Sexual Dysfunction in Adolescent and Young Adult Survivors of Childhood Cancer: Presentation, Risk Factors, and Evaluation of an Underdiagnosed Late Effect: A Narrative Review

Introduction

Sexual dysfunction (SD), which encompasses a wide range of sexual concerns such as lack of desire, arousal difficulties, inability to achieve climax, lack of pleasure, pain with sexual activity, and dissatisfaction, is an important but often unrecognized area of concern for childhood cancer survivors (CCS).^1–3 SD may occur as a result of physiologic changes or psychosexual reasons, such as poor body image, concerns about fertility, and disruption of normal psychosexual development.^2,4–6 CCS, defined as patients diagnosed at 21 years or younger, are at particularly high risk for SD, as normal childhood and adolescent physiologic and psychologic developments are interrupted during cancer diagnosis and treatment.^7,8 CCS are known to report lower related measures of body image, romantic relationships, fertility, and physical and socioeconomic health.^9–11 As such, SD may contribute to this quality of life deficit. Unfortunately, SD in all cancer survivors, including CCS, is widely under-recognized, with both patients and providers reporting low levels of routine screening, despite recommendations by the National Comprehensive Cancer Network (NCCN) and the American Society for Clinical Oncology (ASCO) to address sexuality and cancer during and after therapy.^3,12–15

SD is a fairly common problem across all adult populations, with historic prevalence reports ranging from 10% to 52% of men and 25% to 64% of women.^4,16 Varying definitions for SD exist. Most broad is the World Health Organization (WHO) definition for sexual health, which is “a state of physical, emotional, mental, and social well-being in relation to sexuality, not merely the absence of disease, dysfunction, or infirmity.”¹⁷ Within WHO, the International Classification of Disease (ICD-10) has a more granular definition, encompassing a spectrum of symptoms, including lack of sexual desire, lack of sexual pleasure, failure of genital response, orgasmic dysfunction, premature ejaculation, and dyspareunia.¹⁸ In 2013, the Diagnostic and Statistical Manual of Mental Disorders-5 simplified SD categories, which may be used by clinicians to make official diagnoses, to include female sexual interest/arousal, genito-pelvic pain/penetration (females), erectile disorder, male hypoactive sexual desire disorder, delayed ejaculation, and “other” or “unspecified” SD.¹ Similar to ICD-10 definitions, the National Health and Social Life Survey defines SD using the following domains: (1) lack of desire for sex, (2) arousal difficulties (erection, lubrication), (3) inability to achieve climax/ejaculation, (4) anxiety about sexual performance, (5) climaxing/ejaculating too rapidly, (6) physical pain during intercourse, and (7) lack of pleasure.⁴ Patients who are undergoing or have undergone treatment for cancer have increased risk of SD, as might be expected given that individuals with poorer health, in general, are at higher risk.^5,19,20

SD among CCS is under-recognized, and studies are heterogeneous in their methods and results. Thus, a narrative review was undertaken to summarize current knowledge of the incidence, risk factors, underlying pathophysiology, and effects of SD on quality of life among CCS, as well as to describe current gaps in clinical care and research (Box 1). Two electronic databases (PubMed and Google Scholar) were searched, and 24 publications pertaining to SD and childhood cancer were reviewed, with 19 research articles directly relevant to SD in this population, all of which were included in this review (Tables 1 and 2). Where research in CCS was lacking, we summarized publications on SD in adult cancer survivors.

Epidemiology of SD Among CCS

Studies evaluating SD among CCS demonstrate that roughly 30%–50% of CCS experience SD.^2,6,21,22 Specifically, using four validated sexual functioning questions within a larger quality of life survey, Bober et al. found that roughly 50% of young adult and adult (age 18–57) CCS have at least one SD problem, with 30% reporting two SD problems.^21,22 Similarly, using a single-item question, the Adolescent and Young Adult Health Outcomes and Patient Experience (AYA HOPE) demonstrated that 1 year after diagnosis, 49% of adolescent and young adult (AYA) oncology patients (diagnosed between ages 15 and 39, both on therapy and off therapy) report a negative impact of cancer on sexual function or intimacy, and at 2 years after diagnosis, 43% still report a negative impact on sexual function.⁶ Zebrack et al. reported a 43% incidence of SD among young adult (18–39 years) survivors of childhood cancer (diagnosed at 21 years of age or younger), with 52% of females and 32% of males reporting a problem on the Medical Outcomes Study Sexual Functioning Scale.²¹ Similarly, in a Dutch population of CCS (age 17–39 at time of survey) which did not use a validated measure, there were high rates of sexual symptoms (e.g., 41% experienced no or almost no sexual attraction), although only 18% of survivors reported a limitation in their sexual life due to illness.² These studies provide some insight into the incidence of SD after cancer therapy and demonstrate that SD persists for at least 2 years after therapy, but no study has described the natural history of SD during therapy or in long-term follow-up, such as when it first occurs and how it may or may not change through adulthood.

Rates of SD among CCS appear to be higher than non-CCS populations of similar ages. For example, a U.S. population of male survivors of childhood acute lymphoblastic leukemia age 25–38 years had diminished sexual arousal, sexual behavior, orgasm, and sexual drive compared with age-matched controls, as measured by the Derogatis Interview for Sexual Functioning self-report.¹⁹ Swedish AYA cancer survivors (age 15–29 years at time of treatment) reported lower sexual satisfaction than controls, with females specifically having a lower frequency of orgasm and males having lower sexual desire; this study used a questionnaire with face and content validity but without further psychometric testing.²³ Male CCS, age 20 or older (mean age 39), also have higher rates of erectile dysfunction than sibling controls (12.3% vs. 4.2%).²⁴ In a study focused on all CCS (vs. AYA only) using the validated International Index of Erectile Function, rates of erectile dysfunction among CCS were even greater (29% of participants, median age 31).²⁵ These discrepancies may be explained by the difference in age at diagnosis/treatment, the difference in tools used, or perhaps in that the latter study, through the CCS Study, did not include some patients with malignant diagnoses, such as patients with germ cell tumors, retinoblastoma, or hepatoblastoma, who may be at lower risk due to less intensive or gonadotoxic treatment. Similarly, female CCS, age 18–51 (mean 29), have psychosexual dysfunction, including diminished sexual interest, desire, arousal, and satisfaction than sibling controls, as measured by dichotomized results of the validated Women's Health Questionnaire.²⁰

It is notable, however, that a few studies have yielded more reserved results regarding the impact of childhood cancer on sexual function. A different Swedish cohort of CCS focusing on childhood cancer (age 0–18 at time of treatment) found that male survivors had more dysfunction than controls, while females reported similar dysfunction rates to controls, although with notably fewer reported current romantic partnerships than controls.⁵ In the United States, a web-based questionnaire using the Global Measure for Sexual Satisfaction found that young adult CCS (age 20–40) had comparable relationship status and sexual function to age and sex matched controls; the reason for this study's discrepant findings is unclear, but it is possible that this is due to exclusion of patients with a history of central nervous system (CNS) tumors (see “Pathophysiology” section).²⁶

While all the studies reviewed in this section provide some insight into SD after childhood cancer, they used tools that were either not fully validated or were validated for a general adult population, which may underestimate or inaccurately categorize SD among younger cohorts, who may have different sexual concerns than older adults. Furthermore, several studies have suggested that the inherent linkage of SD to sexual activity level may limit defining and understanding SD in younger patients. This is particularly true for females, as awareness of desire, arousal, and satisfaction are more closely related to partner intercourse, whereas these domains in males are more readily recognized without a partner.²¹ To adequately assess for SD among AYA survivors of childhood cancer, tools should be validated specifically in this population, as these patients may have different concerns than healthy populations.

In studies using adult measurements of SD (both validated and not), patient risk factors for SD among CCS included female sex, older age at diagnosis, being older at time of evaluation, lower income, and concurrent health problems.^{6,19–22,27–31} While most studies seem to agree on these risk factors, the AYA HOPE study did not find that female sex was a risk factor for SD and found increased SD in younger (age 15–20) oncology patients and survivors, and Bober et al. did not find that age at diagnosis was a risk factor.^6,22 In females, ovarian failure at a younger age is also a risk factor for SD.²⁰ As with studies reporting the prevalence of SD after cancer, no studies have prospectively evaluated for risk factors for SD, during or after cancer.

Proposed Underlying Pathophysiology of SD in CCS

CCS experience a wide variety of types of SD, including decreased desire/drive, arousal, behavior, satisfaction, and difficulty with orgasm.^2,5,19,20,22 Reasons for SD have been postulated to be both physiologic and psychosexual in nature (Fig. 1).

While SD has been described across all types of CCS, and in fact cancer type was not a risk factor in several studies, one study did find increased rates in survivors of CNS tumors.^5,21,22 Treatment-related risk factors may include having undergone chemotherapy, retroperitoneal lymph node dissection, cranial irradiation, testicular irradiation, surgery on spinal cord/nerves, or pelvic surgery, although many of these risk factors are specifically related to erectile or ejaculatory function.^{19,20,24,28,29,31–33} The significance of each exposure varied across these studies; furthermore, one study did not demonstrate any relationship between treatment exposures and SD.²² Treatment or supportive care with ovarian function suppression³⁴ and general androgen suppression place survivors at risk for SD. Notably, among adult gynecologic cancer patients and survivors (mean age 54), radiation exposure was not a risk factor for SD, but in other instances, including in CCS, pelvic radiation has been associated with SD.^31,32,35

The effects of chemotherapy and cranial, pelvic, and testicular irradiation on SD are likely related to effects of sex hormone production and balance.³¹ A single study on male CCS demonstrated that lower testosterone levels were associated with erectile dysfunction.²⁵ All other studies on this topic have focused on adult or young adult cancer populations often in testicular cancer and breast cancer survivors where sex hormone changes due to cancer treatment may be important. Several studies in adults, across cancer diagnoses, have shown that lower testosterone levels may be associated with SD in males across domains of interest, satisfaction, and erectile function.^30,36,37 Even mildly increased luteinizing hormone, particularly in the setting of low-normal or low testosterone both in clinical and subclinical hypogonadism, and increased follicle stimulating hormone may be associated with less sexual activity in male survivors.^19,30,36,37 Females with ovarian failure also have SD in both adult and CCS populations, with effects on sexual interest, desire, arousal, satisfaction, pain, overall function, and other domains.^20,38 Furthermore, premature ovarian failure as a result of chemotherapy may have a greater impact on sexual function than normal menopause.³⁵ These studies have largely used self report without the use of validated SD screening tools; because physiologic disturbances could possibly be targeted with hormone replacement or other interventions, clarifying these links using validated instruments is crucial.

As previously mentioned, while physiologic dysfunction due to treatment exposures and cancer types may play a role in the pathophysiology of SD in cancer survivors, they do not seem to explain all SDs among CCS. In fact, multiple studies suggest that psychosocial factors may be more significant.^2,3,30,35,37 Mental health conditions, including depression, anxiety, and higher emotional distress, are associated with increased SD among CCS, and these concerns are generally increased among CCS compared to the general population.^19,28–31 Beyond mental health, psychosexual reasons for dysfunction may include body image issues, concerns about fertility, and disruption of normal psychosexual development.^{2,3,5,6,26,39} Most of these potential underlying psychological causes are hypothesized based on known risk for these issues among CCS and qualitative patient report. For example, it is well-established that CCS have delayed psychosexual milestones compared to peers, demonstrating disruption in development, but linkage between this and SD has not been directly demonstrated.^2,39,40 One study did demonstrate a significant association between body image and SD using validated scales for both.²⁶ Additional research evaluating for links between objectively demonstrated psychosexual concerns and SD are needed.

SD and Quality of Life

Results of studies evaluating the association between quality of life and SD are mixed and are limited by the use of unvalidated assessment tools. Zebrack et al demonstrated a significant association between SD and diminished quality of life (specifically bodily pain, general health, vitality, social functioning, role-emotional status, and mental health) among young adult survivors of childhood cancer, with potentially greater distress among men as a result of sexual symptoms than among women.²¹ Partner status has not been reliably associated with SD risk, with several studies demonstrating increased SD in partnered individuals and other demonstrating increased risk in unpartnered individuals.^20,21,28,29 Notably, those studies that demonstrated being partnered as a risk factor for SD specifically found this to be true in females.^21,28 Patient awareness of SD can be dependent on sexual activity with a partner, particularly in females; self-report screening tools may have varying sensitivity to this dependency, which may explain this variation in associations.²¹ Bober et al. also found correlations between SD and quality of life in all CCS, but demonstrate differences between sexes: SD in females correlated with emotional/mental health, energy/fatigue, and social functioning, whereas SD in males correlated with physical quality of life and pain.²² In contrast, two studies did not find an association between SD and quality of life, one of which was completed in a Dutch CCS population and one of which in a U.S. population. Notably, while in the U.S. population there was no correlation between SD and health-related quality of life, there was an association between increased psychological distress and SD.^2,28 Importantly, no study has evaluated the association between SD and quality of life specifically among AYA CCS.

Interpreting the association between SD and quality of life is challenging not only because of mixed study results but also because it is possible that SD and poor quality of life may have causal links in both directions. Interestingly, two studies (one focusing on AYAs and one on adult cancer survivors) have shown improvement in psychological distress as a result of addressing SD with education and therapy interventions.^41,42 This suggests that, regardless of the direction of causality between SD and diminished quality of life, improvements in SD may result in improvements in quality of life. Larger studies and studies evaluating the association between change in SD and quality of life are needed.

Poor Clinician Recognition of SD

Studies have found that both adult cancer survivors and CCS want help for SD. For example, a survey of adult cancer survivors demonstrated that 24% of women and 21% of men reported wanting help for sexual problems but that that need was largely unmet.¹² AYA cancer patients and survivors place a high priority on addressing sexual health concerns and report a desire to discuss sexual and reproductive health with their oncologists, but report that they are hesitant to initiate conversations and prefer their provider to take the lead.^3,43 In one of these studies, all participants reported inadequate clinical support regarding sexual health.³ AYA patients both on therapy and in survivorship would like their providers to address these issues in a number of ways, including: routinely offering time for private physician–patient conversations, normalizing the topic, engaging in ongoing conversations throughout treatment and survivorship, tailoring communications to individual needs, and engaging in direct in-person communication.⁴³ Importantly, many CCS patients report problems with SD but are unaware of the potential relationship between this problem and their history of cancer, further demonstrating the need for patient education and provider-initiated conversations.⁴⁴

Provider perspectives have also been assessed. Among pediatric oncologists, challenges in meeting sexual health needs include lack of knowledge/experience, lack of resources/referrals, parent/family presence, concerns of patient discomfort, clinical discomfort, lack of rapport, low priority, and limited time.⁴⁵ Unfortunately, these issues persist for survivors outside of their oncology care as well. Among general internists, there is a low rate of screening for SD in cancer survivors, resulting in low recognition and inadequate intervention. For example, at a major academic medical center, 62% of general internists reported that they never or rarely addressed SD among their patients who are cancer survivors, citing inadequate training or perception of patient anxiety/fears about the topic.¹³

This low rate of screening is remarkable given NCCN and ASCO recommendations, which state that the oncology team should initiate discussions of sexuality and cancer both during treatment planning and periodically in follow-up.^14,15 The absence of adequate support for cancer survivors despite these guidelines indicates that provider education is crucial to (1) improve awareness of the risk for SD after cancer and (2) to increase provider comfort with screening for and addressing SD concerns, so that physicians are more likely to ask patients about SD. To this end, Aubin and Perez created a “Clinician's Toolbox” to assess for sexual impacts of cancer in AYA CCS, although implementation and effectiveness have not been tested.⁴⁶ Similarly, the American Society of Pediatric Hematology/Oncology hosted a webinar aimed at educating providers about SD and how to screen for it.⁴⁷ Again, wider dissemination of education efforts have not been published, although the authors of this review are aware of such efforts being underway. While oncologists must be leaders in providing treatment-related late effect screening, other relevant providers, including radiation oncologists, endocrinologists, gynecologists, urologists, and psychologists, caring for these patients should also be aware of the need for, and participate in, the need for education and screening.

Given the sensitivity of this topic, it is possible that screening tools may be a helpful approach to assessing for SD both in clinical and research settings.⁴⁶ Screening tools may be particularly useful in AYA CCS, who may be especially vulnerable to discomfort related to discussions of sexuality and sexual function.^43,48,49 As previously mentioned, no SD screening tools exist that have been validated in the adolescent or specifically young adult general populations, and no screening tools focus specifically on AYA CCS. In contrast, multiple tools exist to screen for SD in adult populations, a number of which have been validated.^50,51 Only the National Institute of Health developed Patient-Reported Outcomes Measurement Information System Sexual Function and Satisfaction (PROMIS^® SexFS) measures, have comparable sex-specific versions, and have been validated in adult (mean age 58.5 years) cancer populations.^52–54

It is unknown whether adult screening tools for SD are relevant for the AYA population; as such, this requires formal assessment. Tools validated in older adult populations may have decreased relevance especially in younger (age 15–24) AYAs, because they do not take into consideration developmental stage. Furthermore, as previously noted, CCS are at unique risk due to the potential for disrupted or delayed sexual development.⁸ Even within this younger AYA age range, the sexual health concerns for a 15-year old may be different than for a 24-year old, particularly based on attainment of psychosexual milestones. Beyond evaluation of such tools in this unique population, their utility for screening or diagnosis, compared with in-person communication, is unknown.

Interventions for SD

As SD is identified, one challenge is having the tools and infrastructure to intervene and support patients.⁴⁵ While pediatric providers have been trained in adolescent health, as subspecialists, they may need continuing education about how to address SD concerns and guidance as to which referrals are appropriate. To this end, the Children's Oncology Group has embarked on a nationwide survey of oncology providers trying to understand their education needs.

Current literature regarding interventions to address SD demonstrates that interventions must be interdisciplinary since the complex pathophysiology of SD may include physical, hormonal, and psychological challenges. For example, approaches that address physiologic limitations (such as with lubricators, dilators, moisturizers, pelvic floor exercises, phosphodiesterase inhibitors, and vacuum erection devices), as well as cognitive behavioral therapy, counseling, and education, are necessary.¹⁵ This may be achieved through multidisciplinary sexual medicine clinics, where available, or through establishing a comprehensive referral network. Several studies focusing on adult cancer survivors have demonstrated improved sexual function in women using such multidisciplinary approaches with a significant education component.^42,55 A similar counseling and education-focused intervention aimed at enhancing psychosexual development in the AYA CCS population demonstrated improved body image and decreased distress, but did not assess for changes in SD.⁴¹

Published interventions involving hormone replacement have involved significant treatment heterogeneity (dosage, form, etc.) and, while largely not promising for treatment of SD, have generally used unvalidated assessments for SD.^20,56–59 It is possible that a study with more specific eligibility and a homogenous hormone replacement intervention would find benefit. The utility of pharmacologic interventions to directly address SD, such as phosphodiesterase type 5 inhibitors for erectile dysfunction or the recently approved bremelanotide for female hypoactive sexual desire disorder, has not been studied in CCS and rarely in adult cancer survivors.

In summary, while several studies demonstrate possible interventions and success of multidisciplinary clinics for addressing SD in adult cancer survivors or general adult populations, no studies have focused on CCS or the AYA population.

Conclusion

CCS have an increased risk of SD as a result of their cancer diagnoses and treatment. While pediatric oncology guidelines do not address the risk for SD among childhood cancer patients and CCS, pediatric oncologists and survivorship clinicians should consider following guidelines for adult cancer patients and survivors.^14,15 ASCO recommends, “That there be a discussion with the patient, initiated by a member of the health care team, regarding sexual health and dysfunction resulting from cancer or its treatment.”¹⁵ Identifying specific risk groups has been challenging because of multifactorial pathophysiology, which may include endocrine dysfunction due to chemotherapy, radiation, or surgery, local effects of radiation and/or surgery, and disruptions in usual psychosocial development. As such, all patients who undergo treatment for childhood cancer warrant education and screening for SD; it is the responsibility of all relevant providers, including oncologists, radiation oncologists, surgeons, endocrinologists, and psychologists, to address this unmet need. In patients where SD is identified, while physiologic or medical issues should be addressed first, “psychosocial and/or psychosexual counseling should be offered to all patients with cancer, aiming to improve sexual response, body image, intimacy and relationship issues, and overall sexual functioning and satisfaction.”¹⁵

In addition to immediate changes in clinical practice, additional research addressing gaps in knowledge, including pathophysiology, preferred screening approaches, age and development-specific screening tools, interventions, and associations between SD and quality of life, is necessary. Ultimately, improved identification and treatment of SD will address an unmet need and may lead to improved quality of life for the vulnerable AYA CCS population.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this work.