We are entering exciting times when it comes to the impact of genetics on the diagnosis and treatment of cancer.
However, my colleagues’ discussions over the past 24 hours about a new test that claims to predict a woman’s risk of breast cancer have me thinking these exciting times may be more like the Old Wild West where “anything goes” than some meaningful movement forward in our battle against cancer.
The stated goal of the test is to enable a woman to predict her lifetime risk of getting cancer. The accompanying material from the company liberally quotes the American Cancer Society and the American Society of Clinical Oncology as references to support the value of the test.
The company says that this test—which is an analysis of genetic code obtained from cells that are either swabbed from the mouth or taken with a blood sample and then analyzed in the company’s laboratory—measures seven genetic changes that “contribute to the incidence of an estimated 60 per cent of all breast cancer cases.”
The test is based on “studies published in major peer-reviewed journals and involving nearly 100,000 breast cancer patients and healthy volunteers, principally of European descent.” The company presents several recently published journal articles and their own proprietary unpublished data as support for the science behind the test.
If you do this test (which costs $1625; please see correction posted 10/10/08), you will receive a calculated estimate of your risk of developing breast cancer. This estimate apparently is obtained by multiplying the increased risk values found in the research studies for the particular genetic changes.
The problem is, no one to our knowledge has ever shown that multiplying all of these newly discovered genetic risk factors together really gets you to a number that has real meaning.
Why should we be concerned about this?
There is a lot we know about breast cancer risk, and there is a lot we don’t know.
There are genetic tests widely available in this country which measure a change in the genes of certain women called BRCA 1 and BRCA 2. These genes, when present, do predict for certain women that they have a very high risk of developing breast and/or ovarian cancer during their lifetimes. For the vast majority of women, however, we don’t have any specific genetic tests to pinpoint their individual risk.
In general, we are left to rely on other factors such as family history, age of menarche, age at first pregnancy, prior breast biopsies, and similar information to give us a gauge of their risk.
Once we know their risk of breast cancer, we can recommend strategies to reduce that risk. For women who are BRCA positive, that includes preventive mastectomies and oophorectomies, for example. For other women, it could mean treatment with tamoxifen or raloxifene.
Along comes this new genetic test making certain claims, and we start to wonder what does it mean?
First, you should know that the test has not been approved by the Food and Drug Administration, according to the company's consent form for the test.
Actually, for laboratory tests, that is not unusual since a single laboratory that develops a test is allowed to market it without FDA approval. However, the FDA is particularly concerned about the burgeoning numbers of genetic tests, how they are performed, and whether they in fact produce the types of results they claim under this so-called “home brew” exemption.
In addition, for this particular test, there are several states that do not allow it according to the informed consent provided by the company. Those states include California, Maryland, New York, and Pennsylvania.
Most importantly, according to my American Cancer Society colleagues, they are not aware of any research that has looked directly at this particular test as a forward-looking assessment of the risk of an individual woman developing breast cancer. They don’t know whether or not it really measures the risk it is supposed to measure. Doing such studies takes time, and costs money (which may be in short supply for this company according to public documents).
Simply stated, this lab has taken advantage of published research, said that they can measure the presence or absence of a particular genetic change in a woman’s DNA as reported in that research, multiplied the risk factors together, and come up with a number.
Whether or not that particular approach translates into real life risk, in the opinion of my colleagues, is uncertain.
And, there is no published data on the test as to whether or not the test itself accurately measures these genetic changes.
But what concerns us the most is the link between our guideline recommendations for MRI breast cancer screening and the test’s promotional materials.
The American Cancer Society and its advisors have not evaluated this test, nor are our guideline recommendations regarding preventive treatment or breast cancer screening with MRI based on the results of this test. It is not appropriate to link the results of this test with any recommendation from the American Cancer Society.
Our guideline recommendations are based on what we call “clinical factors” related to risk, not this particular test. In our opinion, to cross-link this test with our guidelines is not appropriate in the absence of validated clinical research.
I could go on about our concerns, but you would probably become even more confused that you are already. Even I have difficulty trying to sort out the data and the recommendations made by the company’s promotional literature. When my colleagues start to talk about the blurring of relative and absolute risk even my eyes start to glaze over.
I would emphasize that the American Cancer Society encourages research and development of biomarker tests for breast and other cancers. It is our recommendation that these tests must be studied carefully before being offered to the public. They should undergo validation studies and demonstrate their value in the clinic.
We don’t see the evidence that that is the case in this particular situation.
Like I said, we hold out great hope for the role of genetics in the diagnosis and treatment of cancer. We know that the future holds great promise for these types of tests.
The question for today, however, is whether some of these tests that are coming to market really advance our cause in the fight to reduce deaths from cancer in a meaningful, evidence-based and scientifically accurate way.
For now, with respect to this particular test, our opinion is—unfortunately—not a positive one.
