Every year, over 350 000 women in the EU are diagnosed with breast cancer, and more than 90 000 die from the disease. Yet these statistics obscure a great heterogeneity in the malignancy of the tumour and the treatment options available. “Breast cancer is not one disease, it consists of several subtypes, and we need special preventative measures to tackle that,” explains Marjanka Schmidt, project coordinator of the EU-funded B-CAST (Breast CAncer STratification: understanding the determinants of risk and prognosis of molecular subtypes) project. “The end point is to better predict which woman is likely to develop which breast cancer subtype.” B-CAST builds on the EU-funded COGS project, which identified a large number of germ line variants that influence the risk of breast cancer. The development of cancer is determined by a complex interplay of these genetic risk factors with lifestyle and environment. The aim of B-CAST was to identify women with a moderate to high risk of breast cancer, identify the subtype of cancer that is most likely to develop, and define the prognosis of that particular subtype.
To achieve this, epidemiologist Schmidt and her team at the Netherlands Cancer Institute combined large-scale pathologic-molecular analyses of over 20 000 breast tumours with clinical data from 100 000 breast cancer patients. “We also generated a unique data set of 10 000 tumours, containing not just germ line information but also tumour sequence data, copy number aberrations, and so on, to better measure the genetic profile of the tumour,” says Schmidt. The clinical data contained information on both the tumour and the patient, including lifestyle characteristics, details of the diagnosis, and the presence of immunohistochemical markers. Collaborators in Spain helped to isolate the DNA and panel sequence these 10 000 tumours. “We were hoping to see a more complete view of the genetic landscape. This is a larger and more representative sample than anything published before,” notes Schmidt. With this information, Schmidt hopes to uncover patterns that influence the development of particular cancer subtypes. The collaborating team developed a polygenic risk score covering 313 single nucleotide polymorphisms that offers a powerful and reliable predictor of breast cancer risk. The funding also allowed the group to make the BOADICEA cancer prediction model more accessible, translating it into Dutch, French, German and Spanish. Clinicians use this to stratify patients, offering those at high risk further diagnostics such as an MRI scan.
A key question was quantifying the importance of lifestyle factors in breast cancer subtypes, an overlooked part of managing risk. An inventory of national guidelines carried out in the context of B-CAST found that advice for reducing the risk of breast cancer was almost non-existent. “A small percentage of breast cancers can be prevented by reducing alcohol consumption, maintaining a healthy weight, and performing regular physical exercise,” remarks Schmidt. “The advice is there for heart disease guidelines, but not for the most common cancer in women.” She adds that, in the future, risk models like BOADICEA will aid the provision of genetic counselling that is focused on each individual in the family rather than on the proband. The work was supported by the European Research Council, funding Schmidt says was essential. “What we have pulled off with all these partners could never have been done if each partner went to national funders. To coordinate that, so that you’re all successful at the same time, would have been impossible.” Describing the new database as a “goldmine”, Schmidt concludes that it will continue to generate new discoveries and research papers for the next 5 years.
B-CAST, breast cancer, genetic, lifestyle, risk, prognosis, disease, tumour, subtype, BOADICEA