IJE Advance Access originally published online on September 16, 2004
International Journal of Epidemiology 2005 34(1):193-197; doi:10.1093/ije/dyh332
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
IJE vol.34 no.1 © International Epidemiological Association 2004; all rights reserved.
Sexual behaviour, history of sexually transmitted diseases, and the risk of prostate cancer: a case–control study in Cuba
1 Co-ordinator group of the project in Cuba, National Institute of Oncology, Havana, Cuba
2 International Agency on Research on Cancer (IARC), Lyon, France
3 Catalan Institute of Oncology, Barcelona, Spain
* Correspondence: Instituto Nacional de Oncología y Radiobiología, Vicedirección de Investigaciones, Registro Nacional de Cáncer, 29 y F, Vedado, CP 10 400, Ciudad de La Habana, Cuba. E-mail: yaima{at}infomed.sld.cu
 |
Abstract |
|---|
 Top Abstract Patients and methodsResultsDiscussionReferences |
|---|
Background The relationship between the risk of prostate cancer and sexual activity and history of sexually transmitted diseases was investigated in a case–control study conducted in Cuba aimed at assessing the effect of lifestyle and environmental factors, as well as hormonal and genetic factors, on the occurrence of this disease.
Methods During the period 1998–2000, all men up to 84 yr old with newly diagnosed, cytologically and/or histologically confirmed prostatic cancer who were resident in Havana City were identified in nine major hospitals in the area. Controls were resident in the same city, frequency-matched by age (±5 years) and hospital. The study included 273 cases and 254 controls. Information was obtained through an interview.
Results The risk of prostate cancer was increased among men with a history of venereal disease (odds ratio = 1.7, 95% CI = 1.1–2.5). A higher frequency of cases reported having had sex with prostitutes, although the estimate of relative risk did not reach statistical significance. Similarly, a nonsignificant positive association was found with the number of female sexual partners. A significant increased risk was observed in subjects who reported having sexual intercourse more than 7 times per week compared with those who reported a weekly frequency of 3 times or fewer (odds ratio = 2.1, 95% CI = 1.2–3.7). Moreover, a significant trend was demonstrated.
Conclusions The study supports the hypothesis that an infectious factor related to sexual behaviour could be involved in the occurrence of prostate cancer. A role for hormonal factors related to sexual activity cannot be ruled out.
Keywords Case–control studies, prostate cancer, sexual behaviour, sexually transmitted diseases
Accepted 9 August 2004
Cancer of the prostate is the second most frequently diagnosed nonskin cancer in men1 in Cuba (15% of all new cancer cases) and the second most common underlying cause of cancer death, accounting for 
19.4% of all cancer mortality.2 The annual incidence of prostate cancer in Cuba (age-adjusted by world standard population) is 31.3 per 100 000 men, within the range of overall rates from Central America (26.9 per 100 000) and the Caribbean countries (38.6 per 100 000).3 It is substantially higher than the very low rates in Eastern Asia (3.4 per 100 000) and less-developed countries (7.7 per 100 000) but far from the high incidence in more-developed countries (46.6 per 100 000) or Northern America (102 per 100 000).3 The occurrence of prostate cancer in Cuba increased from 20.7 per 100 000 in 1977 to 28.6 per 100 000 in 1999. This increase seems not to have been influenced by diagnostic practices, since screening or case-finding using prostate-specific antigen is not done on the island.
Despite many aetiological studies having been carried out on prostate cancer, no modifiable risk factors have been clearly established. All that is known with certainty is that the incidence increases with age, varies by geographic area, by race or ethnicity, and is higher among men whose father or brother had the disease.4 Several epidemiological studies have suggested that factors related to sexual behaviour and sexually transmitted diseases may be associated with prostate cancer.4 A recent meta-analysis5 has reported slight increases of risk associated with the number of sexual partners and history of sexually transmitted infection. Although the mechanisms are unclear, the search for potential infectious agents involved in prostate carcinogenesis is an important focus of research; identification of an infectious cause would be a major advance in the efforts to prevent prostate cancer.
In this article we report results regarding sexual activity and history of sexually transmitted diseases from a case–control study carried out in Cuba, a country where a significant proportion of the population is of African origin. The aim of the study was to assess the effect of lifestyle and environmental factors, as well as hormonal and genetic factors, on the occurrence of prostate cancer in this population.
|
Patients and methods |
|---|
|
TopAbstractPatients and methodsResultsDiscussionReferences |
|---|
During the period 1998–2000, all men up to 84 years old with newly diagnosed, cytologically and/or histologically confirmed prostatic cancer who were resident in Havana City were identified in nine major hospitals in the area. The city notifies
22% of all new prostate cancer cases in Cuba, and the participating hospitals cover 90% of new prostate cancer cases resident in Havana City according to the National Cancer Registry. In all cases the tumour was graded and staged according to the Gleason and Union Internationale Contre le Cancer (UICC) classifications, respectively.6,7 Controls were selected from patients admitted to the participating hospitals who were resident in the same city and who had never been diagnosed with benign prostatic hyperplasia or cancer. The eligible controls were identified through the list of admissions from all departments and underwent clinical examination to rule out the presence of a palpable tumour in the prostate. The control series was designed to be of equal size to the case series, frequency-matched by age (±5 years) and hospital. Both cases and controls suffering from dementia, Parkinson's disease, or any other condition that did not allow the interview were excluded. Cases and controls were interviewed at the hospital by trained interviewers. The interviews lasted 40 min on average and were based on a structured questionnaire. The data obtained included information about demographic characteristics, marital status, smoking habits, occupational history, physical activity, personal and family medical history, skin colour, first-degree family with prostate cancer, sexual behaviour, and dietary habits. Additionally, height and weight were measured and 20 ml of venous blood were collected from all subjects. In terms of their history of sexually transmitted diseases, subjects were asked whether they had ever been diagnosed with venereal disease, their age at the onset and whether they had received medical treatment. Questions related to sexual behaviour included the age at which they first had intercourse, their weekly frequency of sexual intercourse during adult life, their lifetime number of female sexual partners, and whether they had ever visited prostitutes.
Ethical approval for the study was obtained and an informed consent was signed by each subject who agreed to participate in the study.
Unconditional logistic regression models were used to compute odds ratios (ORs) and to estimate 95% confidence intervals (95% CIs).8 Tests for linear trends were carried out using ordinal or categorical variables as continuous in the logistic regression. Unless otherwise specified, all logistic models included as covariates age group, hospital, educational level, skin colour, and having a relative with prostate cancer.
|
Results |
|---|
|
TopAbstractPatients and methodsResultsDiscussionReferences |
|---|
Of 306 patients with histologically confirmed prostate cancer, 20 could not be interviewed due to their bad health status, 5 had previously been treated, 4 did not return to the hospital after diagnosis, 3 refused to participate, and 1 patient died before we were able to contact him. As a result, 273 cases (89.2% participation) were included in the study. According to the tumour extension, 168 cases (61.5%) presented with local disease, 19 (7%) had regional disease, and 39 (14.3%) had disseminate disease; information on extension was missing for 47 cases (17.2%). From 345 eligible controls, 254 were finally included (73.6% participation). The main reasons for exclusion were the bad health status of 51 subjects (14.8%), refusal to participate by 20 (5.8%), and other reasons, including death, for the remaining 20 (5.8%). Most controls were hospitalized for general examination, hypertension, heart failure, respiratory disorder, cerebrovascular disease, or bone pains.
Table 1 shows certain demographic characteristics of cases and controls. There were no differences by age, marital status, and educational level. No statistically significant association was observed between skin colour and prostate cancer risk, although the proportion of subjects with black skin colour was slightly higher for cases than for controls; the OR (95% CI) for black as compared with white skin colour was 1.32 (0.83–2.08). Similarly, a higher proportion of cases than controls had a family history of prostate cancer, but the difference was not statistically significant: OR (95% CI) = 1.47 (0.72–2.98).
|
The risk of prostate cancer (Table 2) increased among men with a history of venereal disease (OR = 1.7, 95% CI = 1.1–2.5), mainly when it was diagnosed before 30 years of age. Almost all subjects with a history of venereal disease (86 out of 87 cases and 53 out of 54 controls) reported that they had received medical treatment. More than 67% of cases and 59% of controls reported having had sex with prostitutes, producing an increased risk of prostate cancer, although the estimate did not reach statistical significance. Cases tended to report a higher number of female sexual partners, but this was not a risk factor for prostate cancer. About 46% of cases and 36% of controls reported having had sexual intercourse more than 7 times per week during adult life, with a significant increased risk as compared with those who reported a weekly frequency of 3 times or fewer (OR = 2.1, 95% CI = 1.2–3.7); furthermore, this showed a significant trend. When the frequency of sexual intercourse was introduced into a model together with history of venereal disease and of visiting prostitutes, its point estimate and trend remained unchanged (results not shown), suggesting an independent effect.
|
|
Discussion |
|---|
|
TopAbstractPatients and methodsResultsDiscussionReferences |
|---|
We observed a positive association of risk of prostate cancer with a history of prior venereal disease, mainly at young ages, and an increase of risk among men who had sex with prostitutes. Although our study does not involve any specific sexually transmitted disease, our results support the hypothesis that a sexually transmitted agent may be related to the occurrence of prostate cancer. Our results suggest as well a potential effect on prostate cancer associated with high sexual activity during adult life as measured by the frequency of sexual intercourse.
A series of early epidemiological observations suggested that ethnic and geographic differences found in prostate cancer rates might be due in part to variations in male sexual behaviour or sexual activity.9 A review10 stated that there was reasonably consistent evidence suggesting an increased risk of prostate cancer associated with a high level of sexual activity and/or a history of sexually transmitted disease. A more recent review4 summarized the main conclusions of studies issued during the 1990s. Most studies supported a positive association with a self-reported history of sexually transmitted diseases, mainly gonorrhoea and syphilis, although the associations were seldom statistically significant. Age at the time of first intercourse was often associated with an increased risk of prostate cancer, whereas lifetime number of sexual partners was not found to be associated in most studies. The review noted, however, that many studies were small and had methodological limitations. Contrary to these conclusions, a large population-based case–control study11 reported a significant trend of increasing risk with increasing number of female sexual partners; there was also a higher risk for subjects with younger age at first intercourse, but this association became weaker after controlling for the lifetime number of sexual partners. Finally, in a recent meta-analysis5 there was a modest but significant linear increase in the estimated risk of prostate cancer by number of sexual partners, similar in population- and hospital-based studies. There was a lack of association with age at first intercourse, age at first marriage, or multiple marriages. From studies that reported age-specific frequencies of sexual intercourse it was estimated that there was a significant increase in risk with increasing frequency of sexual activity. In terms of sexually transmitted diseases, there were significant positive associations for any sexually transmitted disease, the point estimates being higher for population-based than hospital-based studies. This association was further supported by an increased risk among men who visited prostitutes.
Some sexually transmitted diseases may have been asymptomatic and therefore underreported. Gonorrhoea and syphilis may be sentinels for a sexually transmitted infectious agent that contributes to the aetiology of prostate cancer; in most studies, however, the exposure was based on a self-reported history of infection by the subject. In terms of more direct evidence of an infectious agent, a positive association has been reported of prostate cancer with long-term persistence of antibodies to syphilis in serum,12 whereas no associations have been found with serological response for herpes simplex, cytomegalovirus, and Epstein–Barr virus.4 Given its association with sexual practices as well as its role in anogenital infection and neoplasia,13 a potential aetiological agent to be considered is the human papillomavirus (HPV), mainly the oncogenic types HPV-16 and HPV-18. The first attempts aimed to detect HPV DNA in tumour tissue compared with prostate tissue from men with benign conditions and gave very inconsistent results.4 Two recent large population-based studies14,15 have reported no association between serological evidence of HPV-16 or HPV-18 infections and the development of prostate cancer. In the study of Rosenblatt et al.14 the association was absent after taking into account the extension of the disease at diagnosis or measures of tumour aggressiveness. Furthermore, they did not find any association of HPV-16 or HPV-18 prevalence with number of sexual partners or history of gonorrhoea or syphilis among controls. In the study by Adomi et al.15 there was a significant risk associated with a high level of antibodies against a less common oncogenic type, HPV-33, although the authors could not rule out that it was produced by chance; the result therefore needs to be replicated. These findings are consistent with several other case–control studies, but they differ from those of previous nested case–control studies.16,17 However, these studies collected sera more than 20 years before the diagnosis; although HPV antibodies may persist for decades, it seems that the level of antibodies decreases in individuals who do not have recurrent disease. Thus, it has been postulated that seropositivity for HPV might be a surrogate for sexual activity.
Several studies have reported an association of prostate cancer with frequency of sexual activity,5 suggesting a possible link with sexual hormone level. Nevertheless, mechanisms to explain this relationship are unclear. Androgens are required for the growth and functional activity of the prostate; on the other hand, in some men the frequency of sexual activity seems to be related to hormonal levels. However, although hormonal factors may contribute to the potential relationship between prostate cancer and sexual activity, sexual practices are complex and may not be directly correlated with hormonal levels.
There are several limitations of our study to be considered. First, almost 90% of potential cases were included, whereas only about three-quarters of eligible controls could be interviewed. However, only 5% of them actively refused to participate. Although the possibility of differential characteristics among respondents and nonrespondents can never be discounted, it is unlikely that this fact could have seriously biased our results. Potential for selection bias and differential recall bias must always be kept in mind when controls are selected among hospital patients. In a previous meta-analysis5 there were no differences in many associations according to the origin of the controls; in terms of association with sexually transmitted diseases, the estimates from population-based studies were actually higher than those from hospital-based studies. As in many previous case–control studies our assessment of exposure to a sexually transmitted agent relied on a self-reported history of venereal disease. When subjects were asked whether they were treated because of the disease all but two provided an affirmative answer. Apart from age, ethnicity and familial antecedents of prostate cancer are established risk factors. In our study all these factors were controlled for in the analysis.
In conclusion our study supports the hypothesis that an infectious factor related to sexual behaviour could be involved in the occurrence of prostate cancer. We found a consistent association with a previous history of venereal diseases, but the search for potential agents remains a focus of interest. The role of hormonal factors related to sexual activity, on their own or interacting with sexually transmitted diseases, cannot be ruled out.
KEY MESSAGES
|
|
Acknowledgments |
|---|
This study has been supported by the Catalan Institute of Oncology, Barcelona, Spain and the International Agency for Research on Cancer (IARC), Lyon, France (Collaborative Research Agreement NTR/98/05). The preparation of this manuscript was also supported by a Fellowship (ICRETT UICC Fellowship No. 618) from the International Union against Cancer, awarded to Yaima Galán.
|
Notes |
|---|
Collaborative group: Ivette Portilla, Julia Cruz, Santiago Quintero, Vicente Osorio, Maria Victoria López, Raúl Espinosa, Rosa Jorge, Olga Piera, Wilfredo Domínguez, Mario Coll Ruíz, Nora Cañizares, Sergio Ferrán, Lic. Alexis Janero, Teresa Martí, Celestino Labori, Rosaura Rego, Walfrido Beiries, Blanca Blanco, Francisco Alonso, and Enrique Abraham.
Participating institutions: Instituto Nacional de Oncología y Radiobiología, Hospital Clínico Quirúrgico Hnos Ameijeiras, Hospital Clínico Quirúrgico Enrique Cabrera, General Hospital Calixto García, Hospital Clínico Quirúrgico Freyre Andrade, General Hospital Joaquín Albarrán, Hospital Clínico Quirúrgico Salvador Allende, Hospital Clínico Quirúrgico de 10 de Octubre, and Hospital Clínico Quirúrgico Manuel Fajardo.
|
References |
|---|
|
TopAbstractPatients and methodsResultsDiscussionReferences |
|---|
1 Registro Nacional de Cáncer. Incidencia de Cáncer en Cuba, 1999. La Habana, Cuba: Instituto Nacional de Oncología y Radiobiología, 2003.
2 Dirección Nacional de Estadísticas. Anuario Estadístico, 1999. La Habana, Cuba: Ministerio de Salud Pública, 2003.
3 Ferlay J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2000: cancer incidence, mortality and prevalence worldwide, version 1.0. Lyon: International Agency for Research on Cancer, IARC CancerBase No. 5; IARC Press, 2001.
4 Strickler HD, Goedert J. Sexual behaviour and evidence for an infectious cause of prostate cancer. Epidemiol Rev 2001; 23:144–51.
5 Dennis LK, Dawson DV. Meta-analysis of measures of sexual activity and prostate cancer. Epidemiology 2002; 13:72–79.[CrossRef][Web of Science][Medline]
6 Gleason DF. Histologic grading of prostate cancer: a perspective. Hum Pathol 1992; 23:273–79.[CrossRef][Web of Science][Medline]
7 International Union against Cancer. Prostate (ICD-O C61). In: Sobin LH, Wittekind CH (eds). TNM Classification of Malignant Tumours. 5th Edn. New York: Wiley-Liss, 1997, pp. 170–73.
8 Breslow EN, Day N. Statistical Methods in Cancer Research. Vol 1. The Analysis of Case-control Studies. Lyon: International Agency for Research on Cancer, IARC Scientific Publications No. 32, 1980.
9 Ross RK, Schottenfeld D. Prostate cancer. In: Schottenfeld D, Fraumeni JF, Jr (eds). Cancer Epidemiology and Prevention. New York: Oxford University Press, 1996, pp. 1180–206.
10 Key T. Risk factors for prostate cancer. Cancer Surv 1995; 23:63–77.[Web of Science][Medline]
11 Rossenblatt KA, Wicklund KG, Stanford JL. Sexual factors and the risk of prostate cancer. Am J Epidemiol 2000; 152:1152–58.
12 Hayes, RB, Pottern LM, Strickler H et al. Sexual behaviour, STDs and risks for prostate cancer. Br J Cancer 2000; 82:718–25.[CrossRef][Web of Science][Medline]
13 Schiffman M, Kjaer SK. Natural history of anogenital human papillomavirus infection and neoplasia. J Natl Cancer Inst Monogr 2003; 31:14–19.[Medline]
14 Rosenblatt KA, Carter JJ, Iwasaki LM, Galloway DA, Stanford JL. Serologic evidence of human papillomavirus 16 and 18 infections and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev 2003; 12:763–68.
15 Adami HO, Kuper H, Anderson SO, Bergström R, Dillner J. Prostate cancer risk and serologic evidence of human papillomavirus infection: a population-based case-control study. Cancer Epidemiol Biomarkers Prev 2003; 12:872–75.
16 Dillner J, Knekt P, Boman J et al. Sero-epidemiological association between human-papillomavirus infection and risk of prostate cancer. Int J Cancer 1998; 75:564–67.[CrossRef][Web of Science][Medline]
17 Hisada M, Rabkin CS, Strickler HD, Wright WE, Christianson RE, van den Berg BJ. Human papillomavirus antibody and risk of prostate cancer. JAMA 2000; 283:340–41.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Nonn, D. Duong, and D. M. Peehl Chemopreventive anti-inflammatory activities of curcumin and other phytochemicals mediated by MAP kinase phosphatase-5 in prostate cells Carcinogenesis, June 1, 2007; 28(6): 1188 - 1196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Nonn, L. Peng, D. Feldman, and D. M. Peehl Inhibition of p38 by Vitamin D Reduces Interleukin-6 Production in Normal Prostate Cells via Mitogen-Activated Protein Kinase Phosphatase 5: Implications for Prostate Cancer Prevention by Vitamin D. Cancer Res., April 15, 2006; 66(8): 4516 - 4524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Minerva BMJ, March 5, 2005; 330(7490): 548 - 548. [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Smith Epidemiological Freudianism Int. J. Epidemiol., February 1, 2005; 34(1): 1 - 2. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||