Abstract
Cervical cancer is the second most common type of cancer in women worldwide. Preinvasive disease can be detected by cervical cytology. All currently available cytology technologies rely on the visual analysis of exfoliated cells from the uterine cervix. Improvement of conventional cytological screening has been proposed by the introduction of molecularโbased markers applied to liquidโbased cytology (LBC), the suspension of cells collected from the cervix. DNA methylation changes occur very early in carcinogenesis and identification of appropriate DNA methylation markers in such samples should be able to distinguish highโgrade squamous intraepithelial lesions (HSIL) from nonspecific cytology changes and the normal cervix. To address this potential, we have undertaken a proofโofโprinciple study of methylation status of LBC samples from HSIL cytology cases compared against matched normal controls. Using quantitative methylationโspecific PCR on 28 genes, we found SOX1, HOXA11 and CADM1 to significantly discriminate between the groups analyzed (p < 0.01). Area under the receiver operating characteristic (ROC) curve (AUC) demonstrated that methylation of SOX1, HOXA11 and CADM1 could discriminate between HSIL cases and controls with high sensitivity and specificity (AUC 0.910, 0.844 and 0.760, respectively). The results were further validated in an independent set. This proofโofโprinciple study is the first to validate the results in an independent case/control set and presents HOXA11, a gene that is important for cervical development, as a potentially useful DNA marker in LBC samples. Further assessment of these preliminary estimates will need to be performed in a larger cohort to confirm clinical utility. ยฉ 2009 UICC