Abstract Anti-tumor vaccines have demonstrated clinical benefit in patients with advanced prostate cancer, and the first vaccine approved by FDA as a treatment for human cancer was a vaccine targeting the prostate tumor antigen prostatic acid phosphatase (PAP). Plasmid DNA, as another method of antigen delivery, offers significant potential safety and manufacturing advantages over other immunization approaches. DNA vaccines have been approved by USDA as treatments for diseases in fish, horses, dogs, and non-human primates. However, DNA vaccines have generally been perceived as less immunologically potent than other methods of antigen delivery in human clinical trials, leading to efforts to increase their immunogenicity. Most of these efforts have focused on increasing the delivery of vaccine to antigen-presenting cells using different carriers or by electroporation. Our group has evaluated repetitive immunization with plasmid DNA vaccines in rodent models, and has translated this approach to early clinical trials in men with low-volume recurrent prostate cancer with a DNA vaccine encoding PAP. We have demonstrated that persistent Th1-biased PAP-specific T-cell immune responses can be elicited in men with prostate cancer. Moreover, the development of persistent immunity appeared to be associated with favorable changes in tumor growth rate. A randomized phase II trial using this approach is currently ongoing in men with minimal residual prostate cancer to determine whether immunization can delay the time to metastatic disease progression (NCT01341652). In an effort to further improve the immunogenicity of DNA vaccines, we have explored methods to increase antigen presentation by encoding epitopes with increased MHC class I affinity or by increasing the duration of target antigen expression. While such “optimized” vaccines elicited increased numbers of antigen-specific CD8+ T cells, these T cells had less anti-tumor activity in vivo. We determined that immunization with plasmid DNA elicited antigen-specific IFNγ-secreting T cells that led to increased expression of PD-L1 on antigen-expressing tumor cells. However, immunization with the vaccine encoding high affinity epitopes resulted in antigen-specific CD8+ T cells with higher PD-1 expression. In the presence of PD-L1-expressing tumors these CD8+ were less effective. The anti-tumor activity was restored in the presence of PD-1 or PD-L1 blockade. Conversely, immunization with plasmid DNA engineered to increase the duration of antigen expression led to CD8+ T cells with normal PD-1 expression but higher LAG-3 expression. The anti-tumor activity was similarly restored in the presence of LAG-3 blockade. We are currently studying the mechanisms by which different T cell regulatory ligands are expressed following immunization. Notwithstanding, such findings suggest that subtle differences in the method of immunization can affect these different regulatory mechanisms, and hence the choice of optimal combination strategies may depend on identifying the specific mechanisms of resistance to immunization. In the case of patients immunized with a DNA vaccine encoding PAP, we have recently found that PD-L1 expression on circulating tumor (CD45-EpCAM+) cells was increased following immunization, analogous to our murine studies. In addition, PAP-specific secretion of IFNγ and granzyme B by T cells obtained from patients after immunization was increased in the presence of PD-1/PD-L1 blockade. Finally, using a trans vivo delayed type hypersensitivity (tvDTH) assay, we observed that PAP-specific immune responses in T cells obtained from patients after immunization with a DNA vaccine encoding PAP could be “uncovered” with PD-1 blockade but not blockade with anti-LAG-3 or anti-TIM-3. Collectively, these findings suggest that the PD-1/PD-L1 pathway and mechanism of resistance is most relevant to a DNA vaccine encoding PAP, and the efficacy of this vaccine might be augmented with PD-1 blockade. A clinical trial testing this approach in men with advanced prostate cancer, using this DNA vaccine and pembrolizumab, delivered either concurrently or sequentially, is currently underway (NCT02499835). Citation Format: Brian T. Rekoske, Viswa T. Colluru, Douglas G. McNeel. DNA vaccines as treatment for prostate cancer - understanding mechanisms of resistance. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr CN04-03.