Patients undergoing treatment for breast cancer with ovarian suppression and aromatase inhibitors experience significant hormonal side effects due to hormonal deprivation, such as vaginal dryness and loss of libido. Our study demonstrated the feasibility of delivering low doses of systemic testosterone to improve sexual function without increasing estradiol levels.

The treatment of vaginal atrophy and loss of libido remains a significant challenge. The safety of intravaginal testosterone use has been evaluated in women with breast cancer. Overall, the studies demonstrate no significant effect on serum estradiol levels, despite some degree of systemic testosterone absorption [6, 7]. The improvement associated with its application, however, was mostly limited to local effects, such as vaginal atrophy and sexual function, with no significant impact on systemic effects of ovarian suppression, such as hot flashes, metabolic changes, sleep disturbances, and irritability. Similarly, the use of low-dose intravaginal estriol (0.005%) has also been evaluated in women with breast cancer undergoing adjuvant treatment with AI, with evidence of safety in this replacement therapy, as there was no detection of increased serum estradiol levels or hormonal levels of FSH and LH in treated patients [8]. However, the benefit is also limited to the local application area. Furthermore, the methodology used for analyzing serum estradiol levels was not homogeneous among the studies, limiting the sensitivity of the methods for detecting low concentrations of estradiol. Currently, it is known that the measurement of estradiol levels in these patients should ideally be performed using mass spectrometry. This technique is highly sensitive and provides more accurate results as there is no interference or interaction of other molecules in the assay, such as biotin, leading to more precise results. Therefore, MS is the preferred technique for evaluating breast cancer patients undergoing ovarian suppression combined with AI [9].

Several studies evaluated the impact of hormone replacement in patients with a history of breast cancer. The HABITS study, prematurely terminated due to alarming results, showed that after a 2-year follow-up, there was a more than 2-fold increase in the risk of breast cancer recurrence in women who received hormone replacement therapy with estrogen and progestogen-based compounds [12]. The Stockholm study also evaluated hormone replacement therapy in a similar setting and was halted due to the results of the HABITS study. However, in its update, recurrence rates were not as high as those found in the HABITS study [13]. Regarding hormonal treatment in patients with breast cancer, The LIBERATE study evaluated the use of tibolone in improving vasomotor symptoms caused by hormone therapy [14]. In this study, despite symptom improvement, there was a significant increase in breast cancer recurrence in patients who used the medication.

More recently, a longitudinal study conducted in Denmark evaluated postmenopausal patients with luminal breast cancer who received or did not receive adjuvant hormone therapy and who used some form of systemic or vaginal hormone replacement. This study showed a 39% increased risk of recurrence in women using aromatase inhibitors and vaginal estrogen replacement [15]. However, the absence of significant impact on the mortality makes it unclear whether the small (or absent) systemic absorption by vaginal estrogen replacement, as demonstrated by several studies [8], could negatively impact outcomes in women on aromatase inhibitors, or if the findings merely reflect a group with higher risk of relapse versus patients on tamoxifen or no endocrine therapy.

Systemic testosterone replacement therapy may also pose risks as it is converted to estradiol by aromatase enzyme activity [6]. However, by using an AI, this risk may be mitigated. For this reason, patients were required to be on AI treatment for breast cancer to prevent the conversion of testosterone to estradiol [20]. Our study demonstrated the safety of topical testosterone gel replacement, with undetectable E2 levels measured throughout the 3-month treatment protocol. Furthermore, the data showed a significant improvement in sexual satisfaction symptoms continuously throughout the treatment, with statistical significance observed between the second and third month of replacement. This finding was supported by an increase in the total score of FSFI, with statistically significant benefits observed in almost all domains assessed by the questionnaire, which are not restricted to local effects related to vaginal atrophy.

Among the evaluated patients, there was a small percentage of treatment discontinuation, mainly due to logistical difficulties and the need for outpatient evaluation to complete the questionnaire and collect the testosterone. This percentage was influenced by the fact that the majority of the study took place during the COVID-19 pandemic, which caused anxiety among patients about going to the hospital, in addition to the restrictions imposed by the isolation of infected individuals and their contacts.

Furthermore, one patient experienced a local recurrence of breast tumor in the last month of treatment, requiring surgical intervention followed by systemic treatment. This patient had E2 measurements consistently below 2.7 pg/mL throughout the treatment and was already under observation for a suspicious lesion months before entering the protocol, indicating that her recurrence was not associated with the use of the treatment.

Despite the positive results regarding the safety of testosterone replacement and the significant improvement observed in the total scores and key domains evaluated by the FSFI during the treatment, our study has several limitations. Firstly, the number of treated patients is small due to the pilot nature of the study, aimed at determining the relative safety regarding serum estradiol levels. Secondly, the follow-up period is short to assess possible recurrences and long-term benefits since the patients were treated for only three months per the protocol. Thirdly, the absence of a placebo group precludes a formal analysis of the potential benefits of the medication. However, the data obtained justify the conduction of a larger study with participant blinding and the inclusion of a control group treated with placebo gel. Fourthly, our study did not measure serum testosterone for pharmacokinetic and pharmacodynamic evaluation due to logistical and cost limitations. Lastly, the study focused on the patients’ sexual function quality and did not include validated questionnaires for other domains of quality of life that may be related to hormonal suppression. Nevertheless, the primary outcome of the study was achieved, and promising results were obtained regarding symptom improvement.

The results obtained in the study justify the implementation of larger-scale protocols to assess the role of this treatment in improving the quality of life of breast cancer patients and its safety regarding hormonal levels and potential long-term breast cancer recurrence.