Investigating divergent ER + and ER - breast cancer subtype risks by using oppositely associated genetic variants

Breast cancer is the second most common cause of cancer-related deaths among women in the United States, with its incidence increasing each year. Estrogen receptor-negative (ER - ) breast cancers are particularly aggressive and less responsive to conventional treatments compared to estrogen receptor-positive (ER + ) types, though they exhibit higher immunogenicity and better responses to immunotherapies. This study focuses on identifying single nucleotide polymorphisms (SNPs) that are oppositely associated between ER + and ER - breast cancers, meaning that the presence of a SNP increases the risk of one subtype while decreasing the risk of the other subtype. We identified 481 SNPs from the Breast Cancer Association Consortium (BCAC) that were oppositely associated between subtypes (p < 0.1) and cross-validated these associations using data from breast cancer patients in The Cancer Genome Atlas (TCGA), imputed using the TOPMED imputation server. Our findings revealed 45 SNPs associated with increased ER + risk but decreased ER - risk and 2 SNPs associated with increased ER - risk but decreased ER + risk that were concordant between the BCAC and TCGA datasets. Of the concordant SNPs that increased ER + risk, the majority mapped to a region of chromosome 11 near the known tumor suppressor gene ATM; however, one SNP on chromosome 10 (10:21471086:C/T) has been previously shown to increase circulating IGF-1 and decrease colorectal cancer risk to a greater extent in women than men. Further analysis using the CIBERSORTx tool showed that this SNP was associated with increased infiltration of perivascular-like immature cells in the tumor microenvironment. Of the SNPs that increased ER - risk, one localized to a poorly described antisense non-coding transcript of the DLX2 gene which has been implicated in breast cancer, while the other localized to an intron of the PTPRN2 gene that is implicated in type 1 diabetes mellitus and many cancers. While this study is limited by its focus on European women and the small sample size of ER - cases, the results offer preliminary insight into inherited variants that may differentially influence ER + and ER - disease, supporting future work aimed at clarifying the pathways that contribute to subtype-specific risk, immune infiltration, and treatment response.