Genetic Testing Guidelines "Not Helpful" in Breast Cancer, Study Shows
Three-quarters of patients who did not meet the guidelines for testing had a variant for which established clinical management recommendations were available.
Genomic sequencing in patients based on The National Comprehensive Cancer Network (NCCN) genetic testing guidelines may result in the underdiagnosis of breast cancer in patients with clinically actionable germline variants, according to a study reported at the 2018 San Antonio Breast Cancer Symposium (SABCS), Texas, and concurrently published in the Journal of Clinical Oncology.1,2
Currently, the NCCN guidelines limit which breast cancer patients should be tested on the basis of risk factors, such as family history. The study findings, however, suggest it may be time to update the NCCN guidelines to recommend genetic testing for all patients with breast cancer. Guidelines for certain other cancer subtypes, such as metastatic prostate cancer, have already been changed to recommend testing for all patients.
“All women with breast cancer should be genetically tested,” lead study author Peter Beitsch, MD, cofounder of Targeted Medical Education Breast Care Network, Allentown, Pennsylvania, told Cancer Therapy Advisor. He explained, “The guidelines are not helpful at distinguishing women [with breast cancer] who have pathogenic variants from those who do not.”
The study enrolled patients with breast cancer from 20 academic and community breast cancer practices between April 2017 and September 2018. Patients who had received prior genetic testing were not eligible for study enrollment. A total of 959 patients were included in the study cohort, of which 479 (49.95%) met genetic testing guidelines and 480 (50.05%) did not. All patients underwent genetic testing with an 80-gene panel from Invitae.
Nearly all of the patients in the cohort were women, with the exception of 2 men in the group that met guidelines and 1 man in the group that did not. About two-thirds of the patients in study cohort were considered recently diagnosed (ie, having a diagnosis within 1 year of study consent) and the remaining one-third were not recently diagnosed.
The 80-gene panel detected a variant of unknown significance in about half of individuals enrolled in the study, and a pathogenic or likely pathogenic variant was detected in 8.65% (83 of 959 of individuals). The prevalence was similar across groups, with 9.39% of patients who met guidelines and 7.9% of patients who did not meet guidelines having a pathogenic or likely pathogenic variant (P = .4241).
Of the 83 patients who had a pathogenic or likely pathogenic variant, 45 were from the guidelines group and 38 were from the nonguidelines group. Further, three-quarters of the 38 patients who did not meet the guidelines for testing had a variant for which established clinical management recommendations were available.
“The guidelines aren't nearly as successful in capturing all the patients with hereditary cancer risk as we imagined,” said Harold Burstein, MD, PhD, Dana-Farber Cancer Institute, Boston, Massachusetts, during an interview with Cancer Therapy Advisor. He was not involved in the study. “We've probably been too narrow in how we've thought about which patients need genetic testing and which do not.”