A new test can predict both the risk of breast cancer recurrence and may identify women who will benefit most from chemotherapy, according to research supported by the National Cancer Institute (NCI), part of the National Institutes of Health, and performed in collaboration with the National Surgical Adjuvant Breast and Bowel Project (NSABP) and Genomic Health Inc. These results suggest that almost half of 43,000 U.S. women diagnosed with estrogen-dependent, lymph-node negative breast cancer every year are at low risk for recurrence and may not need to go through the discomfort and side effects of chemotherapy.

The test is based on levels of expression (increased or decreased) of a panel of cancer-related genes. This panel is used to predict whether estrogen-dependent breast cancer will come back, according to a study that will be published online in the New England Journal of Medicine on Friday, December 10, 2004*. Scientists on this study also will present new results on that day at San Antonio Breast Cancer Symposium indicating that the same test can predict which women benefit most from chemotherapy. Women with low risk of breast cancer recurrence-about half of the women in the recent study-do not appear to derive much benefit from chemotherapy.

The researchers used tissue samples and medical records from women enrolled in clinical trials of the cancer drug tamoxifen, which blocks the effect of estrogen on breast cancer cells. These women had a kind of breast cancer defined as estrogen receptor-positive, lymph node-negative. Each year, 43,000 women are diagnosed with this kind of breast cancer, which needs estrogen to grow but has not spread to the lymph nodes. Currently, many women with this type of breast cancer in the United States do receive chemotherapy in addition to hormonal therapy.

Using samples from 447 patients and a collection of 250 genes in three independent preliminary studies, 16 cancer-related genes were found that worked best. The scientists created a formula that generates a "recurrence score" based on the expression patterns of these genes in a tumor sample. Ranging from 1 to 100, the recurrence score is a measure of the risk that a given cancer will recur**.

Prior to this research, analysis of the expression of genes was performed on tumor specimens that were frozen rather than tissue prepared for routine pathologic evaluation (fixed and embedded). The expression analysis depended on measurement of RNA (the molecule necessary for the translation of a gene into a protein), and RNA is altered when tissues are fixed and embedded. Frozen tissues are generally not readily available in routine practice. Researchers at Genomic Health, Inc. developed a method for performing these analyses on tissues embedded in paraffin wax. Their method allows them to use the altered RNA that is found in fixed tissue.

The results published in the New England Journal of Medicine validate the ability of the recurrence score to predict risk of recurrence. Using biopsy tissue and medical records from another NSABP tamoxifen trial, researchers divided 668 women into low, intermediate, and high risk of recurrence groups. Fifty-one percent were in the low risk group (with a score of less than 18); 22 percent were at intermediate risk (recurrence score 18 or higher but less than 31); 27 percent were at high risk (a score of 31 or higher).

These risk group divisions correlated well with the actual rates of recurrence of breast cancer after 10 years. There was a significant difference in recurrence rates between women in the low and high risk groups. In the low risk group, there was a 6.8 percent rate of recurrence at 10 years; in the intermediate and high risk categories these rates were 14.3 and 30.5 percent, respectively. Up to a recurrence score of 50, rates of recurrence increased continuously as the recurrence score increased. These trends held across age groups and tumor size.