Abstract 2283: Ancestral and genetic determinants of pathologic tumor infiltrating lymphocytes in a large multi-ethnic population of women with breast cancer
Abstract Background: There are well-documented racial and ethnic disparities in breast cancer, with Black women having the highest mortality and Asian women having the lowest mortality. While reasons for this disparity are multi-factorial, differences in tumor biology and tumor microenvironment may contribute. Over the past two decades, pathologic tumor-infiltrating lymphocytes (TILs) have emerged as a histologic biomarker of prognosis, where higher TIL levels are associated with better outcomes and overall survival, and a predictive biomarker for cancer treatment. In recent years, there is increasing evidence of genetic determinants of immune response in the tumor microenvironment, which is known to vary by ancestral background and may explain differences in patient survival. We hypothesize that genetic selection and adaptation to distinct pathogenic environments during the evolution of diverse human populations led to ancestral differences in host immunity. Methods: To investigate this hypothesis, we performed a local ancestry-informed GWAS of pathologic TILs using Tractor, a statistical framework and software package. With data from 4, 728 women of European (n=2, 740), African (n=1235), Amerindian (n=318), and East Asian (n=435) ancestry from the Women’s Circle of Health Study (WCHS) and the Pathways Study, we estimated four-way local ancestry with the tool FLARE using reference genomes from 1000 Genomes Project and Human Genome Diversity Project. Given the non-normal distribution of TIL scores, we applied linear regression to the inverse normal transformation of TIL scores to assess genotype-local ancestry interaction effects. Results: Global genetic similarity to African reference populations was significantly associated with increased TILs (P=3.8x10-22). We identified rs17132663 on African haplotype (P=1.7x10-8) and rs6812740 on Amerindian haplotype (P=3.8x10-8) as genome-wide significant SNPs in our analyses. rs17132663 is an intronic variant of AMZ1 and GNA12. rs6812740 is located 2 KB upstream of HPGD and is significantly associated (P=1.6x10-12) with expression of the gene in GTEx version 10. To our knowledge, neither of these SNPs previously have been significantly associated with the tumor immune response. In addition, 20 other loci were significant at P<1x10-6. Conclusion: These findings underscore ancestral differences in breast tumor immune response and highlight the critical need to expand datasets with matched genotype and histology data to further explore these associations. Citation Format: Peter N. Fiorica, Lucas Mendicino, Li Yan, Angela R. Omilian, Elisa V. Bandera, Scott I. Abrams, Lawrence H. Kushi, Thaer Khoury, Christine B. Ambrosone, Song Yao. Ancestral and genetic determinants of pathologic tumor infiltrating lymphocytes in a large multi-ethnic population of women with breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2283.
Authors
- Lucas Mendicino
- Elisa V. Bandera (ORCID: https://orcid.org/0000-0002-8789-2755)
- Song Yao
- Lawrence H. Kushi (ORCID: https://orcid.org/0000-0001-9136-1175)
- Peter N. Fiorica (ORCID: https://orcid.org/0000-0002-1001-1518)
- Scott I. Abrams (ORCID: https://orcid.org/0000-0002-8742-4708)
- Thaer Khoury (ORCID: https://orcid.org/0000-0002-8112-6590)
- Angela R. Omilian (ORCID: https://orcid.org/0000-0001-6348-9411)
- Yan Li (ORCID: https://orcid.org/0000-0003-1963-839X)
- Christine B. Ambrosone
Institutions
- Rutgers, The State University of New Jersey (US)
- Roswell Park Comprehensive Cancer Center (US)
- Rutgers Cancer Institute (US)
Publication Details
- Journal
- Cancer Research
- Published
- 2025-04-21
- DOI
- https://doi.org/10.1158/1538-7445.am2025-2283
- Primary Topic
- Cancer Immunotherapy and Biomarkers
- Type
- article
- Field-Weighted Citation Impact
- 0.00