Improving T-Cell Therapy for Epstein-Barr Virus Lymphoproliferative Disorders

Abstract

More than 90% of the population is infected with Epstein-Barr virus (EBV), which is an enveloped herpesvirus. EBV infection in immunocompetent individuals causes a mild self-limiting illness, but the virus persists lifelong as a latent infection in B cells and spreads in the community by productive replication in B cells and oral epithelial cells. EBV latency varies in the number and type of EBV proteins expressed by B cells. In type 3 latency, all nine latency-associated EBV proteins are expressed. Such B cells express cell adhesion and costimulatory molecules rendering them immunogenic and susceptible to immune-mediated killing by EBV-specific cytotoxic T lymphocytes (CTLs). After hematopoietic stem-cell transplantation (HSCT) without functional CTLs, type 3 latency EBVinfected B cells proliferate unchecked. EBV-infected B cells cause a lymphoproliferative disorder that can progress to lymphoma with increased levels of EBV DNA detected by polymerase chain reaction (PCR). EBV-associated post-transplant lymphoproliferative disease (PTLD) occurs in less than 1% to 25% of HSCT recipients depending on the type of graft, the degree to which the graft is T-cell depleted, and the post-HSCT immunosuppression. Rituximab is an effective monotherapy even in established PTLD after HSCT, with response rates of 55% to 100%. However, this B-cell depleting antibody increases risk of infection, and PTLD can recur if EBV-specific T-cell immunity is not restored. Because of its immunogenicity and occurrence in the context of immunodeficiency, PTLD is highly amenable to immunotherapy with EBVspecific CTLs. In the article that accompanies this editorial, Icheva et al used donor-derived Epstein-Barr nuclear antigen 1 (EBNA1) –specific T cells to treat post-HSCT PTLD. CTLs were generated with antigenpresenting cells that were pulsed either with whole protein or synthetic overlapping peptides to elicit EBNA1-specific T cells, which were then selected by interferon alfa capture (Fig 1). The selected T cells were either transfused directly without in vitro expansion or cryopreserved for future use. These rapidly generated EBNA1 CTLs were effective in seven of 10 cases of EBV-associated PTLD. Immunotherapeutic strategies to restore EBV-specific T-cell function have been used for more than 15 years to improve the outcome of PTLD. Initially, unselected donor lymphocytes from the EBV seropositive stem-cell donors (assumed to have a high frequency of EBV-specific precursors) were used to restore the immune response to EBV. This simple strategy is readily available at any transplant center and yields responses of approximately 70%. However, unselected donor lymphocyte infusions carry significant risk of severe graft versus host disease. Because of this, several groups generated and infused donor-derived EBV-specific T cells in patients after SCT. More than 100 patients have received such EBV-specific T cells, either as prophylaxis or as treatment for EBV PTLD with complete remission rates also in the order of 70%. Thus far, to generate EBV-specific CTLs, investigators have used EBV-transformed B cells (lymphoblastoid cell lines [LCLs]) that express type III latency like PTLD. It takes 10 to 12 weeks to generate LCLs and sufficient EBV-specific CTLs for infusion and release testing. This prolonged manufacturing time limits the ready use of this therapy for the urgent treatment of EBVassociated PTLD. The importance of the study by Icheva et al is that it demonstrates that rapidly generated EBV-specific CTLs can control PTLD similarly to the traditionally manufactured EBVspecific CTLs. Other investigators have successfully reported the rapid manufacture of EBV-CTLs by using HLA-A2 restricted tetramer selection, immunomagnetic selection of interferon-gamma–secreting T cells after stimulation of donor mononuclear cells with a spectrum of class I and II EBV peptides or by using latent membrane protein 2 (LMP2), EBNA1, and BamH1 Z leftward reading frame 1 (BZLF1, an immediate early transcription protein responsible for switching between latent and lytic EBV infection) plasmid transfected dendritic cells to expand EBV-specific T cells. The novelty of the approach by Icheva et al is that they chose (more for regulatory rather than scientific reasons) to target only the immunosubdominant EBV antigen EBNA1 rather than the immunodominant EBV proteins EBNA3, EBNA2, or LMP2 expressed in PTLD. Nevertheless, EBNA1 was shown to be an effective target and, moreover, promises to broaden the applicability of this immunotherapy to other EBV type 1 and 2 latency tumors (including Burkitt’s lymphoma and Hodgkin’s lymphoma, respectively) because all EBVassociated cancers express EBNA1. However, there are still problems to be addressed in additional studies. First, the targeting of a single antigen can increase the likelihood of tumor immune escape. For this reason, the authors intend to target additional EBV antigens such as LMP2 in future studies. In addition, because EBV reactivation and lymphoproliferation usually occurs from endogenous virus, seropositive recipients of transplants from EBV-naive, seronegative donors are at greatest risk for EBV LPD. Although EBV-specific CTLs have been JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 31 NUMBER 1 JANUARY 1 2013