Aim Many cervical cancers occurs among women over 65 and prevalence of HPV genotypes in this age cohort is sparingly studied. One aim of this study was to study the prevalence and distribution of HPV genotypes in women 55–59 years, with normal cytology when exiting the screening program. Secondly, HPV clearance as well as the value of HPV genotyping and/or liquid based cytology as triage tests for identifying histological dysplasia among women with persistent HPV was studied. Methods Women that exited the screening program with normal cytology, between the years 2012–2014, in Örebro County, Sweden, were invited to this study. A total of 2946 samples were analyzed with a broad-spectrum assay to detect both hrHPV and lrHPV in order to investigate the distribution of genotypes. In the consent group, women with a positive hrHPV test were offered a follow-up test and a cone biopsy for histological confirmation, and a follow up sample 6 months post cone. Results The overall prevalence of hrHPV was 7.4% and 59% of them remained hrHPV positive in a follow-up test after 12 months. A total of 99 women had a cone biopsy done, where 19% showed histological dysplasia. HPV 53 was the most common genotype, and among women with histology confirmed LSIL or HSIL, HPV 31 was most common. A positive hrHPV result showed a PPV of 25% for LSIL+ and 12.5%for HSIL+. Using detection of HPV 16/18 genotypes as a triage test for hrHPV positive tests, indicated FNR for histological LSIL+ and HSIL+ of 94% and 87.5% respectively, whilst triage based on cervical cytology had a FNR of 69% for LSIL+ and 37.5% for HSIL+. Conclusion The most common hrHPV genotypes among women 55–59 years of age were non HPV16/18 genotypes, and in this population, these genotypes represented most of the histological verified HSIL lesions. This result does not support the proposition of a HPV 16/18 triaging test after a positive hrHPV test as a marker of histological HSIL+ cervical lesions in women over 55 years of age. Similarly, cytological triage after a positive hrHPV showed no additional benefit in this population. Specific triaging tests should be validated to follow post-menopausal women with a positive hrHPV test.

Purpose Örebro County introduced an updated screening program 2016 with primary HPV test for women over 30 years and prolonged screening, increasing the cutoff age from 56-60 to 64-70. The aim of this study was to investigate the prevalence of HPV genotypes and their correlation to histological changes in women, 10 years after exclusion from the screening program, due to an eventual implementation of a catch-up program including all women aged 60-70. Methods All women in Örebro County, born 1,946 (n = 1,968), were invited to a liquid-based cell sample with primary HPV screening. Samples were analyzed for hrHPV mRNA and positive samples were genotyped. hrHPV positive women were offered to do a conization. Results Out of 809 participants, 31 (3.8%) were hrHPV positive, of these 22 did a conization. Histologically, 5/22 (23%) had LSIL and 5/22 (23%) had HSIL. Normal histology was found in 12/22 (55%). The most prevalent genotypes were HPV 16, 33, 52, 56, and 68. Of the women with HSIL, one case of cervical cancer was confirmed in a recone biopsy after 4 months. Conclusion The study showed considerable prevalence of hrHPV and histologically confirmed LSIL/HSIL. These data led to catch-up screening for women between 60 and 70 years when overlapping two screening strategies.

Postmenopausal women could be offered self-sampling devices to increase screening-program coverage while maintaining test quality.

Highlights In an HPV-based cervical screening programme more HSIL + cases was detected but the clinical effectiveness comes with higher rates of colposcopy and clinical irrelevant findings. In an HPV-based cervical screening programme two thirds of the cost, come from screening visits with sampling. From a health economic perspective, our data emphasize the need for considering alterations in the screening programme including more specific triage analysis and the implementation of self-sampling.

No abstract