麻豆传媒

Women who did not meet breast cancer genetic testing criteria had cancer-associated mutations just as often as women who met the criteria, two separate studies showed.

A study involving 713 patients with newly diagnosed breast cancer showed pathogenic variant rates of 8.6% among women who did not meet National Comprehensive Cancer Network (NCCN) criteria for genetic testing, and 10.2% among those who met the NCCN criteria, a nonsignificant difference.

A study of 4,200 Medicare patients referred for genetic testing showed genetic variants in 10.5% of patients who met Medicare testing criteria versus 9.0% of the women who did not meet the criteria for testing, also a difference that did not reach statistical significance.

"Economically based guidelines for genetic testing of breast cancer patients are anachronistic, miss a significant number of pathogenic variants, and we believe they should be abolished immediately," Peter Beitsch, MD, of TME Breast Care Network in Dallas, said during a press briefing at the meeting in Orlando, Florida.

Current guidelines for genetic testing are based on historically severe clinical presentation of high-penetrant genes and variants, such as BRCA 1 and 2, said genetic counselor Jennifer Axilbund, who reported the study of Medicare patients.

"These criteria do not adequately reflect evolving genetic knowledge, and they miss patients who should have increased cancer screening or should receive different systemic therapy," said Axilbund, of Invitae, a genetic testing company in San Francisco. "Testing criteria should be broadened to include unaffected individuals with a positive family history and affected individuals with less severe presentation."

Judy Boughey, MD, of the Mayo Clinic in Rochester, Minnesota, said the two studies emphasize the need for discussions about genetic testing with patients who have breast cancer.

"With the advances in panel testing for predisposition genes, the likelihood of identifying a pathogenic or likely pathogenic mutation are higher," said Boughey, who was not involved in the studies. "However, we must remember that not all mutations are medically actionable. When evaluating genetic testing, we should consider the likelihood of identification of a mutation, cost of testing, and potential impact of the result on medical management."

In discussing the background of his study, Beitsch said current guidelines for genetic testing for breast cancer as economic constraints, designed to limit use of early-generation tests (primarily for BRCA1/2)that cost about $5,000. New-generation testing, though more comprehensive, costs considerably less, on the order of a few hundred dollars. Moreover, the criteria established a penetrance threshold for variants of 10% for a test to be cost effective.

Beitsch's group hypothesized that genetic testing of all patients with breast cancer would identify significantly more affected patients as compared with guideline-driven testing. To test the hypothesis, investigators in the Universal Breast Cancer Genetic Registry enrolled patients with newly diagnosed breast cancer, and evaluated each patient by means of an 80-gene test panel. The NCCN test criteria specify testing for 11 genes known to be associated with breast cancer.

The study has an accrual goal of 1,000 patients, 500 who met NCCN testing criteria and 500 who did not. Beitsch reported findings for the first 713 patients with complete data, comprising 353 patients who met NCCN testing guidelines and 360 who did not. The initial results showed that a similar proportion of patients in each group had definitely or likely pathogenic variants.

Comparison of the variants identified in the two groups of patients showed a predominance of BRCA1/2 and other common variants among women who met NCCN testing criteria, whereas women who did not meet the criteria had a larger proportion with less common variants that nonetheless have guidelines for clinical management. Patients who did not meet testing criteria also had a higher proportion of variants of unknown significance (VUS).

Had the nonguideline-compliant patients been tested only for the 11 mutations recommended by NCCN, about half of the variants detected by the 80-gene panel would have been missed.

Axilbund reported findings from a cohort of Medicare patients for whom genetic testing had been requested, including BRCA1/2. All the tests involved expanded panels, ranging from nine to 80 genes (median of 37). In each case, the provider ordering the test indicated whether the patient met testing criteria.

Medicare criteria for genetic testing focus on high-risk genes and require the patient to have a cancer diagnosis. For their analysis, Axilbund's group expanded the inclusion criteria to include patients with a personal or family history of breast or gynecologic cancer. They excluded patients with a family history of pathogenic or likely pathogenic variants.

The study population consisted of 3,549 patients who met testing criteria and 647 who did not. The results showed that a higher proportion of patients who met the criteria tested positive for BRCA1/2 (3.2% vs 1.9%). The 1.75 odds ratio for the in-criteria group still did not achieve statistical significance, said Axilbund.

The overall proportion of patients with variants detected in the two groups also did not differ significantly (P=0.26). The proportion of patients with VUS was similar between the two groups.

In the out-of-criteria group, the most commonly detected variants were CHEK2 and BRCA2, both of which are covered by guideline recommendations related to clinical management, Axilbund noted. Use of Medicare criteria to select patients for genetic testing would miss about half of all patients with actionable variants, she added.

Comment