Recent studies suggest that immune-modulating single-nucleotide polymorphisms (SNPs) influence the risk of developing cancer-related infections. Here, we evaluated whether 36 SNPs within 14 immune-related genes are associated with the risk of invasive aspergillosis (IA) and whether genotyping of these variants might improve disease risk prediction. We conducted a case-control association study of 781 immunocompromised patients, 149 of whom were diagnosed with IA. Association analysis showed that the IL4Rrs2107356 and IL8rs2227307 SNPs (using dbSNP numbering) were associated with an increased risk of IA (IL4Rrs2107356 odds ratio [OR], 1.92; 95% confidence interval [CI], 1.20 to 3.09; IL8rs2227307 OR, 1.73; 95% CI, 1.06 to 2.81), whereas the IL12Brs3212227 and IFNγrs2069705 variants were significantly associated with a decreased risk of developing the infection (IL12Brs3212227 OR, 0.60; 95% CI, 0.38 to 0.96; IFNγrs2069705 OR, 0.63; 95% CI, 0.41 to 0.97). An allogeneic hematopoietic stem cell transplantation (allo-HSCT)-stratified analysis revealed that the effect observed for the IL4Rrs2107356 and IFNγrs2069705 SNPs was stronger in allo-HSCT (IL4Rrs2107356 OR, 5.63; 95% CI, 1.20 to 3.09; IFNγrs2069705 OR, 0.24; 95% CI, 0.10 to 0.59) than in non-HSCT patients, suggesting that the presence of these SNPs renders patients more vulnerable to infection, especially under severe and prolonged immunosuppressive conditions. Importantly, in vitro studies revealed that carriers of the IFNγrs2069705C allele showed a significantly increased macrophage-mediated neutralization of fungal conidia (P = 0.0003) and, under stimulation conditions, produced higher levels of gamma interferon (IFNγ) mRNA (P = 0.049) and IFNγ and tumor necrosis factor alpha (TNF-α) cytokines (P value for 96 h of treatment with lipopolysaccharide [PLPS-96 h], 0.057; P value for 96 h of treatment with phytohemagglutinin [PPHA-96 h], 0.036; PLPS+PHA-96 h = 0.030; PPHA-72 h = 0.045; PLPS+PHA-72 h = 0.018; PLPS-96 h = 0.058; PLPS+PHA-96 h = 0.0058). Finally, we also observed that the addition of SNPs significantly associated with IA to a model including clinical variables led to a substantial improvement in the discriminatory ability to predict disease (area under the concentration-time curve [AUC] of 0.659 versus AUC of 0.564; P-2 log likehood ratio test = 5.2 · 10(-4) and P50.000 permutation test = 9.34 · 10(-5)). These findings suggest that the IFNγrs2069705 SNP influences the risk of IA and that predictive models built with IFNγ, IL8, IL12p70, and VEGFA variants can used to predict disease risk and to implement risk-adapted prophylaxis or diagnostic strategies.