Among more than 23,000 women without breast cancer aged 40 to 74 years who underwent criteria-independent genetic testing, nearly 3% carried pathogenic or likely pathogenic variants (PVs) in breast cancer susceptibility genes—and 30% of these carriers would not have qualified for testing under current family history-based guidelines, wrote lead study author Kirkpatrick B. Fergus, MD, of the Cancer Genetics and Prevention Program at University of California, San Francisco, and colleagues.

The findings from this secondary analysis of the Women Informed to Screen Depending on Measures of Risk (WISDOM) randomized clinical trial, published December 12 in JAMA Internal Medicine, represent one of the first large-scale efforts to integrate germline genetic testing into personalized breast cancer screening independent of family history requirements. The WISDOM trial was designed to compare annual screening mammography with personalized risk-based screening, integrating genetic testing as part of the risk assessment strategy. The trial enrolled 30,192 women to the personalized screening arm between August 2016 and February 2023.

Among 23,098 women who completed germline testing for 9 breast cancer susceptibility genes (BRCA1, BRCA2, ATM, CHEK2, PALB2, CDH1, PTEN, STK11, and TP53), 605 women (2.6%) carried a previously undetected PV—higher than previous population-based estimates, though the WISDOM trial population was enriched for family history risk factors.

Pathogenic variants were most common in CHEK2 (337 participants, 1.5%) and ATM (101 participants, 0.4%) but less common in higher-penetrance genes: BRCA1 (33 participants, 0.1%), BRCA2 (82 participants, 0.4%), and PALB2 (44 participants, 0.2%).

The study population showed considerable racial and ethnic diversity: 9% Hispanic (any race), less than 1% American Indian or Alaska Native, Native Hawaiian, or Pacific Islander (combined due to low numbers), 5% Asian, 4% Black, 77% White, and 4% multiracial or other/unknown race. Additionally, 15% reported Jewish ancestry.

Family History Patterns

Notably, 180 of 605 women with PVs (30%) did not report a first-degree or second-degree family member with a history of breast or ovarian cancer, male relative with breast cancer, or Jewish ancestry—the primary criteria that currently determine eligibility for genetic testing.

The prevalence of PVs among women with no family history of breast or ovarian cancer was 2% overall and 2% for high-penetrance and moderate-penetrance genes. In comparison, women with at least 2 female relatives with breast cancer had a PV prevalence of 3%, with 2% having a variant in a high-penetrance or moderate-penetrance gene.

Among women with BRCA1 variants, 21 of 33 (64%) reported no family history of ovarian cancer specifically. Similarly, 72 of 82 women with BRCA2 variants (88%) reported no family history of ovarian cancer.

The research identified 44 of 159 women with BRCA1, BRCA2, or PALB2 PVs (28%) who lacked the composite family history risk factors, as did 135 of 438 women with ATM or CHEK2 variants (31%).

Demographic Associations

Several demographic variables correlated with PV presence. Younger age groups showed higher proportions of high-penetrance variants in breast cancer susceptibility genes compared with older age groups (40 to 49 years: 73 participants, 0.8%; 50 to 59 years: 56 participants, 0.8%; 60 to 69 years: 32 participants, 0.5%; 70 to 74 years: 6 participants, 0.4%), with the important context that women with any personal history of breast cancer were ineligible to participate.

CHEK2 variants showed clear ancestry patterns, with a higher proportion identified in White individuals (308 of 17,770, 2%) compared with all other racial and ethnic groups combined (29 of 5,219, 0.6%)—consistent with founder variants in CHEK2 being associated with European ancestry. The study also detected a higher proportion of PALB2 variants in non-Hispanic Black individuals (6 of 994, 0.6%) compared with other groups (38 of 21,995, 0.2%).

Among the 3,393 women reporting Jewish ancestry (15% of those tested), 119 (4%) had a PV. However, only 53 of these individuals (2%) had a variant in high-penetrance and moderate-penetrance genes, while 57 (2%) carried the founder CHEK2 S428F variant.

Rare Syndromic Variants

Pathogenic variants in genes associated with rare cancer syndromes were uncommon: CDH1 (n = 3), PTEN (n = 0), STK11 (n = 0), and TP53 (n = 5). Among the 5 women with TP53 variants (aged 45 to 69 years), none reported any personal cancer history. One reported multiple relatives with various cancers, and 4 did not report any family history suggestive of Li-Fraumeni syndrome.

Eight participants were heterozygous for 2 different PVs: 2 had ATM and CHEK2 variants; 2 had PALB2 and CHEK2 variants; 3 had BRCA2 and CHEK2 variants; and 1 had BRCA1 and CHEK2 variants. In all 8 cases, the CHEK2 variant was either the I157T or S428F low-penetrance variant.

Implications for Clinical Practice

"The identification of BRCA1, BRCA2, and PALB2 PV carriers without a family history of breast cancer highlights an important implementation gap in testing guidelines," the researchers wrote.

"Lower-than-expected proportions of family history risk factors in these carriers could have several explanations. These genes have incomplete penetrance, and patients may have incomplete or limited family structure (e.g., have fewer first-degree or second-degree relatives surviving beyond age 45 years). Other possibilities include small family size, undisclosed nonpaternity events, male-dominant family structures, competing mortality, and poor communication about cancer diagnoses within families. Criteria-independent testing should thus be seen as a complementary approach to indication-based testing," Dr. Fergus and colleagues wrote.

The researchers noted that "given that women with high-penetrance PVs often develop breast cancer in their 30s, our goal is to identify PV carriers who can benefit from increased surveillance or risk-reducing interventions in this younger age range."

Limitations

The study had several limitations. Variant calls were all determined at a single clinical laboratory. Self-reported prior testing was not confirmed. Family history of cancer was not confirmed with medical records and relied on participant self-report. The study did not collect data on family history of pancreas, prostate, or other cancers. The BARD1, RAD51C, and RAD51D genes were not included in the panel because they were not yet confirmed as breast cancer susceptibility genes when the study was designed. Additionally, the study population was enriched for family history of breast cancer compared with general population rates, which may have influenced the prevalence of pathogenic variants detected.

"These findings support the feasibility and acceptability of criteria-independent genetic testing for a diverse group of women. While we identified high-penetrance PVs in women who otherwise would not have qualified for germline testing, the most common PVs identified in our population were moderate-penetrance or low-penetrance variants, which require a nuanced and holistic risk assessment approach that deserves careful planning prior to implementing unrestricted genetic testing," the researchers concluded.

Disclosures can be found in the study.

Source: JAMA Internal Medicine