The biological basis and clinical correlates of the association between EBV-positive Hodgkin lymphoma and HLA class I

QR355 Virology;RC0254 Neoplasms. Tumors. Oncology (including Cancer)

Farrell, Katrina ; Jarrett, Ruth F.

University:University of Glasgow

Department:Institute of Infection Immunity and Inflammation

关键词: Hodgkin lymphomaEpstein-Barr VirusHuman leucocyte antigenPrognosis;   

Others  :  http://theses.gla.ac.uk/4900/1/2014Farellphd.pdf

来源: University of Glasgow

【 摘 要 】

Classical Hodgkin lymphoma (cHL) is one of the commonest lymphomas in the developed world, frequently affecting young adults. The majority of patients will be cured of their disease, but the toxicities of the therapy required to achieve this can lead to long-term morbidity in survivors. In addition, whilst most patients are cured, there remains approximately 15% -20% of patients who will not respond to primary therapy and may ultimately die of their disease.Approximately one-third of cases of cHL in the developed world are associated with the Epstein-Barr virus (EBV), where this association is believed to be causal. A ubiquitous herpesvirus, persistently infecting more than 95% of the world’s population, precisely how this virus causes malignant disease in a minority of immune competent hosts remains ill-understood. Epidemiological evidence points to inherited genetic factors. Long-recognised to have an association with the class I human leukocyte antigen (HLA) system, recent studies have confirmed that risk of EBV-associated cHL is related to an individual’s HLA-A* allotype, with HLA-A*01:01 being associated with increased risk of disease and HLA-A*02:01 being protective. Heterozygotes are observed to have an intermediate risk. HLA plays a central role in the recognition and cell killing of virally-infected or malignant cells by the cytotoxic T lymphocytes (CTLs) of the cell-mediated immune system. The exact mechanism whereby HLA-A* exerts its effect on risk of cHL unknown, but CTL responses to EBV in this context are hypothesised to be crucial. The CTL response to EBV is well-studied. Immunodominant epitopes restricted through common class I alleles have been described, many directed towards peptides derived from proteins expressed in the lytic cycle of viral infection. In spite of intensive study, no confirmed HLA-A*01:01-restricted EBV-specific CTL responses have been described, raising the possibility that absent or weak CTL responses specifically to EBV might lead to elevated risk of disease. However, particularly given the intermediate risk of disease seen in HLA-A*01:01 heterozygotes, it remains a possibility the HLA-A*01:01-associated risk might be due to qualitative or inhibitory changes to the EBV-specific immune response.The work of this thesis set out to address a number of specific questions regarding the role of HLA class I in the aetiology and clinical outcome of cHL. Firstly, whether an HLA-A*01:01 allele could modify the magnitude of the CTL response to HLA-A*02:01-restricted epitopes was examined. In a study of healthy adults examining CTL responses using interferon-γ ELISPOT, overall HLA-A phenotype did not significantly affect the EBV-specific CTL response restricted through HLA-A*02:01. However, exploratory analysis of cytokine levels in response to stimulation with EBV peptides demonstrated significantly higher secretion of IL-10 (with a nearly 10-fold difference), IL-17 and IL-5 in response to stimulation with EBV peptides in HLA-A*02:01/A*01:01 heterozygotes, compared to other HLA-A*02:01 phenotype groups. This suggests a possible effect of HLA-A*01:01 in HLA-A*02:01/A*01:01 heterozygotes which might begin to explain some of the HLA-associated differences in risk of developing EBV+ve cHL.Secondly, again in a study of healthy EBV-seropositive adults, and using sensitive methodologies, HLA-A*01:01-restricted EBV-specific CTL responses were sought, and, in an exploratory analysis, cytokine responses were examined. No HLA-A*01:01-restricted CTL responses to EBV were detected in this study, however, exploratory analysis demonstrated statistically significant differences in cytokine levels following simulation with EBV, with HLA-A*01:01 homozygotes generating much higher levels of IL-6.Lastly, given the importance of class I HLA in determining risk of developing EBV+ve cHL, a study of 424 patients with cHL was performed to determine if HLA-A*01:01 and A*02:01 alleles are a factor in determining clinical outcome. In this study, HLA-A*02:01 was associated with inferior overall survival (OS) and disease-specific survival (DSS) in EBV+ve cHL, and was independently prognostic in an adjusted analysis. Given the extremely poor outcomes seen in this study in HLA-A*02:01 carriers with EBV+ve disease (61.7% 10-year OS), it is possible that this group of patients is not currently being well-served by standard first-line therapy and may benefit from novel therapies.