Abstract
Background: Women with complete androgen insensitivity syndrome (CAIS) after gonadectomy have complained about reduced psychological wellbeing and sexual satisfaction. The aim of this study was to compare the effectiveness of hormone-replacement therapy with either androgen or oestrogen in women with 46,XY karyotype and CAIS after gonadectomy. Methods: This national, multicentre, double-blind, randomised crossover trial was performed at three university medical centres and three specialised treatment institutions in Germany. Eligible participants were women aged 18–54 years with 46,XY karyotype, genetically diagnosed CAIS, and removed gonads. Participants were randomly assigned (14:12) by a central computer-based minimisation method to either oestradiol 1·5 mg/day for 6 months followed by crossover to testosterone 50 mg/day for 6 months (sequence A) or to testosterone 50 mg/day for 6 months followed by crossover to oestradiol 1·5 mg/day for 6 months (sequence B). Participants also received oestradiol or testosterone dummy to avoid identification of the active substance. All participants received oestradiol 1·5 mg/day during a 2 months' run-in phase. The primary outcome was mental health-related quality of life, as measured with the standardised German version of the SF-36 questionnaire. Secondary outcomes were psychological wellbeing, as measured with the Brief Symptom Inventory (BSI), sexual function, as measured with the Female Sexual Function Index (FSFI), and somatic effects, such as signs of virilisation and effects on metabolic blood values. The primary analysis included all patients who were available at least until visit 5, even if protocol violations occurred. The safety analysis included all patients who received at least oestradiol during the run-in phase. This trial is registered with the German Clinical Trials Register, number DRKS00003136, and with the European Clinical Trials Database, number 2010-021790-37. Findings: We enrolled 26 patients into the study, with the first patient enrolled on Nov 7, 2011, and the last patient leaving the study on Jan 23, 2016. 14 patients were assigned to sequence A and 12 were assigned to sequence B. Ten participants were withdrawn from the study, two of whom attended at least five visits and so could be included in the primary analysis. Mental health-related quality of life did not differ between treatment groups (linear mixed model, p=0·794), nor did BSI scores for psychological wellbeing (global severity index, p=0·638; positive symptom distress index, p=0·378; positive symptom total, p=0·570). For the FSFI, testosterone was superior to oestradiol only in improving sexual desire (linear mixed model, p=0·018). No virilisation was observed, and gonadotrophin concentrations remained stable in both treatment groups. Oestradiol and testosterone concentrations changed substantially during the study in both treatment groups. 28 adverse events were reported for patients receiving oestradiol (23 grade 1 and five grade 2), and 38 adverse events were reported for patients receiving testosterone (34 grade 1, three grade 2, and one grade 3). One serious adverse event (fibrous mastopathy) and 20 adverse events (16 grade 1 and four grade 2) were reported during the run-in phase, and 12 adverse events during follow-up (nine grade 1 and three grade 2). Interpretation: Testosterone was well tolerated and as safe as oestrogen for hormone-replacement therapy. Testosterone can be an alternative hormone substitution in CAIS, especially for woment with reduced sexual functioning. Funding: German Federal Ministry of Education and Research.
| Original language | English |
|---|---|
| Journal | The Lancet Diabetes and Endocrinology |
| Volume | 6 |
| Issue number | 10 |
| Pages (from-to) | 771-780 |
| Number of pages | 10 |
| ISSN | 2213-8587 |
| DOIs | |
| Publication status | Published - 01.10.2018 |
Funding
Clinical trials in rare diseases pose a great challenge, especially in medical treatment. Small numbers of patients and heterogeneous groups of diagnoses impede the generation of statistically significant evidence of efficacy. In this study we gained prospective data of a relatively homogeneous group that was well defined by genetic diagnosis. This trial shows the difficulty of recruiting patient cohorts that are large enough to reach high statistical power in rare diseases. Considering the rareness of CAIS, the study cohort was adequate in size and provides valuable information. With slightly reduced sample sizes and enlarged drop-out rates, the power of efficacy analyses was somewhat reduced. Because of the small sample sizes, the reported exploratory pairwise comparisons of visits are unadjusted for multiple testing. Nevertheless, for the first time, reliable data from a prospective randomised study in the field of disorders of sex development were obtained and trends were visible. Sexual satisfaction was markedly reduced in the study cohort. Higher sexual dysfunction in patients with CAIS seems plausible when compared with healthy controls. 18 Hormone replacement has an effect on sexual functioning in CAIS, but other factors contribute to poor outcomes. The approach to disclosure, social support, and contact to the community for disorders of sex development affects the coping process. Testosterone treatment seems to have a stronger effect on sexual functioning in patients with CAIS than oestrogen treatment does, but the difference was significant only for the desire domain of sexual functioning. Oestradiol treatment did not have any effect on sexual functioning. Although applying an advanced statistical model to rule out confounding variables, testosterone treatment could not be proven to be superior in all aspects of sexual functioning. Yet scores in the FSFI were higher during testosterone treatment than during oestradiol treatment. Central regulation of sexual functioning is complex and involves various distinct brain sites. In both sexes, oestradiol and testosterone affect responsiveness to sexual stimuli and sexual behaviour. 26 However, assuming that the AR gene is non-functioning in patients with CAIS, how might testosterone improve sexual functioning? One possibility, as shown in rodent models, is that site-specific conversion of testosterone in the brain by steroidogenic enzymes such as aromatase or 5-α-reductase into oestradiol and 3α-androstanediol, respectively, are involved in activating sexual behavior, 27 hereby allowing for preserved central effects of testosterone in women with CAIS, despite the absence of a functioning AR gene. Desire is a relevant domain of sexual functioning in patients with CAIS, especially when sexual activity is practiced without a partner. This was the case in some of the study participants. Lubrication, pain, and satisfaction with emotional aspects of partnership are rather unsuitable scales to analyse in these circumstances. Unfortunately, no questionnaires are specific for disorders of sex development. The FSFI is a validated and widely used tool but provides an informative basis mostly for patients in a relationship. Baseline scores in sexually active patients ( appendix ) were higher than in the whole cohort. Yet mean values were below the cutoff value, indicating a risk of sexual dysfunction for many of these participants. This finding underlines the importance of endocrine assessment and psychosexual counselling for patients with CAIS to initiate hormone treatment and improve coping strategies. The SF-36 also is a common patient-reported outcome instrument in clinical studies. The mental health summary score encompasses aspects of vitality, social functioning, role limitations, and psychological wellbeing. MHRQoL was significantly reduced in the study cohort compared with healthy controls, whereas physical HRQoL was significantly increased. Similar results were found in the German clinical evaluation study 28 and in a national registry-based study from Denmark. 29 Additionally, participants had higher psychological distress, as measured with the BSI. 47% of the participants had to be classified as clinical cases before beginning the trial, corroborated by results reported by Bennecke and colleagues in 2017. 28 Some participants were not taking any medication before the study ( table 1 ). In some cases, uncertainty about the best hormone replacement led to refusal to take any medication. The lack of lifelong concepts of care and of experienced caregivers contributes to poor compliance with hormone replacement and difficulty in coping with the diagnosis. Positive effects on mental health, psychological wellbeing, and sexual functioning, to varying extents, could be detected when continuous hormone treatment was implemented with run-in medication. Reliable hormone substitution for patients with CAIS seems important for an improved quality of life. Continuous therapy could also lead to positive long-term effects, which were not measured in this study, such as preservation of bone mineral density. Bone mineral density impairment after removal of the gonads has been reported in patients with CAIS, but the underlying mechanisms are not understood. 30 Taking into account that testosterone was well tolerated and just as safe as oestrogen treatment, we conclude that testosterone can be an alternative hormone substitution for patients with CAIS, especially when sexual satisfaction is reduced. Complementary treatment with oestrogens (in case of low aromatisation) for potential beneficial effects on bone metabolism should be discussed. Long-term follow-up will be crucial to assess effects on physical and psychological wellbeing. Our findings also emphasise the need for life-long concepts of care for patients with CAIS to ensure they are given the best tailored hormone replacement and to reduce relevant psychological distress, improve sexual functioning, and improve mental health-related quality of life. Contributors WB, LM and OH planned the study and coordinated the study centers. WB, LM, KR, BK, AR-U, MKA, and OH were in charge of individual study centers, recruited patients, and did the study. RW did the molecular genetic studies. AK, P-MH, MFH, and SAW did the metabolic studies. AL and SK were in charge of statistical analysis. All authors contributed to the writing of the report. Declaration of interests We declare no competing interests. Acknowledgments This study was funded by the German Federal Ministry of Education and Research (BMBF), project no 01KG1003.
Research Areas and Centers
- Academic Focus: Center for Brain, Behavior and Metabolism (CBBM)
