Risk determination and prevention of breast cancer

Abstract

Breast cancer is an increasing public health problem. Substantial advances have been made in the treatment of breast cancer, but the introduction of methods to predict women at elevated risk and prevent the disease has been less successful. Here, we summarize recent data on newer approaches to risk prediction, available approaches to prevention, how new approaches may be made, and the difficult problem of using what we already know to prevent breast cancer in populations. During 2012, the Breast Cancer Campaign facilitated a series of workshops, each covering a specialty area of breast cancer to identify gaps in our knowledge. The risk-and-prevention panel involved in this exercise was asked to expand and update its report and review recent relevant peer-reviewed literature. The enlarged position paper presented here highlights the key gaps in risk-and-prevention research that were identified, together with recommendations for action. The panel estimated from the relevant literature that potentially 50% of breast cancer could be prevented in the subgroup of women at high and moderate risk of breast cancer by using current chemoprevention (tamoxifen, raloxifene, exemestane, and anastrozole) and that, in all women, lifestyle measures, including weight control, exercise, and moderating alcohol intake, could reduce breast cancer risk by about 30%. Risk may be estimated by standard models potentially with the addition of, for example, mammographic density and appropriate single-nucleotide polymorphisms. This review expands on four areas: (a) the prediction of breast cancer risk, (b) the evidence for the effectiveness of preventive therapy and lifestyle approaches to prevention, (c) how understanding the biology of the breast may lead to new targets for prevention, and (d) a summary of published guidelines for preventive approaches and measures required for their implementation. We hope that efforts to fill these and other gaps will lead to considerable advances in our efforts to predict risk and prevent breast cancer over the next 10 years.

Figure 1

An example of the distribution of visually assessed percentage density of the breast. The sample consists of 50,831 women between 46 and 73 years of age. Density was estimated in two views of each breast on a visual analogue scale, and the four readings were combined to give a single value per woman [54].

Figure 2

Estimation of the effect on the distribution of Tyrer-Cuzick scores by adding the results of 18 or 67 single-nucleotide polymorphisms (SNPs) in 10,000 women [[53]]. Adding SNPs increases the number of women in high- and low-risk groups. ER, estrogen receptor; SNP 18 and SNP 67, distribution using SNPs alone; TC, the Tyrer-Cuzick score alone; TC + SNP67, distribution of the combined score.

Figure 3

Features of the normal breast. (a) Electron micrograph of a ductule of the breast. (b) Section of lobules of the breast showing a relationship with collagenous and fatty stroma. Reprinted with permission from the American Association for Cancer Research [166]. (c) A simplified cartoon of reported potential interactions between three cell types in the stroma and the epithelium of the breast. CSF, colony-stimulating factor; ER, estrogen receptor; IGF1, insulin-like growth factor 1; PR, progesterone receptor; PTH, parathyroid hormone; TDLU, terminal duct lobular unit.

Figure 4

US Institute of Medicine blueprint for lifestyle change. Reprinted with permission from the US Institute of Medicine [192].