The association between certain human papillomaviruses (HPV) and cervical intraepithelial neoplasia (CIN) is well documented, but there is uncertainty about the strength of association and the role of co-factors is unclear. This population-based case-control study in Norwegian women 20-44 years of age included 103 cases with histologically confirmed CIN II-III and 234 age-matched and randomly selected controls. Cytological specimens from the cervix were analyzed using the polymerase chain reaction (PCR). In all, 91% of the cases and 15% of the controls were HPV DNA positive, giving a crude odds ratio (OR) of 67.2 (95% confidence interval: 28.6-157.5). The association between HPV 16 and CIN II-III was even stronger (crude OR = 123.9; 46.7 - 328.5). In logistic regression analysis, additional to HPV, only a high number of sexual partners and a low educational level contributed independently to the risk. The adjusted OR for the association between HPV and CIN II-III was 72.8 (95% CI: 27.6-191.9). The association between HPV and CIN remains very strong even after adjustment for proposed confounding factors. The results therefore support the role of HPV as a causative agent in the development of CIN.

Several recent studies have reported different associations between HLA specificities and human papillomavirus (HPV)associated disease of the cervix. We report the distribution of DQA1 and DQB1 genes and HPV infection in a populationbased case-control study including 92 patients with histologically verified cervical intraepithelial neoplasia grade II-III (CIN II-III) (thus including moderate and severe dysplasia and carcinoma in situ) and 225 control subjects. We found an overrepresentation of the DQA1*0102-DQB1*0602 haplotype among HPV-positive cases compared with controls. The association was even stronger when comparing HPV-16positive cases with HPV-16-positive controls. In addition, among HPV-16-positive individuals, we observed a decreased frequency of DQA1*0102-DQB1*0604 in cases compared with controls. We were not able to detect any association between CIN II-III and DQB1*03. Compared with previous findings in cervical cancer, our data indicate that carrying the DQA1*0102-DQB1*0602 haplotype gives an increased risk of developing CIN when infected with HPV-16, without influencing progression to cancer.

(Genitourin Med 1997;73:131-135)

arstavik, I. Correlation between polymerase chain reaction and cervical cytology for detection of human papilloinavirus infection in women with and without dysplasia APMIS 105: 71-75. 1997.The aim of this study was to compare the ability of two methods, the polymerase chain reaction (PCR) and cervical cytology. to detect HPV infection. The study population included 222 randomly selected women without dysplasia (controls) and 91 women with histologically confirmed dysplasia (CIN 11-111) (cases). In women without dysplasia, 8.6% had cytological signs of HPV infection, whereas 15.3'%1 were HPV DNA positive by PCR. In women with dysplasia, 72.5%) had cytological signs of HPV infection, whereas 90.1'X were HPV PCR positive. The statistical agreement between the two diagnostic methods was low (controls: ~= 0 . 7 6 , cases: K = -0.03). In total, PCR failed to detect 17 of 85 women with cytological signs of HPV infection, whereas cervical cytology failed to detect 48 of I16 HPV PCR-positive women. In women with dysplasia, but not in women without dysplasia, the oncogenic HPV types were associated with cytological signs of HPV infection.

HPV is suspected of being a major cause of cancer of the uterine cervix. To understand the risk of disease in the general population of women, it is important to estimate the prevalence of HPV infection in a random population‐based sample of women without disease. In this study, a total of 231 randomly selected women without dysplasia (controls) were examined, and compared with 103 women with histologically confirmed CIN II–III (patients). The prevalence of HPV DNA in cervical scrapes was determined by general nested PCR, which was expected to detect any relevant HPV type commonly found in cervical samples. The nested positive samples were typed with type‐specific PCR. In the general nested PCR, 15% of the controls were positive, compared to 91% of the patients. In the population‐based sample, 2.2% had HPV types 6 and 11, and 10% had types 16, 18, 31, and 33. In both groups, HPV DNA was observed less frequently in women above than below the age of 30. The results are among the few population‐based figures on the prevalence of HPV in women, and provide a baseline for understanding the risk of developing cancer of the uterine cervix, and determining the proportion of women to be included in intervention studies.