Updated dose–response meta-analysis of sexual activity and prostate cancer risk

Abstract

Background

This systematic review and meta-analysis aimed to update the evidence on how sexual activity factors, including frequency of intercourse, number of female sexual partners, age at first intercourse, and number of ejaculations and masturbations relate to prostate cancer.

Methods

We conducted a comprehensive search across PubMed, Web of Science, and Scopus up to October 19, 2025. We also reviewed existing study references for additional relevant research. To assess study consistency, we used I² statistics. Publication bias was evaluated using Begg and Egger tests. Our analysis reported pooled odds ratios (OR) with 95% confidence intervals using a random-effects model. The research protocol was approved by PROSPERO, ID CRD42024549018.

Results

From 5,311 identified studies, 29 (315,193 participants) met the inclusion criteria. Positive but non-significant associations were observed between prostate cancer and more frequent sexual intercourse (OR = 1.02; 95% CI: 0.88, 1.18), a higher number of female partners (OR = 1.01; 95% CI: 0.90, 1.12), and masturbation frequency (OR = 1.12; 95% CI: 0.82, 1.54). Older age at first intercourse showed a negative but non-significant association (OR = 0.99; 95% CI: 0.90, 1.08), whereas higher ejaculation frequency demonstrated a significant protective effect (OR = 0.83; 95% CI: 0.77, 0.90).

Conclusion

These findings suggest that most sexual behaviors are not significantly associated with prostate cancer risk. However, higher ejaculation frequency appears to have a protective effect. The results highlight the multifactorial nature of prostate cancer and emphasize the need for further research to clarify the role of sexual activity in disease prevention.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-025-15410-3.

Background

Prostate cancer is the second most commonly diagnosed cancer worldwide and the fifth leading cause of cancer-related mortality among men in 2022 [1, 2]. The burden of prostate cancer is projected to nearly double by 2040, from 1.4 million to 2.9 million cases annually, with mortality increasing by 85% to approximately 700,000 [3]. Although prostate cancer is frequent in most countries, geographic variation is evident globally [4]. Countries with a higher Human Development Index (HDI) exhibit a higher incidence rate of prostate cancer, primarily attributed to population aging and increased utilization of prostate-specific antigen (PSA) screening. Conversely, these countries typically report lower mortality rates [2, 5]. Established risk factors for prostate cancer include advanced age, black race, and a familial history of the disease [6, 7]. Evidence for lifestyle factors such as diet, physical activity, occupational exposure, or sexual activity, remains insufficient [8–11].

Sexual activity has been hypothesized to influence prostate cancer risk through several biological mechanisms, such as elevated androgen levels, exposure to sexually transmitted infections, and the potential accumulation of carcinogens within prostatic cells [12]. Sexual activity encompasses a wide range of behaviors related to sexual expression and engagement, including acts involving sexual stimulation or intercourse with oneself or a partner [13]. Despite numerous studies and systematic reviews [14–18], the association between sexual activity and prostate cancer risk remains inconclusive due to differences in study design, exposure definitions, and categorization of sexual behaviors. These inconsistencies have led to conflicting findings across previous studies, emphasizing the need for a more comprehensive and updated synthesis of evidence.

A previous meta-analysis conducted in 2018 included 22 studies and examined three types of sexual activity—ejaculation frequency, number of female sexual partners, and age at first sexual intercourse [15]. However, several additional studies have been published in recent years, providing new data and broader perspectives on this topic. The present meta-analysis extends the evidence base by including 29 studies and evaluating five distinct types of sexual activities, offering a more detailed and contemporary assessment of their potential association with prostate cancer risk. This updated analysis aims to clarify previous inconsistencies and provide more robust and comprehensive insights into how various aspects of sexual activity may influence prostate cancer development.

Methods

Research protocol and registration

The protocol of research was approved by the international prospective register of systematic review (PROSPERO) with identification CRD42024549018.

Eligibility criteria (PICOS)

This meta-analysis was conducted and reported in accordance with the guidelines outlined by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (see Additional file 2) [19].

Eligibility criteria were determined utilizing the Population, Intervention/Exposure, Comparison, Outcome, and Study (PICOS) framework.

Population

The research population comprised the general populace, devoid of any demographic constraints related to race, ethnicity, or nationality.

Intervention/exposure

The exposure of interest was sexual activities such as the frequency of sexual intercourse per week, number of female sexual partners, age at first intercourse, number of ejaculations per week, and number of masturbations per month. In this study, sexual activity was defined based on the data available in the included studies, which primarily referred to heterosexual intercourse between male participants and female partners. Furthermore, sexual intercourse refers to vaginal intercourse between male participants and female partners, which may or may not involve ejaculation.

Comparison

The same population engaging in different sexual activities.

Outcome

The outcome of interest was prostate cancer of any type which was confirmed pathologically.

Studies

We selected all observational studies, encompassing both cohort and case–control designs, that investigated the association between sexual activity and prostate cancer. These studies were selected based on their provision of effect size data, specifically in the form of odds ratios (OR), accompanied by their respective 95% confidence intervals (CI), without regard to publication status or language.

Information sources and search

A comprehensive literature search was conducted across major bibliographic databases, including PubMed, Scopus, and Web of Science, up to October 19, 2025. This search was unrestricted by time or language to ensure inclusivity. The search strategy employed a multifaceted approach, utilizing a combination of following keywords as both Text Words and MeSH terms to maximize coverage and comprehensiveness: (coitus or intercourse or coital frequency or ejaculation or sexual partners or multiple partners) and (prostate or prostatic) and (neoplasms or cancer or malignancy or tumor or carcinoma) to find the relevant studies (Appendix 1). Additionally, we cross-referenced the bibliographies of all included studies to uncover supplementary citations.

Study selection

The retrieved studies were imported into EndNote for management. Duplicate entries were systematically eliminated. Subsequently, two authors (A.R. and M.J.) conducted an independent assessment of the studies based on their titles and abstracts. The full texts of the remaining studies were then thoroughly examined to identify relevant studies. Data extraction from the included studies was performed by two authors (A.R. and S.O.). In both two steps, the authors disagreements were resolved through iterative discussion, with senior author arbitration used when consensus was not achievable. The inter-rater agreement during the screening phase was evaluated using Kappa statistics, yielding a concordance rate of 79.3%.

Data collection process

The required data was extracted and collected using the Stata Software. The extracted variables for analysis included the first author's name, publication year, country of study, age, exposure (sexual activities: frequency sexual intercourse per week, number of female sexual partners, and age at first sexual intercourse), study design (cohort and case–control), sample size, and effect size (OR) with their corresponding 95% confidence intervals (95% CI) for each study.

Study risk of bias assessment

In the final phase, the quality of reporting and potential bias in the included studies were evaluated using the Newcastle–Ottawa quality assessment Scale (NOS) manual (adopted for Cohort and Case–Control studies) [20]. This assessment tool allocates a maximum of nine stars across three domains: selection, comparability, and exposure/outcome ascertainment. Studies achieving seven or more star-items were categorized as having a low risk of bias, while those with six or fewer star-items were deemed to have a high risk of bias. Two investigators (A.R. and M.J.) conducted independent assessments of the methodological rigor of the selected studies, categorizing each as either high or low quality.

Heterogeneity and reporting biases

Heterogeneity was evaluated using the chi-squared test [21], I² statistics, and the tau-square test (τ²) [22]. Additionally, publication bias was assessed through the application of Beggs' [23] and Egger's [24] tests.

Effect measures

This investigation examines the relationship between sexual activity and the incidence of prostate cancer, using Odds Ratio (OR) as summary measure, with results presented alongside 95% CI. A random effects model was utilized to synthesize findings from multiple studies. All statistical analyses were conducted using Stata software, version 14.2 (Stata Corp, College Station, TX, USA), at a significance level of 0.05. Additionally, RevMan software version 5.4 was employed to create forest plots.

Dose–response meta-analysis

To conduct a dose–response meta-analysis, we categorized the exposure variables (frequency of intercourse, number of female sexual partners, age at first intercourse, and number of ejaculations) into logical subgroups. We then used the lowest exposure level as the reference group and compared the risk of prostate cancer across all other subgroups. This Stata-based dose–response analysis allowed us to quantify how the risk changes with increasing exposure.

Subgroup analysis

A subgroup analysis was conducted based on varying levels of sexual intercourse frequency per week, the number of female sexual partners, and different ages at first sexual intercourse.

Certainty assessment

Following the evidence synthesis, we applied the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach [25, 26] to assess the certainty of evidence for each outcome. This framework evaluates five domains that may decrease certainty (risk of bias, inconsistency, indirectness, imprecision, and publication bias) and three domains that may increase it (large effect, dose–response relationship, and plausible confounding).

The rating system was as follows:

• Very confident: rated up two levels (↑2 levels)

• Confident: rated up one level (↑1 level)

• No concern: no rating change (0 score)

• Some concerns: rated down by 0.5 level (↓0.5 level)

• Serious concerns: rated down by one level (↓1 level)

• Very serious concerns: rated down by two levels (↓2 levels)

Based on these ratings, the GRADE system classifies evidence into four levels of certainty:

• Very low certainty: Evidence is insufficient to draw reliable conclusions

• Low certainty: Evidence is limited, with a high risk of bias or inconsistency.

• Moderate certainty: Evidence is reasonably robust but has some limitations in quality or consistency.

• High certainty: Evidence is strong, with minimal risk of bias or inconsistency.

Results

Description of studies

Initially, a total of 5311 studies were identified, with 4629 derived from electronic database searches and 682 from reference lists of included studies. After removing duplicates and excluding ineligible studies, 29 studies involving 315,193 participants were selected for the meta-analysis (Fig. 1). These studies consisted of 26 case–control studies and three cohort study. The descriptive characteristics of the selected studies are detailed in Table 1. Quality assessments indicated that 17 studies [6, 17, 18, 27–40] were of high quality, while 12 studies [16, 41–51] were classified as low quality.

Fig. 1.

Flow of information through the various phases of the systematic review

Table 1.

Characteristics of included studies, sorted by authors' names

ID	Author	Country	Age (yr)	Study	Exposure	Sample size	NOS score	Quality
1	Ahmadi 2011	Iran	70.69	Case–control	[A]	511	6	Low
2	Angeles-Garay 2019	Mexico	65.85	Case–control	[A], [B], [C]	783	7	High
3	Checkoway 1987	USA	> 50	Case–control	[B]	104	5	Low
4	Dimitropoulou 2008	UK	20–50	Case–control	[A], [B], [C], E	840	7	High
5	Ewings 1996	UK	-	Case–control	[B]	484	6	Low
6	Fernández 2005	Cuba	55–84	Case–control	[A], [B], [C]	527	7	High
7	Fincham 1990	Canada	> 45	Case–control	[C]	1,007	6	Low
8	Giles 2003	Australia	< 70	Case–control	[B], [D]	2,338	7	High
9	Hayes 1992	Netherland	50–79	Case–control	[A]	213	5	Low
10	Hayes 2000	USA	40–79	Case–control	A, B, C	2,296	8	High
11	Hennis 2013	India	67.1	Case–control	[B], [C]	1,904	8	High
12	Honda 1988	USA	< 60	Case–control	[A], [B], [C]	432	7	High
13	Ilic 1996	Central Serbia	71.16	Case–control	[A], C	303	7	High
14	Krain 1974	USA	34–90	Case–control	[A], [B]	442	5	Low
15	La Vecchia 1993	Italy	< 80	Case–control	[B], [C]	956	7	High
16	Leitzmann 2004	USA	40–75	Cohort	D	29,342	6	Low
17	Lozano-Lorca 2023	Spain	40–80	Case–control	D	833	8	High
18	Mandel 1987	USA	< 70	Case–control	A, B, C	490	8	High
19	Nair-Shalliker 2017	Australia	> 18	Case–control	[B], [C], [D]	10,796	8	High
20	Oishi 1990	Japan	50–79	Case–control	[A], [B], [C], E	200	6	Low
21	Papa 2017	Australia	65.05	Case–control	[D]	2141	8	High
22	Perez-Cornago 2017	UK	56.5	Cohort	[B], [C]	219,335	7	High
23	Pourmand 2006	Iran	68.76	Case–control	[A]	205	6	Low
24	Rider 2016	USA	40–75	Cohort	D	31,925	7	High
25	Rosenblatt 2001	USA	40–64	Case–control	A, [B], [C]	1,456	6	Low
26	Rosenblatt 2003	USA	40–64	Case–control	[B], [C]	1,212	6	Low
27	Sarma 2006	USA	40–79	Case–control	[A], [B], [C]	832	8	High
28	Spence 2014	Canada	40–79	Case–control	[B], [C]	3,208	7	High
29	Steele 1971	Canada	50–90	Case–control	[B]	78	5	Low

NO1S The Newcastle–Ottawa Scale

A = number of sexual intercourses per week; B = number of female sexual partners; C = age at first intercourse; D = number of ejaculations; E = number of masturbations

[] = study variable that could be pooled in the dose–response meta-analysis

Synthesis of results

Number of sexual intercourses per week

Figure 2a presents the association between the frequency of sexual intercourse per week and the risk of prostate cancer. The included studies are categorized into five subgroups based on different levels of sexual activity per week. Each study reports an OR with a 95%CI, reflecting whether higher or lower sexual frequency influences prostate cancer risk. The overall pooled effect size is also calculated, alongside heterogeneity statistics to assess the variability between studies.

Fig. 2.

Association between the number of sexual intercourses per week and prostate cancer: a Forest plot; b Linear dose–response curve (Solid line: pooled odds ratio; dashed lines: 95% confidence intervals)

The results indicate that the overall OR across all studies is 1.02 (95% CI: 0.88, 1.18), suggesting a slight increase in prostate cancer risk with higher number of sexual intercourses. However, the confidence interval includes 1, meaning the association is not statistically significant (p = 0.11). Among the subgroups, only the ≥ 1 to < 1 times per week category shows a statistically significant increase in prostate cancer risk (OR = 1.14; 1.04, 1.26, p = 0.007). In contrast, other categories do not show significant associations, with the ≥ 3 to < 4 times per week group even suggesting a possible protective effect (OR = 0.93; 0.65, 1.32), though this result is not statistically meaningful. Heterogeneity analysis reveals moderate overall heterogeneity (I² = 60%), indicating some variability across studies.

Figure 2b illustrates the linear dose–response relationship between the number of sexual intercourses per week and the risk of prostate cancer. The solid line represents the pooled odds ratio, and the dashed lines indicate the 95% confidence intervals. Overall, the pooled estimate showed a slight increase in the odds of prostate cancer with higher intercourse frequency; however, the 95% confidence interval suggests that this linear trend was not statistically significant.

Number of female sexual partners

Figure 3a illustrates the relationship between the number of female sexual partners and the risk of prostate cancer, measured by OR. The analysis is divided into five subgroups, each defined by varying thresholds for the number of female sexual partners, which are compared to the corresponding lower categories.

Fig. 3.

Association between the number of female sexual partners and prostate cancer: a Forest plot; b Linear dose–response curve (Solid line: pooled odds ratio; dashed lines: 95% confidence intervals)

The overall OR across all studies is 1.01 (95% CI: 0.90, 1.12), indicating a slight reduction in prostate cancer risk associated with a higher number of female sexual partners. However, since the confidence interval includes 1, this association is not statistically significant (p = 0.85). Within the individual subgroups, none demonstrate a statistically significant reduction in prostate cancer risk. Conversely, some categories indicate an increased risk, although these results also lack statistical significance. Furthermore, the heterogeneity analysis shows a high level of overall variability among the studies (I² = 84%).

Figure 3b illustrates the linear dose–response relationship between the number of female sexual partners and the risk of prostate cancer. The solid line represents the pooled odds ratio, and the dashed lines indicate the 95% confidence intervals. Overall, the pooled estimate shows a slight increase in the odds of prostate cancer with a higher number of female sexual partners; however, the 95% confidence interval suggests that this linear trend is not statistically significant.

Age at first intercourse

Figure 4a examines the association between age at first sexual intercourse and the risk of prostate cancer. The analysis is structured into six subgroups, each representing a different age threshold for first sexual intercourse, compared to those with first intercourse at a younger age. For each study within these subgroups, the plot presents the OR and the corresponding 95% CI.

Fig. 4.

Association between age at first intercourse and prostate cancer: (a) Forest plot; (b) Linear dose–response curve (Solid line: pooled odds ratio; dashed lines: 95% confidence intervals)

The overall pooled OR is 0.99 (95% CI: 0.90, 1.08), suggesting no significant association between age at first sexual intercourse and prostate cancer risk when considering all age thresholds. However, substantial heterogeneity is observed across studies, as indicated by the high I² value of 78%. The test for subgroup differences shows no significant variation in effect sizes across the different age thresholds.

Among the subgroups, none show a statistically significant association with prostate cancer risk. While some individual studies within subgroups show ORs suggesting an increased or decreased risk, these are not consistent across the subgroups, and the overall subgroup pooled estimates do not reach statistical significance.

Figure 4b illustrates the linear dose–response relationship between age at first intercourse and the risk of prostate cancer. The solid line represents the pooled odds ratio, and the dashed lines indicate the 95% confidence intervals. Overall, the pooled estimate shows almost no change in the odds of prostate cancer with increasing age at first intercourse, and the 95% confidence interval confirms the absence of a significant linear trend.

Number of ejaculations per week

Figure 5a presents the association between the number of ejaculations and the risk of prostate cancer. The analysis is divided into three subgroups, each comparing higher versus lower ejaculation frequency. For each study within these subgroups, the plot displays the OR and corresponding 95% CI. The overall pooled OR is 0.83 (95% CI: 0.77, 0.90), indicating a significant inverse association between ejaculation frequency and prostate cancer risk. Moderate heterogeneity was observed across studies (I² = 50%).

Fig. 5.

Association between number of ejaculations and prostate cancer: a Forest plot; b Linear dose–response curve (Solid line: pooled odds ratio; dashed lines: 95% confidence intervals)

Figure 5b illustrates the linear dose–response relationship between the number of ejaculations and the risk of prostate cancer. The solid line represents the pooled odds ratio, and the dashed lines indicate the 95% confidence intervals. Overall, the pooled estimate shows a decreasing trend in the odds of prostate cancer with increasing number of ejaculations, and the 95% confidence interval suggests that this inverse linear trend is statistically significant.

Number of masturbations per month

Figure 6 displays the association between the number of masturbations per month and the risk of prostate cancer. Based on this limited evidence, the overall pooled OR is 1.12 (95% CI: 0.82, 1.54). This finding indicates no statistically significant association between masturbation frequency and prostate cancer risk. No heterogeneity was observed across the included studies (I² = 0%).

Fig. 6.

Forest plot showing the association between the number of masturbations and prostate cancer

Publication bias

Publication bias was assessed using Begg’s and Egger’s tests (Table 2). Begg’s test showed no significant bias for any variable (all p > 0.05). Egger’s test indicated significant bias only for the number of female sexual partners (p = 0.003); no bias was found for other variables, and data were insufficient for number of masturbations per month.

Table 2.

Results of Begg’s and Egger’s tests (P-values) assessing potential publication bias in studies investigating the association of sexual activity with prostate cancer

Variables	Begg test	Egger test
Number of female sexual partners	0.189	0.003
Number of sexual intercourses per week	0.278	0.861
Age at first sexual intercourse	0.424	0.122
Number of ejaculations per week	0.091	0.199
Number of masturbations per month	0.317	Insufficient data

Certainty of evidence

There was very serious concern regarding the overall risk of bias and inconsistency across studies, except for those examining the number of ejaculations, where the evidence was confident. No concerns were identified regarding indirectness. Imprecision and effect size also raised very serious concerns, except for studies on the number of ejaculations, where the evidence was moderate to very confident. There were some concerns about publication bias, and no significant dose–response gradient was observed.

Overall, we are seriously concerned about the evidence supporting an association between sexual activities (including intercourse frequency, number of female partners, age at first intercourse, and masturbation) and prostate cancer risk. However, we are moderately confident that a higher number of ejaculations may reduce the risk of prostate cancer.

Discussion

The systematic review and meta-analysis included 22 studies involving over 250,000 participants, providing a substantial dataset for evaluating the relationship between sexual activity and the risk of prostate cancer. The results indicated no statistically significant association between prostate cancer and various aspects of sexual behavior, such as frequency of sexual intercourse, number of female sexual partners, and age at first sexual intercourse. The observed heterogeneity in some analyses suggests that individual study characteristics may influence outcomes, highlighting the need for careful interpretation of the results.

The meta-analysis revealed a positive but nonsignificant association between higher frequency of sexual activity per week and prostate cancer compared to less frequent activity. This finding aligns with previous research suggesting that frequent sexual intercourse may be linked to increased prostate cancer risk due to potential inflammation or hormonal changes associated with higher sexual frequency [14, 37]. However, the lack of significant associations for higher frequencies of sexual intercourse indicates that the relationship may not be straightforward and could be influenced by confounding factors such as age, lifestyle, and genetic predispositions [28, 52].

Moreover, while some studies indicate that sexual activity may stimulate the prostate and potentially increase the risk of developing cancer, others argue that frequent ejaculation, regardless of sexual intercourse, could help clear carcinogenic substances from the prostate, thereby reducing risk [6, 53, 54]. For example, Rider et al. reported that men who ejaculated more than 20 times per month had about a 20% lower risk of prostate cancer compared to those with less frequent ejaculation [17]. It is important to note that ejaculation frequency and sexual intercourse frequency represent related but distinct constructs, and the effects of each may vary based on individual health profiles and underlying conditions.

The relationship between the number of female sexual partners and prostate cancer risk remains complex and inconclusive. While our analysis did not find a significant association, other studies have reported mixed findings. Some research suggests that having more female sexual partners may increase prostate cancer risk, potentially due to a higher likelihood of sexually transmitted infections (STIs), which have been linked to an elevated risk of prostate cancer [14]. A meta-analysis by Jian et al. in 2018 reported that an increase of 10 female sexual partners was associated with a 1.10-fold increase in prostate cancer risk [15].

In contrast, some studies suggested that increased female sexual partners may offer some protective effect against prostate cancer, possibly due to enhanced sexual health behaviors [40]. However, it is also hypothesized that individuals with multiple female sexual partners might engage in more health-seeking behaviors, such as regular medical check-ups and screenings, which could lead to earlier detection of prostate cancer. This dual perspective underscores the need for careful consideration of how sexual behaviors are contextualized within broader health patterns.

Our analysis revealed nonsignificant associations between the age at first intercourse and prostate cancer risk across various age thresholds. This suggests that delaying sexual initiation may not significantly impact prostate cancer risk, contrary to hypotheses that earlier sexual activity could increase risk due to prolonged exposure to potential carcinogens [28]. A 2018 meta-analysis found a 4% decrease in prostate cancer risk for every five-year delay in first intercourse [15].

Some studies propose that early sexual initiation may lead to risky behaviors, including higher rates of STIs, which could elevate cancer risk [31, 49]. However, inconsistent findings highlight the need for standardized methodologies to clarify these associations. Cultural and societal factors may also influence how age at first intercourse relates to health outcomes, potentially affecting access to healthcare and health literacy [55–57].

Our results align with studies showing no significant relationship between age at first intercourse and prostate cancer risk [14, 40]. While early sexual initiation might be linked to more female sexual partners and increased STI risk, potentially elevating cancer risk, the evidence remains mixed. Further research is necessary to understand these complex interactions and develop targeted interventions.

Several population-based cohort studies have reported an increased incidence of prostate cancer among men with prior erectile dysfunction. For example, nationwide population-based cohorts from Taiwan showed a modestly higher hazard of subsequent prostate cancer among men with erectile dysfunction compared with matched controls (adjusted HR = 1.27 s) [58]. In another large Taiwanese population-based cohort using two concurrent comparison groups, men aged ≥ 40 years with erectile dysfunction had a significantly higher risk of developing prostate cancer during up to 8 years of follow-up. After adjusting for age, socioeconomic status, and comorbidities, the risk of prostate cancer among men with erectile dysfunction was approximately 2.6 times higher than in the general population (adjusted HR = 2.63) [59].

In a large prospective cohort of 31,925 men from the Health Professionals Follow-up Study, higher ejaculation frequency was associated with a reduced risk of prostate cancer over nearly two decades of follow-up. Men reporting ≥ 21 ejaculations per month had a significantly lower risk of developing prostate cancer compared with those reporting 4–7 per month, with adjusted hazard ratios of 0.81 (95% CI: 0.72, 0.92) at ages 20–29 and 0.78 (95% CI: 0.69, 0.89) at ages 40–49. These results support a potential protective role of more frequent ejaculation throughout adulthood against prostate cancer development [17]. Several biological mechanisms have been proposed to explain why more frequent ejaculation may reduce the risk of prostate cancer. One hypothesis is that regular ejaculation facilitates the clearance of prostatic secretions and potentially carcinogenic substances—such as inflammatory mediators, oxidative compounds, and lipid peroxides—from the prostatic ducts, thereby preventing the accumulation of genotoxic agents that could induce DNA damage and malignant transformation [60]. Another plausible explanation involves hormonal and neurophysiological regulation. Frequent ejaculation may help maintain a favorable androgen–estrogen balance and healthy prostate function. Androgen-dependent pathways, particularly those mediated through the binding of dihydrotestosterone (5α-DHT) to androgen receptors, play a key role in regulating prostate growth and maintaining tissue homeostasis. Regular ejaculation may modulate these hormonal pathways, prevent excessive androgenic stimulation and promote normal cellular turnover. Therefore, frequent ejaculation—when not associated with risky sexual behavior—may contribute to a lower risk of prostate cancer by supporting hormonal equilibrium and reducing prostatic hyperactivity [61].

Despite the robust methodology applied in this meta-analysis, several limitations should be acknowledged. First, the reliance on self-reported data for sexual behaviors introduces potential recall bias, as participants may inaccurately recall their sexual history, thereby affecting the validity of the findings. Second, we did not differentiate between various types of sexual intercourse (e.g., oral, anal, or same-sex intercourse) or whether ejaculation occurred during each act, which may limit the precision of our estimates regarding the association between sexual activity and prostate cancer risk. Third, most of the included studies did not explicitly examine sexual behaviors among the LGBTQI + community. Consequently, our findings primarily reflect heterosexual male populations and may not be generalizable to other sexual orientations. Fourth, as with any meta-analysis of observational studies, residual confounding cannot be entirely ruled out. Factors such as age, lifestyle, comorbidities, hormonal status, and other unmeasured variables may influence both sexual behavior and prostate cancer risk. Although some included studies adjusted for key covariates, variations in adjustment strategies could have affected the pooled estimates. Sixth, this meta-analysis was limited by the inability to perform finer categorical analyses because of inconsistent and broadly defined exposure categories across the included studies. Furthermore, this meta-analysis can only identify associations rather than establish causality; therefore, the results cannot definitively confirm that sexual activity causes prostate cancer. Finally, the heterogeneity in study methodologies across different populations may limit the generalizability of our findings.

Although this study recognizes its inherent complexities and limitations, a rigorous search strategy was employed to examine the association between sexual activity and prostate cancer risk. The largely non-significant findings suggest that most forms of sexual activity may not substantially influence prostate cancer risk, except for ejaculation frequency, which appears to have a significant protective effect. Nevertheless, this meta-analysis highlights the importance of further research on lifestyle and other modifiable risk factors to inform public health strategies and preventive interventions. It is also important to acknowledge that chronic diseases, including prostate cancer, do not result from a single risk or protective factor but rather from a complex interplay of multiple determinants [62]. Therefore, these factors should be considered collectively rather than in isolation when evaluating their overall impact on prostate cancer development.

Conclusion

The findings of this meta-analysis highlight the complexity of the relationship between sexual activity and prostate cancer risk. Although some associations were observed, the overall lack of significant relationships—except for ejaculation frequency, which demonstrated a protective effect—combined with the presence of substantial heterogeneity, indicates the need for further research to better elucidate these associations. Understanding the multifactorial nature of prostate cancer etiology is essential for developing effective prevention strategies. Future large-scale cohort studies should aim to clarify the role of sexual behaviors within the broader context of men’s health, taking into account biological, psychological, and social factors that may influence both sexual activity and cancer risk.

Authors’ contributions

Conceptualization: Jalal Poorolajal, Abouzar Raeisvandi, Sanaz Omidi, Massoumeh Javaheri. Data curation: Jalal Poorolajal, Abouzar Raeisvandi, Sanaz Omidi, Massoumeh Javaheri. Formal analysis: Jalal Poorolajal, Abouzar Raeisvandi, Hanieh Moradi. Investigation: Jalal Poorolajal, Abouzar Raeisvandi, Sanaz Omidi. Methodology: Jalal Poorolajal, Abouzar Raeisvandi, Hanieh Moradi. Project administration: Jalal Poorolajal. Software: Jalal Poorolajal, Abouzar Raeisvandi, Sanaz Omidi. Supervision: Jalal Poorolajal. Validation: Jalal Poorolajal, Abouzar Raeisvandi, Massoumeh Javaheri. Visualization: Jalal Poorolajal, Abouzar Raeisvandi. Writing – original draft: Jalal Poorolajal, Abouzar Raeisvandi, Hanieh Moradi. Writing – review & editing: Jalal Poorolajal, Abouzar Raeisvandi, Sanaz Omidi, Massoumeh Javaheri, Hanieh Moradi.

Funding

No source of support was provided for this research.

Data availability

All data utilized in this meta-analysis are publicly accessible and can be retrieved from their respective sources. The search strategy employed to identify pertinent studies is detailed in the manuscript.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.