Abstract
Background: Although allogeneic graft versus tumor responses represent the most potent form of immunotherapy, recurrent malignancy remains a major cause of post-transplant mortality. Thus, strategies to enhance GVT are needed to improve outcomes. Dendritic cell (DC) vaccines can potently enhance anti-tumor immunity in the autologous setting. The impact of alloreactivity on DC vaccine responses is not known, but the potency of the GVT effect suggests that mild alloreactivity might enhance vaccine responses to third party antigens.
Methods: To determine the effect of a minor histocompatibility mismatch on DC vaccine responses, lethally irradiated female C3H.SWxC57BL/6 and C3H.SW recipients were transplanted with T cell depleted (TCD) C3H.SW bone marrow (BM). On days 14 and 28, recipients received donor-type lymph node cells containing mature T cells (DLI, 1 x 10^6, 5 x 10^6, or 10 x 10^6 IV). To immunize against the male antigenic complex (HY), BM-derived, anti-CD40 activated donor strain male DC were administered IP at the same time as the DLI. At Day 35, HY specific immune responses were measured using interferon γ ELISPOT against the dominant (UTY) and subdominant (SMCY) class I epitopes and dominant class II epitope (DBY). In some experiments, rhIL-7 and rhIL-15 were administered (5 mcg/day IP) from day 14 to 35.
Results: In syngeneic recipients, robust HY responses were observed at all DLI doses, consistent with rapid T cell reconstitution via the thymus. Allogeneic recipients demonstrated mild weight loss and decreased splenic cellularity as the DLI dose was increased, consistent with the development of mild GVHD. However, even recipients receiving the highest DLI dose recovered completely and had none of the major physical symptoms of GVHD. Increasing DLI dose from 1 x 10^6 to 10 x 10^6 paradoxically led to a marked decrease in the median total number of T cells/spleen responding to UTY, SMCY, and DBY (9196 at 1 x 10^6 vs 897 at 10 x 10^6, p = .01; 15662 vs 2854, p= .02; 10257 vs 4227, p = .02). To test whether the presence of alloreactive T cells were necessary for this effect, C3H.SW minor antigen reactive T cells were deleted from B6-derived DLI. TCD B6 BM was transplanted into C3H.SWxC57BL/6 mice and at 5 weeks LN cells were collected and transferred as DLI. Strikingly, transfer of 10 x 10^6 tolerized DLI did not result in a loss of HY vaccine responses when compared to transfer of non-tolerized DLI (UTY, 3888 vs 21075, p= .05; SMCY, 7135 vs 21126, p = .018; DBY, 7117 vs 29201, p= .006). Although IL-7 and IL-15 can potently enhance DC vaccine responses in the syngeneic setting, these cytokines did not improve vaccine responses in allogeneic recipients.
Conclusions: Using a minor mismatched allogeneic transplant model, we have shown that mild GVHD can diminish the ability to respond to third party antigens expressed on activated DC vaccines. This suggests that the potency of GVT does not result from the alloreactive milieu, but rather from the nature of the antigens targeted. The mechanisms involved in the suppression of DC vaccine responses in allogeneic recipients are currently under study. These findings have important implications for the use of tumor antigen expressing vaccines as a modality to improve the graft versus tumor effect.
Author notes
Corresponding author
