Gene-based vaccines for immunotherapy of prostatecancer - lessons from the past

摘要

Gene-based vaccination in its current mode of application is effective in breaking tolerance to a self- or tumor- associated antigen, but the response is narrow and restricted to few of the potential epitopes due to immunodominance. In cancer, immunodominance carries the risk of inefficient immune surveillance due to loss of MHC alleles or point mutations in the recognized sequences. We have found that a T cell response to sub-dominant epitopes can be primed with transfected dendritic cells in which the newly expressed antigen is purposefully targeted for proteasomal degradation. Beginning in May 1998, we performed a phase I/II clinical trial for immunotherapy of prostate cancer that targeted the prostate-specific membrane antigen (PSMA). The primary objective of the study was to determine the safety of the described vaccines after repeated intradermal injections (Mincheff et al., 2000a; Mincheff et al., 2000b), since using PSMA as a target could be seriously offset by the development of autoimmunity (Gilboa, 1999b; Overwijk and Restifo, 2000). So far, six years since the study has begun, no patient has experienced any short- or long-term side effects, including anti-DNA antibody. Twenty-nine patients from this random population were treated solely by immunotherapy. Eighteen of them had biochemical recurrence following radical prostatectomy and eleven responded to the therapy with a PSA drop exceeding 50% of pre-therapy value. Patients with advanced disease and distant metastases were not influenced by the immunotherapy despite the fact that they all showed signs of T cell immunity towards PSMA. We found, however, that the post-vaccination T cell response was directed against only two of the potential 4 PSMA epitopes that had high affinity for binding. At least in vitro, priming with one of our vaccines led to a poly-epitope response. Unfortunately, even in such instances, consequent exposure to poly-epitope expressing dendritic cells during re- immunization led to selection of an immunodominant clone. To alleviate immunodominance and decrease tumor evasion due to loss of antigenic determinants, a poly-epitope T cell response would need to be maintained. Ensuring such a cytotoxic T cell response, therefore, would require either construction of separate epitope encoding vectors for boosting, an approach with limited therapeutic application, or identifying conditions during boosting that would restrict immunodominance. CD4 T cell depletion, GITR-L signaling or CTLA-4 all show promise in achieving this goal.