Introduction: Infection with Plasmodium falciparum (Pf) malaria is widely accepted as a risk factor for endemic Burkitt lymphoma (eBL), but whether children with eBL are more likely to have detectable blood-stage Pfmalaria parasites and/or report a history of malaria morbidity compared to location-matched control children without eBL is unknown. We investigated this hypothesis in children with eBL (cases) compared to location-matched children without eBL (controls) from the National Cancer Institute (NCI)-sponsored EMBLEM Study.
Methods: Cases were children with eBL aged 0-15 years presenting to two hospitals in northern Uganda from 11/2010 to 07/2014 with histologically proven, untreated eBL. Controls were children with similar malaria exposure in the region, including children attending village health centers for minor complaints (pilot health-center controls [PHCs]), at home in 12 randomly selected villages (pilot population controls [PPCs]),and at home in 88 randomly selected villages with matching for the age- and sex-distribution of eBL cases (matched population controls [MPCs]). Cases and controls provided a venous blood sample and questionnaire information on exposure to Pf malaria parasites (mosquito bed net use, insecticide sprays, proximity to a river/swamp, history of treatment for malaria). Blood-stage malaria was evaluated microscopically with giemsa-stained thin and thick blood films and with a commercially available histidine-rich protein (HRPII) antigen-based rapid diagnostic test. The log of thick-film malaria parasite count in cases and controls was compared using the Students t-test. Associations were evaluated by calculating odds ratios (ORs) and 95% confidence intervals (95% CIs) using unconditional logistic regression adjusting for sex, age, ownership/ use of mosquito bed net, and in- or out-patient treatment for malaria.
Results:We studied 280 eBL cases (61% male, mean + SD age 7.9 + 3.4 years) and 1619 controls including 171 PHCs (37% male, mean + SD age 7.3 + 4.0 years), 1005 PPCs (48% male, mean + SD age 7.0 + 4.1 years) and 443 MPCs (56% male, mean + SD age 7.5 + 3.3 years). Overall, eBL cases were less likely to own a mosquito bed net than controls (46% versus 67% - 79% in controls, p<0.0001), but among those who owned a mosquito bed net, eBL cases were more likely to have used it the previous night (42% versus 21% - 31% in controls, p<0.0001). Blood-stage malaria infection was detected less frequently in cases compared to controls (Figure 1). In adjusted results, eBL cases were less likely to have current blood-stage malaria infection based on the thin film (OR 0.42 [95% CI 0.26-0.67], p<0.0001) or thick film (OR 0.55 [95% CI 0.38-0.80], p=0.001) and less likely to have had recent infection based on the HRPII rapid diagnostic test (OR 0.31 [95% CI 0.22-0.44], p<0.0001) using all controls combined, with similar results using separate control groups (OR 0.28 – 0.60). Blood-stage malaria parasite count was 0.88 log lower in parasitemic eBL cases than controls (2.24 log versus 3.12 log, p=0.0003). With all controls combined and adjusting for HRPII antigen, the risk of eBL was inversely associated with female sex (OR 0.65 [95% CI 0.46-0.90], p=0.011), ownership of mosquito bed net (OR 0.03 [95% CI 0.01-0.07], p<0.0001), and any inpatient admission for severe malaria (OR 0.56 [95% CI 0.39-0.80], p=0.001) or outpatient treatment for moderate malaria (OR 0.47 [95% CI 0.32-0.69], p<0.0001). The risk of eBL was directly associated with older age (OR 3.1 [95% CI 1.9-5.00] and OR 2.6 [95% CI 1.60-4.40], for 5-9 and 10-15 years versus 0-4 years) and not sleeping under a mosquito net the previous night (OR 10.9 [95% CI 4.34-27.3], p<0.0001). Socioeconomics, spraying insecticides, and proximity to river/swamp did not influence the results.
Conclusions: Cases of eBL were unlikely to have Pf malaria parasitemia despite high exposure to mosquitoes and low ownership of bed nets. These results reject the hypothesis that eBL is associated with current or recent blood-stage Pf malaria. They support an alternative hypothesis that children with eBL have superior immunological control of blood-stage infection (Pf elite controller phenotype) and that eBL might be an accident of robust immunological control of blood-stage malaria infection. Further studies are needed to characterize the molecular spectrum of Pf parasites and other mechanisms that drive eBL genesis.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.