An infectious origin for pediatric Hodgkin lymphoma (HL) has long been suspected and Epstein-Barr virus (EBV) has been detected in ∼50% of tumors. Because EBV infections are ubiquitous, yet only a minute fraction of those infected develops HL, it follows that individual differences in the innate immune system response, including the identification and elimination of virus-infected and tumor cells by natural killer (NK) cells, may contribute to HL susceptibility. NK cell effector responses are modulated by the surface expression of various killer cell immunoglobulin-like receptors (KIR) and their binding to human leukocyte antigen (HLA) class I ligands on target cells. Importantly, the specific KIRs expressed on a given NK cell surface are determined in part by the highly polymorphic KIR genes on chromosome 19q13.4. There is considerable interindividual variability in the type and number of these genes, however, two regions of strong linkage disequilibrium have been identified, defining the centromeric and telomeric haplotype segments, which are flanked on each end by highly conserved genes and joined by a recombination hotspot. Accordingly, the haplotypes for a striking majority (≥93%) of individuals examined to date can be formed combinatorially from 3 common centromeric and 2 telomeric motifs.

KIR gene content has been associated with pregnancy outcomes, response to infection, susceptibility to autoimmune disease, and malignancy. Few epidemiologic studies examining KIR genotype and lymphoid neoplasms (LN) have been conducted in children and adults. Generally, these studies have reported inverse associations between presence of activating KIR genes and susceptibility to LN. Given the strong likelihood for a viral etiology and the known correlations with HLA type, we performed for the first time a case-control study of the association between KIR gene content and pediatric/adolescent HL using archived samples.

Cases (n=171) diagnosed with HL in 1989-2003 at 3-17 years of age at North American Children’s Oncology Group institutions and healthy adult controls (n=332) provided peripheral blood or buccal cell samples. Presence/absence of KIR genes (KIR2DS2, KIR2DL2/2DL3, KIR2DL5, KIR2DS3, KIR2DS5, KIR2DL1, KIR3DS1/3DL1, KIR2DS1, KIR2DS4) was determined using a Luminex DNA-based sequence-specific oligonucleotide (SSO) typing methodology. Each individual’s repertoire of genes was then classified into centromeric and telomeric KIR haplotype groups. Case haplotypes were compared to those of controls via unconditional logistic regression, resulting in odds ratios (ORs) and 95% confidence intervals (CIs) for the presence of one or two copies of a B haplotype (A/B and B/B) versus no copies (A/A), where B haplotypes encompass greater numbers of activating KIR. Associations were examined for HL overall, for the nodular sclerosis histologic subtype, and for the 3-9 and 10-17 year age groups.

The presence of two centromeric B haplotype motifs was associated with development of pediatric and adolescent HL (OR=1.92, 95% CI: 1.12-3.26), but there was no evidence for an association with telomeric haplotypes. Similar results were observed for nodular sclerosis HL (OR=2.13, 95% CI: 1.02-4.43) and for the younger and older age groups (OR=1.80, 95% CI: 0.69-4.66; OR=1.76, 95% CI: 0.91-3.38), respectively.

Our results suggest that activating centromeric KIR may play a role in increasing susceptibility to pediatric and adolescent HL. Although these findings differ from those of prior studies of adult HL and other LN, activating KIR have been positively associated with other virally-driven malignancies, such as nasopharyngeal and cervical carcinomas; underlying mechanisms have yet to be discovered. Examination of compound KIR-cognate HLA ligand genotypes is in progress; additional results will be presented.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution