Cervical cancer prevention Studies on possible improvements

Sammanfattning: Aims: The aim of this study is to target and assess possible improvements for women attending cervical cancer screening programs. Methods: In a randomized study the use of ThinPrep liquid based cytology (LBC) was tested against conventional cytology. 13 484 samples, taken within the screening program at five screening units in the G?teborg area, were evaluated. Main outcome was cervical intraepithelial lesion grade two or more (CIN2+) in histopathology at follow up. A scoring system for colposcopy was constructed containing five parameters and 3 ordered values for each. It was tested in 297 women with abnormal cytology. The relationsship between the scoring units and CIN was modelled with logistic regression. To assess the risk after treatment of CIN3 a cohort of 132 493 women with this diagnose was followed up in the Swedish cancer registry for development of vaginal or cervical cancer. Standard incidence ratios (SIR) and absolute risk difference were calculated with the entire female population as reference and age, time-period of treatment and time since diagnosis were included in a multivariable log-linear regression model. To study the predictive ability of HPV-testing after surgery for CIN2+ a case control study nested in a cohort of women treated for CIN2-3 was performed. Cases were 189 women with recurrence of CIN2+ more than two years after treatment. Exposure was presence of HPV in at least one of two archival smears within two years post treatment. Results: In the LBC-arm >40% more high grade lesions were found and 30% more women needed follow up. The inadequacy rate of smears fell 60%. The scoring system for colposcopy showed very good ability to find and exclude CIN2+ at certain cut off points and the area under the ROC curve for the system was 0.87. Women treated for CIN3 had 2.5 times increased risk for vaginal or cervical cancer. Risk did not decrease substantially after 25 years and was accentuated when treatment was made in women older than 50. Risk also increased with time-period of treatment. Among women treated for CIN2-3 the odds ratio for recurrence was 2.5 when testing positive for HPV 6 ? 12 months post surgery and the sensitivity of the test was 24%. Conclusions: LBC and a new colposcopic scoring system can improve detection of CIN2+. Women once treated for CIN3 constitute a high risk group that needs to be followed up for a long time, More studies are needed to find the best strategies for such follow up.