Secrets of the Cervical Cells

“Cancer development is complex, with both germline genetic and non-genetic influences playing an essential role,” says Martin Widschwendter, explaining the inspiration behind two new papers investigating the DNA methylome of cervical cells as a predictor of breast and ovarian cancer prognosis (1,2). “The underlying and principal drive to our work is to develop novel, easy-to-apply primary and secondary cancer preventive measures,” he continues. “To achieve this, we need to understand who is at high risk of developing cancer.”

The team chose to study epithelial cells in both studies – why? Widschwendter says there were multiple reasons behind the decision. “The matrix that reflects influencing factors is the epigenome – and, metaphorically speaking, these risk factors leave an epigenetic footprint. The tissue from which this epigenetic footprint comes needs to satisfy three main requirements: i) the cells that act as a surrogate for the cells of origin must be easily accessible, because we cannot perform biopsies or surgical procedures to obtain at-risk tissue when the main purpose is to identify at-risk individuals; ii) the cancers of interest originate from epithelial cells and, because the epigenome in epithelial cells differs drastically to that in blood cells, the surrogate tissue must be an epithelial tissue; and iii) surrogate tissue must be hormone-sensitive, because several risk factors for breast and ovarian cancer are hormonal. The only tissue that meets all three of these requirements is that obtained from a cervical smear sample: easy to access, hormone-sensitive epithelial cells.”

In the studies, the researchers developed two indices for predicting risk in breast and ovarian cancer patients. “The Women’s Risk Identification for Breast Cancer (WID-BC) and Ovarian Cancer (WID-OC) indices are epigenetic (DNA methylation) signatures in cervical smear samples,” explains Widschwendter. “They are based on the combination of DNA methylation levels at several CpG sites. WID indices were derived by comparing DNA methylation in samples from women with and without cancer.”

What makes the studies unique is that they selectively included women whose cancers had characteristics already known to be associated with poorer outcomes. “By including women with these types of cancer, we ensured that the WID indices were designed to predict cancers with the worst prognoses,” says Widschwendter. “We know that breast cancer is a heterogeneous disease and that some cancers are overdiagnosed – leading to harms associated with overtreatment. By designing our test to identify cancers with the poorest prognoses, we aim to avoid this type of overdiagnosis.”

The WID indices have great potential for supporting pathologists and laboratory medicine professionals working in cancer prognostics, which could positively impact patient outcomes. Widschwendter says, “We are hopeful that, in the not-too-distant future, a WID test result derived from a cervical smear sample will afford women the opportunity to understand not only their risk for cervical cancer, but also their risk for endometrial, ovarian, and breast cancer. This will lead to tailored advice regarding primary and secondary preventive measures, and we will continue to work with pathologists and laboratory medicine professionals to achieve these goals. There is also great promise in the WID-test approach to be utilized and delivered in a self-sampling setting.