Genetic factors are likely to contribute to low severe malaria case fatality rates in Melanesian populations, but association studies can be underpowered and may not provide plausible mechanistic explanations if significant associations are detected. In preparation for a genome-wide association study, 29 candidate single-nucleotide polymorphisms (SNPs) with minor allele frequencies >5% were examined in a case-control study of 504 Papua New Guinean children with severe malaria. In parallel, an immunological substudy was performed on convalescent peripheral blood mononuclear cells (PBMCs) from cases and controls. Following stimulation with a Toll-like receptor (TLR) 1/2 agonist, effector cytokines and chemokines were assayed. The only significant genetic association observed involved a nonsynonymous SNP (TLR1rs4833095) in the TLR1 gene. A recessive (TT) genotype was associated with reduced odds of severe malaria of 0.52 (95% confidence interval (0.29-0.90), P=0.006). Concentrations of pro-inflammatory cytokines interleukin-1β and tumour necrosis factor α were significantly higher in severe malaria cases compared with healthy controls, but lower in children with the protective recessive (TT) genotype. A genetic variant in TLR1 may contribute to the low severe malaria case fatality rates in this region through a reduced pro-inflammatory cellular phenotype.