Hierarchy of resistance to cervical neoplasia mediated by combinations of killer immunoglobulin-like receptor and human leukocyte antigen loci

Abstract

Killer immunoglobulin-like receptor (KIR) recognition of specific human histocompatibility leukocyte antigen (HLA) class I allotypes contributes to the array of receptor-ligand interactions that determine natural killer (NK) cell response to its target. Contrasting genetic effects of KIR/HLA combinations have been observed in infectious and autoimmune diseases, where genotypes associated with NK cell activation seem to be protective or to confer susceptibility, respectively. We show here that combinations of KIR and HLA loci also affect the risk of developing cervical neoplasia. Specific inhibitory KIR/HLA ligand pairs decrease the risk of developing neoplasia, whereas the presence of the activating receptor KIR3DS1 results in increased risk of disease, particularly when the protective inhibitory combinations are missing. These data suggest a continuum of resistance conferred by NK cell inhibition to susceptibility involving NK cell activation in the development of cervical neoplasia and underscore the pervasive influence of KIR/HLA genetic variation in human disease pathogenesis.

Figure 1.

Independent associations between HLA ligands for KIR and CIN3/cancer outcome in three independent studies. Odds ratios (circle) and 95% confidence intervals (dashes) are shown separately for HLA-Cw group 2 and HLA-B Bw4, for the Eastern U.S., Costa Rica, and Portland studies, followed by estimates for data from all three studies merged.

Figure 2.

Effect of compound genotypes KIR 3DS1/ HLA-Cw group 2 and KIR3DS1/HLA-B Bw4 on CIN3/cancer. Odds ratios (circle) and 95% confidence intervals (dashes) are shown separately for KIR 3DS1 and HLA-Cw group 2, and for KIR3DS1 and HLA-B Bw4. The referent groups, KIR3DS1 ⁺ /Cw group 2 ⁻ (left) and KIR3DS1 ⁺ /HLA-B Bw4 ⁻ (right), are those with which all other genotypically defined groups are compared, and the odds ratio for each referent group is set at 1. Middle genotypes in both analyses were grouped together to generate the P trend values because we have no functional criteria to suggest that one genetic factor (HLA-Cw group 2, HLA-B Bw4, or KIR3DS1) is dominant over the others. We list the various genotypes separately and ordered by decreasing odds ratios as a means to compare odds ratios of individual genotypes.

Figure 3.

Combined effect of activating KIR3DS1 and HLA ligands (HLA-Cw group 2 and HLA-B Bw4). Odds ratios (circle) and 95% confidence intervals (dashes) are shown for each combination. In this analysis, the genotype of the referent group (for which the odds ratio is set at 1) was KIR3DS1 ⁺ /HLA-Cw group 2-/HLA-B Bw4 ⁻. As in Fig. 2, middle genotypes were grouped together to generate the P trend values (Fig. 2).

Figure 4.

Model of the spectrum of KIR-mediated NK phenotypes associated with risk of developing cervical neoplasia. Resistance to cervical neoplasia increases when genotypes are ordered by their ability to confer the most activation (susceptibility) to the most inhibition (protection). (a) Presence of KIR3DS1 and absence of ligand for inhibitory KIR results in strong activation (most susceptibility). (b) Presence of KIR3DS1 and presence of one or the other ligand for the inhibitory KIR results in weak activation. (c) Absence of the activating KIR3DS1, and absence of ligands for inhibitory KIR result in relative neutrality. (d) Presence of KIR3DS1 along with ligands for one or both inhibitory KIR confers weak inhibition. (e) Presence of ligands for inhibitory KIR in the absence of KIR3DS1 results in strong inhibition (most protection). Note that expression of KIR3DS1 and a potential ligand for this receptor have not been defined, and that its hypothetical expression and binding to a ligand are proposed in this model based solely on the data presented herein. Red shapes, activating receptor and ligand; Green shapes, inhibitory receptors and ligands.