Risk factors for cervical neoplasia in Denmark.

With the overall goal of elucidating the risk factor pattern for cervical neoplasia, two case-control studies and a prospective cohort study were conducted. The first case-control study focused on female lifestyle risk factors. It was designed to include all women (aged 20-49 years) in Greater Copenhagen, diagnosed with invasive cervical cancer or carcinoma in situ (CIS) from January 1985 to December 1986. They were identified from the Danish Cancer Registry. An age-stratified control group was randomly selected from the study area by means of The Danish Central Population Register. Information on risk factors was collected using a self-administered questionnaire. The study, which included 586 women with CIS, 59 women with cervical cancer, and 614 control subjects, confirmed that CIS and invasive cervical cancer share similar risk factors. Both disease entities were strongly associated with sexual and venereal factors. This applied especially to lifetime number of sexual partners and age at first episode with genital warts (proxy measure for human papillomavirus (HPV)), supporting that HPV infection in the adolescent cervix is associated with a higher risk of cervical neoplasia compared with such an infection later in life. Our results also suggested that parity, oral contraceptive use, and smoking may be important risk factors. In the second case-control study, we identified all women with one lifetime sexual partner based on the questionnaire information obtained in the first case-control study. To investigate the role of the "male factor", the women were invited to participate in the study together with their husband. In all, 41 case couples and 90 control couples were enrolled. Data collection included a personal interview, blood samples, and penile swabs from the males. The most significant risk determinants of cervical neoplasia were a history of genital warts in the male and non-use of condoms, emphasizing the venereal nature and pointing to HPV as an important agent. Genital warts are usually associated with the low-risk HPVs (types 6 and 11) rather than with the high-risk HPV types. However, an explanation for the observed relationship between risk of cervical neoplasia and genital warts in the woman herself and in her male partner could be, that they are more likely also to harbour the high-risk HPV types. Only 2 case husbands and no control husbands had HPV DNA detected in the penile swabs (ViraPapR, ViraTypeTM). As the number of cells in the swab always exceeded 3 x 10(4), the result may reflect shortcomings in the test kit used. From our population-based prospective cohort study of 11,088 women, we selected the prevalent cases (199 women with LSIL/HSIL(low-grade/high-grade squamous intraepithelial neoplasia), 131 women with ASCUS (atypical squamous cells of undetermined significance)), and 1000 random controls (women and normal cervical cytology). At enrollment, the women were personally interviewed and had a gynecological examination including cervical swabs for HPV testing and a Pap smear. HPV DNA detection was done using polymerase-chain-reaction methods. Cervical HPV infection (especially with the high-risk types) was the out-standing risk factor for all grades of neoplasia, the association being strongest for HSIL. Women with high-risk HPV infection had a nearly 33-fold increased risk of HSIL compared to HPV-negative women. Possible risk factors for cervical neoplasia in HPV-positive women included smoking, non-use of barrier contraceptives and parity. If analysis was restricted to histologically confirmed high-grade lesions, the proportion of cases that could be attributed to HPV infections was 80%. The importance and urgent need for studies which include HPV as an adjunct to cervical cytology is emphasized. Greater effort should be made to determine the usefulness of this modality (HPV diagnostics) in cervical cancer screening or in the management of cervical neoplasia, especially ASCUS and LSIL.