Linda Bird Randolph
PENN Health Magazine
Last Modified: November 1, 2001
These are confusing times for women who are concerned about their chances of developing breast cancer. Now that they have relatively easy access to genetic tests, many women -- especially those with a family history of cancer -- are trying to decide whether they should be tested. But which groups of women should have the test is not clear. In addition, once a test is done, it is difficult for physicians to interpret the result and to make recommendations to patients, because the research community is constantly uncovering more information on this relatively new discovery.
To help women sift through some of the conflicting reports on genetic testing for breast cancer, and to provide some food for thought, Mildred Cho, Ph.D., assistant professor at Penn's Center for Bioethics, gave a presentation on "The Ethical and Social Implication of Genetic Breast Cancer Testing." The talk was a part of a well-attended, day-long symposium called "Life After Breast Cancer," held on April 29. The object of the program -- sponsored by the University of Pennsylvania Cancer Center and now in its sixth year -- is to provide information and support to survivors of breast cancer and their loved ones.
Genetic testing for hereditary breast and ovarian cancer has raised numerous ethical and social issues, Cho said. These tests can be used on women without symptoms to detect predispositions for cancer that might occur well into the future -- or not at all. And although tests can reveal mutations, they do not tell the whole story or take into account other possible contributing factors, such as lifestyle, family history, or other genes that might come into play.
Because information about test performance and the efficacy of treatment is still incomplete, experts generally agree that, currently, it is only appropriate to screen women who come from high-risk families. Yet tests are available to women outside the research protocol, and commercial companies are marketing the tests to women who may not be at high risk.
Cho emphasized that women considering whether to be tested must take into account the potential risks and lack of information. According to Cho, before anyone takes any type of medical test, he or she should know certain things: how accurate and predictive a test is; what the possibility is of false positives or negatives; and whether there is anything the patient can do to prevent or treat the disease.
For the time being, women have to make their decisions without all of the answers. "One of the reasons for this limited amount of information," Cho said, "is the difficulty of getting data about the performance of tests that detect late-onset conditions." For such data, she explained, "you'd have to follow up people for a long time. And we just haven't had that much time since the development of the test."
Perhaps the most important thing that women should know before testing for BRCA1 or BRCA2, the mutated genes, is that a positive test does not mean a patient will definitely get cancer. "There are false positives and false negatives," Cho said. "In fact, BRCA1 and BRCA2 are not, as diagnostic tests go, terribly accurate."
According to Cho, most articles in the medical literature and the lay press about BRCA1 say that the risk of breast cancer with a positive genetic test is 85 percent. But, she said, there is recent evidence that this risk figure may actually be much lower in individuals who do not have a very strong family history of cancer. (See editor's note.)
Cho asked the audience to bear in mind that disease-associated genes are usually identified first in families that are at very high risk. With BRCA1 and BRCA2, she said, "many of the women who were studied initially came from families in which multiple relatives have breast or ovarian cancer. The present data on the accuracy and predictive value of the tests comes mostly from these very-high-risk populations. Once a gene is identified that is associated with a disease, it takes a long time and many samples from lower-risk populations to determine whether the test performs as well in those situations."
In general, genetic tests are less predictive in populations with a lower incidence or lower severity of disease. "Without systematic and long-term data collection," Cho said, "we'll not know what the predictive value of BRCA1 and BRCA2 testing really is in actual clinical use if it's used in people who do not have a strong family history of the disease, and that's generally not considered appropriate right now."
Not only does it appear that genetic tests may have varying significance in higher- and lower-risk populations, Cho said, but the data on the efficacy of preventive measures and treatments may not apply equally to those with strong or weak family histories of cancer. "Many studies on prevention such as prophylactic mastectomies or mammography were not necessarily designed to take into account factors relevant to hereditary breast and ovarian cancer, which may have a very different kind of mechanism," she explained. "For example, while an increased frequency of mammography may sound like a good idea for people who test positive for BRCA1 and BRCA2, there is the question about whether that is a good idea if this group of people is potentially more likely to be sensitive to radiation damage."
At present, said Cho, regulation of genetic testing is minimal, which contributes to the lack of pooled information. The Food and Drug Administration does not regulate the tests; and although the Health Care Financing Administration regulates laboratory quality, it does not determine whether a diagnostic test is safe or effective.
The long consent forms and dense brochures that many genetic testing programs have developed "may deliver too much information," said Cho. In her view, these written "informed consent" materials should be supplemented with one-on-one discussions with the health care provider. But studies indicate that these talks do not always happen. Cho told of a recently published study on commercial genetic testing for familial colon cancer that showed that only 18.6 percent of patients received genetic counseling before they took the test. Only 16.9 percent of patients provided written informed consent before testing. "Furthermore," Cho added, "nearly a third of the physicians ordering the test misinterpreted the test results."
According to Cho, patients have to keep in mind that knowing their medical family tree could have unexpected consequences. For some patients, finding out that they have the mutation could be "toxic knowledge" that might affect their outlook on life. "When you're evaluating the ethics of a genetic test, the medical risks and benefits have to be placed in the larger context of social and psychological risks." One risk she cited was job discrimination if an employer knows an applicant's medical history. And women who test positive but who do not yet have the disease -- and who may never develop it -- face the same possibility of social and psychological disturbances and discrimination.
Cho referred to a study published last year in Science magazine that surveyed 332 members of support groups for genetic disorders, not necessarily for breast cancer: 25 percent of the respondents or their affected family members believed they had been refused life insurance because of their genetic disorder. In addition, 22 percent believed that they were refused health insurance and 13 percent believe that they were denied a job on the same basis.
Another study Cho cited showed that some women have unexpected responses to their test results. Some women may find relief or empowerment from their results, whether positive or negative. However, a small but significant proportion may actually become more anxious or depressed after either a positive or a negative test result.
As Cho put it, "The onus -- for better or for worse -- is on patients to be alert and proactive about getting their own information about testing." She emphasized that patients must keep in mind that genetic testing for breast cancer is voluntary and that each patient has the right to decide whether to test, despite any pressure from their health care providers. "And practitioners," Cho added, "need to be aware of the potential downside of testing and to be educated on how to counsel and interpret test results. Or know when to refer to specialists who have this expertise already."
Editor's note: Shortly after Cho's talk, a Penn research team led by Barbara Weber, MD, associate professor of medicine and genetics, published findings in The New England Journal of Medicine that showed that BRCA1 mutations were found in only 16 percent of women with a family history of breast cancer. As the study notes, the percentage is "less than the 45 percent predicted by genetic-linkage analysis." The figure drops to 7 percent in the case of women from families with a history of breast cancer but not ovarian cancer. Among women with family histories of both breast and ovarian cancer, the risk of having BRCA1 mutations was higher. For example, "BRCA1 mutations were identified in . . . 18 of 45 families (40 percent) with members with both breast and ovarian cancer." Weber's study concluded: "Even in a referral clinic specializing in screening women from high-risk families, the majority of tests for BRCA1 mutations will be negative and therefor uninformative."