Management of sexual dysfunction in breast cancer survivors: a systematic review

Female sexual dysfunction occurs frequently in midlife breast cancer survivors (BCS) and encompasses problems with sexual desire, interest, arousal, orgasm and genitopelvic pain. Although common, sexual problems are under-diagnosed and under-treated in BCS. The objective of this review was to assess primary studies that intervene on sexual dysfunction in BCS. In February 2015, PubMed, SCOPUS, CINAHL, COCHRANE and Web of Science databases were systematically searched for randomized controlled clinical trials (RCTs) of vaginal (lubricants, moisturizers, estrogens, dehydroepiandrosterone [DHEA], testosterone, vibrators, dilators), systemic (androgens, anti-depressants, flibanserin, ospemifene), physical therapy (physical activity, pelvic floor training), counseling and educational interventions on sexual function in BCS. Observational studies of vaginal interventions were also included due to the paucity of RCTs. The search yielded 1414 studies, 34 of which met inclusion criteria. Both interventions and outcomes, measured by 31 different sexual function scales, were heterogeneous, and therefore data were not pooled. The review found that regular and prolonged use of vaginal moisturizers was effective in improving vaginal dryness, dyspareunia, and sexual satisfaction. Educational and counseling interventions targeting sexual dysfunction showed consistent improvement in various aspects of sexual health. No consistent improvements in sexual health were observed with physical activity, transdermal testosterone or hot flash interventions. There was a lack of BCS-specific data on vaginal lubricants, vibrators, dilators, pelvic floor therapy, flibanserin or ospemifene. Overall, the quality of evidence for these studies was moderate to very low. Because each of the interventions with BCS data had limited efficacy, clinical trials to test novel interventions are needed to provide evidence-based clinical recommendations and improve sexual function in BCS.


Introduction
In the United States, there are more than 2.3 million female cancer survivors who are younger than age 60; 40 % of these women are survivors of breast cancer [1]. Most midlife breast cancer survivors (BCS) undergo surgery, chemotherapy, radiation and/or endocrine therapy for cancer treatment. Receiving a breast cancer diagnosis and undergoing associated treatments including long term endocrine therapy can impair sexual function via a number of mechanisms, including disrupting ovarian function, body image, intimacy and relationships [2][3][4][5][6][7]. In turn, impaired sexual function contributes to lower quality of life in survivorship [8,9].
Female sexual dysfunction has been classified into three categories: sexual interest or arousal disorder, orgasmic disorder, and genitopelvic pain or penetration disorder. A women is diagnosed with sexual dysfunction if she experiences persistent symptoms that last at least six months and cause marked distress, as detailed in the Diagnostic and Statistical Manual 5 th Edition (DSM-5) [10] (Table 1). A population-based cohort study of recently diagnosed BCS showed 65 % reported that they were sexually active; 52 % of sexually active women described problems with two or more areas of sexual function [11]. At 5 and 10 years after cancer diagnosis, prevalence of sexual problems remained significant, 26 and 19 %, respectively [12]. These findings that BCS are sexually active and experience sexual dysfunction that persists throughout survivorship have been replicated in multiple cohorts [9,[13][14][15].
Sexual health is often under-addressed in survivorship care, and only a minority of BCS receives information and education about sexual function from oncology professionals [16]. Among primary care providers at a university-based medical center, 62 % self-reported never or rarely discussing sexual issues with cancer survivors [17]. Providers who perceived having adequate preparedness to evaluate late effects or formal training in survivorship care were more likely to address sexual health considerations. Conversely, lack of knowledge in healthcare providers was a significant barrier to discussions on sex [18]. Moreover, patients may be reluctant or embarrassed to raise sexual concerns with healthcare providers [19]. Only 50 % of BCS thought their providers were knowledgeable about cancer care follow-up and even fewer (41 %) felt that their providers were equipped to treat their cancer therapy-related symptoms [20]. Hence, disseminating evidence-based information on managing sexual concerns to healthcare providers is a critical aspect of improving sexual health care after breast cancer.
Multiple pharmacologic and behavioral treatments have been tested to improve sexual health after breast cancer. We present a systematic review of primary research on managing sexual dysfunction in breast cancer survivors to generate evidence-based content for improving knowledge on sexual health for BCS and their healthcare providers.

Search strategy
This systematic review was conducted in accordance with PRISMA guidelines [21]. In February 2015, we systematically searched the following databases: PubMed (1966( -February 2015, SCOPUS (1966-February 2015, CINAHL (Cumulative Index to Nursing and Allied Health Literature) (1981( -February 2015, COCHRANE (all years), and Web of Science (1900( -February 2015. We screened the bibliographies of all included studies for additional references. We sought peer-reviewed articles examining interventions on sexual health among female BCS. We included studies on female breast cancer patients without age restriction and excluded studies on males, nonhumans and other female cancer patients. We included studies on sexual dysfunction, including problems with dyspareunia, sexual pain, vaginismus, vaginal dryness, sexual arousal, desire, and orgasm. For types of interventions, we included vaginal (lubricants, moisturizers, estrogens, dehydroepiandrosterone [DHEA], testosterone, vibrators and dilators), systemic (androgens, anti-depressants, flibanserin, ospemifene), physical therapy (physical activity, pelvic floor training), counseling and educational interventions. We did not include studies on systemic estrogen interventions. For physical therapy, systemic, and counseling

Outcome measures
The primary outcome of this systematic review was sexual function. Measures of sexual function varied widely among studies and are summarized in Table 2.

Data collection
Three review authors (SS, SD, IS) independently screened the titles and abstracts of all search citations using the inclusion and exclusion criteria. Discrepancies among authors were resolved via consensus. Two of the three review authors (SS, SD, or IS) independently abstracted data on included articles. Data extracted included participants, interventions, sexual health outcome measures, results, and risks of bias (randomization, allocation concealment, blinding, sample size and analysis approach). Risk of bias for all included studies was assessed independently by two review authors (SD and IS) using the Cochrane risk of bias assessment tool [22]. Discrepancies were resolved by discussion. Studies were evaluated for the following: selection bias (random sequence generation and allocation concealment); performance blinding (blinding of participants and personnel); detection bias (blinding of outcome assessment); attrition bias (incomplete outcome data); reporting bias (selective reporting); and other bias. Each bias criteria was assigned a high, low or unclear risk of bias rating. Additionally, we evaluated the quality of each study using the following GRADE criteria: study limitations (i.e., risk of bias); consistency of effect; imprecision; indirectness and publication bias. RCTs were first classified as high quality, and observational studies were first classified as low quality. All studies were downgraded in quality for any of the following problems: serious limitation to study quality; important inconsistency; uncertainty about directness; imprecise or sparse data; or high probability of reporting bias.

Results
After searching PubMed (n = 637), SCOPUS (n = 665), CINAHL (n = 276), COCHRANE (n = 220) and Web of Science (n = 186) and hand picking (n = 14), 1984 articles were retrieved, leaving 1414 articles after removing duplicates. Forty-two full-text articles were accessed, from which 8 were excluded, leaving 34 articles included in this review. The PRISMA flow diagram details study selection results (Fig. 1). No article was excluded because of non-English language.
A total of 31 different sexual health outcome measures were used to assess intervention effects across the 34 papers ( Table 2). The Female Sexual Function Index (n = 4 studies) and Cancer Rehabilitation Evaluation System (n = 3 studies) were the most commonly used measures. The Vaginal Maturation Index, Vaginal Health Index, and Sexual Activity Questionnaire were each used in 2 separate studies. All other outcome measures were used by single studies. Because of heterogeneity in both intervention and outcome measures, we were unable to pool estimates for a meta-analysis or derive strengths of recommendations based on the GRADE approach.

Vaginal products interventions
We searched for studies on vaginal lubricants, moisturizers, estrogens, DHEA, testosterone, vibrators and dilators. Eleven studies met inclusion criteria (Tables 3 and 4, Fig 2a). No studies were found on lubricants, DHEA, vibrators and dilators. There were 3 RCTs and 8 single-arm prospective cohorts with no controls. All participants had genitourinary symptoms, experienced ≥ 6 months of amenorrhea, and completed primary breast cancer treatment. The studies occurred in Australia, Belgium, Germany, Italy, Korea, and the United States. The polycarbophil-based moisturizer Replens® was tested in 4 studies involving 133 participants, one in combination with olive oil and pelvic floor muscle relaxation [23][24][25][26]; compounded testosterone cream was tested in 2 studies involving 34 participants [27,28]; pH balanced lactic acid gel was used in 1 study of 98 participants [29]; and vaginal estrogens were used in 5 studies involving 47 participants [24,[30][31][32][33]. Outcomes included patientreported vaginal symptoms, such as dryness, dyspareunia and itching, and vaginal exam-based pH and cytology.
In women using Replens® , vaginal dryness decreased in the first week of use [23], with significant additional improvement in dryness, dyspareunia, sexual satisfaction and frequency by 4 and 12 weeks of use [23,25,26]. Compared with local vaginal estrogens (estriol or estradiol), Replens® appeared less effective at decreasing vaginal symptoms and improving vaginal histology. However, women who used vaginal estrogens experienced an increase in their serum estradiol levels or decline in gonadotropins, both evidence of systemic absorption [24,[30][31][32][33]. At steady state, women on aromatase inhibitors using 25 microgram estradiol tablets twice weekly had low levels of serum estradiol (median 1.3 pg/mL) [33]. However, 12 h after insertion of the tablet, median peak estradiol reached approximately 28 pg/mL [33]. A pH-balanced gel (pH 4.0) decreased vaginal dryness and dyspareunia more than the placebo gel with a higher pH [29]. Across products, vaginal irritation occurred in 12-50 % of participants, but whether this symptom persisted was not well described. Menopausal Symptom Score [80] Adapted 7-item scale measuring study-specific menopausal symptoms • 5-point Likert scale (0 -4); total score 0 -28 • Higher scores indicate more symptoms.
Psychological Adjustment to Illness Scale (PAIS) [82] 46-item clinical interview with Sexual Relationships (PAIS-SR) and Sexual Problems Subscales measuring psychological and social adjustment to illness • Higher scores indicate poorer adjustment.
Study-Specific Scales [50] Study-specific scales measuring frequency of sexual desire, intercourse, masturbation, orgasm, initiative for sex, and relationship satisfaction • Higher scores indicate less impairment.
Study-Specific Scales [51] Study-specific items measuring sexual satisfaction, relationship satisfaction, dyspareunia, and comfort with sexuality Two studies without control participants intervened with vaginal compounded testosterone in BCS on aromatase inhibitors [27,28]. Compared to baseline measures, 4 weeks of vaginal testosterone improved all domains of the Female Sexual Function Inventory (FSFI) and vaginal atrophy symptoms. One study found 10 % (n = 2) of women had detectable serum estradiol levels after testosterone, though both estradiol levels were very low, <8 pg/mL [28].

Systemic therapy interventions
We sought studies using systemic androgens, antidepressants, ospemifene and flibanserin to intervene on sexual function (Tables 5 and 6, Fig 2b). No studies on ospemifene or flibanserin were found. Three randomized, double-blind cross-over trials on androgens and antidepressants were included. All participants completed primary cancer treatment. The studies were conducted in Brazil, Netherlands, and the United States. In the single study on applying daily testosterone cream to the skin for one month, testosterone in postmenopausal cancer survivors did not result in greater sexual desire, pleasure or function than placebo cream [34]. This study accepted all cancer types, with 73 % of the 150 participants on tamoxifen or aromatase inhibitor, suggesting that they are breast cancer survivors. No increases in estradiol were noted while on testosterone cream, consistent with prior studies in women without history of breast cancer [35][36][37][38][39][40]. Two additional trials involving 115 participants intervened on hot flashes as the primary outcome with venlafaxine,  [85] Includes 8-item Sexual Desire and 9-item Sexual Arousal Subscales • 6-or 7-point Likert scale (0 -5 or 0 -6) • Higher scores indicate less impairment.
Sexual Problems Frequency [45] Adapted subscale from BIRS measuring frequency of sexual problems • Higher scores indicate greater impairment.
Sexual Satisfaction Scale (SSS) [61,87] Study-specific items measuring relational sexual satisfaction for male and female partners • Higher scores indicate less impairment.
Sexual Satisfaction Scale (SSS) [88] 17-item study-specific scale measuring sexual satisfaction among Korean women • Higher scores indicate less impairment.
Visual Analog Scale for Vaginal Dryness and Dyspareunia [29] Linear rating scale measuring vaginal dryness and pain • 10-point scale (0 -10) • High scores indicate greater impairment.
• A score of < 50 indicates vaginal atrophy.
a All scales were designed as self-report questionnaires (unless otherwise reported as an examination, interview, or visual analog scale) clonidine or bupropion and examined if sexual function differed by these medications [41,42]. There were no differences in sexual function between women treated with venlafaxine compared to clonidine nor with women treated with bupropion versus placebo [41,42].

Physical therapy interventions
Three RCTs tested physical activity interventions on the primary outcomes of hot flashes, lymphedema, or physical strength and measured sexual health secondarily (Tables 5 and 6, Fig 2c). All participants completed primary breast cancer treatment. There were no studies on pelvic floor physical therapy. Included studies were conducted in the Netherlands, Sweden and United States. A home-based, self-directed exercise program intervened on 422 BCS and did not improve sexual habit, frequency or discomfort as measured by the Sexual Activity Questionnaire [43]. In the two arms with cognitive behavioral therapy, with or without exercise, there was a modest effect on improving sexual health habit at 24 weeks when compared to waitlist controls. Strength training over one year in the second trial of 295 participants was associated with a small improvement in self-perceptions of appearance and sexuality [44]. Finally, a general physical training and coping skills intervention in 199 cancer survivors (80 % with breast cancer) did not directly address sexual health and did not find change in frequency of sexual problems [45].

Counseling and educational interventions
Seventeen RCTs delivered counseling and/or educational interventions and measured sexual health outcomes in a total of 2,494 participants (Tables 5 and 6, Fig 2d). Participants were studied at various stages of cancer treatment. Studies were conducted in Australia, Finland, Greece, Korea, Netherlands, United Kingdom, and United States. Nine studies targeted sexual health as the primary outcome [46][47][48][49][50][51][52][53][54]. There was considerable heterogeneity on intervention and outcome measurements. Twelve studies intervened on the individual, while 5 studies intervened on the couple. The majority delivered in-person interventions, many with additional telephone-support [46,53,[55][56][57][58]. Two recent studies tested web-based interventions [46,54]. Counseling strategies varied widely, from problem-solving therapy to sexual therapy to cognitive behavioral therapy. Most interventions were delivered by nurses, psychologists, social workers, or peers. Several findings were consistent. In studies designed specifically to intervene on sexual health, improvements in sexual function were observed in the intervention group compared to controls [46,[48][49][50][51], but effect sizes were generally modest and of unclear clinical significance. For example, a 4-month trial tested behavioral and non-estrogen replacement pharmacologic interventions on menopausal symptoms in 76 BCS [48]. The           Duijts (2012) [43] To test cognitive behavioral therapy and physical exercise on menopausal symptoms       Vulvovaginal symptoms occur in 20 to 50 % of healthy women of midlife and older as a result of estrogen deprivation [62]. BCS are at heightened risk of these symptoms because chemotherapy, oophorectomy and/or endocrine therapies further decrease estrogen exposure. The clinical trial data show improvements in vaginal dryness, dyspareunia, sexual satisfaction and frequency, and vaginal pH with regular use at least 2-3 times weekly of a polycarbophil-based vaginal moisturizer. Compliance for at least twelve weeks is important, because major symptom gains occurred between 1-3 months and recur after stopping use, similar to data in the general population [63]. Vulvovaginal symptom relief from regular use of other moisturizers is likely, and pH balance in products may be important [23,29]. Among available vaginal moisturizers, BCS should consider preferentially using products with evidence of efficacy.
Use of minimally absorbed local vaginal estrogens and androgens provide vaginal symptom relief, with local estrogens appearing more effective than non-hormonal moisturizers [24,64]. Even at low doses, estradiol tablets and creams and compounded testosterone are systemically absorbed [24,28,[30][31][32][33]. Unfortunately, there are no clinical trial data on adverse breast cancer outcomes with extended use. Nor are there studies in BCS that compare 7, 10 and 25 micrograms of vaginal estradiol for symptom control and systemic absorption. Whether risk of breast cancer recurrence or death would be higher in estrogen-responsive tumors is also unknown. As local estrogens and androgens are not FDA-approved for use in BCS, these medications are prescribed offlabel and use requires careful discussion between BCS and their healthcare providers.
There was a lack of evidence to support incorporating systemic interventions or physical therapy into the treatment paradigm for sexual dysfunction. The single trial on transdermal testosterone did not demonstrate greater sexual desire compared to the placebo cream after 1 month of use [34]. These findings stand in contrast to several trials in women without prior breast cancer in which androgen therapy improved sexual desire, potentially because these trials were longer in duration (12-24 weeks) and provided supplemental estrogen [35][36][37][38][39][40]. Notably, there were no clinical trials on treating sexual dysfunction related to serotonin receptor uptake inhibitors in BCS.
Multiple counseling and educational strategies, particularly those targeting sexual dysfunction, have been shown to improve sexual health in BCS. Marriage and family therapists, sex therapists, sexual counselors or psychologists offer counseling interventions. With the aid of online resources, BCS can look for providers who are appropriately educated, credentialed or have significant prior experience with sexual health after cancer. Excellent online resources are found on sites for the American Association of Sexuality Educators, Counselors, and Therapists, the International Society for the Study of Women's Sexual Health, and the American Cancer Society. A number of investigators have designed educational interventions using printed materials, CDs, and websites for content with healthcare provider or peer support [46,48,[52][53][54]. This approach is important to study further, as it has the potential advantage of being delivered remotely to extend access to BCS who do not have specialized care locally.
The strength of this review is the systematic approach to identifying and grading current evidence on sexual health interventions specific to breast cancer survivors.
This approach enabled us to identify the gaps in data. Several interventions that have shown promise in women without a history of breast cancer have not undergone clinical trials in BCS. These include ospemifene and systemic DHEA for the treatment of vulvovaginal symptoms and flibanserin for the treatment of arousal and sexual interest disorders [65,66]. The primary limitation was heterogeneity of interventions and outcome measures that restricted the ability to pool data from studies of limited sample size. A recent systematic review sought to evaluate the psychometric properties of sexual dysfunction screening tools and the extent to which they measure DSM-5 aspects of sexual dysfunction for BCS [67]. The review found 31 different scales measuring sexual function, of which the Arizona Sexual Experience Scale, Female Sexual Function Index, and Sexual Problems Scale were determined to meet criteria for acceptable psychometric properties while incorporating DSM-5 areas of sexual dysfunction. Future studies in BCS should carefully consider these outcome measures in study design.
This review demonstrated that current evidence on interventions for improving sexual interest, orgasm and genitopelvic pain in BCS of midlife is limited in quantity and moderate to low in quality. From these data, we recommend prolonged and regular use of non-hormonal vaginal moisturizers to alleviate vulvar and vaginal dryness symptoms and dyspareunia. We also recommend seeking educational and counseling interventions. A number of online resources on sexual health after breast cancer may be useful for BCS and their providers (Table 7). Because each of these interventions have limited efficacy, clinical trials to test novel interventions such as ospemifene are needed in breast cancer survivors.