GENE-47. EVALUATING GLIOMA RISK ASSOCIATED WITH EXTENT OF EUROPEAN ADMIXTURE IN AFRICAN-AMERICANS AND LATINOS. (6th November 2017)

• Record Type: Journal Article
• Authors: Ostrom, Quinn; Egan, Kathleen; Nabors, Burt; Amos, Christopher; Armstrong, Georgina; Bernstein, Jonine; Chowdhary, Sajeel; Claus, Elizabeth; Eckel-Passow, Jeanette; Gerke, Travis; Houlston, Richard; Il'yasova, Dora; Jenkins, Robert; Johansen, Christoffer; Lachance, Daniel; Lai, Rose; LaRocca, Renato; Lau, Ching; Merrell, Ryan; Olson, Jeffrey J; Olson, Sara; Sadetzki, Siegal; Schildkraut, Joellen; Shete, Sanjay; Thompson, Reid; Wrensch, Margaret; Wiencke, John; Melin, Beatrice; Bondy, Melissa; Barnholtz-Sloan, Jill
• Is Part Of: Neuro-oncology. Volume 19(2017)Supplement 6
• Page Start: vi102
• Page End: vi103
• Publication Date: 2017-11-06
• Subjects: Brain Neoplasms -- Periodicals; Brain -- Tumors -- Periodicals; Brain -- Cancer -- Periodicals; Nervous system -- Cancer -- Periodicals; 616.99481
• DOI: 10.1093/neuonc/nox168.420
• Languages: English
• ISSNs: 1522-8517

Abstract

Abstract: Glioma incidence is highest in non-Hispanic Whites, where it occurs ~2x as frequently compared with other race/ethnicity groups. Glioma GWAS have included European-ancestry (EA) populations only, and is unknown whether identified variants are associated with glioma in non-EA populations. African Americans (AA) and Hispanics are admixed populations with varying proportions of EA. Here we attempted to assess whether excess EA at risk loci (Melin et al, Nature Genetics, 2017) was associated with glioma risk in non-EA populations. We identified 197 AA cases, 168 AA controls, 269 Hispanic Whites and 123 Hispanic Whites controls within the Glioma International Case-Control Study and GliomaSE Study. Local ancestry (LA) was estimated using RFMix with four 1, 000 genomes project continental reference populations (African, American, East Asian, and European). Global ancestry was calculated by averaging LA. We evaluated differences in local EA between cases and controls using logistic regression within 500kb of 26 previously identified risk variants (766 SNPs with minor allele frequency≥0.01) for all glioma, GBM and non-GBM. Results were considered significant at p<1.1x10 -5 (adjusted for 4, 596 tests). All analyses were performed stratified by self-reported race/ethnicity. AA cases had higher global EA (GEA) than controls (28.3% vs 23.6%, p=0.0009), there was no significant difference in GEA in Hispanic White cases and controls. There was a nominal association in AA between increased local EA (LEA) and all glioma at 3p14.1 (pminimum =0.001), and with GBM at 5p15.33 (pminimum =0.002). In Hispanic Whites, there was a nominal association between increased LEA and GBM at 12q23.33 (pminimum =0.008), and at 17p13.1 (pminimum =0.007). EA populations show increased susceptibility to glioma, and it is possible that increased EA increases risk of glioma in admixed populations. These results suggest that glioma risk may be associated both with increased GEA in AA and increased LEA at some previously identified variants in AA and Hispanic whites.