ASCL1 and LHX8 methylation in HPV-positive self-collected samples 101 4 This combined strategy nonetheless obviates the need for a recall for a clinician-collected cervical sample for cytology triage, which may counterbalance the increased referral rate due to the drop in specificity, and integrates high adherence to triage. The total impact of screening on the detection of cervical lesions is not only related to the performance of the test, but also dependent on the compliance with follow-up procedures. In the Netherlands, about 10-20% of women with an hrHPV-positive self-collected sample does not show up for cytology triage 35. Therefore, alternative triage methods, such as host-cell DNA methylation analysis either or not combined with HPV genotyping, may be considered which do not require an extra visit to the clinician. The decrease in specificity of the combined strategy is expected to gradually fade when HPV-vaccinated women will enter the screening program. The first vaccinees will reach screening age in the Netherlands in 2023. It should furthermore be noted that the performance of cytology as triage strategy is largely dependent on the quality of cytology, that varies widely among countries and is high in the Netherlands. The full molecular strategy directly applicable to self-samples would be particularly beneficial in settings without a quality-assured cytology infrastructure or low-resource settings where cytology screening is limited. The unique study design of the IMPROVE study allowed for direct comparison of methylation analysis on self-collected and clinician-collected samples from the same hrHPV-positive women. Though promising, the results also highlight that the performance for implementation of the methylation assay in an organised screening setting as single triage test on self-collected samples would require further improvement. Alike for cytology triage which currently requires a retest at 6-12 months in case of normal cytology at baseline to ensure sufficient protection, one could consider to improve the NPV by offering repeat testing to those women that screen hrHPV-positive but methylation negative at baseline testing of their self-collected sample. Alternatively, depending on country preferences, a strategy with baseline testing only could comprise recall of women that screen hrHPV-positive but methylation negative on their self-collected sample for an additional cervical sample at the clinician’s office for e.g. methylation analysis or cytology testing. For the current study cohort, these latter algorithms would have resulted in NPVs of 96.1% (95% CI 93.5-98.8%) and 98.6% (95% CI 97.1-100.2%), respectively (data not shown). The methylation markers used in this study were discovered using a genome-wide screen on self-collected samples of non-attendees 19. In an initial validation series, these markers showed good clinical performance for CIN3 detection in both hrHPV-positive
RkJQdWJsaXNoZXIy MjY0ODMw