Contributions
Abstract: P428
Type: Poster Sessions
Abstract Category: Pathology and pathogenesis of MS - Genetics/Epigenetics
Differences in multiple sclerosis (MS) prevalence and clinical presentation across populations provide some evidence that allele and locus heterogeneity may exist. Currently, 200 non-MHC autosomal risk variants in 153 loci have been identified as conferring risk for MS in >100,000 ancestral Europeans. To evaluate the distribution of risk alleles across populations, we computed a genetic risk score of the 200 variants in 1000 Genomes populations of Europe (N=441), Africa (N=323), Asia (N=166), and the Americas (N=258). European populations include CEU: Utah, IBS: Spain, GBR: Britain, FIN: Finland, TSI: Italy; African include ACB: Barbados, ASW: SW US, LWK: Luhya, YRI: Yoruba; Asian include CHB/CHD: China, JPT: Japan; and the Americas include CLM: Colombia, MXL: Mexico, PEL: Peru, PUR: Puerto Rico. The mean risk score ranges from 18.8 in Asians to 19.8 in Africans (~13 more risk alleles in Africans), with Europeans and people of the Americas demonstrating intermediate scores of 19.4 and 19.3 respectively. When comparing the mean risk score across the four populations, differences (p≤0.05) were observed for each of the pairwise combinations except Europeans and people of the Americas; indicating differences in the distribution of MS risk variants across ancestral groups. We further evaluated the genetic risk score in admixed (3 or 2-way European, African, and Native American ancestry) Hispanics (1398 cases; 1386 controls) and African Americans (1305 cases; 1155 controls). The mean risk score was 19.4, 20.0 (p=8.12E-58) and 19.7, 20.03 (p=3.51E-23) in Hispanic and African American controls, cases. The increase in cases indicates the utility of the risk score in characterizing risk among diverse populations. In each data set, we compared risk scores in the top 5th percentile of the distribution to those in the interquartile range and found that being in the top 5th percentile indicates greater odds of disease in Hispanics than in African Americans (3.16 vs. 2.39) despite higher mean risk scores observed in African Americans. Utilizing local ancestry analyses to compute of the odds of MS for those with 0, 1, and 2 alleles of European, African, and Native American ancestry (derived from Karitiana, Maya, Pima, and Surui peoples of HGDP) for each risk variant; we find evidence that for some risk loci, haplotype structures differ across populations and the identified risk variant may not be the best tag for the true causal variant in all populations.
Disclosure: Ashley Beecham: nothing to disclose
Lilyana Amezcua: nothing to disclose
Angel Chinea: nothing to disclose
Gary Beecham: nothing to disclose
Silvia Delgado: nothing to disclose
David Conti: nothing to disclose
Jorge Oksenberg: nothing to disclose
Jacob McCauley: nothing to disclose
Abstract: P428
Type: Poster Sessions
Abstract Category: Pathology and pathogenesis of MS - Genetics/Epigenetics
Differences in multiple sclerosis (MS) prevalence and clinical presentation across populations provide some evidence that allele and locus heterogeneity may exist. Currently, 200 non-MHC autosomal risk variants in 153 loci have been identified as conferring risk for MS in >100,000 ancestral Europeans. To evaluate the distribution of risk alleles across populations, we computed a genetic risk score of the 200 variants in 1000 Genomes populations of Europe (N=441), Africa (N=323), Asia (N=166), and the Americas (N=258). European populations include CEU: Utah, IBS: Spain, GBR: Britain, FIN: Finland, TSI: Italy; African include ACB: Barbados, ASW: SW US, LWK: Luhya, YRI: Yoruba; Asian include CHB/CHD: China, JPT: Japan; and the Americas include CLM: Colombia, MXL: Mexico, PEL: Peru, PUR: Puerto Rico. The mean risk score ranges from 18.8 in Asians to 19.8 in Africans (~13 more risk alleles in Africans), with Europeans and people of the Americas demonstrating intermediate scores of 19.4 and 19.3 respectively. When comparing the mean risk score across the four populations, differences (p≤0.05) were observed for each of the pairwise combinations except Europeans and people of the Americas; indicating differences in the distribution of MS risk variants across ancestral groups. We further evaluated the genetic risk score in admixed (3 or 2-way European, African, and Native American ancestry) Hispanics (1398 cases; 1386 controls) and African Americans (1305 cases; 1155 controls). The mean risk score was 19.4, 20.0 (p=8.12E-58) and 19.7, 20.03 (p=3.51E-23) in Hispanic and African American controls, cases. The increase in cases indicates the utility of the risk score in characterizing risk among diverse populations. In each data set, we compared risk scores in the top 5th percentile of the distribution to those in the interquartile range and found that being in the top 5th percentile indicates greater odds of disease in Hispanics than in African Americans (3.16 vs. 2.39) despite higher mean risk scores observed in African Americans. Utilizing local ancestry analyses to compute of the odds of MS for those with 0, 1, and 2 alleles of European, African, and Native American ancestry (derived from Karitiana, Maya, Pima, and Surui peoples of HGDP) for each risk variant; we find evidence that for some risk loci, haplotype structures differ across populations and the identified risk variant may not be the best tag for the true causal variant in all populations.
Disclosure: Ashley Beecham: nothing to disclose
Lilyana Amezcua: nothing to disclose
Angel Chinea: nothing to disclose
Gary Beecham: nothing to disclose
Silvia Delgado: nothing to disclose
David Conti: nothing to disclose
Jorge Oksenberg: nothing to disclose
Jacob McCauley: nothing to disclose