Background <p>First-void urine (FVU) collection for high-risk human papillomavirus (HPV) testing may reach un(der)-screened women in cervical cancer screening programs. This cross-sectional study investigated the clinical performance of DNA methylation markers <i>ASCL1</i> and <i>LHX8</i> in HPV-positive FVU for detecting high-grade cervical intraepithelial neoplasia and cancer (CIN2+ and CIN3+). Additionally, results were compared with paired HPV-positive clinician-collected cervical samples (CS) and HPV genotyping.</p> Methods <p>Paired FVU and CS samples were collected from 286 women aged 23–64&#xa0;years referred for colposcopy or cervical excision. Histological endpoints included 123 ≤ CIN1 (no dysplasia and CIN1), 38 CIN2, and 123 CIN3/AIS and 2 cancers. Samples were tested for HPV DNA and <i>ASCL1</i>/<i>LHX8</i> methylation. Methylation test performance was evaluated by area under the curve (AUC) and logistic regression. Differences between paired samples and across methylation, HPV16/18, and extended HPV16/18/31/33/52 genotyping in FVU were tested using McNemar’s test.</p> Results <p><i>ASCL1</i> and <i>LHX8</i> methylation levels in HPV-positive FVU increased significantly with disease severity. Methylation testing in FVU yielded an AUC of 0.76 (95% CI: 0.70–0.82) for CIN3 + and 0.73 (95% CI: 0.67–0.79) for CIN2 + , with sensitivities of 79.2% (95% CI: 71.0–85.9%) and 75.5% (95% CI: 68.1–81.9%), respectively, and a specificity of 57.0% (95% CI: 47.8–65.8%) for ≤ CIN1. In CS, methylation testing yielded higher AUCs of 0.84 (95% CI: 0.79–0.89) for CIN3 + and 0.80 (95% CI: 0.75–0.85) for CIN2 + , with significantly higher sensitivities (<i>p</i> ≤ 0.02) but lower specificity (<i>p</i> = 0.04) than FVU. In women aged ≥ 30&#xa0;years, CIN3 + sensitivity and specificity for ≤ CIN1 were similar between FVU and CS (both <i>p</i> = 0.09). No significant differences in accuracy were observed between methylation testing and extended genotyping (<i>p</i> ≥ 0.35)&#xa0;in FVU, while methylation testing showed higher sensitivity (<i>p</i> &lt; 0.01) but lower specificity (<i>p</i> &lt; 0.01) than HPV16/18 genotyping.</p> Conclusions <p><i>ASCL1/LHX8</i> methylation testing in HPV-positive FVU showed promise for detecting high-grade cervical disease. Its performance in FVU did not differ significantly from extended HPV genotyping and, in women aged ≥ 30&#xa0;years, was comparable to performance in CS. This supports methylation testing as a complementary triage test alongside HPV genotyping in urine-based cervical cancer screening, removing the need for follow-up cervical sampling. Further validation in HPV-positive FVU from un(der)-screened populations is warranted.</p> Trial registration <p>ClinicalTrials.gov: NCT05065853.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Evaluating DNA methylation markers and extended HPV genotyping in first-void urine for detecting high-grade cervical lesions in HPV-positive women: a cross-sectional study

  • Mette Tranberg,
  • Severien Van Keer,
  • Albertus T. Hesselink,
  • Pia Nørgaard,
  • Rikke Brøndum,
  • Chunsen Wu,
  • Line Winther Gustafson,
  • Anne Hammer,
  • Pinar Bor,
  • Karen Omann Binderup,
  • Christina Blach,
  • Alex Vorsters,
  • Renske Steenbergen

摘要

Background

First-void urine (FVU) collection for high-risk human papillomavirus (HPV) testing may reach un(der)-screened women in cervical cancer screening programs. This cross-sectional study investigated the clinical performance of DNA methylation markers ASCL1 and LHX8 in HPV-positive FVU for detecting high-grade cervical intraepithelial neoplasia and cancer (CIN2+ and CIN3+). Additionally, results were compared with paired HPV-positive clinician-collected cervical samples (CS) and HPV genotyping.

Methods

Paired FVU and CS samples were collected from 286 women aged 23–64 years referred for colposcopy or cervical excision. Histological endpoints included 123 ≤ CIN1 (no dysplasia and CIN1), 38 CIN2, and 123 CIN3/AIS and 2 cancers. Samples were tested for HPV DNA and ASCL1/LHX8 methylation. Methylation test performance was evaluated by area under the curve (AUC) and logistic regression. Differences between paired samples and across methylation, HPV16/18, and extended HPV16/18/31/33/52 genotyping in FVU were tested using McNemar’s test.

Results

ASCL1 and LHX8 methylation levels in HPV-positive FVU increased significantly with disease severity. Methylation testing in FVU yielded an AUC of 0.76 (95% CI: 0.70–0.82) for CIN3 + and 0.73 (95% CI: 0.67–0.79) for CIN2 + , with sensitivities of 79.2% (95% CI: 71.0–85.9%) and 75.5% (95% CI: 68.1–81.9%), respectively, and a specificity of 57.0% (95% CI: 47.8–65.8%) for ≤ CIN1. In CS, methylation testing yielded higher AUCs of 0.84 (95% CI: 0.79–0.89) for CIN3 + and 0.80 (95% CI: 0.75–0.85) for CIN2 + , with significantly higher sensitivities (p ≤ 0.02) but lower specificity (p = 0.04) than FVU. In women aged ≥ 30 years, CIN3 + sensitivity and specificity for ≤ CIN1 were similar between FVU and CS (both p = 0.09). No significant differences in accuracy were observed between methylation testing and extended genotyping (p ≥ 0.35) in FVU, while methylation testing showed higher sensitivity (p < 0.01) but lower specificity (p < 0.01) than HPV16/18 genotyping.

Conclusions

ASCL1/LHX8 methylation testing in HPV-positive FVU showed promise for detecting high-grade cervical disease. Its performance in FVU did not differ significantly from extended HPV genotyping and, in women aged ≥ 30 years, was comparable to performance in CS. This supports methylation testing as a complementary triage test alongside HPV genotyping in urine-based cervical cancer screening, removing the need for follow-up cervical sampling. Further validation in HPV-positive FVU from un(der)-screened populations is warranted.

Trial registration

ClinicalTrials.gov: NCT05065853.