels of estrogen effect), pH values, a questionnaire and observations at vaginal examination. It is important to note however that measurements such as maturation value are not reflected in clinical symptoms. In a study by Davila et al., 15 there were no or weak associations between symptom scores and maturation value, vaginal health at clinical examination and age in both symptomatic and asymptomatic women. The same study did demonstrate a moderate correlation between vaginal health on examination and maturation value and age. Changes to outcome measures reported in ERC trials such as maturation index and pH have not been considered in this review as they are not routinely used in routine clinical practice to assess treatment efficacy and are therefore less relevant than symptom scores.
Only one question from the Atrophy Symptom Questionnaire 15 was used as a primary outcome in ERC trials. Whilst this symptom questionnaire as a whole has demonstrated repeatability and correlation with other symptom scores, 16 both ERC-231 and 238 used only one question from this five question symptom score as a primary outcome. This has not been validated, causing concern when drawing conclusions from this data. Given the potentially complex psychological, social and cultural factors behind symptoms such as dyspareunia it may also be difficult to adequately represent patient symptoms using such a brief assessment tool.
In Barton et al.’s trial, the primary outcome measure was change to bothersome symptom of dryness or dyspareunia with secondary outcome measures of Female Sexual Function Index (FSFI) 17 and a quality of life (QOL) measure. 18 The later measures are objective and validated.17,19 The authors note the primary outcome measure was chosen for being comparable to previously discussed data in ERC trials. Drop-out rates were comparable for placebo (81%) and DHEA 0.5% (75%). 14
Efficacy of DHEA on symptoms of VVA
Statistically significant improvement in dyspareunia has been demonstrated with both placebo (vaginal moisturiser) and DHEA 0.5% in patients with and without a previous diagnosis of cancer. A clinically significant improvement in symptom score was seen across trials with placebo, with self-rated symptoms improving by between 0.9 and 1.5 out of 3.8,9,14
In women without a previous diagnosis of cancer, improvement in dyspareunia with DHEA seems clinically unconvincing, with an improvement in symptom score of between 0.35 and 0.37 out of 3 above placebo, see Supplementary Table 1, irrespective of age, time since menopause and previous use of hormone replacement therapy 20 As discussed, the outcome measure is also not validated which limits the meaning of the results.
In Barton et al.’s trial of women with a history of a gynaecological cancer, an additional 0.3 point reduction above the reduction seen with placebo was observed with DHEA for the primary outcome of ‘most bothersome symptom’ (vaginal dryness or dyspareunia).
*A score of 26 is reported to be a cut-off point to distinguish sexual dysfunction (lower scores indicate sexual dysfunction). 21
Significant numbers of these women were also using medications such as aromatase inhibitors which may cause more severe GSM symptoms compared to those not taking these medications. Thus, conclusions on impact on symptoms from this trial are limited to this specific patient group.
There was a decrease from baseline for most bothersome symptom in all arms of the trial, with no significant difference between arms at 12 weeks. 14
It should be noted only women receiving DHEA reported post-treatment mean scores in the FSFI normal range (≤3) in the dyspareunia subscale. 21 Further analysis of vaginal dryness and dyspareunia showed intravaginal DHEA did not affect one symptom more than the other. 14
Safety
Systemic levels of sex steroids was a secondary outcome measure in ERC-231, 238 and Barton et al., and a primary outcome measure for the safety study ERC-230, as a surrogate marker of safety. Levels increased but remained within normal postmenopausal limits with intravaginal DHEA see Table 3. This is perhaps not conclusive of safety, however, as it is not a measure of adverse clinical diagnoses, relevant when considering the possibility of intravaginal DHEA playing a role in hormone sensitive malignancies. These trials specify normal postmenopausal levels of sex steroids, reflected in its license for postmenopausal, rather than perimenopausal, women whose baseline hormone levels may be different. Women taking systemic hormone replacement therapy (HRT) were excluded from the ERC trials, limiting the applicability of this safety data to women taking systemic HRT.
Change in sex steroid levels after 52 weeks of nightly intravaginal DHEA in 429 participants.
Free testosterone ranges from 30–190pg/mL. 23
A pooled analysis of ERC-210, 230, 231 and 238 demonstrated serum sex steroid data in DHEA groups (n = 723) was increased from baseline and relative to placebo (n = 266) at week 12 but remained within normal postmenopausal values. 22
Barton et al. measured androgen concentrations in 345 participants 24 finding they were significantly increased in intravaginal DHEA users compared to placebo but remained within the normal postmenopausal range.
Neither estradiol nor oestrone concentrations changed in women taking AIs, though they experienced the same level of improvement in vaginal symptoms as those not on AIs. The short follow up period of 12 weeks is insufficient to demonstrate safety outcomes such as recurrence of cancer however.
Endometrial safety was evaluated in 457 women at baseline and after receiving DHEA 0.5% for 52 weeks. 421 (92%) women had endometrial atrophy or inactive endometrium, 36 (8%) women had insufficient tissue for diagnosis. 12
Side effects
The most common adverse effect in ERC trials was application site discharge occurring in >5% women. Less than 1% women discontinued DHEA due to this. 23 In >1% women, increased adverse effects compared to placebo were weight fluctuations and cervical dysplasia (1.8% vs 1.3%, rates consistent with postmenopausal women smear abnormalities 25 ). 5 Barton et al. found statistically significant differences occurred in women receiving DHEA reporting worsening voice changes and, in the DHEA 0.25% group only, headaches. 14
Comparison to standard therapy (intravaginal estrogen)
There is more robust data on the benefit and risks of intravaginal estrogen therapy. A Cochrane meta-analysis in 2016 included 30 RCTs (6235 women) and compared efficacy and safety of different intravaginal estrogenic preparations with each other and placebo. 26 This showed improvement in symptoms with intravaginal estrogen compared to placebo, and no evidence of difference in adverse events between estrogenic preparations and placebo. Vaginal estrogen tablets had an odds ratio of 12.5 compared to placebo in participant assessed symptom improvement, in 841 vs 797 women. The main trial 27 in this part of the systematic review found in 659 women with dyspareunia, symptoms improved in 10% women using placebo, and 34% women using vaginal estrogen tablets, demonstrating benefit to an additional 24% women compared to placebo.
Prescribing intravaginal DHEA over standard therapy is more expensive. The monthly cost of intravaginal estrogen tablets is £5 per month and £16 per month for intravaginal DHEA. Over one year this equates to a £65 per year for intravaginal estrogen vs £336 for intravaginal DHEA, a difference of £271. 28 Given the numbers of women requiring treatment for VVA there would be significant cost implications to routine prescription of intravaginal DHEA.
Conclusion
Intravaginal DHEA is a new treatment that has been licensed for moderate to severe vulvovaginal atrophy. Trial data needs close scrutiny and this review has shown it is of limited quality, due to the use of an unvalidated primary outcome measure and significant differences in follow-up rates between placebo and intervention groups in some of the trials. Efficacy data regarding dyspareunia is especially limited because patient’s symptom scores may not correlate with clinical examination findings.5,29 To improve meaningful research it is important that more comprehensive symptom assessment tools are used to reflect the complex nature of VVA symptoms. 30
The data so far does not demonstrate a convincing enough clinical effect to warrant routine prescribing of intravaginal DHEA given the potential for exposing women to side effects and increased financial burden on healthcare systems. It is also important to note that women using systemic hormone replacement (HRT) were excluded from safety data.12,23 As women using systemic HRT are likely to have a different baseline sex steroid profile to the women assessed in these trials, our ability to apply trial safety data is limited. Women who have current and previous comorbidities such as previous VTE were also not included in these trials, and this should also be discussed carefully with patients as needed. Larger independent studies using comprehensive validated symptom scores with longer follow up are needed to establish the true benefit and safety of intravaginal DHEA for VVA.
In women with a history of hormone sensitive malignancy, data that vagina DHEA improves sexual health and quality of life measures is limited to one trial. 14 Intravaginal DHEA appears not to effect systemic estradiol levels in those on aromatase inhibitors 24 It could therefore be an option for these women suffering severe VVA symptoms of the menopause after a multidisciplinary discussion. Larger studies investigating recurrence of malignancies in this group to establish longer term safety for these women are needed.
The use of a 3 month trial of nightly vaginal moisturisers (likely to reproduce the statistically significant placebo effect found in all the trials discussed) is perhaps an underused treatment option that should be suggested before menopause specialists consider prescribing intravaginal DHEA. Intravaginal DHEA (prasterone) is locally metabolised to estrogens and androgens and is licensed for the treatment of moderate to severe vulvovaginal atrophy (VVA) in postmenopausal women. Available trial data shows minimal additional symptom improvement in dyspareunia with intravaginal DHEA over placebo, a several month trial of a nightly (as opposed to 2 to 3 times per week) vaginal moisturiser with a hydrating component could be tried initially. Further independent trials using more sophisticated, validated symptom scores in clinically representative populations of women are needed to establish more robust efficacy and safety data for intravaginal DHEA before it is routinely prescribed. Naumova I, Castelo-Branco C. Current treatment options for postmenopausal vaginal atrophy. International Journal of Women’s Health. 2018:10 387-395. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6074805/ Pitsouni E, Grigoriadis T et al. Efficacy of vaginal therapies alternative to vaginal estrogens on sexual function and orgasms of menopausal women: A systematic review and meta-analysis of randomised controlled trials. European Journal of Obstetrics and Gynaecology and Reproductive Biology. 2018:229 45-56. https://pubmed.ncbi.nlm.nih.gov/30103082/ Lundh A, Lexchin J et al. Industry sponsorship and research outcome. Cochrane Database of Systematic Reviews. 2017, Issue 2. Art No MR000033. DOI: 10.1002/14651858.MR000033.pub3 30 year old patient, previously using estrogens cream. ‘I’ve been using it for around a year, but I know I don’t use it frequently enough. It’s more messy than the [estrogen] cream, especially the next morning, it feels greasy and uncomfortable. It’s ok to use at night, but it would be better if the pessary was less oily. I probably use it 4 times a month at the moment. I don’t think it’s improved my libido or pain during sex, but it’s hard to know as I haven’t used as frequently as I should and there have been other things going on more broadly that might have affected those things. I feel that it’s more effective than the [estrogen] cream for dryness. One comment I would make is that the reason for using it is so it’s less painful and more enjoyable to have sex, but once you’ve used it that night you couldn’t have sex anyway due to the pessary itself and the unpleasant greasy feeling that it leaves, which seems counterproductive. So in summary it’s ok but I should use it more frequently to have a more definitive view of the benefits’.Practice Points
Additional educational resources
Patient account
Supplemental Material
Supplemental Material - Intravaginal dehydroepiandrosterone for genitourinary symptoms of the menopause: Is the evidence sufficient?
Supplemental Material for Intravaginal dehydroepiandrosterone for genitourinary symptoms of the menopause: Is the evidence sufficient? by Katherine Kearley-Shiers, Debra Holloway, Janice Rymer¸ Deborah Bruce in Post Reproductive Health
Footnotes
Acknowledgements
I would like to thank the patient who has provided her account for this review
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Guarantor
KKS
Authors’ contribution
KKS performed the literature search and wrote the first draft of the review. DB edited the article. All authors reviewed and edited the manuscript and approved the final version.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
