[Skip to Navigation]
Sign In
Comment & Response
December 2018

Cancer Risk Estimates for Study of Multiple-Gene Testing After Diagnosis of Breast Cancer—Reply

Author Affiliations
  • 1Department of Medicine, Stanford University, Stanford, California
  • 2Department of Health Research and Policy, Stanford University, Stanford, California
  • 3Department of Health Management and Policy, School of Public Health, University of Michigan, Ann Arbor
  • 4Division of General Medicine, Department of Internal Medicine, University of Michigan, Ann Arbor
JAMA Oncol. 2018;4(12):1788. doi:10.1001/jamaoncol.2018.4959

In Reply Our results demonstrate that multiple-gene panel testing has largely replaced BRCA1/2-only testing in the United States.1 We agree that this marked surge in multiple-gene testing may be in advance of the evidence supporting the clinical use of more vs less comprehensive genetic testing, and we have commented on this problem in our prior work.2 Thus, there is need for studies that will address the consequences of this sea change on the evaluation and management of cancer risk. Our recent results addressed 1 aspect of potential harm from multiple-gene panel testing vs BRCA1/2-only testing: the possibility of unwarranted use of more extensive surgery (bilateral mastectomy) owing to finding more variants of uncertain significance or pathogenic variants in genes whose risks are less well characterized than those of BRCA1/2. We found that bilateral mastectomy rates in patients with variants of uncertain significance or with pathogenic variants in non-BRCA1/2 genes were no higher than those in patients with negative test results,1 which provides some reassurance. We and others have reported increasing use of bilateral mastectomy and that patient preference plays a role in this trend.3 However, our recent results offer no evidence that testing more vs fewer genes fuels patient preference for more extensive breast surgery. The suggested analysis (ie, generating specific risk estimates for contralateral breast cancer for the entire study population) is infeasible; validated breast cancer risk models, such as the Tyrer-Cuzick model, are indicated only for women who have never been previously diagnosed with breast cancer,4 and considerable uncertainty remains about the contralateral breast cancer risks associated with pathogenic variants in breast cancer susceptibility genes other than BRCA1/2.

Add or change institution