Communications Biology (Jan 2024)

Human leukocyte antigen-DQA1*04:01 and rs2040406 variants are associated with elevated risk of childhood Burkitt lymphoma

  • Zhiwei Liu,
  • Yang Luo,
  • Samuel Kirimunda,
  • Murielle Verboom,
  • Olusegun O. Onabajo,
  • Mateus H. Gouveia,
  • Martin D. Ogwang,
  • Patrick Kerchan,
  • Steven J. Reynolds,
  • Constance N. Tenge,
  • Pamela A. Were,
  • Robert T. Kuremu,
  • Walter N. Wekesa,
  • Nestory Masalu,
  • Esther Kawira,
  • Tobias Kinyera,
  • Isaac Otim,
  • Ismail D. Legason,
  • Hadijah Nabalende,
  • Herry Dhudha,
  • Leona W. Ayers,
  • Kishor Bhatia,
  • James J. Goedert,
  • Nathan Cole,
  • Wen Luo,
  • Jia Liu,
  • Michelle Manning,
  • Belynda Hicks,
  • Ludmila Prokunina-Olsson,
  • George Chagaluka,
  • W. Thomas Johnston,
  • Nora Mutalima,
  • Eric Borgstein,
  • George N. Liomba,
  • Steve Kamiza,
  • Nyengo Mkandawire,
  • Collins Mitambo,
  • Elizabeth M. Molyneux,
  • Robert Newton,
  • Ann W. Hsing,
  • James E. Mensah,
  • Anthony A. Adjei,
  • Amy Hutchinson,
  • Mary Carrington,
  • Meredith Yeager,
  • Rainer Blasczyk,
  • Stephen J. Chanock,
  • Soumya Raychaudhuri,
  • Sam M. Mbulaiteye

DOI
https://doi.org/10.1038/s42003-023-05701-5
Journal volume & issue
Vol. 7, no. 1
pp. 1 – 9

Abstract

Read online

Abstract Burkitt lymphoma (BL) is responsible for many childhood cancers in sub-Saharan Africa, where it is linked to recurrent or chronic infection by Epstein-Barr virus or Plasmodium falciparum. However, whether human leukocyte antigen (HLA) polymorphisms, which regulate immune response, are associated with BL has not been well investigated, which limits our understanding of BL etiology. Here we investigate this association among 4,645 children aged 0-15 years, 800 with BL, enrolled in Uganda, Tanzania, Kenya, and Malawi. HLA alleles are imputed with accuracy >90% for HLA class I and 85-89% for class II alleles. BL risk is elevated with HLA-DQA1*04:01 (adjusted odds ratio [OR] = 1.61, 95% confidence interval [CI] = 1.32-1.97, P = 3.71 × 10−6), with rs2040406(G) in HLA-DQA1 region (OR = 1.43, 95% CI = 1.26-1.63, P = 4.62 × 10−8), and with amino acid Gln at position 53 versus other variants in HLA-DQA1 (OR = 1.36, P = 2.06 × 10−6). The associations with HLA-DQA1*04:01 (OR = 1.29, P = 0.03) and rs2040406(G) (OR = 1.68, P = 0.019) persist in mutually adjusted models. The higher risk rs2040406(G) variant for BL is associated with decreased HLA-DQB1 expression in eQTLs in EBV transformed lymphocytes. Our results support the role of HLA variation in the etiology of BL and suggest that a promising area of research might be understanding the link between HLA variation and EBV control.