Detection of hypermethylated genes in women with and without cervical neoplasia

DNA methylation changes are an early event in carcinogenesis and are often present in the precursor lesions of various cancers. We examined whether DNA methylation changes might be used as markers of cervical intraepithelial neoplasia (CIN) and invasive cervical cancer (ICC).

We used methylation-specific polymerase chain reaction (PCR) to analyze promoter hypermethylation of 20 genes, selected on the basis of their role in cervical cancer, in 319 exfoliated cell samples and matched tissue biopsy specimens collected during two studies of Senegalese women with increasingly severe CIN and ICC (histology negative/atypical squamous cells of undetermined significance [ASCUS] = 142, CIN-1 = 39, CIN-2 = 23, CIN-3/carcinoma in situ [CIS] = 23, ICC = 92). Logic regression was used to determine the best set of candidate genes to use as disease markers. All statistical tests were two-sided.

Similar promoter methylation patterns were seen in genes from exfoliated cell samples and corresponding biopsy specimens. For four genes (CDH13, DAPK1, RARB, and TWIST1), the frequency of hypermethylation increased statistically significantly with increasing severity of neoplasia present in the cervical biopsy (P<.001 for each). By using logic regression, we determined that the best panel of hypermethylated genes included DAPK1, RARB, or TWIST1. At least one of the three genes was hypermethylated in 57% of samples with CIN-3/CIS and in 74% of samples with ICC but in only 5% of samples with CIN-1 or less. The estimated specificity of the three-gene panel was 95%, and its sensitivity was 74% (95% confidence interval [CI] = 73% to 75%) for ICC and 52% (95% CI = 49% to 55%) for CIN-3/CIS. By extrapolation, we estimated that, among Senegalese women presenting to community-based clinics, detection of the DAPK1, RARB, or TWIST1 hypermethylated gene would reveal histologically confirmed CIN-3 or worse with a sensitivity of 60% (95% CI = 57% to 63%) and a specificity of 95% (95% CI = 94% to 95%).

Aberrant promoter methylation analysis on exfoliated cell samples is a potential diagnostic tool for cervical cancer screening that potentially may be used alone or in conjunction with cytology and/or human papillomavirus testing.