The Tyrer-Cuzick model predicts the risk of developing breast cancer in a certain age interval, i.e. the probability of developing breast cancer until a certain age if one has not yet developed breast cancer. The model is based on the population survival function and the survival functions of BRCA1 and BRCA2 gene mutation carriers. Segregation part of the model includes age and genetic risk factors obtained from the individual's family history of breast and ovarian cancer or results of genetic tests. The regression part of the model considers the other personal risk factors such as age at menarche, age at first childbirth, menopausal status, receiving hormone replacement therapy, breast density, etc.
In the program IBIS Risk Evaluator v8, authors of Tyrer-Cuzick model included also the incidence of breast cancer in Slovenian population. In this master's thesis, we were interested whether changing population survival function and survival functions of BRCA1 and BRCA2 gene mutation carriers brings clinically significant differences in the calculated risk of developing breast cancer.
We have studied two articles on the Tyrer-Cuzick model, where several parts of the model are not explained in detail. We implemented the model in R software environment and compared risk values according to the IBIS Risk Evaluator v6 program with our calculations. The implementation of the model gave us the ability to change the estimates of the parameters within the model, more precisely the population survival function and the survival functions of the BRCA1 and BRCA2 gene mutation carriers. Differences in risk calculation were examined on a sample of 350 individuals who were included in a preventive screening at the Breast Disease Center and had not yet been diagnosed with breast cancer at the time of screening. We compared the distributions of risk of developing breast cancer and the distribution of individuals into the general, moderate, and high risk category for breast cancer. With the use of the package Shiny, we created an interactive web application in the R software environment for calculating the risk of developing breast cancer.
We showed that differences in risks are clinically significant if we change the population survival function from British to Slovenian, whereas the change in the survival functions of BRCA1 and BRCA2 gene mutation carriers did not lead to clinically significant differences in the distribution of risk or the distribution of individuals into risk categories.
Our application for calculating the risk of developing breast cancer enables entering information about personal risk factors for breast cancer and family history of breast and ovarian cancer. The application returns an estimate of the risk of developing breast cancer in a 10 year period and the probability that the individual is a BRCA1 or BRCA2 gene mutation carrier. Data on the individual woman and her risk can be saved in an Excel file, which enables further data analysis.