Résumé
Human African Trypanosomiasis (HAT) or sleeping sickness is a Neglected Tropical Disease. Long regarded as an invariably fatal disease, there is increasing evidence that infection by T.b. gambiense can result in a wide range of clinical outcomes, including latent infections, which are long lasting infections with no parasites detectable by microscopy. The determinants of this clinical diversity are not well understood but could be due in part to parasite or host genetic diversity in multiple genes, or their interactions. A candidate gene association study was conducted in Cote d'Ivoire using a case-control design which included a total of 233 subjects (100 active HAT cases, 100 controls and 33 latent infections). All three possible pairwise comparisons between the three phenotypes were tested using 96 SNPs in 16 candidate genes (IL1, IL4, IL4R, IL6, IL8, IL10, IL12, IL12R, TNFA, INFG, MIF, APOL1, HPR, CFH, HLA-A and HLA-G). Data from 77 SNPs passed quality control. There were suggestive associations at three loci in IL6 and TNFA in the comparison between active cases and controls, one SNP in each of APOL1, MIF and IL6 in the comparison between latent infections and active cases and seven SNP in IL4, HLA-G and TNFA between latent infections and controls. No associations remained significant after Bonferroni correction, but the Benjamini Hochberg false discovery rate test indicated that there were strong probabilities that at least some of the associations were genuine. The excess of associations with latent infections despite the small number of samples available suggests that these subjects form a distinct genetic cluster different from active HAT cases and controls, although no clustering by phenotype was observed by principle component analysis. This underlines the complexity of the interactions existing between host genetic polymorphisms and parasite diversity.