Abstract 4996: A real-time PCR-based approach to quantitatively assess tumor immune profiles and immune responses

Exciting breakthroughs in tumor immunology have led to the discovery and development of several promising immunotherapeutic strategies for cancer patients. Thus far, immune checkpoint inhibitors have been the most encouraging for solid tumors, however, treatment responses are observed in only a subset of patients and these appear to be primarily dependent on a tumor’s baseline immune profile. Additionally, tumors respond differently to immunotherapy and criteria to assess treatment responses are still being refined. We have identified a set of immune-related genes and developed a scoring system to profile the tumor’s immune status and assess immunological responses. In this study, we used a qRT-PCR-based approach to assess this panel and determine baseline tumor immune-profiles in an immunocompetent mouse model of Pten-null prostate cancer. We also assessed and compared tumor immune responses following androgen withdrawal (via surgical castration) and anti-PD-L1 immune checkpoint blockade. A total of 96 genes were selected for this focused panel which consisted of cell-type, immuno-responsive and housekeeping control genes. Core modules were designated based on functional gene associations which included antigen presentation, tumor inflammation, effector cells, immunomodulatory cells, immunosuppressive signaling and immune checkpoints. Core modules were further subdivided in to appropriately relevant sub-modules. A score for each submodule was calculated and a weighed score was then assigned for each core module. Tumor immunophenotypes based on core and sub-core immune scores (IS) were corroborated with immunohistochemical and/or flow cytometric analyses. Mouse castration-naïve prostate tumors exhibited low scores indicating a ‘”cold tumor” phenotype with little variation between individual mice. Androgen withdrawal induced cancer cell death in these tumors that in turn promoted immune cell infiltration. Core and sub-core IS of individual mice identified varied signatures reminiscent of immune excluded and inflamed “hot tumor” phenotypes. Short-term treatment with anti-PD-L1 blockade (clone D265A, mouse/IgG1 kappa) elicited an increased immune response signature in tumors from surgically castrated mice, but not in tumors from intact mice. Notably, discernable differences were noted among individual responders supporting the notion that responses to immune checkpoint may indeed be dependent on baseline tumor immune-profiles. In summary, androgen withdrawal