Pathogenic Variant Frequency in Late-Onset Breast Cancer May Inform Future Screening Practices

Data collected from several large studies of women with breast cancer over the age of 65 years at diagnosis show how frequently certain known pathogenic variants occur in different patient subsets.

An analysis of data from population-based studies including over 25,000 women with breast cancer who were diagnosed over the age of 65 years suggests that the diagnosis of triple-negative or estrogen receptor (ER)–negative disease warrants genetic testing in all cases, according to results published in the Journal of Clinical Oncology.

Additionally, investigators recommend that women over 65 years of age with BRCA1/2 and, in some cases, PALB2 and CHEK2 pathogenic variants (PVs) be considered for MRI screening.

“Although up to 10% of women with breast cancer with a young age of diagnosis or a [family history] of breast or ovarian cancer have PVs in predisposition genes, little is known about the frequencies of PVs in women who are older than age 65 years when diagnosed with breast cancer,” wrote the investigators, who were led by Nicholas J. Boddicker, PhD. “Importantly, women with breast cancer diagnosed over age 65 years with no other significant risk factors…are not recommended by NCCN to receive genetic testing on the basis of a presumed low yield [translating to a lower than] 2.5% probability of detecting a high-penetrance PV.”

A total of 26,707 women older than 65 years were examined, of whom 51.5% had breast cancer and the rest were age-matched unaffected participants. All patients were tested for germline PVs and outcomes were stratified by age group (66-75 vs 75-85 years), race or ethnicity, ER status, and family history.

Family history of breast cancer in a first-degree relative was present in 25.6% of patients with breast cancer and 17.9% of the controls. ER status in patients with available tumor pathology was negative in 14.0% of cases and positive in 86.0%. The mean age of diagnosis in patients over the age of 65 who had available multigene hereditary genetic testing was 71.4 years, of whom 18.4% had ER-negative disease.

PVs in 12 genes established as being correlated with risk of breast cancer were present in 3.18% of cases vs 1.48% of controls. A lower frequency of PVs was observed in Asian American women vs non-Hispanics White or African American women (1.24% vs 3.4%). Patients with breast cancer had a high frequency of PVs in CHEK2 (0.92%), BRCA2 (0.71%), ATM (0.57%), and PALB2 (0.36%), with a lower incidence of BRCA1 (0.28%). In the control group, the highest frequency of PVs was in CHEK2 (0.39%) and ATM (0.39%).

Similar frequencies of PVs were seen in age-stratified groups with breast cancer, except BRCA1, which was more common in patients aged 66 to 75 years (0.37%) vs 75 to 85 years (0.06%). No differences in PV frequencies were noted in the control patients for BRCA2 (0.24% v 0.23%), CHEK2 (0.44% v 0.39%), and PALB2 (0.13% v 0.10%).

ER status was associated with the presence of PVs, with 2.8% of those with ER-positive disease, 5.24% of those with ER-negative tumors, and 4.79% of triple-negative breast cancers harboring these genetic features. Corresponding rates of high-risk genes—defined as BRCA1, BRCA2, or PALB2—were 1.00%, 3.42%, and 3.01%, respectively. For diagnoses over age 65 years, PVs in PALB2CHEK2, and BRCA2 translated to a moderately increased risk of ER-positive breast cancer, and PVs in BRCA1BRCA2, and PALB2 were associated with a higher risk of ER-negative tumors.

BRCA1 PVs (OR, 3.37; 95% CI, 1.68-7.51) as well as those in BRCA2 (OR, 2.64; 95% CI, 1.78-0.02), PALB2 (OR, 3.09; 95% CI, 1.71-5.98), and CHEK2 (OR, 2.13; 95% CI, 1.53-3.02) resulted in a moderate risk of breast cancer, whereas PVs in ATM were not found to be associated with risk of breast cancer in women over age 65 years (OR, 1.38; 95% CI, 0.96-2.00; P = .09). Evaluation of other candidate breast cancer predisposition genes showed no association with an increased risk of malignancy.

In women with breast cancer, family history vs no history was also correlated to a higher presence of PVs (4.5% vs 2.7%). The most common PVs in women with breast cancer and a family history were BRCA1, BRCA2, and PALB2; in those without family history, most common PVs were the same, but occurred at lower frequencies overall. Patients with ER-negative and a family history were more likely to harbor PVs (8.8%) vs those without a family history (4.2%).

Importantly, remaining lifetime risk was calculated based on the presence of key PVs. From ages 66 to 85 years, the remaining lifetime risk was 18.4% for BRCA1 carriers, 18.7% for BRCA2, 15.9% for PALB2, 14.9% for CHEK2, and 9.9% for ATM. Stratifying by ER status, ER-positive breast cancer lifetime risk was 15.0% for BRCA2, 14.4% for CHEK2, and 13.3% for PALB2 whereas the remaining lifetime risk was lower for ER-negative disease (5.2% for BRCA1 and 5.1% for BRCA2). Of note, the absolute remaining risk of breast cancer in patients 66 to 85 years is 6.8%.

Limitations of the study include that it was underpowered to investigate other potential PVs such as BARD1, RAD51C, and RAD51D.

“The results presented in this study will inform cancer screening guidelines and risk management for women over age 65 years in the general population, especially for those with TNBC diagnosed over age 60 years or ER-negative disease diagnosed over age 65 years,” the investigators concluded. “Furthermore, because women over age 65 years with PVs in BRCA1BRCA2CHEK2, and PALB2 continue to be at increased risk of breast cancer…MRI screening should be considered for those with PVs in BRCA1 and BRCA2 and might be considered for those with PVs in PALB2 and CHEK2.”


Boddicker NJ, Hu C, Weitzel JN, et al. Risk of Late-Onset Breast Cancer in Genetically Predisposed Women. J Clin Oncol. Published online July 22, 2021. doi:10.1200/JCO.21.00531