This study was carried out in the hospital, Dr. Burhan Nalbantoglu Devlet Hastanesi, Lefkosa, North Cyprus. This hospital treats all BC cases in North Cyprus. Ethical approvals were obtained from the Near East University scientific research evaluation ethics community and Dr. Burhan Nalbantoglu Devlet Hastanesi’s ethics community before the research was carried out. All methods were performed following the relevant guidelines and regulations.
Sampling size was based on the following calculations:
Equation 1
Equation 1 was used to determine how many women are needed in order to get results that reflect the target population as precisely as possible. Were, N = 121257 (Women Population Size) This is the total number of women in the whole population, t = t-table value = 1.96 (α=0.05) t is the number of standard deviations a given proportion is away from the mean, p= (prevalence rate) = 91/100000 = 0.00091 (Expected Frequency) This is the proportion of the population affected with BC, q= 1-p = 0.99909 This is the proportion of the population not affected with BC, d= (Acceptable margin of Error) =0,001 The margin of error is the amount of error that can be tolerated. Lower margin of error requires a larger sample size. Following the calculations the required sample size= 317.8 women.
Study group:
This was a study consisting of 655 women that were separated into two groups as follows: Case group = 318 women with confirmed cases of BC were included. The patients with BC were registered with the center’s database and diagnosed based on pathological report according to the international classification of diseases for oncology 3rd edition (C50.0 – C50.9)(22). Hospital-based control groups = 337 women without BC. Women with history of lobular or ductal carcinoma in-situ were excluded from the controls. Only participants between the ages of 30 to 84 years were included in the study group. Informed consent to participate was obtained after the aim of the study was explained by a medical professional.
Data collection:
Retrospective medical and demographic information of case group and hospital-based control group was collected from medical records and the information not available in the medical records was collected through one-on-one interviews.
A retrospective data was used because of the long latency period to BC manifestation and the dynamic nature of the population, thus making it difficult for follow-up.
The Information collected included: age, age at diagnosis, age at menarche, age at first delivery, menopausal status, presence or absence of benign breast disease, history of BC in first-degree relatives (FDR) or other relatives, BRCA 1 and 2 mutation, BMI, history of hormone replacement therapy (HRT) including estrogen/progestin and breast density.
Breast cancer risk assessment:
The information collected was used in the models to predict the risk of BC. The information on BRCA 1 and 2 mutations status could not be provided by the participants so it was excluded.
The IBIS model is a computer-based program that provided a woman’s overall risk of BC by incorporating genetic determinants such as the BRCA 1 and 2 genes (19), Details about breast/ovarian cancer among family members, personal risk factors such as age, BMI, age at menarche, parity, age at first child, menopausal status, breast density, age at menopause, and benign breast disease (19). The IBIS model accommodates residual familial correlation by incorporating a latent common autosomal dominant low-risk gene (23). The IBIS or Tyrer-Cuzick BC risk evaluation tool version 8.0b used is available at (http://www.ems-trials.org/riskevaluator/). The performance of the IBIS model was measured by estimating the BC risk for each individual. The 10year risk was divided by 2 to obtain the 5year risk. Though BC risk increases with age dividing the 10year risk gave an approximate value for the 5year risk. The BOADICEA model calculated 5year risk of BC in the women based on their age, family history, BRCA 1 and 2 carrier probabilities and includes a polygenic component, which allows for the familial correlation that is not captured by mutations in BRCA1 or 2. The BOADICEA risk calculation was carried out using BWAv3 (http://ccge.medschl.cam.ac.uk/boadicea/). The National Cancer Institute’s online version of the breast cancer risk assessment tool (BCRAT) also known as the Gail model available at (http://www.cancer.gov/bcrisktool/) was also used and has questions about the 5 year BC risk based on age, age at menarche, age at first life birth, FDR with breast cancer, previous breast biopsies with or without atypical hyperplasia, BRCA mutation and race. White race/ethnicity (Caucasians) variables was used for all the women in this study in estimating their risks. For the Gail model five-year risk assessment, a rate of less than 1.67% was defined as low risk while a rate of 1.67% or more was defined as high risk (23). Based on the NCCN (2016) recommendation for prophylactic treatment for women the cut of value of 1.67 % for 5year risk was used for all they models while categorizing high and low risk women (33).
Statistical analysis:
The Receiver Operating Characteristic Curve (ROC) plots, was utilized to measure the model’s discriminative capacities. The c-statistics ranges from 0.5 (No discriminative ability) to 1.0 (Perfect discrimination). This determines whether the models will yield a higher risk for BC cases and lower risk for hospital-based controls. The predicted scores were used to distinguish between high and low risk. Sensitivities and specificities of the models were estimated for 5year BC risk at 1.67%. The sensitivity test has the diagnostic ability to detect true positives and the specificity, the diagnostic ability to detect true negatives. A cut off value of 1.67% was used to predict the risk of BC.
The predictive accuracies of correctness (PAC) and the Nagelkerke’s R square of the models were analyzed by logistic regression of the scores predicted by the models. The PAC measures how well the models fit the samples. The Nagelkerke’s R square is an adjusted version of the Cox and Snell R square that adjust the scale of the statistic to cover the full range from 0 to 1. A perfect model has a theoretical maximum value of less than 1.All statistical analysis was done using SPSS version 24.0 analytical software.