The risk of breast cancer - one of the most important malignant tumors of modern society - is continually increasing across different societies of the world, with rates in developing countries (which were inherently lower) catching up with the higher rates observed in the Western world.
Although cure rates are improving, not all of this can be ascribed to mammography screening. Thus the true value of population-based screening is still a subject of debates. Consequently, enthusiasm for risk-stratified screening is on the rise.
Estimating an absolute risk for individual women is pivotal for proper decision making about screening and preventive actions. During the last three decades, different sophisticated models have been developed that can assist clinicians in more precise risk estimation. These models enable quantification of risk for breast cancer development by combining multiple risk factors into a coherent risk expression.
One of these models is the Gail model, which is a statistical risk assessment tool used to estimate the probability of breast cancer development in presently healthy women within the next five years and during their lifetime. It represents one of the most commonly employed risk prediction models for breast cancer in the clinic.
The Development of the Gail Model
The Gail model was developed by Dr. Mitchell Gail and his colleagues at the National Cancer Institute (NCI) in the US with the aim to calculate an individual's combined risk of developing invasive carcinoma, ductal carcinoma in situ, or lobular carcinoma in situ over a specific period of time when undergoing annual mammographic screening.
This model was generated with the help of data from 4,496 matched pairs of cases and controls in the Breast Cancer Detection Demonstration Project that primarily used non-genetic factors to predict breast cancer risk for women without a personal history of breast cancer. Factors included in the model are current age, age at menarche, age at first birth or nulliparity, number of previous breast biopsies, number of first degree relatives with breast cancer, as well as the presence of atypical hyperplasia on biopsy.
This was the initial Gail model (also referred to as the Gail model 1), which was later modified by statisticians from the Breast Cancer Prevention Trial to better project the risk of invasive breast carcinoma. This revised model (also known as the Gail model 2) uses age-specific invasive breast cancer incidence rates and has a high predictive value, regardless of the factors that place an individual woman at risk.
Clinical Use
The Gail model represents an extremely useful tool for counseling women about their level of breast cancer risk, as it can estimate individualized absolute risk. Furthermore, it is also established as the basis for sample size calculation for the prevention trials with the drug tamoxifen (a selective estrogen receptor modulator used in the treatment of breast carcinoma).
The model was also adapted to identify and predict the risk of individual women but with certain precautions, as there is not enough precision needed to provide recommendations for individual patients. It has also been used exclusively in women with atypia in order to assess the risk of developing invasive forms of breast cancer.
Shortcomings of the Model
An important limitation of the Gail model (but also other existing models) is that it does not include an exhaustive set of risk factors for breast cancer development. For example, certain established risk factors such as increased estrogen levels, adiposity (most notably for postmenopausal women), the use of hormone replacement therapy, a sedentary lifestyle, and alcohol abuse are absent.
Furthermore, this model has not been validated in all population subgroups, such as women with genetic predispositions for breast cancer or younger women. The Gail model can even underestimate the absolute risk for breast cancer in women who carry predisposing genetic changes (such as mutant BRCA1/BRCA2 gene or Cowden multiple hamartoma syndrome). Accordingly, the Gail model should not be used if the patient's family history is the principal source of risk.
Further Reading
Bio-Rad launches first ultrasensitive multiplexed digital PCR assay for breast cancer mutation detection in clinical research