Abstract 5225: Expression- and splicing-based multi-tissue transcriptome-wide association studies identified multiple genes for estrogen receptor-negative breast cancer

Abstract

Abstract Although several transcriptome-wide association studies (TWAS) have been performed to identify genes associated with overall breast cancer risk, few genes were found to be associated with estrogen receptor-negative (ER-) breast cancer. These studies were based on gene expression prediction models trained mainly in breast tissue, and they did not account for alternative splicing of genes. In this study, we utilized two approaches to perform multi-tissue TWASs of ER- breast cancer: 1) an expression-based TWAS that combined TWAS signals for each gene across multiple tissues and 2) a splicing-based TWAS that combined TWAS signals of all excised introns for each gene across tissues. To perform these two TWASs, we first obtained gene expression prediction models that were trained in 11 tissues from the Genotype-Tissue Expression including breast, ovary, uterus, vagina, EBV-transformed lymphocytes, whole blood, spleen, liver, subcutaneous adipose, visceral adipose, and cell-cultured fibroblasts. Furthermore, we obtained GWAS summary statistics for ER- breast cancer by performing a meta-analysis of 21,468 ER- cases and 105,974 controls from the Breast Cancer Association Consortium and 9414 BRCA1 mutation carriers who were breast cancer cases and 9494 BRCA1 mutation carriers who did not have breast cancer from the Consortium of Investigators of Modifiers of BRCA1 and BRCA2. For our expression-based TWAS, we utilized the Aggregated Cauchy Association Test (ACAT) to collate TWAS signals across all 11 tissues for each gene using equal weights for each tissue, while for our splicing-based TWAS, we utilized ACAT to collate TWAS signals for all excised introns in all tissues for each gene. Overall, we identified 63 genes in 29 loci that were significantly associated with ER- breast cancer, including 23 genes that were identified using both approaches, 27 that were uniquely identified in the expression-based TWAS, and 13 that were uniquely identified in the splicing-based TWAS. Of the 63 genes, 50 genes have not been previously reported in TWAS studies of breast cancer. Using a weighted ACAT method, in which breast tissue was given a 5-fold higher weight, we obtained similar results. In summary, our joint, multi-tissue TWAS corroborated previous GWAS loci for both ER- and overall breast cancer while highlighting how incorporating TWAS signals from multiple tissues and alternative splicing allowed us to discover new susceptibility genes for ER- breast cancer. Citation Format: James L. Li, Julian C. McClellan, Guimin Gao, Dezheng Huo. Expression- and splicing-based multi-tissue transcriptome-wide association studies identified multiple genes for estrogen receptor-negative breast cancer. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5225.