Background
Endemic Burkitt Lymphoma (eBL) is an aggressive B-cell non-Hodgkin lymphoma that is associated with endemic
Plasmodium falciparum malaria [
1]. Thus, the incidence of eBL correlates with the endemicity of
P. falciparum malaria [
1‐
3] and eBL incidence is highest in malaria endemic countries in sub-Saharan Africa [
4], where eBL cases account for 50–75% of childhood cancers in some countries [
5]. The role of malaria is supported by significant associations of eBL risk with high antibody titers of markers of long-term exposure to
P. falciparum infection [
6‐
8] and inverse associations with antibodies that are associated with protection from severe
P. falciparum infection [
6,
9]. In addition, there is support from indirect evidence based on inverse associations with on carriage of genetic variants that are associated with resistance to severe malaria morbidity [
10‐
12], especially the sickle cell trait [
13]. However, these results, although important because they are not affected by reverse causality, have not been consistent because they were non-significant in some studies [
14,
15] and null in at least one study [
16]. The conflicting results may be due to small and under-powered studies or reliance on hospital-based studies or selection bias of controls.
Whether the relationship between malaria and eBL is related to malaria morbidity and circulating malaria parasite burden and inflammation [
17], for which morbidity is a surrogate of, is unknown. The severity of clinical malaria (severe malaria anaemia, hyperparasitaemia, cerebral malaria, malaria prostration, moderate malaria, and mild malaria) is directly related to parasite burden and associated host response [
18], but the correlation between clinical malaria, which is a surrogate for uncontrolled parasite burden [
17], has not been investigated.
This paper reports an investigation to assess the patterns of age and selected malaria-related laboratory measures (parasite density, haemoglobin, platelet count, and white cell count (WBC) count) in children with eBL, asymptomatic parasitaemia/antigenaemia, and clinical malaria in Uganda, Tanzania, and Kenya using primary data from a case–control study or secondary data extracted from papers published in malaria endemic areas.
Discussion
The current study was done to evaluate whether eBL risk might be connected with high parasite density using acute malaria conditions as surrogates of moderate or high parasite density. The results show clustering of eBL and asymptomatic parasitaemia/antigenaemia in children aged > 5 years who have normal or near normal values of parasite density, haemoglobin, platelet, and WBC counts. These results suggest that eBL could be a complication of asymptomatic parasitaemia/antigenaemia and cast doubt on the idea that it is a complication of high parasite density or the associated high inflammation, as was originally hypothesized. High parasite density was a feature of three sub-clusters of clinical malaria, all found in children below 5 years. These early age-onset malaria syndromes were characterized by markedly abnormal values of parasite density, haemoglobin, platelet counts and WBC counts. These syndromes included children with severe malaria anaemia and hyperparasitaemia, those with cerebral malaria and moderate malaria, and those with malaria prostration and mild malaria and markedly abnormal values of some or all of parasite density, haemoglobin, platelet counts and WBC counts. The findings in this study are likely valid because ~ 9.5% of eBL cases and 6.7% of the controls in the EMBLEM study were 0–3 years, but even in this age range the children with eBL and asymptomatic parasitaemia/antigenaemia still showed malaria-related laboratory measures that were different from those in children with acute malaria conditions. Moreover, children with eBL had significantly lower parasite density than those with asymptomatic parasitaemia/antigenaemia suggesting that the similarity of eBL with asymptomatic parasitaemia/antigenaemia observed in overall results is established at a very early age. These results are consistent with a recent review by Quintana et al. [
24] where they propose an explanation of how malaria exposure may precipitate the malignant transformation of a B-cell clone that may progress to eBL. Together, the accumulating evidence suggests that eBL develops in children who control malaria parasitaemia well, perhaps due to acquired age-related immunity to malaria [
25], and that the capacity to control malaria parasitaemia precedes and continues after eBL onset.
Antigenic stimulation from malaria is widely accepted as an insult that leads to B cell proliferation in the germinal center and increases the chance of translocation of
c-
MYC into the vicinity of immunoglobulin enhancer elements and progression to eBL [
26]. However, it has remained unclear whether uncontrolled malaria parasite proliferation, i.e., high parasite density, plays a key part in antigenic stimulation from malaria that triggers eBL development. The analysis presented suggests otherwise because asymptomatic parasitaemia/antigenaemia is associated with well-controlled malaria infection. If so, then children with eBL have a well-controlled malaria infection at the time of their diagnosis. The results from EMBLEM are notable because previous analyses suggest that eBL cases were more likely to be exposed to heavy malaria, based on being more likely to live in a village near surface water, to report inpatient or outpatient malaria history > 13 months before enrollment, than geographically matched controls [
12], thereby underscoring the likelihood that immunity to malaria is likely to develop prior to eBL onset. The current analysis suggests that immunity to malaria appears to be established from a young age because children below 5 years with eBL had lower haemoglobin than children with asymptomatic malaria, consistent with heavy exposure to malaria, but significantly lower parasite density, consistent with a capacity to control malaria parasitaemia in young children. In view of the fact that malaria is the pre-eminent cause of early mortality in areas where eBL is common [
17], the overlap patterns observed support a speculation that eBL occurs in children who are adapted to heavy exposure to malaria and it may be a trade-off exchange for a high risk for death from acute malaria. Because children with eBL are less likely to carry the classical polymorphisms that protect from severe malaria [
10‐
12], such as the sickle cell trait [
12], the trade-off for the much rarer, albeit deadly at the individual level, like involves mechanisms of resistance to early malaria mortality that are currently unknown but worth investigating [
12].
Asymptomatic malaria is typically associated with low-, rather than high-parasite density malaria parasitaemia. Thus, these results also suggest that eBL risk may be associated with low density malaria infection. Furthermore, because an asymptomatic infection, particularly in children above 5 years, is also likely to be a clonally diverse infection [
27], these results raise the hypothesis that a clonally diverse malaria antigenic stimulation is related to eBL development. This hypothesis is consistent with previous findings that concurrent infection with multiple malaria genotypes was correlated with eBL [
28] and that eBL cases were more likely to have a higher
P. falciparum genetic diversity score than controls [
29]. This hypothesis is also consistent with findings in the EMBLEM study that children with eBL were less likely to report clinical malaria up to 12 months before eBL diagnosis and less likely to have detectable malaria parasite/antigens at enrolment [
19] because children aged > 5 years with clonally diverse malaria infection had a decreased risk for clinical malaria during a follow up of 90 or more days [
30]. The hypotheses that parasite density and genetic diversity are related to eBL could be tested in the era of next generation sequencing or proteomic technologies [
28,
29].
This study has some limitations, notably, reliance on cross-sectional data. Thus, the correlations should not be interpreted as evidence for causality. Second, the reliance on published literature to increase complement assessment in EMBLEM is a strength, but the literature may be biased or incomplete. For example, studies of asymptomatic parasitaemia/antigenaemia may under-sample children below 5 years, or studies of severe malaria may oversample children below 5 years, which would lead to erroneous mean age patterns. However, the results from the EMBLEM study, which enrolled representative, population-based eBL cases and controls and uniformly tested subjects for malaria-related laboratory measures, support the overall patterns. This study lacks direct measures of immunity to malaria, including anti-malaria antibodies, which limits the ability to compare the immune status of eBL cases with that of children with asymptomatic parasitaemia/antigenaemia or clinical malaria. The strengths of this study include the large sample size and detailed data from the EMBLEM study and published data about children with malaria from malaria endemic areas.
Acknowledgements
We thank the study population and communities for their participation. We are grateful to Mr. Josiah Magatti of Shirati Health Education and Development Foundation for his contribution to the design of the EMBLEM study, securing ethical permission to conduct the study in Tanzania, and leading the implementation of fieldwork in Tanzania up to the time of his passing in 2014. We thank Ms. Janet Lawler-Heavner at Westat Inc, (Rockville, MD, USA) and Mr. Erisa Sunday at the African Field Epidemiology Network (Kampala, Uganda) for managing the study. We are grateful to the leadership of the collaborating countries and institutions for hosting local field offices and laboratories and supporting the fieldwork. We thank Ms. Laurie Buck, Dr. Carol Giffen, Mr. Greg Rydzak and Mr. Jeremy Lyman at Information Management Services Inc. (Calverton, MD, USA) for coordinating data, and preparing data analysis files.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.