Association of HLA class II genes with clinical hyporesponsiveness to trivalent inactivated influenza vaccine in children.

BACKGROUND

The primary prevention measure for influenza infection has been the use of influenza vaccines. However, even when the vaccine and circulating strains are well-matched, some healthy children do not respond to the vaccine, likely due to a genetic basis for immune hyporesponsiveness. The primary objective of this study was to identify HLA class II genes associated with clinical hyporesponsiveness after trivalent inactivated influenza vaccine (TIV) in children.

METHODS

We conducted a case-control study nested within a retrospective cohort of children that were screened at birth for HLA-DR,DQ genotypes by the Diabetes Autoimmunity Study in the Young (DAISY) and were subsequently followed for up to 8 years by Kaiser Permanente, Colorado (KPCO). Hyporesponsiveness was clinically defined as the occurrence of influenza or influenza-like illness (ILI) in peak influenza weeks in children fully vaccinated with TIV. Each child with clinical hyporesponse (n=252) was matched to 4 randomly selected controls (n=1006) by age and season. Children with clinical hyporesponse to TIV were identified using the Kaiser electronic clinical and immunization databases. Fully vaccinated children within the KPCO-DAISY cohort who did not have a diagnosis of ILI during the entire influenza season were eligible to be controls for that season. Class II HLA-DRB1 and HLA-DQB1 genes were the primary exposure variables. We used conditional logistic regression to calculate the matched odds ratios.

RESULTS

In non-Hispanic white children, HLA-DR7/4,DQB1 0302 genotype was significantly associated (OR=5.15; 95% CI=1.94, 13.67; p=0.001), while in Hispanic children, HLA-DRB1 15 or 16 allele (OR=0.31; 95% CI=0.14, 0.69; p=0.004) and HLA-DR7/Y (DRB1 11, DRB1 13 and DRB1 14) genotype (OR=5.84; 95% CI=1.68, 20.28; p=0.006) were significantly associated with clinical hyporesponsiveness after TIV.

CONCLUSIONS

HLA class II genes are associated with clinical hyporesponsiveness to TIV. This finding is important as it may help identify a group of children who are not protected by the commonly used TIV and may require alternative preventive strategies.