Population Screening for Inherited Cancer-Related Gene Mutations

Treating IEN, Not Chemoprevention

News from the War on Cancer

Virtual Colonoscopy...At the Forefront of Colorectal Cancer Screening

Virtual Colonoscopy...Is Not Yet an Acceptable Option for Colorectal Cancer Screening

Cancer Prevention Working Group Convenes to Discuss Speeding Cancer Prevention Agents to Patients

Cancer Prevention Clinical Trials

Cancer Prevention Legislation(State)

Cancer Prevention Legislation(Federal)

Victor R. Grann, MD, MPH
Clinical Professor of Medicine and Epidemiology

Judith S. Jacobson, DrPH, MBA
Assistant Professor of Epidemiology
Columbia University College of Physicians and Surgeons
New York, New York

Over the past five years, a number of tests for germline (heritable) gene mutations associated with high risk for certain cancers, such as breast cancer, have become available (Table 1). Appropriately used, such tests may lead to reductions in morbidity and mortality. However, the value of screening and preventive treatments for individuals with cancer-related mutations in different populations depends on a number of factors. In addition to sensitivity, specificity, and predictive value, key criteria for screening include: mutation prevalence (how common the mutation is in a given population), mutation penetrance (the cancer risk that the mutation confers in that population), disease mortality, age at screening, and the potential effects of screening and primary preventive measures on disease incidence, quality of life, and health care costs.

Screening is the use of a medical test in a defined population of asymptomatic individuals to identify those who have undetected disease or an elevated risk of developing disease.(1) Common cancer screening tests, such as mammography and sigmoidoscopy, differ from tests for cancer-related mutations in several respects (Tables 2 and 3). Most cancer screening tests detect but do not predict disease.(2) For that reason, individuals who have had a negative (normal) test result usually need to be tested again; the recommended interval between tests depends mainly on what is known about the latency of the disease.

Cancer screening tests provide information only about the individual tested; they identify conditions that occur often enough, or are serious enough, to justify testing all asymptomatic individuals in a defined population (e.g., every adult over age 50 years for colorectal cancer).(1) The tests usually involve sampling or imaging the target tissue or organ. Some cancer screening tests, such as colonoscopy and Pap smears, can lead to the identification and removal of premalignant lesions, thereby reducing the patient's risk of developing invasive cancer. But more often, these and other screening tests, such as mammography, do not prevent the development of the disease; they serve only to detect early stage disease, when treatment is more likely to prevent death.

In contrast, genetic tests can identify cancer-related germline mutations in cells from any conveniently sampled tissue (e.g., blood) in disease-free individuals or in patients with the disease who have unaffected family members who may be at risk and also can be tested. Disease-free individuals who test positive for cancer-related mutations are candidates for aggressive primary preventive measures.(3,4) To the extent that such measures are available and effective, genetic testing can prevent people at high risk for cancer from developing the disease. In addition, genetic tests need to be done only once per individual.

Germline mutations that cause significant morbidity and mortality at a relatively young age are generally quite rare in the population for obvious evolutionary reasons. Tests for cancer-related germline mutations should therefore not be used in the general population; candidates for testing should meet specific guidelines, usually involving family or personal medical history or membership in an ethnic group in which the mutation is known to be common.(5) Sensitivity, specificity, and predictive value, the common criteria by which screening tests are evaluated, are relevant to genetic screening but have a special meaning in the genetic context.

The predictive value of a positive nongenetic test is the probability that an individual with a positive test result will have the disease.(1) Strictly speaking, the predictive value of a positive genetic test is the probability that an individual who tests positive will have the mutation.(1,6) The effects of penetrance, or the probability that an individual who tests positive will develop the disease, are shown in Figure 1 as it relates to breast cancer.

Ideally, randomized clinical trials or observational studies would be in progress now to assess the costs, quality of life, and survival associated with genetic testing and the use of preventive strategies by those who tested positive. A few observational studies are in progress, but their results are not yet available. Meanwhile, patients, physicians, and policy makers still need to make decisions about testing.

Familial cancer risk is a sensitive issue. In our preference rating survey, individuals contemplating genetic testing were less concerned about their personal risk for cancer than about the possible increased risk for their children.(7) An individual who tests positive for a cancer-related germline mutation has a moral obligation to notify at-risk relatives, but it is not pleasant to have to tell one's relatives that they may be at risk for a serious disease because of a germline gene mutation. Family members often differ in their willingness to find out whether they have the mutation or to share that information with others. Genetic counseling can help individuals and families to deal with these issues and may be essential to the success of a genetic screening program.(8-10)

The future holds the promise of rapid, inexpensive, and accurate tests that may empower individuals to make preventive treatment decisions that enhance their lives. However, even if genetic tests become very inexpensive, they will not be appropriate for general population screening. If genetic screening policy is based on sound epidemiological principles, rather than on consumer demand and third-party willingness to pay, it holds the potential to enhance the public health and the public good.