IMPROVE-study

To evaluate whether clinician-based cervical sampling for HPV testing can be replaced by self-sampling in primary screening, we will perform a randomized pilot implementation trial of self-sampling in primary cervical screening setting. In the intervention (i.e. self-sampling) arm, women are invited for self-collection of (cervico-)vaginal material at home, whereas in the control (i.e. clinician-based sampling) arm women are invited to visit their general practitioner for a cervical (LBC) sample. In both arms, women with hrHPV DNA-positive test results are further triaged with cytology. To enhance the power of the design for assessing the relative sensitivity of self-sample-based HPV testing, women positive for HPV self-sampling (intervention arm) will also be tested by an clinician-based HPV test 2-4 weeks later, and women positive for HPV on a clinician-based sample (control arm) will also be asked to perform HPV self-sampling 2-4 weeks later. This type of design is a randomized paired screen-positive design and has been put forward by Alonzo and Kittelson (2006). In this trial, it will be determined whether HPV self-sampling is not inferior to HPV testing on a clinician-based sample in terms of detection of CIN3+ and CIN2+. To assess non-inferiority of HPV self-sampling in comparison with a HPV test on a clinician-based sample, 30,000 women will be invited for study enrolment.

In the IMPROVE-study, the clinical performance in terms of clinical sensitivity and clinical specificity for high-grade CIN lesions and cervical cancer (CIN2+/CIN3+) of primary HPV screening via two different sampling methods will be compared: 1. self-sampling (i.e. by women themselves), and 2. clinician-based sampling (i.e. cervical smear taken by GP). The hypothesis is that HPV self-sampling is at least as good as HPV testing on a clinician-based sample for the detection of CIN2+/CIN3+ in women attending the national cervical screening program.

N/A

Women who are eligible for the national cervical screening program receive an invitation to participate in this study. When women return a signed informed consent form, they will be randomized between self-sampling (intervention arm) and a clinician-based sample (control arm). Both samples will be tested for the presence of high-risk Human Papillomavirus (hrHPV) DNA by a validated assay.

Women whose samples (self-sample or clinician-based sample) test hrHPV DNA negative are referred to the next screening round after 5 years. Women whose sample test positive for hrHPV DNA will undergo cytology triage (based on the new screening program starting 2016). For women in the intervention arm, cytology triage will be done on a newly obtained clinician-based smear taken within a month. Accordingly, these women are asked to visit the GP. For women in the control arm, the remaining material of the cervical sample can be used for (reflex)cytology testing.

Additionally, all hrHPV-DNA positive women (both arms) will be asked to perform similar sampling as done in the cross arm 2-4 weeks after the first analysis, i.e., women in the intervention arm will go to their GP for a clinician-based sample (this is already needed to perform cytology triage) that will be tested for hrHPV DNA; women in the control arm will be asked to perform additional self-sampling for hrHPV DNA testing. This cross-testing is performed to see if individual HPV results match between the self-sample and the clinician-based sample.
In both arms, if baseline cytology triage shows abnormalities (¡ÝBMD cytology), women will be referred for colposcopy. If baseline cytology triage shows no abnormalities (normal cytology), women will be invited for repeat cytology testing at 6 months. Again, if this cytology test shows abnormalities (¡ÝBMD cytology), women will be referred for colposcopy.



If necessary, women will be treated according to national guidelines. If 6-month cytology shows no abnormalities (normal cytology), women will receive a new invitation for screening in the next screening round (approx. after five years).