Breast cancer is an increasingly common disease, particularly amongst younger women. Although new treatments developed over the past few decades mean that most survive the disease, for women with some types of breast cancer, such as triple negative breast cancer, the prognoses remain poor. Women with a germline mutation in the BRCA1 genes have an excessively high risk of developing breast cancer and the majority of these cancers are triple negative breast cancers. Their risk of ovarian cancer is also high. Risk reducing surgery through removal of both breasts, ovaries and fallopian tubes, is currently the only option available to these women at the highest risk as it is not currently possible to predict which women with a mutation will develop cancer.
Role of progesterone in cancer formation
Previous studies in mice have shown that increased progesterone levels trigger increases in the protein RANKL in breast tissue and leads more frequent cell division in non-hormone responsive luminal progenitor cells. This more frequent division of the luminal progenitor cells drives cancer formation.
We studied daily progesterone levels in saliva and urine samples from women with and without BRCA mutations. We found that the progesterone levels were higher throughout the menstrual cycle in women with a BRCA1 mutation compared to controls, particularly during the luteal (post-ovulation) phase. Hence, blocking progesterone signalling in women at high risk of breast cancer could prevent of the disease.
We developed an epigenetic signature in breast tissue – the WID-Breast29 – that identifies the proportion of luminal progenitor cells that have undergone a high number of divisions (or in other words, cells with a high replicative age) and showed that the WID-Breast29 is much higher in breast cancer tissue compared to normal tissue.
Our in vitro studies also highlight the role of progesterone in controlling the cytotoxicity of NK cells in BRCA1 mutation carriers and the role of hypoxia in cancer development.
Mifepristone for breast cancer prevention
Mifepristone is a drug that stops progesterone from binding to its receptor and thereby should interfere with the cascade of events involving RANKL leading to an increased number of luminal progenitor cells and cancer. In our studies, we showed that mifepristone reduced the number of luminal progenitor cells with a high replicative age, as measured with the WID Breast29.
Precision prevention
These new data, as well as the wider literature, support the use of mifepristone for breast cancer prevention. Clinical trials are warranted, but are challenging to perform due to the long-term follow up required to demonstrate the preventive effect.
Our studies show that monitoring breast cancer risk in tissue using the WID-Breast29 could indicate whether breast cancer prevention with mifepristone is working or whether risk reducing surgery would be needed. However, regular tissue sampling is invasive and it would be preferable to monitor risk in a surrogate tissue, such as a cervical sample or cheek swab.
Evidence generated during the BRCA-ERC study support the hypothesis that epigenetic signatures in easy-to-access, hormone sensitive, cervical tissue are reflective of cancer field defects at distant sites, such as in the breast. These data complement other data in surrogate tissue generated during the project for example showing that a tissue-independent epigenetic clock measured in cervical samples (the WID-relative epigenetic age) reflects changes in breast tissue from pre-menopausal women with a BRCA mutation; and a signature in cervical cells – the WID-qtBC – that combines information on methylation status and underlying genetic risk (single nucleotide polymorphisms) identifies women at increased risk for breast cancer. Both the WID-REA and the WID-qtBC indicated a reduction breast cancer risk after treatment with mifepristone.
Early cancer detection
Novel tests for the detection of women’s cancer complement our approach to prevention. Here, we expanded our findings on cell-free DNA methylation (cfDNAme) through analysis in diagnostic samples and early detection (preceding diagnosis; samples from the UK Familial Ovarian Cancer Screening Study). In the diagnostic set, the specificity of cfDNAme was 97.6%. High-risk cancers were detected with a sensitivity of 80%. Combination of cfDNAme and CA125 increased sensitivity to 94.4% for high-risk cancers. We also delivered new studies using an epigenetic test for uterine cancers (the WID-qEC), which outperformed imaging (ultrasound and MRI) in a cohort study.
Next steps
While our ultimate aim is to prevent cancers from occurring, early detection of cancer remains an important target and we have also generated new data on tests for ovarian, endometrial and cervical cancers during this project. Together these insights support clinical investigation of the antiprogestin mifepristone for breast and ovarian cancer prevention. Our ERC Proof of Concept award (BRCA-PREVENT; project number 101113534) will build the necessary consensus to take this forward.
