Session: Late-Breaking Abstracts
The incidence of endemic Burkitt lymphoma (eBL) is high in areas where Plasmodium falciparum (Pf) malaria is endemic, which suggests a role of malaria in eBL etiology. Previous data suggest that children with eBL are 5-12 times more likely to have elevated antibody titers to the whole schizont extract, a surrogate of exposure to malaria, compared to controls of comparable age and sex without eBL. However, the corollary studies to understand the role of protective malarial antibodies in eBL have not been conducted. We hypothesized that the risk for eBL might be different according to the breadth and strength of protective immunity to clinical malaria in children exposed to Pf malaria parasites. We investigated this hypothesis in children with and without eBL cases using samples from the National Cancer Institutes (NCI) Ghana Burkitt Lymphoma Study.
Cases were children aged 0-15 years enrolled at the Korle-Bu Teaching Hospital, Accra, Ghana, during 1965-1994. Controls were children enrolled contemporaneously from the same villages as the cases or children who were referred to Korle-Bu as BL but diagnosed with benign or a non-hematologic malignancy. Antibodies to recombinant Pf serine repeat antigen 36 (SE 36) and merozoite surface protein-1 (MSP-1), which are blood stage vaccine candidates, and antibodies to histidine-rich protein-II (HRP-II), an exposure antigen expressed during the blood stage, and the peptide 6NANP, which is a circumsporozoite protein (CSP) expressed in the pre-hepatic stage, were measured using sub-class-specific enzyme-linked absorbent immunoassays (ELISAs). Antibodies to tetanus toxoid were measured as an irrelevant antigen control. Markers were included if the within (W)- and between (B)-plate coefficients of variation for the sub-class-specific IgG results was <30% (Figure 1). The independent association of each malaria marker with eBL was determined by calculating the odds ratio (ORs) and 95% confidence intervals (95% CIs) using unconditional multivariable logistic regression adjusted for sex, age, calendar year, and for all the other malaria markers, which were hypothesized to be contributory.
In adjusted results, eBL was inversely associated with IgG1 seropositivity to SE36 (OR 0.54 [95% CI 0.34-0.86], p=0.01) and positively associated with HRPII (OR 1.47 [95% CI 1.06-2.02], p=0.019). The ORs for eBL were significantly decreased for low, medium, and high titers, but without a trend (0.44, 0.47, and 0.58 for low, medium, and high, respectively [ptrend=0.216]) (Figure 2: Odds ratios and 95% confidence intervals of association of eBL with different malaria markers). They increased significantly with increasing titers of IgG3 antibodies to HRPII (ORs 1.83, 1.91, to 2.25 for low, medium, and high titers, respectively [ptrend<0.002]) and showed a trend. Having antibodies to 6NANP was associated with eBL (OR 1.48 [95% CI 0.90-2.43]), but among the positives, having medium and high IgG3 antibodies to 6NANP as opposed to being sero-negative was associated with decreased risk of BL (ORs 0.79 and OR 0.60, respectively [ptrend=0.002]). Models with three markers (IgG1 to SE36, IgG3 to HRP-II, and IgG3 to 6NANP) predicted eBL better than models with just one of the markers.
These data suggest children with eBL in Ghana had decreased SE36 IgG1 and increased HRPII IgG3 antibodies compared to children without eBL from the same Pf endemic areas. These results fit with the hypothesis that eBL risk increases with greater exposure to Pf malaria parasites. They also provide the first confirmation of the hypothesis that antibodies elicited by antigens targeted by protective immunity might be protective for eBL. However, they also highlight that the relationship between eBL and Pf antibodies is complex as it appears to depend on whether the antibodies reflect exposure, protection, or both. A better understanding of the specific contribution of immune response to malaria in eBL risk should be the priority of efforts to discover a biomarker profile for eBL.
Disclosures: No relevant conflicts of interest to declare.