Study population
From July 1st 1990 to October 15th 1992, all women referred to the HPV outpatient clinic at UNN Tromsø, having an HPV test at baseline, were included in a prospective study. They were referred from their primary care physician, mainly within Troms and Finnmark counties. The reason for referral was symptoms associated with genital condylomas, cervical- and vulvar dysplasia, resistant vaginal discharge, vulvodynia, vulvar lichen planus or pruritus. Six hundred forty-two women were eligible for study participation. Age at inclusion was between 16 and 70 years. A clinical and colposcopic examination was performed in all patients. Paired samples for HPV testing were taken from cervix and vulva (cytology and biopsy) and visible lesions. The cervical scrapings and biopsy for HPV testing were taken from the endocervical canal and the transformation zone on the ectocervix. Patients with a histological confirmed CIN2+ were offered treatment. HPV results were not used to direct patient management.
HPV DNA testing
HPV-testing was performed by a two-step nonradioactive Southern blot DNA hybridization method using biotinylated HPV-specific probes (ONCOR, Medscan AB PO Box 20,047 S-100 74 Malmø, Sweden). HPV-types identified were 6, 11, 16, 18, 31 and 33 causing 78.3% of all cervical cancer [3]. In addition, a polymerase chain reaction (PCR) method using degenerate L1 consensus primers (MY09/MY11) was performed on samples shown to be adequate for PCR amplification by separate β-globulin PCR with PCO4 and GH20 primers as described [20].
PCR amplifications were done in a GeneAmp PCR System 9600 (Perkin Elmer Cetus, Emeryville, CA), with the following modifications. PCR reaction condition with each primer at 0.25 μM of each primer, 100 μM of each deoxynucleotide and 1 U Taq polymerase in a final reaction volume of 50 μl were used. To obtain reproducible HPV PCR-results it was necessary to use hot-start PCR to avoid primer-dimer artefacts, adding Taq-polymerase after heating the samples at 80 °C for 2 min. The cycling conditions were 1 min at 94 °C, 30 s at 55 °C, 30 s at 72 °C; and a final three-minute extension interval at 72 °C. An aliquot of 15 μl of the PCR-products was analysed on ethidium bromide-stained 2% agarose gels. All experiments were done in parallel with positive (diluted cloned HPV-DNAs) and negative controls (ddH2O). PCR-products were further identified with separate HPV 6, 11, 16, 18, 31 and 33 dot blot hybridizations using specific alkaline phosphatase-labelled oligonucleotides as earlier described [20]. HPV amplicons not identified by any of these oligoprobes were not further examined. The test result was defined as positive when HPV was identified in one or both methods. HR-HPV genotypes (16, 18, 31 and 33) were classified as HPV positive, whereas LR genotypes (6 and 11) in addition to absent HPV were classified as HPV negative. Results were then categorized hierarchically according to a priori ordering of established cancer risk [3]: HPV 16 positive, else HPV 16 negative and HPV 18 positive, else HPV 16 and 18 negative and HPV 33 positive, else HPV 16 and 18 and 33 negative and HPV 31 positive, else HR-HPV negative including HPV 6 and 11. Based on this prior knowledge, we further merged HPV 16 with 18 and HPV 33 with 31, respectively.
Data were saved in a file with the specimen as a unit of registration. Within this file, each woman was given a pseudonymous number. SymPathy is the laboratory information system used in Norway for clinical pathology and cytology. Within this system, the HPV data were retrospectively linked to the correct cytology results.
Follow-up
Until December 31th 2020 a single cohort, in which women were classified for their HPV status at baseline underwent passive follow-up through the NCCSP. Two hundred twenty-three women were HPV positive and 419 were HPV negative. Women with abnormal cytological findings had a cervical biopsy performed according to national guidelines. The Department of Clinical Pathology at UNN Tromsø receives and analyses all cervical cytology and cervical histology specimens from Tromsø and Finnmark counties. The existence of the National Cancer Register makes it possible to conduct studies with virtually no loss to follow-up. No national guidelines for HPV testing existed before 2005. The histopathological endpoint was defined as CIN3+ (CIN3, ACIS - adenocarcinoma in situ, or cancer). During the follow-up period, all incidents of CIN3+ were detected and compared with the HPV status at baseline.
Statistical analyses
Statistical package SPSS, version 22.0, was used for the statistical analyses. The Chi-square test was used to determine whether there was a statistically significant difference within our cohort. Univariate analyses were done using the Kaplan-Meier method, and statistical significance between survival curves was assessed by the log rank test. To measure the uncertainty associated with the results we calculated the 95% confidence interval (CI). The Kaplan-Meier survival curves with 95% CI in the supplemental are computed using the survfit-function in R (version 4.0.3).