Simple Screening Tool Identifies Risk of BRCA Mutations

ROCKFORD, Ill., Sept. 11 -- Women who might benefit from tamoxifen chemoprevention because of an increased risk of hereditary breast cancer can be identified by a five-item assessment at a mammography clinic, suggested researchers here.

A score of eight or higher on the pedigree assessment tool had a sensitivity of 100% and a specificity of 93% for identifying women at risk for BRCA1 or BRCA 2 breast cancer susceptibility mutations, reported Kent F. Hoskins, M.D., of Saint Anthony Center for Cancer Care here, and colleagues, online in Cancer.

Moreover, the pedigree assessment tool score performed better than the Gail model, a tool that is commonly used for calculating individualized risk estimates to identify women for tamoxifen therapy, Dr. Hoskins and colleagues said. Their findings will appear in the Oct. 15 issue of the journal.

The pedigree assessment tool system assigns arbitrary numerical values to pedigree findings. For example, a family history of breast cancer after age 50 is scored at three points, while cancer younger than 50 receives four points. The score for a male relative with breast cancer is eight, Ashkenazi Jewish heritage adds four points and ovarian cancer at any age tallies five points.

The points are added up for both the maternal and paternal pedigrees and the woman's pedigree assessment tool score is the higher of the two scores.

Dr. Hoskins and colleagues tested the pedigree assessment tool assessment with 3,906 women who had screening mammography at a community hospital. The women had no personal history of breast cancer.

The researchers used a risk assessment model available online, known as the Frank model, that relies on prevalence data to identify women at risk for BRCA mutations, with high risk defined as a 10% or greater risk of BRCA. Using that model, 86 women were identified as a high BRCA possibility.

Dr. Hoskins then analyzed the ability of the PAT score and the Gail model to identify these same 86 women.

The tests were compared by plotting a receiver operating characteristic (ROC) curve for the pedigree assessment tool test and two Gail tests-the five year estimate and the lifetime estimate. The ideal area under the ROC curve would equal 1.0, so the test that comes closest to that ideal is the superior test.

The area under the ROC curve for pedigree assessment tool was 0.9625 versus 0.389 for the Gail 5-year estimate and 0.5861 for the Gail lifetime estimate.

The Gail model was not designed as a specific assessment for hereditary risk. But the authors said that other assessment tools designed specifically for that purpose are not suitable for use in a population-based screening setting such as a mammography clinic. The Gail model, however, works well in that setting, so it was selected as the comparator arm.

Another potential limitation was the authors' decision to use the Frank model as a gold standard against which they measured pedigree assessment tool and the Gail models. But Dr. Hoskins pointed out that the Frank model has the largest published data set, and it used the most accurate genetic testing technique. The Frank population was "most similar to our population."

The authors noted that the FDA-approved recommendation for tamoxifen chemoprevention is based on a Gail risk estimation.

But they conclude that combining the Gail model with a tool like the pedigree assessment tool would improve comprehensive breast cancer screening. It could "serve a triage function to identify women at moderate levels of risk who do not require a formal genetic evaluation, but would be candidates for chemoprevention."