“Prevalence of sexual dysfunction and narrative synthesis of associated factors in patients with different type of cancer: a systematic review and meta-analysis”

Abstract

Background

Sexual dysfunction (SD) is a prevalent but underrecognized consequence of cancer and its treatment, affecting patients’ quality of life, psychological well-being, and intimate relationships. Despite its clinical importance, the literature on SD remains fragmented, with previous reviews focusing primarily on specific cancer types. This systematic review and meta-analysis aimed to estimate the overall prevalence of SD across various cancer types and to identify demographic, clinical, psychosocial, and treatment-related factors associated with its occurrence.

Method

A systematic search was conducted in both English and Persian databases, including PubMed, Scopus, Web of Science, Embase, CINAHL, Google Scholar, and regional platforms (SID, Magiran, IranMedex) for observational studies reporting on prevalence of SD and its determinants in cancer patients. The inclusion criteria were developed based on the SPIDER framework, targeting quantitative studies that focused on the prevalence and risk factors associated with SD. Eligible studies were screened independently by two reviewers and assessed using the Observational Studies Quality Evaluation (OSQE) tool. A random-effects meta-analysis was used to calculate pooled prevalence estimates 39 included studies. Subgroup analyses and meta-regression were conducted to explore potential sources of heterogeneity using Comprehensive Meta-Analysis version 4 software. Additionally, a narrative synthesis of associated factors was performed based on 37 included studies.

Results

A total of 39 studies involving 17,970 participants were included. The overall pooled prevalence of SD was 54.0% (95% CI: 50.4%–58.8%), with significant heterogeneity (I² = 95%). Subgroup analysis revealed the highest prevalence among patients with gynecological (93.7%), breast (59.2%), and colorectal (57.4%) cancers. Meta-regression identified publication year, cancer type, and geographical region as significant moderators. However, a large portion of heterogeneity remained unexplained due to insufficient reporting of key socio-demographic and clinical variables. Narrative synthesis identified age, treatment modality, menopausal status, body image, psychological distress, and social factors as major determinants of SD.

Conclusion

These findings underscore the urgent need for standardized assessment tools, individualized care strategies, and theory-informed interventions that integrate sexual health into routine oncology care. Future research should focus on longitudinal designs and individual-level data to better understand the complex interplay of factors contributing to SD in cancer patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-026-15604-3.

Introduction

Cancer remains a major global public health challenge and is the second leading cause of death worldwide, accounting for nearly one in six deaths [1–3]. Among the most prevalent malignancies are breast, skin, colorectal, lung, endometrial, thyroid, cervical, ovarian, vulvar, and vaginal cancers, with breast cancer ranking as the most common globally [4–7]. While the physical and psychological burden of cancer is well established, its impact on sexual health remains an underrecognized aspect of oncology care [8].

Sexual dysfunction (SD), defined as persistent or recurrent problems with sexual response, desire, orgasm, or pain during intercourse, affects a substantial proportion of individuals with cancer [9–11]. The World Health Organization emphasizes that sexual health is integral to overall well-being and quality of life, influencing psychological resilience, body image, intimate relationships, and life satisfaction [12–14]. Cancer treatments, including surgery, chemotherapy, radiotherapy, hormone therapy, and ostomy creation, can disrupt hormonal balance and body image, resulting in decreased sexual desire, arousal, and satisfaction [15–18]. Consequently, SD manifests in diverse forms such as hypoactive sexual desire, erectile dysfunction, vaginal dryness, orgasmic disorders, and dyspareunia, with prevalence varying by cancer type, treatment, age, and psychosocial context [19, 20].

Despite its high prevalence, SD is often underrecognized and insufficiently addressed in cancer care [21]. Barriers include patient reluctance to discuss sexual issues, healthcare providers’ lack of training, and misconceptions that sexual health is nonessential [22, 23]. Reported prevalence rates range from 86 to 91% in prostate, breast, and gynecological cancers to 53–79% in head and neck, hematologic, and colorectal cancers [24, 25]. For instance, Brugger et al. (2019) found that only 2% of patients with advanced cancer were sexually active despite 62% desiring intimacy, illustrating the persistent psychological and functional barriers to sexual well-being [26].

As advances in cancer treatment have improved survival, attention has shifted toward long-term quality of life, with sexual health emerging as a critical yet neglected component [27–31]. However, existing literature on SD in cancer survivors remains fragmented. Differences in study design, sample characteristics, cancer sites, and assessment tools contribute to inconsistent estimates of SD prevalence and risk factors [32–35] Prior reviews have also been limited by narrow scope, focusing on specific cancers, single genders, or selected measurement tools [36–39], and have rarely examined moderators such as treatment modality, time since treatment, or menopausal status [40].

To address these limitations, the present systematic review and meta-analysis comprehensively examines the prevalence of SD across diverse cancer types and identifies modifiable and non-modifiable determinants. By integrating data from observational studies across geographic and clinical contexts, this study aims to provide evidence-based insights to inform clinical practice, survivorship care, and health policy. Addressing sexual dysfunction is not merely an aspect of patient satisfaction but a cornerstone of holistic, patient-centered cancer care.

Method

Protocol and registration of the study

This systematic review and meta-analysis was designed and conducted in accordance with the methodological recommendations outlined in the Cochrane Handbook for Systematic Reviews of Interventions, version 6.5 (2024) [41]. The reporting of this review follows the updated PRISMA 2020 (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [42].

Search strategy and selection criteria

A comprehensive literature search was conducted to identify relevant studies published in both English and Persian. The following English-language databases were systematically searched: PubMed, Scopus, Web of Science, CINAHL, Embase, and Google Scholar search engine. For Persian-language sources, searches were conducted in SID, Magiran, and IranMedex.

The search strategy was designed to maximize sensitivity and coverage, utilizing a combination of keywords and controlled vocabulary terms related to the phenomenon of interest and outcome measures, including terms such as “prevalence”, “associated factors”, “cancer”, and “SD” and their related Persian equivalents for Persian data bases. Synonyms and variations of these terms were also incorporated to ensure comprehensiveness, and no restrictions were placed on publication date. The complete list of search terms is provided in Supplementary material file 2, Sect. 1.

Boolean operators (“AND”, “OR”) were applied to construct the search strings effectively and broaden the scope of results. In addition, the reference lists and citations of all included articles, as well as relevant organizational websites, were manually screened to identify any additional eligible studies. The initial search was carried out on May 13, 2024, followed by an updated search on June 19, 2024, during which no new eligible studies were found. A third search was subsequently conducted on September 3, 2025, which also did not yield any additional eligible studies. Full details of the search strategies tailored to each database are provided in Supplementary Material file 2, Sect. 2, ensuring transparency and reproducibility of the review process.

In developing the research question for this study, particular attention was given to ensuring it was clearly defined, coherent, and methodologically structured. To guide this process, we adopted the SPIDER framework [43, 44] Sample, Phenomenon of Interest, Design, Evaluation, and Research Type which has been recommended for use in mixed-methods and qualitative or quantitative evidence syntheses. This decision was informed by the work of Alison Cooke et al. [45], who advocate for the SPIDER tool, particularly in systematic reviews involving a broad range of study designs and research methodologies.

Accordingly, the components of the SPIDER framework in this review were defined as follows: Sample (S): Patients diagnosed with or surviving any type of cancer, regardless of cancer type, age, or treatment modality. This inclusiveness reflects the study’s primary aim of providing a comprehensive estimate of the prevalence of SD across cancer populations. Phenomenon of Interest (PI): The prevalence and determinants of SD in individuals with cancer. Design (D): All quantitative observational study designs, including cross-sectional, case-control, and cohort studies. Evaluation (E): The primary outcome was the prevalence of SD. Secondary outcomes included determinants or risk factors associated with SD. Research Type (R): Quantitative studies published in peer-reviewed journals in English and Persian.

Study selection and data extraction

The study selection process was conducted in two sequential stages. In the first stage, two independent reviewers (AD and MH) screened the titles and abstracts of all retrieved records to assess preliminary eligibility. In the second stage, the same reviewers independently examined the full texts of studies identified as potentially relevant. Any disagreements at either stage were resolved through discussion; if consensus could not be reached, a third reviewer (MHS) was consulted to achieve a final decision. The entire screening and selection process was managed using EndNote software (version 21) to facilitate reference organization and eliminate duplicates.

Data extraction was independently carried out by the same two reviewers (AD and MH) using a pre-designed and standardized data extraction form, developed in accordance with the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions [46]. Discrepancies in data extraction were resolved through discussion and consensus. Extracted data included: first author’s name, year of publication, type of cancer, country or region of study, study design, assessment tools used, total sample size, health status of participants (e.g., during or post-treatment), participant characteristics potentially influencing SD, and key results.

Assessment of methodological quality

The methodological quality and risk of bias of the included studies were assessed using the Observational Studies Quality Evaluation (OSQE) tool developed by Drukker et al. [47]. This tool is specifically designed to evaluate observational study designs, including cohort, case-control, and cross-sectional studies, and allows for both quantitative scoring and qualitative appraisal.

Each study was appraised independently by two reviewers (AD and MH) according to the criteria outlined in the OSQE. Studies were rated as having low risk of bias (good quality), moderate risk of bias (medium quality), or high risk of bias (poor quality), based on their overall score and adherence to the quality indicators.

Statistical analysis

Pooled prevalence estimates of SD were calculated from the included studies using a random-effects model, given the high heterogeneity (I² = 95%). The meta-analysis was performed using Comprehensive Meta-Analysis (CMA) software, version 4, and a p-value < 0.05 was considered statistically significant.

Due to substantial heterogeneity observed across studies, as indicated by the I² statistic (with values ≥ 50% interpreted as high heterogeneity) [48, 49], a meta-regression analysis was initially conducted to identify potential sources of variability. The following covariates were considered based on their availability and relevance: Sample size, year of publication, proportion of female participants, mean age of participants, educational attainment (up to high school), employment status, menopausal status, smoking status, having a sexual partner, type of cancer, study design, geographical region (continent), measurement tool, and health status of the participants.

Following the identification of significant moderators through meta-regression, subgroup analyses were subsequently performed based on year of publication, type of cancer, and continent. This approach enabled the estimation of pooled prevalence within relevant subgroups, offering more granular insights into patterns of SD across different populations and study characteristics.

Sensitivity analyses were conducted using the leave-one-out method, whereby each study was sequentially removed to assess its influence on the pooled estimate. This was implemented only when the number of included studies was seven or more, in line with methodological guidance [50].

To assess publication bias, a combination of methods was used when the number of studies exceeded ten. These included visual inspection of funnel plot symmetry, as well as statistical tests such as Begg and Mazumdar’s rank correlation and Egger’s regression test.

Finally, a narrative synthesis was employed to summarize the determinants of SD, drawn from 37 studies. Due to the heterogeneity in study design, measurement tools, and outcome definitions, a meta-analysis of determinants was not feasible. Instead, findings were categorized thematically by cancer type to offer a structured summary of the associated factors. The meta-analysis of prevalence was based on 39 studies.

It should be noted that for methodological details, including Ethical approval and assessing the quality of included studies see Supplementary Material File 2 Sect. 3.

Results

Study selection process

Study selection followed the PRISMA 2020 framework (Fig. 1). The database search across nine electronic sources (PubMed, Web of Science, CINAHL, Scopus, Embase, Google Scholar, IranMedex, Magiran, and SID) yielded 8,649 records. After removing 4,588 duplicates, 4,061 titles and abstracts were screened. A total of 267 full texts were sought, of which 95 were retrieved and assessed for eligibility. Fifty-six studies were excluded for reasons such as irrelevance, non-English publication, absence of prevalence data, or publication as conference abstracts. Ultimately, 37 studies were included through database searches. An additional two studies were identified via manual and citation searches, resulting in a final pool of 39 observational studies included in the systematic review and meta-analysis.

Fig. 1.

PRISMA flow diagram for systematic literature search

Characteristics of included studies

The 39 included studies comprised 25 cross-sectional, 11 cohort, and 3 case-control designs, published between 1996 and 2024 (Table 1), encompassing a total of 17,970 participants. Sample sizes ranged from 42 to 7985. Studies covered a wide range of cancer types, including breast, gynecological, colorectal, prostate, cervical, bladder, testicular, lymphoma, ovarian, anal, and mixed cancers.

Table 1.

Characteristics of included studies

First Author / (year)	Type of cancer	Country / City	Study design	Assessment tools	Total sample size	Health status of participants	Results
Topaloglu Ören / 2024 (120)	Gynaecological Cancer (included cancer types: Endometrium cancer, Cervix cancer, Ovary cancer, Vagina-vulvar cancer)	Türkiye / Izmir	Case- control study	The Sexual Quality of Life Questionnaire score (SQOL) and participants characteristics form	65 patients	Patients with histological confirmed Gynaecological cancer	Mean score of sexual function: 60.12 ± 9.01 Factors affecting the sexual function: Working (yes): (B = 4.981, p = 0.016), Smoking (yes): (B = -4.590, p = 0.035), Receiving any type of treatment (yes): (B = -4.722, p = 0.028)
Kyei / 2023 (73)	Cervical cancer	Ghana / Accra	Longitudinal study (April to July, 2021)	The Female Sexual Function Index (FSFI) and participants characteristics form	44 patients	Patients with Cervical cancer	Number of participants who have SD: 17 (38%) Factors affecting the sexual function: Being menopausal (yes): (P = 0.001)
Franzoi / 2024 (34)	Breast Cancer	France / 26 French cancer care centers	Longitudinal study (1–4 year after diagnosis)	The European Organization for Research and Treatment of Cancer (EORTC), Quality of Life Questionnaire (QLQ)-BR23, and participants characteristics form	7985 patients	Patients with Breast cancer	Number of participants who have SD: 4575 (57.3%) Factors affecting the sexual function: Income (Low): (P = 0.011), Treatment with Chemotherapy (yes): (P = 0.029), Endocrine therapy (yes): (P = 0.032), Insomnia (yes): (P = 0.048)
Dahouri / 2024 (35)	Colorectal cancer	Iran / Tabriz	Cross-sectional	International Index Erectile Function (IIEF), Female Sexual Function Index (FSFI), and participants characteristics form	256 patients	Patients with colorectal cancer	Number of participants who have SD: 226 (88.28%) Factors affecting the sexual function: Having colostomy (with): (B = 7.13, P < 0.001), Type of treatment (Chemotherapy-surgery): (B= -6.08, P < 0.001), House type (rent): (B = 8.35, P < 0.001), Working (yes): (B= -5.50, P < 0.001), Comorbid condition (yes): (B = 3.01, P = 0.047), Education (above high school): (-11.09, P = 0.026), Age (more than 40): (B = 17.07, P = 0.02), Time of last chemotherapy (5–10 weeks or more): (B=-5.24, P = 0.018)
Marissa / 2023 (77)	Colorectal cancer and anal squamous cell carcinoma	United States / San Francisco	Longitudinal study (6–12 months after treatment completed)	Female Sexual Function Index (FSFI), and participants characteristics form	97 female patients	Patients with colorectal or anal squamous cell carcinoma	Number of participants who have SD: 34 (71%) Factors affecting the sexual function: Tomur site (being ion rectum): (B= -0.7, P = 0.07), Menopause status (Being menopause): (B= -0.6, P = 0.13)
Victoria / 2023 (51)	Various cancers, including breast, gynecologic (like cervical, ovarian, and endometrial), and testicular cancers	United States / Chapel Hill	Cross-sectional	PROMIS Sexual Function and Satisfaction (SexFS) and participants characteristics form	284 patients who were survived from various cancer	Adolescent and young adult (AYA) cancer survivors	Number of participants who have SD: 77 (27%) Factors affecting the sexual function: Physical activity (active): (P = 0.02) Mental health (good health): (P = 0.04)
Wettergren / 2022 (52)	Breast, Gynecological, Lymphoma, Brain	Sweden / Uppsala, Stockholm, Huddinge, Danderyd, Örebro, Umeå	Population-based cross-sectional design	PROMIS Sexual Function and Satisfaction (SexFS) and participants characteristics form	694 young women	Women after being diagnosed with any cancer type.	Number of participants who have SD: 459 (66.13%) Factors affecting the sexual function: Age, Marital status (having), cancer type (Lymphoma), Mental health, Body image disturbance (P < 0.05)
Smedsland / 2022 (53)	Breast cancer patients	Norway / Oslo	Cross-sectional	The Sexual Activity Questionnaire (SAQ) and participants characteristics form	1307 women	Women who had Breast cancer	Number of participants who have SD: 680 (52%) Factors affecting the sexual function: Not interested: P < 0.001, No partner: P < 0.001, Too tired: P < 0.05, Physical problems: P < 0.05
Lin / 2022 (54)	Various gynecological cancer such as Endometrium, Ovary, and Cervix	Taiwan / Kaohsiung Chang Gung	Cross-sectional	Female SD (based on DSM-5) and participants characteristics form	126 women	Women who were gynecological cancer survivors	Number of participants who have SD: 55 (43.7%) Factors affecting the sexual function: Type of cancer (Endometrial cancer): (P = 0.020)
Gan / 2022 (72)	Lymphoma cancer	Malaysia / Kuala Lumpur, Sarawak	Cross-sectional	International Index of Erectile Function (IIEF), and participants characteristics form	106 patients	Men who were Lymphoma cancer survivors	Number of participants who have SD: 35 (33%) Factors affecting the sexual function: Type of cancer (Endometrial cancer): Age (Higher): (P < 0.005)
Du / 2022 (67)	Breast Cancer	China / Hefei.	Cross-sectional	Body Image Scale (BIS), Female Sexual Function Index (FSFI), and participants characteristics form	276 patients	Women who were Breast cancer survivors	Number of participants who have SD: 175 (63.41%) Factors affecting the sexual function: Body Image (Poor): (P < 0.05)
Seow-En / 2021 (76)	Colorectal cancer	Singapore / Singapore	Longitudinal study (From February 2017 to July 2019)	European Organisation for Research and Treatment of Cancer; QLQ (EORTC QLQ-C30) and Quality of life questionnaire (QLQ-CR29), and participants characteristics form	400 patients	Colorectal cancer survivors	Number of participants who have SD: 200 (50%) Factors affecting the sexual function: Age (Being older): (P = 0.00), Education level (Having higher degree): (P = 0.004), Rectal (vs. colon) cancer (Type of cancer): (P < 0.05)
Ooi / 2021 (55)	Breast Cancer	Malaysia / Kelantan	Cross-sectional	Demographic and clinical information, together with the Malay Version of the Breast Impact of Treatment Scale and the Malay Version of Female Sexual Distress Scale-Revised	94 Female	Female Breast cancer survivors	Number of participants who have SD: 69 (73.4%) Factors affecting the sexual function: Family history of breast cancer: P = 0.040, Duration of marriage: P = 0.046, Frequency of sexual intercourse: P = 0.002
López / 2021 (68)	Breast Cancer	Spain / Cádiz	Prospective	Female Sexual Function Index (FSFI), EORTC QLQ-BR23, State-Trait Anxiety Inventory (STAI), and participants characteristics form	168 patients	Postmenopausal women with Breast cancer who were on Adjuvant Aromatase Inhibitor therapy	Number of participants who have SD: 80 (47.60%) Factors affecting the sexual function: Type of treatment (Adjuvant Inhibitor Vs. Non-adjuvant Inhibitor): (P = 0.000)
Yerramilli / 2019 (78)	Anal Cancer	United States / Boston	Prospective (1–97 months)	Female Sexual Function Index (FSFI), the International Index of Erectile Function (IIEF)	42 patients	The participants in the study were individuals with localized anal squamous cell carcinoma (SCC) who had undergone definitive chemoradiation (CRT)	Number of participants who have SD: 42 (100%) Factors affecting the sexual function: There was insufficient sample size to evaluate the potential factors.
Jeffrey / 2020 (79)	Bladder Cancer	United States / Los Angeles	Prospective (From October 2017 to January 2019)	The International Index of Erectile Function (IIEF-5), and participants characteristics form	111 patients	Male patients with bladder cancer who had undergone radical cystoprostatectomy (RC)	Number of participants who have SD: 62 (56.4%) Factors affecting the sexual function: BMI (Increasing BMI): (P = 0.0025), Surgical treatment (Radical Cystoprostatectomy): (P = 0.0487)
Dahl / 2020 (75)	Cervical Cancer	Norway / South-Eastern and Northern Regions	Prospective (from 2000 through 2007)	EORTC QLQ-CX24, Fatigue Questionnaire, Hospital Anxiety and Depression Scale (HADS), Patient Health Questionnaire-9 (PHQ-9), Work Ability Index, Eysenck Personality Questionnaire, Charlson’s Comorbidity Index	523 patients	Long-term Cervical cancer survivors	Number of participants who have SD: 214 (40.91%) Factors affecting the sexual function: Age (Getting older): (P < 0.001), Education (till high school): (P = 0.022), Having partner (Having): (P < 0.001), Income status (working): (P < 0.001), Obesity (Being obese): (P = 0.005), Mood status (Being depressed): (P < 0.001), Fatigue (Chronic fatigue): (P = 0.002)
Soleimani / 2018 (56)	Gynecological Cancer that included Breast, Ovary, Uterus, Breast and Ovary, and Breast and Cervical Cancers	Iran / Qazvin	Cross-sectional	Demographic questionnaire, the Female Sexual Function Index (FSFI), and the Female Sexual Distress Scale-Revised (FSDS-R)	387 patients	Female with Gynecological cancers	Number of participants who have SD: 387 (100%) Factors affecting the sexual function: Cancer stage (p = 0.023), cancer type (p = 0.025), duration of disease (β = −0.10, 95% CI [-0.17, – 0.02], p = 0.017) and social support (β = 0.53, 95% CI [0.24, 0.83], p < 0.001).
Ljungman / 2018 (57)	Breast Cancer	Sweden / Solna, Örebro, and Uppsala (Cities reported based on authors affiliation)	Cross-sectional	The Swedish National Quality Registry for Breast Cancer, PROMIS Sexual Function and Satisfaction version 2.0 (SexFS), Reproductive Concerns After Cancer (RCAC), EORTC QLQ-C30 (version 3.0)	181 patients	Diagnosed with breast cancer approximately 2 years previously	Number of participants who have SD: 123 (68%) Factors affecting the sexual function: Type of treatment (Current endocrine treatment and previous chemotherapy), Body image (Negative), Reproductive concerns (High level concern), Education (above high school): (P < 0.05)
Attaallah / 2017 (80)	Rectal Cancer	Turkey / Istanbul	Prospective, single-arm cohort study (Between November 2013 and January 2016)	International Index of Erectile Function (IIEF-5), Index of Female Sexual Function (FSFI), and participants characteristics form	187 patients	Patients who diagnosed with Rectal cancer and underwent radical curative surgery (proctectomy).	Number of participants who have SD: 102 (54%) Factors affecting the sexual function: Type of treatment (Type of curative surgery (P < 0.001), Post operative chemotherapy (P = 0.03)), Being metastatic cancer (P = 0.007)
Oberguggenberger / 2017 (58)	Breast Cancer	Austria / Innsbruck	Cross-sectional	Sexual Interest and Desire Inventory-Female (SIDI-F), Sexual Activity Questionnaire (SAQ), Body Image Scale (BIS), Hospital Anxiety and Depression Scale (HADS), Menopause-Specific Quality of Life Questionnaire (MENQOL)	105 Patients	Breast Cancer survivors	Number of participants who have SD: 73 (68.8%) Factors affecting the sexual function: Higher depressive symptoms (P = 0.008), higher age (P = 0.039), lower partnership satisfaction (P < 0.001), and Grade of tumor (P = 0.008)
Chirani / 2017 (59)	Breast Cancer	Iran / Sari	Cross-sectional	Demographic Characteristics Questionnaire, Female Sexual Function Index (FSFI)	104 patients	Breast cancer survivors with mastectomy	Number of participants who have SD: 83 (78.5%) Factors affecting the sexual function: Age (Being older): (P = 0.001), Income (Low): (P = 0.05)
Pinar / 2016 (60)	Gynecological Cancer which included: Endometrium, Ovarian, Cervical, Vulvar	Turkey / Ankara	Cross-sectional	Patient Information Forms, Index of Female Sexual Function (IFSF)	230 patients	Female with different gynecological cancers	Number of participants who have SD: 184 (80%) Factors affecting the sexual function: Age (Being older): (P = 0.031), Education (Primary school): (P = 0.004), Type of treatment (Surgical interventions): (P = 0.001)
Öztürk / 2016 (61)	Breast Cancer	Turkey / Istanbul	Cross-sectional	Individual characteristics form and Female Sexual Function Index (FSFI)	100 patients	Female with Breast cancer	Number of participants who have SD: 100 (100%) Factors affecting the sexual function: Type of treatment (Surgical intervention, Mastectomy): (P = 0.001), Age (Older): (P = 0.002)
Lin / 2016 (62)	Prostate Cancer	Iran / Tehran, Qazvin, Ahvaz, Guilan and Tabriz	Cross-sectional	Premature Ejaculation Diagnostic Tool (PEDT), International Index of Erectile Function (IIEF-5), Mini-Mental State Examination (MMSE), and participants characteristics form	1058 patients	Male with Prostate cancer	Number of participants who have SD: 700 (66.2%) Factors affecting the sexual function: Work (Not have): (P = 0.07), Smoking (Yes): (P < 0.001), Stage of tumor (Stage 4): (P = 0.008), Premature Ejaculation (Low score): (P = 0.04)
Safarinejad / 2012 (83)	Breast cancer	Iran / Tehran	Case-control	Quality of life was assessed by Short Form-36 Health Survey (SF-36), Female Sexual Function Index (FSFI), and participants characteristics form	186 patients	Younge women with early stage of Breast cancer 1 year after lumpectomy	Number of participants who have SD: 100 (53.76%) Factors affecting the sexual function: Type of treatment (Hormone-treated and Combination therapy): (P = 0.006, P < 0.05, respectively)
Ong / 2014 (74)	Prostate Cancer	Australia / Melbourne	prospective longitudinal cohort design (Between 1998 and 2011)	International Index of Erectile Function (IIEF-5), and participants characteristics form	366 patients	Males with prostate cancer who were under seed brachytherapy treatment	Number of participants who have SD: 194 (53%) Factors affecting the sexual function: Age (High): (P = 0.004), Medical comorbidities (Yes): (P = 0.02), Gleason score (< 7.7): (P < 0.001)
Kedde / 2013 (69)	Breast cancer	Netherlands / Utrecht and Groningen	Cross-sectional	The short version of the Questionnaire for Screening SDs, The Consumer Quality Index (CQI) Mammacare, and participants characteristics form	365 patients	Younge women with Breast Cancer	Number of participants who have SD: 198 (54.5%) Factors affecting the sexual function: Receiving any type of treatment (P < 0.01), Experiencing the joint pain (P < 0.05), Axillary lymph node dissection (P < 0.05), Early menopause (P < 0.01)
Kim / 2012 (84)	Testicular Cancer	United states of America / New Haven, Connecticut, and Rockville, Maryland	Case-Control	Brief Male Sexual Function Inventory (BMSFI), Self-Administered Questionnaire, and participants characteristics form	246 patients	Males who have Testicular cancer	Number of participants who have SD: 93 (25.47%) Factors affecting the sexual function: Type of testicular cancer (Nonseminoma): (P < 0.05), Type of treatment (Combination therapy): (P < 0.05)
Tsai / 2010 (28)	Cervical Cancer	Taiwan / Taipei	Cross-sectional	Female Sexual Function Index (FSFI), Demographic and Disease Characteristics Questionnaire	105 patients	Females with Cervical cancer	Number of participants who have SD: 70 (66.67%) Factors affecting the sexual function: Age (Older): (P = 0.00), Education (Low): (P = 0.00), Menopause (Yes): (P = 0.03), Sexual counselling service (No): (P = 0.00), Stage of tumor (2 or above): (P = 0.00)
Cleary / 2011 (85)	Gynecological Cancer that included: Cervical, Ovarian, Endometrial, and Vulvar	Southern Ireland	Cross-sectional	Body Image Scale, Sexual Esteem Scale, Sexual Self-Schema Scale, Intimate Relationship Scale, Arizona Sexual Experience Scale	109 patients	Females with diagnosis of and treatment for various gynecological cancers	Number of participants who have SD: 109 (100%) Factors affecting the sexual function: Age (Older): (P < 0.001), Type of cancer (Endometrial): (P = 0.007), Completion of treatment (Completed patients): (P = 0.015)
Brédart / 2010 (70)	Breast cancer	France / Paris	Cross-sectional	EORTC QLQ-C30 and QLQ-BR23 Questionnaires, Body Image Scale (BIS), Sexual Activity Questionnaire (SAQ), French Sexual Behaviour Survey (CSF)	378 patients	Post treatment females with early stage of Breast cancer	Number of participants who have SD: 109 (29%) Factors affecting the sexual function: Age (Older): (P = 0.04), Having partner (Not have): (P < 0.001), More recent Breast Cancer diagnosis (Yes): (P = 0.01), Menopause (Yes): (P = 0.06), Comorbid disease (Have): (P = 0.01), History of mental health (Yes): (P = 0.01), Body Image (Poor): (P < 0.001), Emotional separation between partners (Yes): (P < 0.001), Quality of life (Poor): (P = 0.08)
Recklitis / 2010 (63)	Hodgkin’s Lymphoma	United States of America / Boston	cross-sectional	Global Sexual Satisfaction Index, and participants characteristics form	465 patients	Survivor patients with Hodgkin’s Lymphoma	Number of participants who have SD: 252 (54.2%) Factors affecting the sexual function: Age (Older): (P < 0.05)
Bruheim / 2010 (64)	Rectal Cancer	Norway / Oslo	cross-sectional	Sexual Function and Vaginal Changes Questionnaire (SVQ), and participants characteristics form	172 patients	Rectal cancer patients without recurrence or metastases at the time of the study	Number of participants who have SD: 62 (36%) Factors affecting the sexual function: Vaginal changes: P = 0.02, Vaginal bleeding during intercourse: P = 0.001, Reduced vaginal dimension: P = 0.001
van der Aa/ 2009 (65)	Bladder Cancer	Netherlands / Rotterdam, Leiden, Amsterdam	Cross-sectional	Demographic characteristics, a visual analogue scale, and validated subset of questions on sexual function and performance derived from QLQ-BLS-24	150 patients	Males who were patients recently diagnosed with primary or recurrent non-muscle-invasive bladder cancer	Number of participants who have SD: 63 (42%) Factors affecting the sexual function: Sex (Being woman): (P = 0.02), Age (Older): (P = 0.02), Perception of overall healthy (Lower): (P = 0.03)
Eberhard / 2009 (66)	Testicular Cancer	Sweden / Lund, Malmö, Stockholm	Cross-sectional	Patient Characteristics form, Measures of Sexual Functions/Dysfunctions	129 Males	Males with Testicular Germ Cell Cancer	Number of participants who have SD: 15 (12%) Factors affecting the sexual function: There were no potential factors found
Taylor / 2004 (71)	Ovarian Cancer	United States of Amera / Houston	Cross-sectional	Sexual Activity Questionnaire (SAQ), Memorial Symptom Assessment Scale, Eastern Cooperative Oncology Group (ECOG) Performance Status Rating, Center for Epidemiologic Studies Depression Scale (CES-D), Spielberger State-Trait Anxiety Inventory, Demographic and Medical Questionnaires	232 patients	Females who diagnosed with epithelial ovarian cancer	Number of participants who have SD: 116 (50%) Factors affecting the sexual function: Having partner (Not have): (P = 0.000), Age (Older): (P = 0.000), Any type of treatment receives (Yes): (P = 0.004), Time since diagnosis (Recently): (P = 0.104), Liking appearance of body: (P = 0.004)
Beckendorf / 1996 (81)	Prostate Cancer	France / Vandoeuvre-les-Nancy, Montpellier, Clermont-Ferrand, Nice, Reims	prospective observational (From January 1992)	Repeated Questionnaires, Clinical Examinations,	67 patients	The participants were men diagnosed with prostate cancer	Number of participants who have SD: 27 (40.29%) Factors affecting the sexual function: Age (Older): (P < 0.05)
Frankland/ 2020 (82)	Colorectal Cancer	United Kingdome / 29 hospitals across the across the country	Prospective longitudinal (from pre-surgery baseline to 60 months post-surgery, with eight assessment points)	Quality of Life in Adult Cancer Survivors (QLACS) Scale, Center for Epidemiological Studies Depression Scale (CES-D), State-Trait Anxiety Inventory (STAI), Medical Outcomes Study Social Support Survey (MOS-SSS), European Organisation for Research and Treatment of Cancer (EORTC QLQ-C30) and EORTC QLQ-CR29	790 patients	Patients who diagnosed with non-metastatic colorectal cancer (Dukes’ stages A–C) and treated with curative intent.	Number of participants who have SD: 403 (51%) Factors affecting the sexual function: Having a stoma, high levels of depression, high self- efficacy (very confident), full social support, and higher quality of life (P < 0.05)

Geographically, studies originated from Asia, Europe, North America, Australia, and Africa, with the largest representation from Asia and Europe. A variety of validated instruments were employed to assess sexual function, most commonly the Female Sexual Function Index (FSFI) for women and the International Index of Erectile Function (IIEF) for men. Other tools included the PROMIS Sexual Function and Satisfaction (SexFS) scale, Sexual Activity Questionnaire (SAQ), and Sexual Interest and Desire Inventory-Female (SIDI-F). Many studies also incorporated broader psychosocial measures such as the EORTC QLQ-C30, QLQ-BR23, QLQ-CR29, Body Image Scale (BIS), and Hospital Anxiety and Depression Scale (HADS).

Participants represented diverse age groups and clinical stages, including patients under active treatment and survivors. Most studies reported a high prevalence of sexual dysfunction, frequently exceeding 50%. Commonly associated factors included older age, menopausal status, comorbidities, lower socioeconomic status, chemotherapy or hormone therapy, and negative body image. Psychosocial correlates such as depression, anxiety, fatigue, and lack of partner support were also consistently identified, underscoring the multifactorial nature of SD in oncology. Detailed participant characteristics are provided in Supplementary Material File 2, Table 4.

Quality assessment of included studies

The methodological quality of the 39 included studies (cross-sectional, cohort, and case-control) was evaluated using the OSQE tool (Supplementary Material File 2, Tables 1, 2 and 3). Based on predefined percentage thresholds, studies were classified as good, moderate, or poor quality.

Among the 25 cross-sectional studies, 18 (72%) [28, 35, 51–66] were rated as good quality, 5 (20%) [67–71] as moderate, and 1 (4%) [72] as poor. Common methodological limitations included the absence of a published protocol, inadequate handling of missing data, and insufficient justification of sample size.

Of the 11 cohort studies, 6 (55%) [34, 73–77] were rated as good and 5 (45%) [78–82] as moderate quality. Frequent concerns involved lack of assessor blinding, absence of protocol registration, and incomplete follow-up reporting.

All three case-control studies (100%) [33, 83, 84] were assessed as good quality, though they frequently lacked information on non-response rates, missing data strategies, and sample size calculations. One cross-sectional study was [85] classified as poor quality due to multiple critical flaws.

Overall, most included studies demonstrated acceptable methodological quality; however, recurring issues—particularly missing data management, sample size adequacy, and protocol transparency—underscore areas requiring improvement in future observational research on sexual dysfunction in cancer populations.

Prevalence of SD

The pooled prevalence of SD among cancer patients, based on a random-effects meta-analysis of 39 studies involving 17,970 participants, was estimated at 54.0% (95% CI: 50.4%–58.8%), with substantial heterogeneity observed across studies (I2 = 95%) (Fig. 2). Sensitivity analysis, conducted by sequentially removing one study at a time, indicated that no single study significantly influenced the overall pooled prevalence estimate of SD (pooled prevalence = 54.0%; 95% CI: 50.4%–58.8%) (Fig. 3). Assessment of publication bias using visual inspection of the funnel plot (Fig. 4), along with Begg and Mazumdar’s rank correlation test (P = 0.281) and Egger’s regression test (P = 0.846), indicated no significant evidence of publication bias among the included studies.

Fig. 2.

The prevalence of sexual dysfunction. (Random effect model)

Fig. 3.

Sensitivity analysis for the prevalence of sexual dysfunction. (Random effect model)

Fig. 4.

Publication bias (funnel plot) for prevalence of SD

Meta-regression

Meta-regression analyses were performed to identify study-level factors associated with the pooled prevalence of SD (Table 2). Among continuous moderators, year of publication significantly influenced prevalence estimates (β = 0.03, p = 0.034), with more recent studies reporting higher SD rates. Other variables—including sample size, gender composition, mean age, education level, employment status, menopausal status, smoking, and marital/partner status—were not statistically significant. Substantial residual heterogeneity (I² ≈ 95%) remained after adjustment, suggesting unmeasured sources of variability (Supplementary Material File 2, Figs. 1–9).

Table 2.

Random effect meta-regression analysis of moderator variables predicting the overall prevalence of SD

Moderator variable (type)	Number of studies	Regression coefficient (SE)	95% CI	Z-value	P-value	QM, df, P-value	QE, df, P-value
Sample size (Continuous)	39	0.0000 (0.0001)	-0.0001-0.0002	0.15	0.884	0.02, 1, 0.8842	761.96, 37, p < 0.001
Year of study (continuous)	39	0.0304 (0.0143)	0.0023–0.0584	2.12	0.0341	4.49, 1, 0.0341	743.78, 37, p < 0.001
Being female (continuous)	39	0.0000 (0.001)	-0.0001-0.0002	0.19	0.852	0.03, 1, 0.8524	766.65, 37, p < 0.001
Mean age of participants (continuous)	29	-0.0103 (0.0109)	-0.0317-0.0110	-0.95	0.3410	0.91, 1, 0.3410	585.92, 27, p < 0.001
Number of participants that have education level till high school (continuous)	21	-0.0001 (0.0002)	-0.0004-0.0002	-0.38	0.701	453.08, 19, 0.701	453.08, 19, p < 0.001
Number of participants who have job (continuous)	19	-0.0023 (0.0012)	-0.0046-0.0000	-1.92	0.055	3.68, 1, 0.0551	347.73, 17, p < 0.001
Being menopause (continuous)	10	-0.0000 (0.0001)	-0.0003-0.0003	-0.20	0.844	0.04, 1, 0.8441	150.78, 8, p < 0.001
Being smoker (continuous)	6	0.0001 (0.0001)	-0.0002-0.0003	0.46	0.6451	0.21, 1, 0.6451	18.04, 4, 0.0012
Having sexual partner (continuous)	22	0.0000 (0.0001)	-0.0002-0.0002	0.18	0.8543	0.03, 1, 0.8543	492.58, 20, p < 0.001
Type of cancer (Categorical, Gynecological was considered as reference category)	39	1.53306 (0.3636)	0.8179–2.2433	4.21	P < 0.001	30.09, 8, 0.0002	630.67, 30, p < 0.001
Type of study (Categorical, cross-sectional studies were considered as reference category)	39	0.1909 (0.3988)	-0.5907-0.9726	0.48	0.6321	2.92, 2, 0.2322	774.64, 369, p < 0.001
Continent (Categorical, America was considered as reference category)	39	-0.0220 (0.2394)	-0.4912-0.4471	-0.09	0.9266	19.5, 4, 0.0007	692.65, 34, p < 0.001
Health status of participants (Categorical, survivors Vs. under treatment)	39	0.1167 (0.1954)	-0.2663-0.4996	0.60	0.5505	0.36, 1, 0.5505	763.53, 37, p < 0.001
Tool of measurement the SD (Categorical, FSFI & IIEF considered as reference category)	39	0.2563 (0.1500)	-0.0377-0.5502	1.71	0.0875	15.24, 11, 0.1719	528.74, 27, P < 0.001

Cl: Confidence intervale, SE: Standard error, QM: Q for moderators, df: Degree of freedom, QE: Q for errors

Categorical meta-regression revealed that cancer type (QM = 30.09, p < 0.001) and continent (QM = 19.15, p = 0.0007) were significant moderators of prevalence, whereas study design, participant health status (survivor vs. under treatment), and measurement tool were not (Table 2; Supplementary Material File 2 Figs. 10, 11, 12, 13 and 14). SD prevalence was higher among patients with gynecological cancers and in studies conducted in Asian countries.

Subgroup analysis

To further explore these moderators, subgroup analyses were conducted (Table 3). By year of publication, SD prevalence increased from 40% in studies before 2010 to nearly 64% during 2011–2020, declining slightly thereafter. By cancer type, prevalence was highest in gynecological (93.7%), breast (59.2%), and colorectal (57.4%) cancers, and lowest in testicular (22.4%) and lymphoma (43.8%) populations. By continent, the highest pooled prevalence was observed in Asia (69.0%), followed by Australia (53.0%) and America (48.5%).

Table 3.

Subgroup analysis of SD based on continent, year of study, and type of cancer

Moderator Variable	Category	Event rate (95% CI)
Year of publication	Less than 2000	40.3% (29.3%-52.4%)
	2001–2019	40.0% (29.1%-52.1%)
	2011–2020	63.9% (56.8%-70.4%)
	More than 2020	54.0% (47.6%-60.3%)
Type of cancer	Blader cancer	48.7% (35.5%-62.1%)
	Breast cancer	59.2% (53.3%-64.9%)
	Cervical cancer	49.0% (31.7%-66.6%)
	Colorectal cancer	57.4% (44.2%-69.6%)
	Gynecological cancer	93.7% (78.8%-98.3%)
	Lymphoma cancer	43.8% (24.8%-64.7%)
	Prostate cancer	54.4% (41.2%-66.9%)
	Testicular cancer	22.4% (6.0%-56.3%)
	Various cancers	45.3% (21.5%-71.4%)
Continent	Africa	47.9% (35.9%-58.4%)
	America	48.5% (39.5%-57.6%)
	Asia	69.0% (60.6%-76.3%)
	Australia	53.0% (47.9%-58.1%)
	Europe	46.6% (40.0%-53.2%)

CI: Confidence interval

SD determinants

A narrative synthesis (Table 4) identified demographic, clinical, psychosocial, and treatment-related determinants of SD across cancer types. Demographic factors: Older age, lower education, and unemployment were consistently associated with greater SD risk. Clinical and treatment-related factors: Advanced disease stage, chemotherapy, surgery (particularly mastectomy and cystoprostatectomy), early menopause, and comorbidities were frequent predictors. Psychosocial factors: Negative body image, depressive symptoms, poor relationship satisfaction, and lack of sexual counselling were strong correlates across several cancers. Protective factors included physical activity, good mental health, and supportive partnerships. These findings highlight the multifactorial etiology of sexual dysfunction among cancer populations, emphasizing the need for comprehensive biopsychosocial assessment.

Table 4.

Predictive factors associated with SD based on cancer type

Type of cancer	Predictors	Type of cancer	Predictors
Blader cancer	Demographic and Biological Factors: • Sex (Being woman): P = 0.02 • Age (Older): P = 0.02 • Body Mass Index (Increased BMI): P = 0.0025 Clinical and Treatment-Related Factors: • Surgical treatment (Radical cystoprostatectomy): P = 0.0487 • Perception of overall health (Lower): P = 0.03	Colorectal cancer	Demographic and Socioeconomic Factors: • Age (More than 40 / Being older): P = 0.02, P = 0.00 • Education (Above high school / Having higher degree): P = 0.026, P = 0.004 • House type (Rent): P < 0.001 • Working (Yes): P < 0.001 Clinical and Treatment-Related Factors: • Having colostomy: P < 0.001 Type of treatment: • Chemotherapy-surgery: P < 0.001 • Curative surgery: P < 0.001 • Postoperative chemotherapy: P = 0.03 • Time since last chemotherapy (5–10 weeks or more): P = 0.018 • Tumor site (Rectum vs. colon): P = 0.07, P < 0.05 • Metastatic cancer (Yes): P = 0.007 • Menopausal status (Being menopausal): P = 0.13 Sexual Health and Symptoms: • Vaginal changes: P = 0.02 • Vaginal bleeding during intercourse: P = 0.001 • Reduced vaginal dimension: P = 0.001 Psychosocial Factors: • Comorbid conditions (Yes): P = 0.047 • Depression (High levels): P < 0.05 • Self-efficacy (High confidence): P < 0.05 • Social support (Full): P < 0.05 • Quality of life (Higher): P < 0.05
Breast cancer	Demographic factors: • Age (Older): P = 0.001, P = 0.002, P = 0.039, P = 0.04 • Income (Low): P = 0.011, P = 0.05 • Marital status (Not having partner): P < 0.001 • Education (Above high school): P < 0.05 • Work (Not employed): P = 0.07 • Smoking (Yes): P < 0.001 • Duration of marriage: P = 0.04 Psychological and Psychosocial Factors: • Insomnia: P = 0.048 • Higher depressive symptoms: P = 0.008 • History of mental health issues: P = 0.01 • Body image disturbance / Negative body image / Poor body image: P < 0.001, P < 0.05 • Reproductive concerns (High concern): P < 0.05 • Lower partnership satisfaction: P < 0.001 • Emotional separation between partners: P < 0.001 • Poor quality of life: P = 0.08 • Not interested in sex: P < 0.001 • Too tired: P < 0.05 • Physical problems: P < 0.05 Clinical Factors: • Tumor grade (High): P = 0.008 • Tumor stage (Stage 4): P = 0.008 • Comorbid disease (Yes): P = 0.01 • Premature ejaculation (Low score): P = 0.04 • Early menopause: P < 0.01 • Menopause (Yes): P = 0.06 • Family history of breast cancer: P = 0.040 • More recent breast cancer diagnosis: P = 0.01 Treatment-Related Factors: • Chemotherapy (Yes / previous): P = 0.029 • Endocrine therapy: P = 0.032 • Current endocrine treatment and previous chemotherapy: P < 0.05 • Hormone-treated therapy: P = 0.006 • Combination therapy: P < 0.05 • Adjuvant inhibitor vs. non-adjuvant: P = 0.000 • Receiving any type of treatment: P < 0.01 • Surgical intervention / mastectomy: P = 0.001 • Axillary lymph node dissection: P < 0.05 • Experiencing joint pain: P < 0.05	Gynecological cancer	Demographic and Socioeconomic Factors: • Age (Older): P < 0.001, P = 0.031, P = 0.000 • Education (Primary school): P = 0.004 • Working (Yes): P = 0.016 Clinical and Cancer-Related Factors: • Type of cancer (Endometrial cancer): P = 0.020, P = 0.007 • Cancer type (Unspecified): P = 0.025 • Cancer stage: P = 0.023 • Duration of disease: P = 0.017 • Completion of treatment (Completed patients): P = 0.015 • Time since diagnosis (Recently): P = 0.104 Treatment-Related Factors: • Receiving any type of treatment (Yes): P = 0.028, P = 0.004 • Type of treatment: • Surgical interventions: P = 0.001 Psychosocial and Behavioral Factors: • Having a partner (Not having): P = 0.000 • Smoking (Yes): P = 0.035 • Social support (Higher): P < 0.001 • Liking appearance of body: P = 0.004
		Various cancers	• Physical activity (Active): P = 0.02 • Mental health (Good mental health): P = 0.04 • Age (Older): P < 0.05 • Marital status (Having a partner): P < 0.05 • Cancer type (Lymphoma): P < 0.05 • Mental health (Poor): P < 0.05 • Body image disturbance (Present): P < 0.05
Cervical cancer	Demographic and Socioeconomic Factors: • Age (Getting older): P < 0.001, P = 0.00 • Education (Low or up to high school): P = 0.022, P = 0.00 • Income status (Working): P < 0.001 • Having a partner (Yes): P < 0.001 • Obesity (Being obese): P = 0.005 Clinical and Psychological Factors: • Menopausal status (Being menopausal): P = 0.001, P = 0.03 • Mood status (Being depressed): P < 0.001 • Fatigue (Chronic fatigue): P = 0.002 • Stage of tumor (Stage 2 or above): P = 0.00 • Lack of sexual counselling services: P = 0.00	Lymphoma cancer	Demographic factors: • Age (Higher): P < 0.005, P < 0.05
Prostate cancer	• Age (Higher): P = 0.004, P < 0.05 • Medical comorbidities (Yes): P = 0.02 • Gleason score (< 7.7): P < 0.001	Testicular cancer	• Type of testicular cancer (Nonseminoma): P < 0.05 • Type of treatment (Combination therapy): P < 0.05

Discussion

This systematic review and meta-analysis aimed to estimate the prevalence and identify the key determinants of SD among individuals with various types of cancer. Drawing on data from 39 studies comprising a total of 17,970 participants, the pooled prevalence of SD was calculated at 54%, underscoring its widespread and persistent impact across cancer populations. However, the high heterogeneity (I² = 95%) and varying risk of bias across included studies, particularly differences in study design, measurement tools, and follow-up timing, may affect the precision and applicability of this estimate. Consequently, while the pooled prevalence provides a useful overall indicator, interpretation should account for these limitations.

These findings reflect the multifactorial etiology of SD in cancer patients, involving not only biological and treatment-related factors but also psychosocial, cultural, and relational influences. Notably, prior meta-analyses have primarily focused on specific cancer types such as gynecological cancers, without providing an overall pooled prevalence in different type of cancers. For instance, Hosseini et al. [36] reported a pooled prevalence of 60% among patients with gynecological cancer, while Qian et al. [86] documented an even higher rate of 80% in the same population. Compared to earlier findings, the present review offers a broader perspective by not only synthesizing data across diverse cancer types but also providing an overall pooled prevalence of SD across different cancer populations. Collectively, these results suggest that SD is a common yet often underrecognized and underreported concern in oncology care.

This review makes several novel contributions beyond previous systematic reviews and meta-analyses. Unlike earlier reviews that focused narrowly on single cancer types (e.g., gynecological or testicular cancers [36, 37], our study synthesized data across diverse cancer populations to provide the first comprehensive pooled prevalence of SD. In addition, by conducting subgroup analyses and meta-regressions, we systematically evaluated moderators such as cancer type, geographic region, and publication year, offering new insights into sources of heterogeneity. Our integration of multiple theoretical frameworks further advances the conceptual understanding of SD in oncology by moving beyond descriptive prevalence estimates to explain how biological, psychological, and social factors interact to shape outcomes.

Our subgroup analyses revealed a significantly higher prevalence of SD among individuals with gynecological (93.7%), breast (59.2%), and colorectal (57.4%) cancers. These elevated rates are likely attributable to the direct involvement of sexual and reproductive organs, the impact of intensive treatments such as chemotherapy, radiotherapy, and surgery, and the associated psychosocial burden these patients often experience. Previous meta-analyses have reported somewhat varied prevalence estimates. For example, a pooled prevalence of 80% was observed among patients with cervical cancer [86], while rectal cancer was associated with a prevalence of 35% [87].

Such variation in prevalence across cancer types and studies can be attributed to several methodological and clinical factors. First, differences in assessment tools used to measure SD (e.g., FSFI, IIEF, non-standardized questionnaires) contribute to inconsistency in prevalence estimates [88]. Second, timing of assessment, whether conducted during treatment, immediately after, or years into survivorship, can significantly influence reported outcomes [37, 89]. Third, cultural and regional factors may shape both patient willingness to disclose sexual concerns and healthcare providers’ readiness to assess them, thus affecting prevalence rates across populations [90–92]. Additionally, variability in treatment protocols, such as the extent of surgical resection or use of hormone therapies, directly influences sexual functioning, especially in cancers affecting the pelvic region [93, 94]. Lastly, sample characteristics, including age, menopausal status, relationship status, and baseline sexual activity, can also contribute to observed differences [95, 96]. Studies that fail to control for these factors may overestimate or underestimate the true burden.

Another notable finding from our subgroup analysis was the variation in SD prevalence across geographical regions. Studies conducted in Asia reported the highest prevalence (69.0%), while those from Europe reported the lowest (46.6%). This is consistent with prior meta-analytic findings, where the highest prevalence was observed in Africa (75%) and the lowest in Europe (43.0%) [36]. These regional differences may reflect cultural norms and reticence surrounding discussions of sexuality, variations in patient-provider communication, or differences in healthcare system priorities and screening practices related to sexual health.

Furthermore, our subgroup and meta-regression analyses indicated a temporal trend, with the prevalence of SD increasing in more recent studies. This rise may be partially explained by greater awareness of sexual health issues, enhanced methodological sensitivity, and the increased use of standardized, validated assessment tools over time. These findings suggest that SD is becoming more widely recognized in oncology research and clinical practice, although it remains insufficiently addressed in many settings.

Despite conducting extensive meta-regressions, heterogeneity remained high (I² = 95%), which is consistent with the limitations of ecological analyses and may reflect unmeasured variables, differences in assessment tools, timing post-treatment, or variation in definitions of SD. Such residual heterogeneity is also a recognized issue in meta-analyses involving patient-reported outcomes [97, 98].

In addition to estimating the pooled prevalence, our review also synthesized evidence on the determinants of SD across various cancer populations. The narrative synthesis revealed that SD is influenced by a broad spectrum of demographic, clinical, psychosocial, and treatment-related factors, underscoring its multifactorial nature. Across multiple cancer types, older age consistently emerged as a key demographic predictor, likely reflecting age-related physiological changes and the cumulative burden of comorbidities that compromise sexual function [52, 99]. Low educational attainment, unemployment, and low income were also associated with higher rates of dysfunction, suggesting that socioeconomic disadvantage may exacerbate the risk of sexual health impairments by limiting access to information, support services, and resources for coping [100–102].

From a clinical standpoint, patients undergoing more aggressive or invasive treatments, such as radical in bladder cancer, colostomies in colorectal cancer, and mastectomy or endocrine therapy in breast cancer, were more likely to report SD [35, 80, 103]. These procedures and interventions often have direct physiological consequences on sexual organs or indirectly affect hormonal pathways and body image [104, 105]. In cancers involving the pelvic region, such as gynecological and colorectal cancers, the impact on sexual function was particularly pronounced, with factors like tumor stage, active treatment status, and type of surgical intervention significantly influencing outcomes [106, 107]. Similarly, in cervical cancer, advanced disease stage, menopausal status, and lack of access to sexual counseling were strongly associated with impaired sexual function [73, 108].

Psychosocial variables played a pivotal role across cancer types. Factors such as poor body image, negative partner relationships, depression, fatigue, and emotional distress were frequently cited as significant contributors to SD, particularly in breast and gynecological cancers. Conversely, protective factors such as being married, having good mental health, engaging in physical activity, and absence of body image concerns were associated with lower prevalence of dysfunction in studies involving heterogeneous cancer populations. These findings align with the biopsychosocial model, which emphasizes the integration of biological disruptions, psychological responses, and social context in shaping health outcomes [109, 110].

Our narrative synthesis further highlights the distinction between common risk factors observed across multiple cancer types and cancer-specific predictors. Common demographic and psychosocial risk factors included older age, low educational attainment, low income, depression, fatigue, and poor body image. These factors were consistently reported across heterogeneous populations, suggesting that they represent general vulnerabilities to sexual dysfunction in cancer survivors.

In contrast, certain risk factors were more specific to particular cancers. For example, in bladder cancer, radical cystoprostatectomy and lower perceived overall health were strongly associated with SD. Colorectal cancer patients experienced higher SD prevalence related to colostomy, chemotherapy-surgery combinations, tumor site, and postoperative complications. In breast cancer, interventions such as mastectomy, axillary lymph node dissection, endocrine therapy, and hormone-related treatments were notable contributors, alongside psychological stressors like body image disturbance and depressive symptoms. For gynecological cancers, advanced disease stage, menopausal status, and lack of sexual counseling were frequently reported. Similarly, in cervical cancer, older age, low education, and obesity were important contributors, while in prostate and testicular cancers, age, comorbidities, Gleason score, and type of treatment emerged as specific determinants.

This distinction between common and cancer-specific factors is critical for guiding clinical practice and intervention design. While common factors indicate areas for broad supportive strategies applicable across cancer populations, cancer-specific factors highlight the need for tailored interventions that address the unique physiological and treatment-related challenges faced by patients with particular malignancies. Future research should aim to quantify these associations in a meta-analytic framework where feasible, using standardized effect measures to allow direct comparison across cancer types.

A deeper understanding of the complexity and variability of SD across cancer types can be achieved by situating the findings within Engel’s Biopsychosocial Model [111]. This framework is particularly well suited because it views health outcomes as the result of interconnected biological, psychological, and social influences. Our results map directly onto these domains: biological factors such as surgical procedures, endocrine therapy, and tumor location; psychological processes including depression, fatigue, and body image concerns; and social determinants such as relationship quality, cultural expectations, and socioeconomic disadvantage. Together these illustrate that sexual dysfunction is rarely the result of a single cause but emerges from the interplay of multiple domains, which reinforces the need for integrated, multidisciplinary approaches to care.

Other theoretical perspectives provide additional nuance to this interpretation. The Lazarus and Folkman’s Transactional Model of Stress and Coping [112], for example, helps explain why some individuals are more resilient than others in adjusting to cancer-related changes in sexual health, depending on their appraisals and coping strategies. Similarly, the Health Belief Model [113] and the Theory of Planned Behavior [114] clarify why sexual concerns are often underreported: patients’ perceptions of severity, cultural stigma, and low self-efficacy may deter them from initiating conversations about sexual health. Concepts from the sexual response cycle witch first developed by Masters and Johnson [115] and later expanded by Kaplan [116] also highlight how treatment may disrupt specific phases of sexual functioning, while self-determination theory [117, 118], underscores the importance of autonomy, competence, and supportive relationships in promoting sexual wellbeing.

By grounding the interpretation primarily in the biopsychosocial framework while drawing selectively on these complementary models, the discussion offers a coherent theoretical lens through which the observed prevalence patterns and determinants of sexual dysfunction can be understood. This integrated approach not only strengthens the conceptual foundation of the review but also underscores the value of addressing biological, psychological, and social dimensions simultaneously in both research and clinical practice.

Limitations, Strengths, and future directions

This study has several strengths. It is among the few meta-analyses to comprehensively evaluate the prevalence of SD across diverse cancer types while examining a broad range of potential moderators through meta-regression. By incorporating demographic, clinical, psychosocial, and geographic factors, the review provides an in-depth understanding of heterogeneity sources. Moreover, the use of a validated risk of bias tool (OSQE) enhanced methodological rigor, transparency, and reproducibility.

However, several limitations should be acknowledged. First, substantial heterogeneity across studies limits the generalizability of pooled estimates. Although subgroup and meta-regression analyses were conducted, much of this heterogeneity remained unexplained, partly due to insufficient reporting of key sociodemographic and clinical moderators. Second, important individual-level determinants of SD, such as psychological distress, body image, cultural factors, relationship quality, and treatment regimens—were inconsistently reported, restricting their inclusion in analyses. Complex interactions among moderators (e.g., age with treatment type or menopausal status with cancer site) likely contributed to variability but could not be adequately modeled. Third, reliance on self-reported measures may have introduced bias, as stigma or cultural taboos could lead to underreporting and underestimation of SD prevalence.

Future studies should employ standardized, validated instruments and ensure comprehensive reporting of sociodemographic and clinical variables. Individual participant data (IPD) meta-analyses are recommended to capture nuanced moderators and interaction effects. Longitudinal designs are also needed to track the trajectory of SD over time. Finally, culturally sensitive approaches are essential to address stigma and encourage open discussions of sexual health in cancer care.

Conclusion

This systematic review and meta-analysis demonstrate that sexual dysfunction is highly prevalent among individuals with cancer, affecting more than half of patients across diverse cancer types. The burden is greatest among those with gynecological, breast, and colorectal cancers, with notable variation by region and time. A wide range of demographic, clinical, psychosocial, and treatment-related factors contribute to this issue, emphasizing its multifactorial nature and the need for comprehensive, patient-centered care. Persistent heterogeneity suggests that unmeasured individual and contextual factors remain influential. Overall, the findings highlight sexual health as a critical yet often neglected component of quality cancer care.

Implications for research and clinical practice

Clinically, these results underscore the importance of routine sexual health screening and multidisciplinary support within oncology settings. Integrating sexual health assessment into survivorship care plans and improving patient–provider communication can help address unmet needs.

Future research should adopt standardized measurement tools, longitudinal approaches, and individual-level data to better understand moderators and causal pathways. Greater attention to cultural sensitivity and underreported populations will further enhance the accuracy and applicability of evidence on sexual health in cancer care.

Authors’ contributions

The study was conceived and designed by AD, with contributions from MH, MHS. AD contributed to the conception and design of the study. AD conducted the literature search, while AD & MH was responsible for screening the studies and extracting data. Data analysis was performed by AD. AD and MH assessed the quality of the included studies. The manuscript was drafted by AD and MH, with revisions carried out by MHS and AD. All authors critically reviewed and approved the final manuscript.

Data availability

This systematic review and meta-analysis utilized data and materials from publicly accessible sources, including academic databases and repositories. All referenced studies and sources are fully listed in the manuscript’s reference section for easy access by readers. A comprehensive description of the search strategy and inclusion criteria employed to identify relevant studies can be found in the Methods section. Furthermore, the datasets analyzed in this study are available from the corresponding author, Mr. Amirmohammad Dahouri, upon reasonable request (amirmohammaddahouri61@gmail.com).

Declarations

Ethics approval and consent to participate

for this study was granted by the Research Ethics Committee of Tabriz University of Medical Sciences (Approval ID: IR.TBZMED.VCR.REC.1403.133). The certificate of ethical approval is provided in the Appendix file 1. As this study is a systematic review and meta-analysis, it involved the analysis of data from previously published studies; hence, no new patient data were collected. All methods and analyses were conducted in strict accordance with the relevant guidelines and regulations, including the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Proper referencing and citation of all included studies have been ensured to maintain academic integrity and transparency.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.