LMP2-Specific TCR Gene Therapy for Hodgkin Lymphoma - Design of a Dominant TCR Construct for Immunotherapy.

Abstract

Hodgkin lymphoma (HL) is the second most commonly diagnosed cancer in the 15–29 year old population. EBV-positive Hodgkin lymphoma typically demonstrates latency II antigen expression, characterised by loss of most EBV antigens except for the latent membrane protein (LMP) 1 and 2 and the EBNA-1 protein. LMP2 is expressed in Reed Sternberg cells and may serve as a target for antigen-specific immunotherapy. However, LMP2 is poorly immunogenic and it is often difficult to generate autologous LMP2-specific cytotoxic T lymphocytes (CTL) for adoptive immunotherapy. T cell receptor (TCR) gene transfer using retroviral vectors containing the TCR alpha and beta chain genes can reproducibly redirect the antigen specificity of a given population of T cells. Such an approach has been used here to generate LMP2-specific CTL independent of the immuno-competence of the patient. The goal of this study was to generate a retroviral TCR construct suitable for rapid and efficient production of LMP2-specific CTL. Retrovirally introduced TCRs compete with endogenous TCRs for a limited pool of CD3 molecules required for assembly of the TCR complex. Competition for CD3 molecules may limit surface expression of the introduced TCR resulting in a transduced T cell with poor functional avidity. In an attempt to generate a ‘highly competitive' LMP2-TCR the following modifications were made to the retroviral vector construct:i.nucleotide sequences were codon optimised for efficient translation in human cells;ii.the TCR chain constant regions were altered to contain murine sequences to enhance CD3 binding; andiii.the TCR alpha and beta chain genes were linked by a self-cleaving 2A sequence from the porcine teschovirus to enhance equimolar expression of both TCR chains.The unmodified HLA-A2-restricted LMP2-specific TCR was poorly expressed in primary human T cells, suggesting that it competed inefficiently with endogenous TCR chains for cell surface expression. Very few CD8+Vβ13+ T cells were detectable after LMP2-TCR transduction (up to 2.5% of viable CD3+ T cells, as detected by FACs analysis using monoclonal anti-Vβ13 antibodies), which included 1.9% CD8+ T cells expressing endogenous Vβ13+ TCRs as quantified in mock-transduced control cells. Poor expression of the wild type LMP2-TCR was consistently observed in independent transduction experiments. However, transduction with the modified LMP2-TCR construct resulted in cell surface expression of the TCR in 55–65% viable CD3+ T cells. HLA-A2/LMP2 pentamer binding was demonstrated in 36–39% CD8+ CTL cells immediately post transduction. The transduced cells showed peptide-specific IFNγ and IL2 production and killed target cells displaying the LMP2 peptide. Of major importance, expression of the introduced LMP2-TCR completely suppressed the cell surface expression of almost the entire repertoire of endogenous TCR combinations, including ‘mis-paired' TCRs in the transduced primary human T cells. ‘Mis-paired' TCRs contain an introduced alpha chain paired with an endogenous beta chain and vice versa. The antigen specificity of such mispaired TCRs generated after transduction is unknown and could lead to unwanted side effects. The design of vectors containing modified TCR sequences, which produce ‘dominant' TCRs may improve the efficacy of TCR gene therapy and reduce the risk of potential auto-reactivity of endogenous and ‘mis-paired' TCR combinations. We have shown that LMP2-specific T cells can be readily generated by TCR gene transfer with minimal risk of autoreactivity.