TOPLINE:

In a meta-analysis, DNA methylation markers such as JAM3, C13ORF18, and PAX1 were associated with high diagnostic accuracy, with high sensitivity and specificity observed for detecting high-grade cervical intraepithelial neoplasia (CIN) and cervical cancer.

METHODOLOGY:

• Researchers conducted a systematic review and meta-analysis including 125 studies examining DNA methylation in archived specimens from 49,242 diseased and non-diseased women reported up to March 2025.
• Overall, 81 cohort studies (64.8%), 44 case-control studies (35.2%), and seven biobank-based studies (6.5%) were included, with 51 studies reporting results in high-risk human papillomavirus (hrHPV)-positive women.
• The primary outcome was the evaluation of DNA methylation as a triage test in hrHPV-positive women, including 13 markers consisting of nine individual human genes, one viral gene, and three mixed panels for CIN3+, CIN2+, and invasive cervical cancer (ICC).
• The secondary outcome was the evaluation of DNA methylation as a diagnostic test among all women regardless of hrHPV status, with 32 methylation markers analysed.

TAKEAWAY:

• For the detection of CIN3+ in hrHPV-positive women, JAM3 methylation demonstrated the highest diagnostic accuracy (diagnostic odds ratio [DOR], 20.82; 95% CI, 14.41-30.08), with a sensitivity of 74% (95% CI, 64%-81%) and a specificity of 88% (95% CI, 84%-92%), followed by C13ORF18 (DOR, 19.50; 95% CI, 11.65-32.65) with the highest specificity of 93% (95% CI, 89%-95%) and PAX1 (DOR, 17.13; 95% CI, 9.20-31.89).
• EPB41L3 demonstrated the lowest negative likelihood ratio of 0.26 (95% CI, 0.19-0.35) for CIN3+ in hrHPV-positive women, achieving posttest probabilities of 1.35% in low-risk settings, 2.51% in moderate-risk settings, and close to 5.00% in high-risk settings.
• In all women regardless of hrHPV status, SOX1 reported the highest accuracy (DOR, 29.72; 95% CI, 12.84-68.81) for CIN3+, with a sensitivity of 93% (95% CI, 86%-96%) and a specificity of 70% (95% CI, 59%-79%), followed by JAM3 and PAX1.
• For the detection of ICC in all women, the GynTect panel showed a perfect sensitivity of 100%, with a corresponding specificity of 68% (95% CI, 48%-83%); however, APC demonstrated the highest specificity of 98% (95% CI, 42%-100%), with a corresponding sensitivity of 26% (95% CI, 13%-44%).

IN PRACTICE:

"DNA methylation markers hold significant promise as diagnostic tests for high-grade CIN, and cervical cancer. We found that in hrHPV positive women, several methylation markers performed well, including some that are not commercial assays, and would be of use in screening either as a primary screening method or as a triage test," the authors wrote.

SOURCE:

This study was led by Laura Burney Ellis, Imperial College London, and Jack Tighe, Imperial College Healthcare NHS Trust, both in London, England. It was published online on May 27, 2026, in the European Journal of Cancer.

LIMITATIONS:

A very low number of studies assessed methylation in the context of a screening population, and more than 90% of studies were rated as a high or unclear risk of bias for the included populations. About one third of studies specified a methylation cutoff before the analysis; however, many determined cutoffs post hoc to optimise diagnostic performance, which likely inflated sensitivity and specificity estimates. High heterogeneity was observed across studies. Publication bias assessment suggested that the pooled diagnostic accuracy may not have fully reflected true performance. Very few studies used self-samples, and none included long-term data beyond 10 years, limiting the assessment of methylation markers' ability to predict incident disease rather than cross-sectional accuracy.

DISCLOSURES:

This study was funded by Imperial Health Charity. One author reported receiving funding through a Cancer Grand Challenges Fellowship from Cancer Research UK, and another reported receiving funding from a National Institute for Health and Care Research academic clinical lectureship, the Wellcome Trust, the National Institute for Health and Care Research, and Wellbeing of Women. Additional disclosures are noted in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.