Abstract 2327: Genetic correlation analysis identifies glioma heritability enrichment in immune cell types and novel protective associations with auto-immune conditions

Abstract

Abstract Introduction: Prior epidemiological studies in glioma have identified 25 germline risk variants, as well as risk association with exposure to ionizing radiation and protective association with history of allergies and aspirin use. In this analysis we applied LDscore regression methods, which leverages single SNP associations and known patterns of linkage disequilibrium (LD) to estimate the genetic correlation between phenotypes without bias for population structure, to confirm prior associations, and to identify novel phenotype associations for traits not previously assessed in any glioma study. We also used LDscore regression to partition heritability for glioma subtypes by immune cell type. This novel method may improve genetic prediction for glioma. Methods: We used the summary statistics for all glioma, glioblastoma (GBM), and non-GBM from a prior meta-analysis (Melin, et al.). We obtained the summary statistics for autoimmune, and atopic traits from the GWAS catalog and UK biobank. We generated pairwise genetic correlations (rg) between these phenotypic traits using LDscore regression as implemented in LDSC, and associations were considered significant at p<0.05. LDSC was also used to generate partitioned heritability using GTex (brain regions) and ImmGen (immune cell types) references sets. Results: Heritability partitioned by immune cell type for glioma was significantly enriched in myeloid cells, natural killer cells, and t cells. We identified significant negative correlations between glioma and primary biliary cirrhosis (rg=-0.24, p=0.0190), and between celiac disease and glioblastoma only (rg=-0.31, p=0.0128), and we identified a female-specific association for rheumatoid arthritis (rg=-0.68, p=0.0225). Pleiotropic effects were identified for variants in telomere-associated regions (TERC, TERT, RTEL1) for glioma and multiple autoimmune traits. Conclusions: This analysis identified significant enrichment for heritability for glioma in immune cell types as well as novel associations between auto-immune traits. It did not validate a genetic basis for previously identified protective effects from allergic rhinitis, suggesting that it may be more strongly influenced by the environment. Further studies are needed to confirm these associations and identify the mechanism through which increased immune activity may lower risk of glioma. Citation Format: Quinn T. Ostrom, Jacob Edelson, Jinyoung Byun, Younghun Han, GLIOGENE Consortium, Kyle Walsh, Christopher Amos, Melissa Bondy. Genetic correlation analysis identifies glioma heritability enrichment in immune cell types and novel protective associations with auto-immune conditions [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2327.