Objective. Our previous work revealed that host genes ZNF582, PTPRR, PAX1, and SOX1 are highlymethylated in cervical intraepithelial neoplasias grade 3 orworse (CIN3+). In this study,we used a standardized testing assay to evaluate the clinical efficacy of these biomarkers in the triage of cytological diagnoses of low-grade squamous intraepithelial lesions (LSILs), and compared the performance with human papillomavirus (HPV) testing.
Methods. This 2-year multicenter prospective study examined a population of 230 women from 12 medical centers who were diagnosed with LSILs on cervical cytology. Cervical scrapings were obtained prior to a colposcopy-directed biopsy for quantitative methylation analysis of ZNF582, PTPRR, PAX1, and SOX1, and HPV testing. Using logistic regression and receiver operating characteristic curve analyses, the abilities of methylated genes and HPV to predict CIN3+ were assessed.
Results. Fifteen (6.5%) of the 230 women with a cytological diagnosis of LSIL were confirmed to have CIN3+ after a colposcopy-directed biopsy. Among the 4 methylated genes, ZNF582 was found to be the best biomarker for detecting CIN3+. The sensitivities for methylated ZNF582 and HPV testing were 73% and 80%, and the specificities were 71% and 28%, respectively. The odds ratio for predicting CIN3+ using methylated ZNF582 was 6.8 (95% confidence interval (CI) 2.1-22.1), which was much better than HPV testing (OR = 1.6, 95% CI 0.4-5.8).
Conclusion. This is the first study to show that ZNF582 methylation analysis of cervical swabs may be a promising choice in the positive triage of cytological diagnoses of LSILs.
All Science Journal Classification (ASJC) codes