Synergistic Tumor Immunity Induced by Chemotherapy and Agonist Anti-GITR Antibody.

Ligation of GITR (glucocorticoid-induced tumor necrosis factor receptor) can both co-stimulate effector CD4 and CD8 T cells (Teff) and abrogate suppression by CD4+foxp3+ regulatory T cells (Tregs). This may be beneficial for the purposes of tumor immunotherapy, and we and others have demonstrated that the agonist anti-GITR antibody DTA1 can enhance both vaccine-induced and naturally-arising tumor immunity in murine models (Turk et al, JEM 2004; Cohen et al, Cancer Res 2006). In this study, we assessed the efficacy of combining GITR ligation with cyclophosphamide (CTX), a cytotoxic chemotherapeutic with immunomodulatory properties, to treat established, poorly immunogenic tumors. C57BL/6 mice received 50,000 B16 melanoma intradermally and were treated on day 6 with CTX 250 mg/kg intraperitoneally (ip), followed 1 day later by 1mg DTA1 or control rat IgG ip. In repeated experiments, 0–20% of mice treated with DTA1 alone or CTX + IgG had long-term tumor-free survival, compared with 60–80% long-term tumor-free survival in mice treated with CTX + DTA1. Starting CTX + DTA1 treatment on day 10 led to 40% tumor-free survival, with no survivors seen with either treatment alone. Synergy was lost with lower doses of CTX, or when CTX was given prior to tumor inoculation, indicating a likely requirement of tumor cell death and cross-presentation of tumor antigens to T cells. Consistent with this, the proliferation and activation of naïve pmel-1 CD8+ T cells (specific for the melanoma antigen gp100) was significantly enhanced when transferred into B16-bearing mice treated 1 day earlier with CTX, compared with untreated mice. This was not simply due to homeostatic proliferation in a lymphopenic state, as no differences were seen when cells were transferred into mice without tumors. In addition, increased GITR expression on both Teff and Tregs was observed for up to 4 days after in vivo CTX treatment, particularly on the proliferating (Ki67+) fraction, providing a greater target for GITR ligation by DTA1. Analysis of T cell populations in spleen, tumor-draining lymph node (DLN), and tumor following CTX + IgG treatment showed a relative decrement in Treg frequency in the first week, followed by a strong rebound, such that by day 14 after CTX roughly 50–60% of the tumor-infiltrating CD4+ cells were foxp3+. This rebound was largely abrogated, however, in the CTX + DTA1-treated mice, with only 5–15% of tumor-infiltrating CD4+ cells expressing foxp3. This led to a dramatic increase in the ratio of CD8+ to CD4+foxp3+ in the tumor (40:1 vs. 5:1 for CTX + IgG-treated mice), with lesser increases seen in the spleen and DLN as well, without significant changes in overall cellularity. Granzyme B expression was also increased in CD8+ and, to a lesser extent, CD4+foxp3- T cells from CTX + DTA1-treated mice, demonstrating increased cytolytic potential by Teff. In sum, CTX + agonist anti-GITR antibody can induce rejection of an aggressive syngeneic tumor, at a stage when either agent alone is ineffective. This synergy likely involves enhanced cross-priming and co-stimulation of Teff, with concomitant decrease in tumor-infiltrating Tregs, leading to a more effective anti-tumor immune response. This combination warrants further evaluation as an immunotherapeutic strategy for cancer.