washingtonpost.com
Firm Says Test Judges Risk For Common Breast Cancers

By Rob Stein
Washington Post Staff Writer
Wednesday, October 8, 2008

A biotech company today will begin offering the first genetic test to assess a woman's risk for the most common forms of breast cancer, reigniting debate about the growing number of unregulated genetic tests.

The test by Decode Genetics of Reykjavik, Iceland, a respected pioneer in genetic research, promises to determine a woman's risk through a simple blood sample or cheek swab. Previously, the only tests for breast cancer risk were for relatively rare genes, leaving most women with no way to assess their individual genetic predisposition.

"What this does for women is allow them to assess their personal risk for the common forms of breast cancer," said Kári Stefánsson, Decode's chief executive. "That's what you need to do to make early diagnoses or take preventive measures. This test will most definitely save lives."

While welcomed by some patient advocates and doctors, the $1,625 test raises concerns among others. Some questioned its reliability, while others worried that the results could either lull women into a false sense of complacency or needlessly alarm them, prompting them to take unnecessary tests or even undergo unneeded surgery.

"There is at least a significant chance this test will could falsely reassure some women and alarm others," said Eric Winer, a breast cancer expert at Harvard Medical School and a spokesman for the American Society of Clinical Oncology. "I fear for many women the results could be quite misleading."

"I wouldn't recommend to anyone that she have such a test. I certainly wouldn't want my daughter to have such a test," said Mary-Claire King, a geneticist and breast cancer expert at the University of Washington. "It's meaningless, and it could very easily introduce real confusion."

The test comes as concern has been rising about the proliferation of genetic tests. Hundreds of laboratories are now offering genetic testing for more than 1,200 conditions. The Food and Drug Administration does not regulate such tests when they are performed by the labs offering them.

"There is concern about the validity of many genetic tests that are being offered," said Joan Scott, deputy director of the Center for Genetics and Public Policy at Johns Hopkins University. "Without an external review, the consumer -- be it the physician or the patient directly -- is not going to know which ones have been validated."

An FDA spokeswoman said the agency had just become aware of the new test and could not say whether it would take any action, though she noted that in August the agency sent a letter expressing concern to another company marketing a questionable genetic test for ovarian cancer.

"I can't speculate on this one," said Karen Riley, the spokeswoman.

Breast cancer strikes more than 180,000 women each year in the United States and kills more than 40,000, making it the most common form of cancer after skin cancer and second leading cancer killer among women.

Doctors can estimate a woman's chances of developing breast cancer based on a variety of factors, including her age and family history, but most women have no way of knowing their precise risk. Women with a strong family history can get tested for two genes that sharply increase their predisposition, but they account for only 1 to 3 percent of all breast cancers.

The new test looks for seven subtle genetic variations (known as single nucleotide polymorphisms or SNPs) that appear to be associated with increased risk, based on studies by Decode and other researchers in the last few years. None of the variations alone is responsible for a significant proportion of breast cancers, but together they account for about half the cases in the United States and other industrialized countries, Stefánsson said.

Although insurance companies do not cover the new test, Stefánsson said he expected many eventually would. Any doctor can order the test and send either a blood sample or a swab of the inside of the cheek containing DNA to the company for analysis. Results will arrive within about three weeks.

Studies indicate the test results will range from telling a woman she has less than half the average lifetime risk for breast cancer to as much as four times the average risk, which the American Cancer Society estimates at 12 percent, Stefánsson said.

"The question is, 'What can you do as a woman if you undergo this test and you turn out to be in the category of highest risk?' " Stefánsson said. "There are several things you can do."

For example, the American Cancer Society recommends women whose risk is 20 percent or more should receive annual MRI screenings in addition to mammograms, he said. Because the test predicts which women are more likely to develop breast cancer fueled by the hormone estrogen if they do develop cancer, the results could also encourage them to take drugs, such as tamoxifen, that block estrogen. Such women may also be more motivated to make positive lifestyle changes, Stefánsson and others said.

"A lot of women are afraid of breast cancer. They just don't know what their risk is," said Kay Wissmann of the Breast Cancer Network of Strength, a Chicago-based advocacy group. "For those women who choose it, this test could provide information that could potentially help women make better decisions. It could empower them."

Because the underlying research has primarily involved women of European descent, it remains unclear how accurate it would be for women of other backgrounds. The company is studying that, Stefánsson said.

King and others, however, questioned whether the markers used in the test have been definitively linked to an increased risk for breast cancer.

"The idea of introducing into medical practical a test whose predictive value is completely unvetted biologically is frightening," she said.

The test should be subjected to much more testing before being made widely available, Winer agreed. In addition to validating its predictive value, studies need to be done to determine whether women whose risk is elevated would benefit from additional screening or drugs such as tamoxifen.

"Women told on the basis of this test that they have a lower-than-average risk may feel they don't have to worry about breast cancer and can forgo standard screening tests. We don't know that's the case," Winer said. "For the women who find out they are at increased risk, I am worried there may be a great deal of misinterpretation and potentially unnecessary alarm."

Debbie Saslow of the American Cancer Society said, "Women should know this test is not recommended by any organization."

But Stefánsson dismissed such criticism, arguing that the test had been validated by studies conducted by the company and others involving nearly 100,000 breast cancer patients and healthy volunteers from many countries.

"This has been better clinically validated than any test than I know," Stefánsson said. "We have extraordinarily good statisticians who have probably the best track record in human genetics today."

View all comments that have been posted about this article.

© 2008 The Washington Post Company