Search

Abstract 5118 Introduction Efficient methods for the reliable in vitro expansion of tumor-reactive T cells will surely broaden the applicability of adoptive T cell therapy in cancer. In this study we investigated the antigen-independent stimulation and expansion of human T cells in peripheral blood mononuclear cells (PBMC) and in long-term cultured tumor-reactive CD8+ T cell lines using superparamagnetic beads coated with antibodies to CD3 and the costimulatory molecules CD28 and CD137. Methods T cell numbers were measured in healthy donor PBMC after in vitro stimulation with Dynabeads® coated with CD3/CD28/CD137 versus Dynabeads® coated with CD3/CD28 (all beads +/- 100 U/mL IL-2) versus IL-2 alone at different bead/cell ratios (3:1, 1:1). Expansion was also analyzed in human renal cell carcinoma-reactive CD8+ T cell lines after restimulation with tumor cells (weekly), CD3/CD28 beads and CD3/CD28/CD137 beads, respectively (bead/cell ratio of 1:5, 100 U/mL IL-2 added). Expanded T cell lines were phenotyped for expression of activation, differentiation and homing molecules (i.e. CD27, CD28, CD45RA, CD45RO, CD57, CD62L, CD137, CCR7) and were also tested for function. Results T cells in PBMC showed an increased expansion rate of up to 17-fold during a 2-week culture period using beads with IL-2 added versus IL-2 alone (p Conclusion Antigen-independent in vitro expansion of T cells in PBMC was equally efficient using CD3/CD28 beads or CD3/CD28/CD137 beads, respectively. In contrast, we observed an increased growth rate for tumor-reactive CD8+ T cell lines when activated with CD3/CD28/CD137 beads compared to CD3/CD28 beads. Antitumor reactivity of T cell lines was maintained during the antigen-independent stimulation step. Bead activation was associated with increased expression of the lymph node homing receptor CD62L on antitumor CD8+ T cell lines, which indicates a central memory phenotype. Our data suggest that the conjugation of anti-CD137 antibodies to the traditionally used CD3/CD28 beads improves their expansion capacity for antitumor CD8+ T cell lines. Disclosures No relevant conflicts of interest to declare.

Although current allo-transplantation therapy can induce considerable graft-versus-tumor (GvT) effects in RCC patients, it is also accompanied by severe, even life-threatening side effects, mainly due to graft-versus-host disease (GvHD). Efforts aiming to improve the specificity and efficiency of allogeneic cell therapy in this disease (e.g. specific donor lymphocyte infusion or vaccination) will certainly benefit from the identification of potential anti-tumor effector mechanisms and their corresponding target structures. We recently demonstrated that RCC-reactive cytotoxic T-lymphocyte (CTL) clones can be isolated from peripheral blood of healthy donors matched with the RCC stimulators for HLA-class I. These CTL were found to recognize a broad panel of RCC antigens with restricted or ubiquitous tissue expression (Doerrschuck A, et al., Blood July 1, 2004, Epub). We now extended our analyses on peripheral blood mononuclear cells (PBMC) of further HLA-matched healthy sibling (1) and unrelated individuals (4) and compared these results with available autologous patient PBMC. While mixed lymphocyte/tumor-cell culture (MLTC) responders derived from allogeneic donors showed a robust antigen-dependent proliferation over several weeks, a weak if any proliferative response was seen with autologous MLTC populations. By analysing the fine specificity of MLTC-derived clonal CTL the majority of allogeneic effectors recognized exclusively their RCC stimulators, but not corresponding lymphoblastoid-cell lines or natural killer target K562. These CTL were restricted by various HLA-A, -B or -C molecules. We further isolated CTL clones that exhibit an extraordinary strong recognition of RCC and various epithelial tumor-cell lines. Antibody blocking experiments provided clear evidence that these CTL are restricted by a not yet defined HLA-Ib molecule and, simultaneously, by a NKG2D-dependent mechanism. Other rapidly proliferating CD3+ CD8+ CTL clones were obtained that showed a non-HLA-restricted reactivity against RCC and a minor but consistent reactivity against targets with low or absent HLA-class I expression (e.g. K562). In conclusion, our results demonstrate that a heterogeneous panel of RCC-reactive HLA-Ia/Ib-restricted CTL can be isolated from PBMC of HLA-class I-matched healthy individuals. Alternatively, CTL recognizing RCC in a non-HLA restricted manner can be obtained. Our observations might reflect the superior ability to activate and expand RCC-directed T cells from PBMC of allogeneic healthy donors compared to the autologous setting. At this point, we cannot conclude whether these various CTL populations contribute to effective anti-RCC immune responses occurring in vivo. Answering this question will certainly require to identify further CTL-defined target structures at the molecular level. This would allow us to analyse the expression of candidate antigens and the frequency of specific CTL in RCC patients after allogeneic blood stem-cell transplantation, and to correlate these findings with clinical GvT and GvH events.