| dc.description.abstract | Etiological mechanism underlying endemic Burkitt lymphoma (eBL), remain unknown despite
the long-standing epidemiological link between Plasmodium Jalciparum and Epstein Barr
Virus (EBV) co-infections. Although past studies have linked malaria and EBV in eBL
etiology none has investigated the role of sickle cell trait (HbAS) and malaria transmission on
the quality of T-cell memory phenotype and function. Furthermore, there, is paucity of data on
the role of common infectious agents in eBL etiology. Previous studies~have shown an agedependent
deficiency in EBV-specific CD8+ T-cell-mediated IFN-y immunosurveillance in
children from malaria holoendemic regions. Further, deficiency in Epstein-Barr nuclear
antigen-l (EBNA-I)-specific IFN-y responses has been associated with eBL diagnosis in
children aged 4 to 9 years. EBNA-I is the sole antigen expressed by EBV-infected B cells as
well as EBV-associated turnours. This case-control study hypothesized that qualitative
differences in EBNA-I- specific T-cell phenotype and function as well as serology existed in
malaria-exposed children and, thus in part, explains their increased risk for eBL. Moreover,
the study examined the relationship between HbAS and EBV titers and how it relates to anti-
EBV immunity. Thus to test this hypothesis, healthy children from western Kenya with
divergent malaria exposure histories: Kisumu (n=24), a malaria holoendemic region and Nandi
(n=23), a hypoendemic region; were age-matched to eBL patients (n=31). T-cell memory
phenotype and function was evaluated by multiparameter flow cytometry against T-cell
lineage markers and functional phenotype based on IFN-y, IL-lO, IL-17 and PD-I expression
after ex vivo stimulation with EBNA-l peptides. Multiplex ELISA was used to determine
serological profiles. EBV titers were determined by qPCR, while PCR-based Restriction
Fragment Length Polymorphism (RFLP) was used to determine HbAS genotype. Pairwise
comparisons were done using Mann-Whitney U test while multiple comparisons were done
using Kruskal-Wallis tests. Logistic regression analysis was used to determine the association
between HbAS trait and viral titers while Chi square was used to test for differences in
proportions of HbAS genotypes. Results show that EBNA-l-specific IFN-y responses were
significantly lower in eBL and Kisumu compared to Nandi children (p=0.01). IFN-y was
generated from CD4+ effector memory T cells and a heterogeneous combination of CD8+ T
cells (jJ<0.0001). The frequencies ofPD-1 expressing T cells were low in eBL patients and
highest in Kisumu children than Nandi children for both CD4+ T cells (p=0.0237) and CD8+
T cells (p=0.0423). The expression of PD-I and IFN-y in response to EBNA-I was mutually
exclusive. Further, Kisumu and eBL children had higher cellular and plasma EBV titers
compared to Nandi children (p=0.000 1). Although HBAS genotype was more common in
Kisumu children than other groups (X~ 23.42, df 2, p<O.OOOl), it was neither associated with
EBV titers nor eBL (OR=0.5473, 95% CI, 0.2835-1.057, p=0.0996). This suggests that
EBNA-I-specific deficiencies in eBL may be due to a deletion of EBNA-l-specific T-cell
population as a consequence of T-cell exhaustion. These results show that HbAS trait does not
offer protective advantage against eBL and hence all children in malaria holoendemic regions
should be treated as vulnerable to eBL risk irrespective of HbAS genotype. Further, results
show that EBV peptide such as EBNA-I can be exploited further as a possible vaccine
candidate against EBV associated turnours. This study recommends a paradigm shift from use
of total EBV viral loads as a prognostic biomarker to use for cellular viral loads in clinical
management of eBL. In addition, there is need for development of novel immunotherapeutic
agents such as adoptive autologous EBNA-I specific T cells as well as PD-I mediated
blockade in clinical management of eBL. | en_US |
