Genetic Risk Models Miss Many BRCA Mutation Carriers from Small Families

NEW YORK JUN 19, 2007 (Reuters Health) - Among women with early-onset breast cancer, those with few surviving female relatives are at greater risk of harboring deleterious BRCA gene mutations than those from larger families, according to a report in the Journal of the American Medical Association for June 20. 

The research team has found that models used for cancer risk counseling tend to underestimate risk among single cases of early onset breast cancer.

Beginning in 1997, Dr. Jeffrey N. Weitzel and associates at the City of Hope in Duarte, California, set up a prospective hereditary cancer registry.  Registrants" family history was sought for three to five previous generations. 

Among 1,543 women who underwent genetic cancer risk assessment and BRCA gene testing, 306 (19.8%) had breast cancer before they were 50 years old and had no first- or second-degree relatives with breast or ovarian cancers.  Thus, the ability to predict their risk was compromised.

The team also analyzed risk among those with "limited family structure," defined as fewer than two first- or second-degree female relatives surviving beyond age 45 years in either lineage.  Half of all of the "single cases" had a limited family structure.

BRCA gene mutations were detected in 13.7% women with a limited family structure and in 5.2% of those with adequate family structure (odds ratio 2.8, p = 0.02).

"This practical observation is consistent with autosomal dominant inheritance of a trait with sex-limited expression," they write.

Three predictive models widely used in cancer risk counseling clinics overestimated the likelihood of a mutation among subjects with an adequate family structure and underestimated that of subjects with a limited family structure.

"Our findings support BRCA gene mutation testing for women with early onset breast cancer when the results will influence medical management, regardless of mutation estimates from existing models," the research team advises.  They also recommend that family structure should be incorporated as an actual variable in probability models.

In light of these limitations, Dr. Noah D. Kauff and Dr. Kenneth Offit, from Memorial Sloan-Kettering Cancer Center in New York, caution that risk assessment models represent only one aspect of the process of determining a patient"s risk for deleterious mutations.

They note that clinical factors, such as pathologic and immunohistochemical features of tumors, and "consideration that breast cancer risk may be conferred by an as yet unidentified cancer susceptibility gene," should be included in the risk assessment.

Moreover, the editorialists add, it may not be appropriate to use currently available models "to triage individuals for genetic testing or...to provide cancer risk prediction and guidance of care in the absence of genetic testing."

SoURCE:

• JAMA 2007;297:2637-2639,2587-2595.