Background: Oral contraception (OC) has been proclaimed by the IARC as a risk factor of cervical cancer (CC), on prolonged use by high-risk human papillomavirus (HPV) positive women. However, the available data are far from complete, and more evidence is necessary on the potential confounding effects of sexual behavior and HPV infection. The aim of the present was study to analyse the risk estimates for OC users in order to develop several intermediate end-point markers in cervical carcinogenesis.
Patients and methods: A cohort of 3,187 women, enrolled in a multi-center screening trial in three New Independent States (NIS) of the former Soviet Union (the NIS Cohort Study), was stratified into three groups according to their contraception modes: i) non-users of contraception, ii) non-OC users and iii) OC users. These groups were analysed forpredictors of three outcome measures: a) exposure to HR-HPV; b) progression to high-grade cervical intraepithelial neoplasia (CIN2/3 and HSIL); and c) persistence/clearance of HR-HPV and cytological abnormalities during a prospective follow-up.
Results: All three groups had an identical prevalence of HR-HPV (HCII and PCR), Pap smear abnormalities and CIN histology, but differed significantly (p=0.0001) with regard to all key variables of sexual behaviour, known as risk factors for CC. Predictors of HR-HPV, CIN2/3 and HSIL were different in the three groups, reflecting these different sexual preferences. Use of OC was not a significant predictor of CIN2/3 or HSIL in HPV-positive or HPV-negative women. Outcomes of cervical disease and HR-HPV infection were unrelated to contraception. In a multivariate regression model mode of contraception was of no predictive value for either HR-HPV or high-grade CIN.
Conclusion: Sexual behaviour is different among OC users, non-OC users and in nonusers of contraception; these risk factors predispose women to HR-HPV, high-grade CIN, and determine the outcome of their cervical disease/HR-HPV infection. The use of OC is not an independent risk factor for any of these intermediate end-point markers of cervical carcinogenesis. Failure to record these epidemiological data inevitably leads to erroneous conclusions about the role of OC as an independent risk factor of cervical cancer.