Response to Dr Simes and Dr Snabes

I thank Dr Simes and Dr Snabes for the information regarding the longer-term safety study of testosterone (T) gel (LibiGel) clarifying that by February 2011 data from some 2800 women-years of exposure have been gathered giving an average per subject exposure of 12 months. I note that enrollment is expected to be complete in the first half of 2011 and women will be followed for 5 years. Very appropriately the women have cardiovascular risk factors that are commonly present in postmenopausal women including hypertension, smoking, and diabetes.

It is not clear to me whether the women are also supplemented with estrogen, nor could I find this information on the company website or in a recent abstract from the annual conference of the International Society for the Study of Women’s Sexual Health. However, I will assume not. Nevertheless, issues of safety involve both androgen and estrogen from the aromatization of testosterone. Current data support the ‘critical window’ hypothesis [Whitmer et al. 2011] such that early postmenopausal estrogen therapy (ET) protects against cardiovascular disease, dementia, and cancer, whereas ET in later life increases these risks [Shumaker et al. 2003] and continued ET from mid to later life offers no protection [Shumaker et al. 2003]. Time since surgical or natural menopause, or duration of any prior ET, may also have an important influence on the safety of supplemental T.

It would be useful to know whether the safety study includes accurate baseline studies of androgen activity using androgen metabolites, principally androsterone glucuronide (ADT-G). It is of note that at any given age there is a broad range of ADT-G amongst women [Labrie et al. 2006] and that risks from T supplementation may vary with endogenous T activity. It remains important to recall that women carefully diagnosed with hypoactive sexual desire disorder (HSDD) have ADT-G levels comparable to carefully assessed controls [Basson et al. 2010]. Recent research identifies certain hormonal milieus that predispose to, or conversely protect from, progression of carotid artery intima media thickness during ET [Karim et al. 2008]. Assessment of hormonal milieu during sex hormone treatment is advocated [Karim et al. 2008].

The expectation is that in some 5 years time much more will be known on the safety of marked alteration to the androgen:estrogen ratio in women. Compared with premenopausal women this ratio becomes highly nonphysiological in T-supplemented women. The theoretical concern is of course that when endogenous androgens are high, cardiovascular risk increases [Wild, 2007].

The major issue with any hormonal supplementation for sexual dysfunction is that the duration of requirement is indefinite: women continue their sexual lives as long as there is an available sexually competent partner. If lack of androgen activity is assumed to be causing the disorder, this lack is only going to increase with age, given the continued reduction of intracrine production of androgen [Labrie et al. 1997]. Although a 5-year study of women with cardiovascular risk factors is an excellent start, the statement that ‘determination of the long-term cardiovascular and breast safety of testosterone will be known once the blind is broken and the data analyzed, which should occur in the next twelve months’ is misleading. The company will know in the next 12 months the safety of exposure to 12 months of supplemental T in a cohort of women with variable prior exposure to postmenopausal ET and variable (possibly unknown) endogenous androgen activity. In a further 5 years they will have data on exposure of a proportion of the initial 2800 women who continued T for the full 5 years. The debacle over postmenopausal hormonal therapy in general stemming from the Women’s Health Initiative study in 2002 [Clarkson, 2008; Writing Group for the Women’s Health Initiative Investigators, 2002] encourages caution and it encourages full discussion with our patients concerning the limited knowledge of the safety of hormonal therapy with testosterone.