The Role of HPV and DNA Methylation in the Development of Precancerous Cervical Lesions

Biological factors associated with cervical intraepithelial neoplasia (CIN) development may be utilized to improve the efficiency of cervical cancer screening. The present study examined human papillomavirus (HPV) genotypes and DNA methylation of imprinted genes in a survival analysis, using two endpoints: CIN1 regression and progression to CIN2+. The CIN Cohort Study (CINCS) was conducted at 10 Duke University clinics in North Carolina. Women ages 18-65 years with enrollment colposcopies following an abnormal Pap test had follow-up cytology/histology for 3 years with up to 5 visits. DNA was extracted from exfoliated cervical cells for methylation at differentially methylated regions (DMR) of imprinted genes and for HPV genotyping. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated using Cox regression models to estimate the association between aberrant methylation and CIN1 regression (normal epithelia) versus persistence/progression (CIN1+). HRs/95% CIs were estimated to quantify genotype-specific risk of progression to CIN2+. Of 1303 participants, 472 normal/CIN1 cases with HPV data and 151 CIN1 cases with HPV/methylation data were included in the prospective analyses. In the analysis of genotype-specific risk of progression, hrHPV-66/51/16 were most prevalent in multiple-type infections among 364 HPV-unvaccinated women; 16/52/35 in single-type. Among 108 HPV-vaccinated women, hrHPV-66/39/51 were most prevalent in multiple-type infections; hrHPV-51/66/52/58 in single-type. Over 3 years, there were CIN2+ events (11%;38% normal;62% in CIN1). HrHPV-16/51 had a 3-fold risk of progression (HR=3.2,95% CI=1.5-7.2;HR=3.2,95% CI=1.3-7.6) versus non-16/51 infections. HPV-16/51 were most predictive of progression to CIN2+. In the CIN1 regression analysis, one-third regressed to normal epithelium (n=53;35.1%). Median time-to-regression was 12.6 months (range:4.5-24.0). The probability of CIN1 regression was negatively correlated with increased methylation at IGF2AS CpG 5 (HR=0.42,95% CI=0.23-0.77;p=0.005) and PEG10 DMR (HR=0.78,95% CI=0.63-0.96;p=0.02). HPV-51 was positively associated with increased risk of progression; aberrant methylation of PEG10/IGF2AS reduced the likelihood of CIN1 regression. PEG10/IGF2AS methylation may serve as potential biomarkers for screening, given further characterization of tumorigenesis pathways related to dysregulation of imprinted gene expression. HPV-51 may be useful for CIN2+ risk stratification. If confirmed in other populations, implementation of these novel biomarkers may improve LSIL management and patient care.Doctor of Philosoph