Effect of HIV infection on the acute antibody response to malaria antigens in children: an observational study
Muema DKM., Ndungu FM., Kinyanjui SM., Berkley JA.
Abstract Background In sub-Saharan Africa, the distributions of malaria and HIV widely overlap. Among pregnant and non-pregnant adults, HIV affects susceptibility to malaria, its clinical course and impairs antibody responses to malaria antigens. However, the relationship between the two diseases in childhood, when most deaths from malaria occur, is less clear. It was previously reported that HIV is associated with admission to hospital in rural Kenya with severe malaria among children, except in infancy. HIV-infected children with severe malaria were older, had higher parasite density and increased mortality, raising a hypothesis that HIV interferes with naturally acquired immunity to malaria, hence with little effect at younger ages (a shorter history of exposure). To test this hypothesis, levels of anti-merozoite and schizont extract antibodies were compared between HIV-infected and uninfected children who participated in the original study. Methods IgG responses to malaria antigens that are potential targets for immunity to malaria (AMA1, MSP2, MSP3 and schizont extract) were compared between 115 HIV-infected and 115 age-matched, HIV-uninfected children who presented with severe malaria. The children were classified as high and low responders for each antigen and assigned antibody-response breadth scores according to the number of antigens to which they were responsive. A predictive logistic regression model was used to test if HIV was an effect modifier on the age-related acquisition of antibody responses, with age as a continuous variable. Results Point estimates of the responses to all antigens were lower amongst HIV-infected children, but this was only statistically significant for AMA1 (P = 0.028). HIV-infected children were less likely to be high responders to AMA1 [OR 0.44 (95%CI, 0.2-0.90) P = 0.024]. HIV was associated with a reduced breadth of responses to individual merozoite antigens (P = 0.02). HIV strongly modified the acquisition of antibodies against schizont extract with increasing age (P < 0.0001), but did not modify the rate of age-related acquisition of responses to individual merozoite antigens. Conclusions In children with severe malaria, HIV infection is associated with a lower magnitude and narrower breadth of IgG responses to merozoite antigens and stunting of age-related acquisition of the IgG antibody response to schizont extract.