Associations of a Breast Cancer Polygenic Risk Score With Tumor Characteristics and Survival

Abstract

PURPOSE A polygenic risk score (PRS) consisting of 313 common genetic variants (PRS313) is associated with risk of breast cancer and contralateral breast cancer. This study aimed to evaluate the association of the PRS313 with clinicopathologic characteristics of, and survival following, breast cancer. METHODS Women with invasive breast cancer were included, 98,397 of European ancestry and 12,920 of Asian ancestry, from the Breast Cancer Association Consortium (BCAC), and 683 women from the European MINDACT trial. Associations between PRS313 and clinicopathologic characteristics, including the 70-gene signature for MINDACT, were evaluated using logistic regression analyses. Associations of PRS313 (continuous, per standard deviation) with overall survival (OS) and breast cancer–specific survival (BCSS) were evaluated with Cox regression, adjusted for clinicopathologic characteristics and treatment. RESULTS The PRS313 was associated with more favorable tumor characteristics. In BCAC, increasing PRS313 was associated with lower grade, hormone receptor–positive status, and smaller tumor size. In MINDACT, PRS313 was associated with a low risk 70-gene signature. In European women from BCAC, higher PRS313 was associated with better OS and BCSS: hazard ratio (HR) 0.96 (95% CI, 0.94 to 0.97) and 0.96 (95% CI, 0.94 to 0.98), but the association disappeared after adjustment for clinicopathologic characteristics (and treatment): OS HR, 1.01 (95% CI, 0.98 to 1.05) and BCSS HR, 1.02 (95% CI, 0.98 to 1.07). The results in MINDACT and Asian women from BCAC were consistent. CONCLUSION An increased PRS313 is associated with favorable tumor characteristics, but is not independently associated with prognosis. Thus, PRS313 has no role in the clinical management of primary breast cancer at the time of diagnosis. Nevertheless, breast cancer mortality rates will be higher for women with higher PRS313 as increasing PRS313 is associated with an increased risk of disease. This information is crucial for modeling effective stratified screening programs.

Development and testing of a polygenic risk score for breast cancer aggressiveness

“…of 1.13 (95% CI 1.06–1.21) in meta-analysis. In contrast, higher PRS overallBC was associated with better survival in the UK Biobank, which is consistent with the results of a large study including nearly 100,000 women with breast cancer 10 . Thus, the associations of the ROR-P PRS and PRS ER-/ER+ with worse survival suggest that it is possible to reconfigure PRS to predict aggressive tumors with worse prognosis.…”

“…For 51 of 76 SNPs in the ROR-P PRS, the breast cancer risk allele, as annotated by the original GWAS, was associated with lower ROR-P (Supplementary Table 6 , Fig. 2b ), which is consistent with the observation that the PRS for overall breast cancer risk is associated with better survival 10 . Six of 76 SNPs in our final ROR-P PRS had nominally significant associations with ROR-P, though none met significance after Bonferroni correction.…”

“…In UK Biobank, we first evaluated the association between a PRS for overall breast cancer risk (PRS overallBC ) and breast cancer-specific survival. We hypothesized that since prior studies have shown that the overall breast cancer PRS is associated with favorable characteristics in breast cancer patients 10 , then higher overall breast cancer PRS among cases would be associated with more favorable survival. As expected, PRS overallBC was inversely associated with breast cancer mortality, with a hazard ratio per standard deviation (HR per S.D.)…”

“…This is a critical limitation since breast cancer encompasses a variety of subtypes ranging from indolent to aggressive, with the latter defined as having increased proliferation or metastatic potential and poor prognosis 8 . However, case-only analyses have found that PRSs for overall breast cancer risk are associated with more favorable clinicopathologic 9 and prognostic features 10 , as well as lower risk of interval versus screen-detected cancer 11 , 12 . Efforts to fit PRSs to subtypes of breast cancer have focused on estrogen receptor (ER) status given that ER-negative breast cancers tend to be more proliferative and are associated with earlier risk of relapse 2 , 13 .…”

Estimation of age of onset and progression of breast cancer by absolute risk dependent on polygenic risk score and other risk factors

“…Isheden et al 24 found no association between the rate of lymph node spread, a measure of tumor aggressiveness, and PRS. Cardozo et al 25 and Li et al 26 found that interval cancers were associated with lower PRS. Our results show agreement with Ishden et al 24 and Cardozo et al 25 because our results show little difference in the sojourn time between low risk and high risk based on PRS.…”