General discussion and future perspectives 179 8 demonstrated the ability to not only predict the risk of cervical cancer, but also the risk of endometrial, breast, and ovarian cancer 14, 95. These findings underscore the potential for additional applications of methylation testing, including the detection of various types of tumours. 8.2 FUTURE PERSPECTIVES 8.2.1 TECHNICAL IMPROVEMENTS OF METHYLATION ANALYSIS: AUTOMATION OF WORKFLOW Population-based cervical cancer screening involves the quality-assured testing of large numbers of samples with a preferred short turn-around-time, asking for automated systems. Currently, most methylation analysis protocols involve isolating genomic DNA from a clinical specimen and performing sodium bisulphite treatment prior to PCR amplification, which is relatively time consuming and requires substantial hands-on time. Therefore, in Chapter 5, we evaluated a direct cell conversion protocol, without the need for prior DNA isolation and normalisation. The direct cell conversion protocol achieved a high success rate and exhibited good analytical performance when applied to cervical samples. Further automation of this protocol is currently under evaluation. Rausch et al. reported on an automated solution for the purification of bisulphite-converted DNA derived from direct conversion of plasma and urine 96. It is anticipated that effective high-throughput DNA methylation analysis is achievable in the near future in populationbased cervical cancer screening settings. 8.2.2 IMPROVEMENT OF THE PERFORMANCE OF METHYLATION MARKER ANALYSIS ON SELF-COLLECTED SAMPLES Chapter 4 revealed a moderate correlation between methylation levels in self-collected samples and clinician-collected cervical samples from the same woman. This outcome aligns with the findings of Klischke et al., who investigated the concordance of methylation analysis between self-collected samples and clinician-collected cervical samples. This study reported a consistent pattern of lower methylation levels in self-samples 74. It is well known that each sample type has its own cellular composition with a different fraction of indicator cells exfoliated from CIN3+, which may affect the performance of methylation tests. Self-collected samples likely have a more random collection of vaginal and cervical indicator cells compared to clinician-collected samples targeting the cervix. This difference in sampling method contributes to a likely lower fraction of methylationpositive cells representative for CIN3+. This may explain the slightly lower sensitivity and
RkJQdWJsaXNoZXIy MjY0ODMw