Complete Tumor Responses in Lymphoma Patients Receiving Autologous Cytotoxic T Lymphocytes Targeting Epstein Barr Virus (EBV) - Latent Membrane Proteins

Abstract 956

The tumor cells of EBV-associated Hodgkin's Lymphoma (HL) and of some non-Hodgkins lymphoma (NHL) express the subdominant EBV antigens LMP1 and LMP2 (Type II viral latency), and these may serve as target antigens for immunotherapy. We hypothesized that cytotoxic T lymphocytes (CTLs) enriched for effector cells specifically targeting LMP antigens would have efficacy in patients with EBV+ lymphoma. We prepared antigen presenting cells (dendritic cells for the initial stimulation, EBV- lymphoblastoid cell lines (LCLs) for the subsequent) that expressed transgenic and inactive LMP1 (ΔLMP1) and/or LMP2, following transduction with the adenoviral vectors Ad5f35LMP2 (n=16) or Ad5f35ΔLMP1-I-LMP2 (n=27). All LMP-CTL lines were polyclonal as measured by flow cytometry CD4+ (mean 17+/−18%; range 1–92%) and CD8+ (mean 74 +/− 25%; range 1–99%) T-cells. Analysis of memory markers revealed mixed populations of CD45RA- CD62L- T-cells (mean 45+/−15%; range 31–63%) and CD45RA- CD62L+ T-cells (mean 34+/−5%; range 28–41%). CTL lines had specificity for CD4+ and CD8+ restricted LMP2 epitopes alone (median 1 epitope; range 0–7) or both LMP1 and LMP2 epitopes (median 3 epitopes; range 0–9) per CTL line, as determined using overlapping LMP1 and LMP2 peptide pools to stimulate the T cells in γ-IFN ELISPOT assays. 43 patients with EBV+ HL (n= 18) and NHL (n=25) have been treated and received between 4×10⁷/m² and 3.2×10⁸/m² T cells; 16 received LMP2 CTLs and 27 were given LMP1/2 CTLs. No immediate toxicity was observed. Within 1 to 4 weeks after CTL infusion, the numbers of LMP-specific T-cells in peripheral blood rose 2 to 70 fold, and elevated levels persisted for up to 3 months. Lymph node biopsies from 3 patients taken 3–6 months post CTL showed selective accumulation of LMP-specific T-cells in lymph nodes compared to peripheral blood. 21/22 high-risk and/or multiply relapsed patients who received LMP-CTL as adjuvant treatment remain in remission for a median of 2.5 years after CTL with a PFS of 80% at 3 years. 21 patients had detectable disease at the time of CTL infusion. Following treatment, 4 had progressive disease by 8 weeks, 1 is too early and 16 had clinical responses. The median duration of the clinical responses is 1.5 years with 1 stable disease, 3 partial responses and 13 complete responses. The 3 year progression free survival for patients treated with disease is 60%. We compared outcomes between patients who received LMP2 CTL alone versus LMP1 and 2-specific CTL. For high-risk patients treated with CTLs as adjuvant, the 3 year PFS is 70% for LMP2 CTL recipients versus 100% for recipients of LMP1/2-CTLs (p=0.06). The 3 year PFS in relapsed patients receiving LMP2-CTL was 50% versus 62% for recipients of LMP1/2 CTL (p=0.3). In conclusion, immunotherapy with CTLs targeting LMP antigens is well tolerated in patients with EBV+ lymphoma. Infused LMP-CTL accumulate at tumor sites and have induced complete and sustained clinical responses in 13/20 patients with relapsed disease. We are now implementing a simpler LMP-specific CTL expansion protocol to enable a definitive efficacy study.