Prostate cancer (PCa) is the most commonly diagnosed malignancy in the U.S. in men and is the second leading cause of cancer-related mortality. While androgen axis directed therapies such as abiraterone and enzalutamide (ADT) have become standard of care for men with metastatic prostate cancer, disease progression to metastatic castration-resistance prostate cancer (mCRPC) remains inevitable. Immune checkpoint blockade (ICB) of PD-1/PD-L1 with monoclonal antibodies have produced durable anti-tumor responses in subsets of patients across a multitude of cancers. However, these approaches have been largely unsuccessful in mCRPC clinical trials. These data suggest that additional immunosuppressive mechanism(s) are at play to drive resistance to ICB in mCRPC. PTEN loss-of-function (LOF), which occurs in 50-75% of mCRPC patients, is known to correlate with disease progression and poor clinical outcomes. PTEN LOF leads to hyperactivation of the PI3K pathway and has been shown to induce an immunosuppressive tumor microenvironment (TME), resulting in a paucity of T cell infiltration and de novo resistance to ICB. Thus, understanding the mechanism of PTEN LOF-induced immunosuppressive TME is critical to developing novel therapeutic approaches to eradicate these tumors.

Prior in vitro studies have shown that the androgen/AR signaling axis can suppress the expression of an additional B7 superfamily member, B7-H3 (CD276). However, IHC analysis on human tissues has demonstrated that B7-H3 expression positively correlates with Gleason score and progression towards mCRPC. Recent studies have also demonstrated activation of PI3K/AKT/mTOR signaling led to the upregulation of B7-H3 in non-small cell lung cancer. By elucidating how B7-H3 is regulated, and likewise functions in the context of PCa disease progression (PTEN LOF; ADT), novel immuno-oncology strategies can be applied in order to effectively treat PTEN-deficient mCRPC. The central hypothesis of this proposal is that PTEN LOF enhances B7-H3 expression via hyperactivation of the PI3K pathway, leading to an adaptive resistance mechanism that results in an immunosuppressive, pro-tumorigenic TME in PTEN-deficient mCRPC.

Aim 1: Elucidate the mechanisms by which PTEN and AR signaling regulate B7-H3 expression.

Aim 2: Assess the impact of tumor cell intrinsic and host B7-H3 knockdown on the immune TME in PTEN-proficient vs. PTEN-deficient PCa murine models.

Aim 3: Determine the anti-tumor efficacy of targeting of B7-H3, singly and in combination with androgen deprivation therapy and/or PD-1 inhibition in PTEN-proficient vs. PTEN-deficient castrate-sensitive and castrate-resistant murine PCa models.

Aim 1: Our hypothesis for this aim is that PTEN LOF leads to hyperactive PI3K signaling and decreased AR signaling (via reciprocal feedback), and together these signaling pathways drive B7-H3 expression. As a first step towards elucidating the mechanistic link between PTEN LOF, low AR and B7-H3 expression, we will initially utilize a panel of human PCa cell lines of varying PTEN and AR status. Following modulation of PTEN expression via both transient siRNA and stable CRISPR/CAS9-mediated knockdown approaches, we will measure B7-H3 expression at the mRNA and protein level. To determine whether hyperactivated PI3K signaling downstream of PTEN loss drives B7-H3 expression, we will also assess B7H3 mRNA and protein levels in the presence or absence of copanlisib (pan-PI3K inhibitor). To elucidate whether PTEN loss regulates B7-H3 expression via transcriptional, post-transcriptional, translational or post-translational mechanisms, we will perform qRT-PCR, actinomycin D (DNA-dependent RNA Pol II inhibitor) cycloheximide (protein synthesis inhibitor) and MG-132 (proteasome inhibitor) experiments, respectively. We will also incorporate culture conditions that mimic castration (charcoal stripped serum) or hyperactivated AR (R1881) signaling to elucidate the relevant contribution of AR signaling to regulation of B7-H3 expression downstream of PTEN loss.

To validate our bioinformatic findings demonstrating a negative correlation of AR signaling and PTEN expression with B7-H3 expression, we will retrospectively evaluate PTEN, PSA and B7-H3 expression by immunohistochemistry (IHC) from 100 primary prostate cancer specimens, and correlate with low vs. high Gleason scores. In addition, we will evaluate B7-H3 expression in pre- vs. 4-week on-treatment metastatic biopsies and in CD8+ T cells from patient matched PBMCs from an investigator-initiated trial (IIT) of rucaparib and nivolumab in 12 mCRPC patients, which was prematurely terminated due to lack of efficacy. The goal of these biomarker studies in metastatic biopsies is to determine whether the observed lack of efficacy is due in part to increased B7-H3 expression following treatment (relative to baseline biopsy). Taken together, these studies will shed light on the mechanism(s) for regulation of B7-H3 expression in PTEN-deficient PCa.

Aim 2: B7-H3 is constitutively expressed on immune cells, specifically antigen-presenting cells (APCs), and APC-expressed B7-H3 has been reported to promote tumor progression by inhibiting CD8+ T and NK cell function. Furthermore, a recent study exploring the role of intrinsic/host expression of B7-H3 in ovarian cancer models found that intrinsic B7-H3 expression, and not APC-expressing B7-H3, regulated tumor progression by attenuating the expansion and cytotoxicity of CD8+ T cells. The central hypothesis for this aim is that B7-H3 expression within both tumor cells and host immune/stromal cells drives immunosuppression within the TME in PTEN-deficient PCa. As a first step towards elucidating the mechanistic role of B7-H3 within the global TME, we will utilize our isogenic PTEN Myc-Cap and B6-Myc syngeneic models to survey B7-H3 expression within tumor cells and immune cells, as a function of PTEN status. Given our preliminary findings that B7-H3 is upregulated in PTEN-deficient PCa models relative to PTEN-proficient PCa models, we will knockout B7-H3 expression via CRISPR/CAS9 deletion in syngeneic PTEN isogenic Myc-CAP and B6-Myc cells in FVB and C57BL6 backgrounds, respectively, and assess the impact on the tumor immune landscape via flow cytometry to look for global changes across multiple immune cell lineages and their functional/activation states. The data collected will be analyzed by conventional flow cytometry software such as FlowJo and SPADE. Conversely, we will also obtain C57BL6 mice deficient for B7-H3 (CD276-/-) to determine the relative contributions of tumor cell intrinsic B7-H3 vs. host B7-H3 to the immune landscape within PTEN-proficient and PTEN-deficient syngeneic models. Furthermore, we will also perform cytokine array profiling in vitro to assess changes in the secretome as a function of B7-H3 knockdown in PTEN-proficient vs. PTEN-deficient Myc-CAP and B6-Myc cells. To assess the relative impact of B7-H3 knockdown in tumor cell intrinsic vs. tumor extrinsic gene expression programs, we will also perform parallel bulk RNA-seq on PTEN isogenic cell lines (+/- B7-H3) and their corresponding engrafted tumors in vivo. To elucidate spatial relationships of B7-H3 expressing tumor and/or immune cells within the TME, we will also perform multiplex immunofluorescence on harvested tumors and stain for tumor infiltrating lymphocytes (TILs) (CD45+ cells: CD4+ cells, CD8+ cells, CD4+/FOXP3+, CD8+/FOXP3+ Treg cells, CD19+ B cells), macrophages (CD11b+, F4/80+) and myeloid derived suppressive cells (CD11b+, GR1+). Taken together, these studies will shed light on the relative impact of B7-H3 expression on the tumor immune microenvironment in PTEN-proficient vs. PTEN-deficient PCa.

Aim 3: In this aim, we propose immuno-oncology (IO) combination strategies that build off of our prior observations that degarelix treatment (chemical castration) induces an immune infiltrate within the immune desert tumor microenvironment in syngeneic Myc-CAP transgenic mice, and increases PD-L1 expression within CD45- tumor cells and CD45+ immune cells (particularly tumor-associated macrophages) in Myc-CAP syngeneic tumors. However, castration-induced increased immune infiltration and PD-L1 expression within the TME is insufficient to sensitize this tumor model to PD-1 blockade, suggesting additional mechanism(s) of immunosuppression are contributing to ICB resistance. The central hypothesis of this aim is that targeting B7-H3 will reprogram the immunosuppressive TME, thereby enhancing responsiveness of PTEN-deficient PCa to castration + PD1 therapy, and delaying time to development of CRPC.

We will evaluate the following parallel therapeutic strategies to enhance T cell infiltration within the TME: 1) anti-B7-H3 alone; 2) anti-B7-H3 + chemical castration (Degarelix); 3) anti-B7-H3 + Degarelix + PD-1 antibody, and compare treatment cohorts to Degarelix alone or Degarelix + PD-1 antibody. PTEN-proficient vs. PTEN-deficient Myc-CAP mice treated with the listed IO combinations, singly and in combination, will be tested under both castrate-sensitive and castrate-resistant conditions, to assess whether there is a differential therapeutic impact of these combinatorial strategies along the disease spectrum, as a function of PTEN status. We will non-invasively monitor anti-tumor responses in the syngeneic model by tumor caliper measurements and mice will be euthanized when the tumors reach 2 cm in long-axis diameter. We will also test the impact of these IO combinations in our prostate-specific PTEN/p53 double knockout genetically engineered mouse model (Pb-Cre; PTENfl/fl; p53fl/fl GEMM), which develop aggressive autochthonous prostate tumors. Tumor development in the PTENfl/fl; p53fl/fl GEMM will be monitored non-invasively by ultrasound, and enrolled into treatment arms when the tumor volume reaches 150 mm3. Following the initiation of treatment, the tumor volume will be serially monitored by non-invasive MRI, and longitudinal tumor growth data will be analyzed. In addition, we will also determine the impact of sequential (A->B or B->A) vs. concomitant administration of combinations with ICB administration on therapeutic efficacy. Tumor sections at the time of euthanasia will be harvested for flow cytometry (TILs analysis), Western blot analysis, RNA-seq and IHC. Taken together, these co-clinical studies will provide the mechanistic foundation for the next wave of immunotherapeutic strategies to eradicate PTEN-deficient mCRPC.

All required software for data analysis will be provided by the lab/mentor.

The student will have the opportunity to present their work in our weekly lab meetings, Hem/Onc seminar series, and poster/oral presentations at national/international conferences