- Project 1: GLIPR1–ΔTM Protein Therapy for Prostate Cancer
- Project 2: Osteocrines in Therapy Resistance of Prostate Cancer Bone Metastasis
- Project 3: Src as a Therapeutic Target in Prostate Cancer Bone Metastases
- Project 4: Targeting FGF Signaling in Prostate Cancer Progression
- Project 5: MicroRNA in Prostate Cancer Progression: Genetic Variation, Genotype-Phenotype Correlation, and Circulating Biomarker
- Core A: Administrative Core
- Core B: Biostatistics and Bioinformatics Core
- Core C: Specimen Core
Timothy C. Thompson, Ph.D., Leader
Christopher J. Logothetis, M.D., Co-Leader
Extensive studies have defined GLIPR1 (glioma pathogenesis-related protein) as a secreted cytostatic/pro-apoptotic tumor suppressor protein that is down-regulated during prostate cancer progression through epigenetic mechanisms. Mechanistic studies have shown that GLIPR1 manifests tumor suppressor functions through coordinated cell type specific activities, including direct, tumor cell selective, pro-apoptotic activities mediated through reactive oxygen species (ROS)-c-jun-NH2 kinase (JNK) signaling. Recently we showed that GLIPR1 expression leads to down-regulation of specificity protein 1 (Sp1). Additional analysis showed that GLIPR1 expression suppressed c-myc through transcriptional repression that was dependent on Sp1-responsive GC/GT sites in the c-myc promoter and resulted in down-regulation of additional Sp1 target genes, including copper/zinc superoxide dismutase (CuZnSOD/SOD1) and manganese superoxide dismutase (MnSOD/SOD2). These data are in agreement with previous findings that Sp1 directly stimulates expression of multiple anti-oxidant proteins including CuZnSOD, MnSOD, and extracellular superoxide dismutase (ECSOD/SOD3). Western blotting analysis of c-myc targets showed that GLIPR1 overexpression resulted in significant suppression of key cell cycle regulatory proteins and also γ-glutamyl-cysteine synthetase, which catalyzes the first rate-limiting step in the synthesis of glutathione. Overall, GLIPR1 suppression of Sp1 activities represents a molecular switch that debilitates the anti-oxidant mechanisms/pathways that prevent cancer cells from ROS-mediated “self-destruction” and inhibits c-myc-mediated cancer cell proliferation. In preclinical studies we have found that recombinant GLIPR1 protein treatment results in tumor cell selective growth arrest and/or apoptotic cell death in multiple prostate cancer cell lines in vitro. Further preclinical studies using VCaP and/or PC-3 xenograft models demonstrated that recombinant GLIPR1 protein suppressed tumor growth and increased tumor cell apoptosis when administered intratumorally or intraperitoneally. In addition, effects on stromal cells were observed in treated tumors, including significant suppression of angiogenesis and macrophage infiltration.
Our first step in developing GLIPR1 protein therapy for prostate cancer is to test in situ delivery of a modified GLIPR1 protein (GLIPR1-ΔTM). This Phase Ib clinical trial will accomplish two important goals:
- Establish the safety of this therapeutic protein in a clinical setting (intraprostatic treatment prior to radical prostatectomy)
- Establish proof of principle for systemic use of GLIPR1-ΔTM
Use orthotopic prostate cancer xenograft models to test the efficacy of GLIPR1-ΔTM treatment and to characterize the molecular and cellular changes that accompany suppression of primary tumor growth and lymph node (LN) metastasis induced by intratumoral (IT) or intraperitoneal (IP) delivery of GLIPR1-ΔTM protein.
- Test selected prostate cancer cell lines and established prostate cancer xenografts for suppression of primary tumor growth and LN metastasis following IT GLIPR1-ΔTM treatment.
- Test selected prostate cancer cell lines and established prostate cancer xenografts for suppression of primary tumor growth and LN metastasis following repeated IT and IP GLIPR1-ΔTM treatment.
- Analyze tumor cell cytotoxicity (apoptosis, necrosis, autophagy), alterations in tumor microenvironment, including angiogenic activities, immune cell infiltration, and bone marrow derived cell infiltration, and serum markers following IT or IP GLIPR1-ΔTM treatment (Aims 1.1 and 1.2) by using immunohistochemical and biochemical techniques. Changes in stromal cell activities will be correlated with suppression of tumor growth and LN metastasis.
- Analyze changes in stromal (mouse) and tumor cell (human) gene expression profiles and identify interactive stromal–epithelial genetic pathways following IT or IP GLIPR1-ΔTM treatment.
Determine selected pharmacokinetics and toxicological parameters of GLIPR1-ΔTM and submit IND for clinical trial.
- Determine specific pharmacokinetics parameters of GLIPR1-ΔTM following IT or IP administration.
- Conduct FDA-required preclinical studies and submit an IND.
Conduct a Phase Ib preoperative clinical trial for in situ GLIPR1-ΔTM in men who will undergo radical prostatectomy for prostate cancer.
- Assess the feasibility and safety of intraprostatic GLIPR1-ΔTM treatment.
- Assess the effect of GLIPR1-ΔTM in index cancer.
- Analyze the effect of GLIPR1-ΔTM on the tumor microenvironment.
Project 2: Osteocrines in Therapy Resistance of Prostate Cancer Bone Metastasis
Sue-Hwa Lin, Ph.D., Leader
Christopher J. Logothetis, M.D., Co-Leader
A distinct feature of prostate cancer (PCa) with lethal potential is the ability of tumor cells to survive in a castrated environment and develop metastases in bone with a bone-forming phenotype. While several new targeted therapeutics have improved patient survival, resistance to targeted therapy invariably develops. The current theory whereby patients acquire resistance is adaptation of tumor cells in response to therapeutic agents. Some common examples of “adaptive resistance” include a second mutation in the target protein; mutations in a gene leading to an altered protein downstream of the target; or mutations that activate compensatory pathways. However, our novel work, on which this proposal is based, reveals that the tumor-educated bone microenvironment provides “pre-existing” resistance before “adaptive resistance” occurs. These new findings challenge the existing paradigm as how resistance is developed. Understanding the mechanism of “microenvironment-induced resistance” will lead to therapeutic strategies that overcome resistance and improve survival.
Metastatic cancer in bone is, in general, resistant to therapy as responses to therapies are usually incomplete and of short duration. Thus, the unique bone environment in osteoblastic PCa metastases likely contributes to resistance to multiple therapies. To study resistance specifically due to osteoblastic metastases, we use a bone forming PCa xenograft MDA-PCa-118b (PCa-118b), which was derived from a bone metastasis sample from a patient, as a model to decipher mechanisms of therapy resistance caused by PCa-induced bone formation. We have used this model to study resistance to cabozantinib, an oral multi-kinase inhibitor with potent activity against p-MET and p-VEGFR-2. Cabozantinib demonstrates striking clinical and radiologic responses in mCRPC patients with bone metastasis. However, as with other therapeutic agents, resistance to cabozantinib rapidly occurs in both our model system as well as humans, as evidenced by the progression of disease after initial response. Using the PCa-118b xenograft, our preliminary results showed that (1) While cabozantinib reduces PCa-118b tumor growth, a “resistance niche” is present, in which viable p-MET positive tumor cells are found in proximity to newly formed bone; (2) Treatment of PCa cells with osteoblast conditioned medium confers upon them properties associated with cell survival and a cancer stem cell-like phenotype. To analyze potential proteins responsible for the “resistance niche”, we did a secretome analysis of tumor-associated bone and identified 57 bone-secreted proteins. Many of these bone secreted proteins signal through integrins and several have been shown to induce cancer stem cell-like properties in other solid tumor types. We term those bone-secreted factors that are involved in therapy resistance “osteocrines”. Thus, we hypothesize that the resistance niche consists of osteocrines secreted from tumor-induced bone. These osteocrines activate PCa cell survival pathways and/or confer PCa cells with stem cell-like properties. Our objectives are to (1) identify specific osteocrines that confer resistance to cabozantinib treatment; and (2) perform preclinical and then clinical trials to identify inhibitor combinations that will overcome resistance to cabozantinib. While resistance to cabozantinib is used as a study model, the results from the proposed study will be able to apply to other therapies, e.g docetaxel or cabazitaxel.
Identify osteocrines in tumor-associated bone conditioned media that mediate therapy resistance of PCa bone metastasis in vitro. We will determine a set of the bone secretory factors for their roles in therapy resistance.
Examine the osteocrine-mediated signal transductions that lead to therapy resistance. We will examine the effects of osteocrines on integrin activation and/or induction of a “stem cell program” that confers therapy resistance to PCa cells.
Examine the effect of osteocrines on therapy resistance of PCa in xenograft mouse models. We will examine whether blocking the osteocrines themselves, their receptors (integrins), or their signaling pathways (e.g. activation of p-FAK) can overcome therapy resistance in vivo.
Examine the clinical relevance of osteocrines in therapy resistance of human PCa bone metastasis. We will determine whether osteocrine levels are increased in the bone marrow supernatants of patients with bone metastasis compared to those without bone metastasis and whether integrin signaling is activated in tumor cells in bone metastases.
Project 3: Src as a Therapeutic Target in Prostate Cancer Bone Metastases
Gary Gallick, Ph.D., Leader
John C. Araujo, M.D., Ph.D., Co-Leader
Current strategies for treatment of prostate cancer metastases in the bone that target mainly the primary tumor or preserve bone have only modestly affected survival. However, signaling pathways mediated by the Src family of tyrosine kinases affect not only tumor growth and invasion, but functions of osteoclasts and endothelial cells that contribute to metastatic growth. This project will test the central hypothesis that therapeutic strategies using Src inhibitors currently in clinical trial will prove efficacious in the treatment of prostate cancer metastases in the bone. The rationale for this hypothesis is that Src inhibitors affect tumor cell/microenvironment interactions that contribute to the “vicious cycle” of bone formation, degradation, and tumor growth. The studies are based, in part, and is based, in part, on the promise of an ongoing phase I/II clinical trial for metastatic prostate cancer using the combination of dasatinib (an SFK/Abl inhibitor) and docetaxel. This project will combine mechanistic-based strategies in preclinical mouse model systems with a phase III clinical trial using dasatinib plus docetaxel to examine the specific contributions of Src in the host and tumor cell contributing to metastatic to growth in the bone.
Determine the role of SFK inhibition in tumor cells, host cells, and both in affecting growth of PCa cells following intratibial injection into nude mice.
Determine molecular alterations in the tumor and host correlating with the effectiveness of dasatinib.
Integrate this knowledge with a phase III trial using dasatinib in combination with docetaxel in select patients with castrate-resistant prostate cancer and bone metastases and correlate changes in molecular markers of Src and bone preservation with clinical course of the disease.
The trial will serve as a platform to associate markers of Src activation with baseline and serial change(s) in bone turnover markers. Thus, the experiments in Project 3 are novel in that they build on a promising therapeutic strategy, which we also view as reiterative, where our increased knowledge of Src’s effects in tumor/bone interaction will help dictate clinical trial design, and the clinical trials will help refine modeling the disease in preclinical studies.
Project 4: Targeting FGF Signaling in Prostate Cancer Progression
Nora M. Navone, M.D., Ph.D., Leader
Paul Corn, M.D., Ph.D., Co-Leader
Wallace McKeehan, Ph.D., Subcontract Leader
Prostate cancer is the second leading cause of cancer death in men in the United States. Localized prostate cancer can be cured by androgen ablation, but when the disease escapes the confines of the gland, the prospects for cure decrease drastically and the disease becomes “castrate resistant.” Bone is the primary site of castrate-resistant disease progression, which is associated with a poor prognosis. The fibroblast growth factor (FGF)/FGF receptor (FGFR) complex, a signaling axis involving multiple FGF ligands and receptors, mediates tumor–stromal interactions and is one of the most commonly altered signaling pathways during prostate cancer progression. Expression of FGFR1, multiple FGF ligands, and FGFR adaptor FRS2a has been observed in prostate cancer epithelial cells. Our recent studies have defined a mouse model of prostate cancer highly dependent on FGF signaling and have implicated FGF9 in the osteoblastic progression of human prostate cancer cells in bone. The results of our preliminary studies support a notion that during bone metastasis, the prostate cancer cells that aberrantly express both FGF and FGFRs create a new “compartment” in bone as source and recipient of additional FGF-mediated signaling, thus subverting homeostasis.
The implication of the FGF axis in prostate cancer progression suggests that FGFR blockade represents a new therapeutic opportunity for men with castrate-resistant prostate cancer. Recently, TKI258, a receptor tyrosine kinase inhibitor (TKI) with strong activity against FGFR1-3 (IC50 < 40 nM), has become available and is being used as an experimental new drug for solid tumors. The main goal of this proposed project is to establish the feasibility of using TKI258 to modulate FGF signaling in men with castrate-resistant prostate cancer and to correlate FGF signaling modulation with clinical disease progression. We will assess the effect of TKI258 on human prostate cancer xenografts growing in the prostate and bone of castrated immunodeficient male mice (Aim 1) and also on mouse models of prostate cancer (Aim 2) to identify markers of response to TKI258 therapy directly related to FGF signaling. We will then perform a proof-of-principle clinical study with TKI258 in men with castrate-resistant prostate cancer and bone marrow infiltration (Aim 3). The study will create an annotated tissue resource and will permit validation of FGF signaling responsive markers emerging from Aims 1 and 2. This will be the first clinical study to assess the effect of TKI258 in prostate cancer.
Project 5: MicroRNA in Prostate Cancer Progression: Genetic Variation Genotype-Phenotype Correlation, and Circulating Biomarker
Xifeng Wu, M.D., Ph.D., Leader
Jeri Kim, M.D., Co-Leader, Clinical
Jian Gu, Ph.D., Co-Leader
Prostate cancer (PCa) has been increasingly detected and treated earlier, leading to a 5-year survival rate of nearly 100%. However, the majority of PCa are indolent and slow growing and patients often die from other causes than PCa. Clinical variables such as stage, Gleason score and serum PSA level are not sufficient to accurately distinguish indolent from aggressive tumors, leading to overtreatment of many indolent, slow growing tumors; conversely, some patients are undertreated because they have more aggressive disease but postpone aggressive adjuvant therapy. This uncertainty is problematic across populations, but is particularly devastating for African Americans (AAs), because AAs are 1.5 times more likely than European Americans (EAs) to develop PCa and 2.3 times more likely to succumb to PCa. Thus, aggressive treatment may be necessary for an increased fraction of AA men diagnosed with PCa. However, the PSA levels of AA men tend to be higher than those of EA men, potentially favoring over-treatment. Exploration of novel biomarkers and creation of an integrated risk stratification approach to determine who does not need treatment and who needs more aggressive treatment is urgently needed for localized PCa patients.
Built upon the large existing PCa patient population with epidemiologic and clinical data, blood collection, and long follow up at MD Anderson Cancer Center, we propose to construct a well-characterized cohort of 2,000 EA and 500 AA PCa patients receiving radical prostatectomy or radiotherapy. We will perform a comprehensive screening and validation approach to identify genetic variations in microRNA (miRNA) genes, miRNA processing genes, and miRNA target genes in major PCa pathways as prognostic markers for PCa patients. Somatic alterations of miRNA expression have been observed in almost all cancers. No study has evaluated genetic variations in miRNA related genes as predictors of progression for PCa patients receiving definitive therapy. Circulating miRNAs have attracted tremendous interest in recent years as valuable biomarkers for early detection, diagnosis, and prognosis of many different cancers. One of the bottlenecks in this regard has been the heterogeneous patient populations, small sample sizes, and technical reproducibility. Capitalizing on our existing large patient cohort, availability of baseline pre-therapy plasma samples, long follow up time, and expertise in serum miRNA profiling, we will evaluate the utility of circulating miRNAs as predictive biomarkers for progression in localized PCa patients. There are three specific aims:
To systematically screen and validate SNPs in miRNA gene, miRNA processing genes, and miRNA target sites in major PCa relevant genes, as predictors of PCa progression (biochemical failure).
1a) In the discovery phase, we will use 1,000 EAs patients to screen a custom-designed SNP panel in miRNA genes, miRNA processing genes, and miRNA binding sites in key molecular pathways involved in the PCa development and progression (RB, PI3K/AKT, RAS/RAF, TGFβ/BMP, Wnt/β-catenin and androgen receptor signaling pathways). We expect to genotype ~1,200 SNPs in this phase. 1b) in the validation phase, we will validate the top 60 SNPs in an additional 1,000 EAs patients. 1c) we will explore the association of these 1,200 SNPs with the risk of progression in 500 AAs PCa patients.
To assess the functional significance of the validated SNPs.
SNPs in miRNA genes and miRNA biogenesis genes are expected to affect miRNA expression. We will determine global miRNA expression profiles in 50 pairs of tumor and adjacent normal tissues from surgically resected PCa patients to correlate validated SNPs (from Aims 1) with miRNA expression in normal tissues and changes from normal to tumor tissues. This genotype-phenotype correlation will provide biological plausibility for the observed associations.
To identify circulating miRNAs as predictors of progression.
3a) In the testing set, we will profile global miRNA expression in the plasma of 150 pairs of localized PCa patients receiving radical prostatectomy (100 pairs of EAs and 50 pairs of AAs patients) with or without progression (biochemical failure). 3b) in the validation set, we will use an additional 150 pairs of PCa patients (100 pairs of EAs and 50 pairs of AAs patients) receiving radical prostatectomy with or without progression to validate the top 20 miRNAs that are predictive or progression from the testing set in each ethnic group. We will also compare the significant miRNAs between EAs and AAs.
This project incorporates inherited genetic variations, somatic molecular alterations, and circulating biomarkers. The incorporation of inherited genetic variations and somatic profiling will provide biological insights into the role of genetics in PCa progression. Minimally invasive circulating biomarkers will have enormous clinical impact on targeted therapy for localized PCa patients.
Core A: Administrative Core
Christopher J. Logothetis, M.D., Director
Timothy C. Thompson, Ph.D., Co-Director
The principal role of the Administrative Core of the MD Anderson Cancer Center Prostate SPORE is to expand the integration of investigators from diverse scientific disciplines who have joined the translational research effort in prostate cancer. It is the purpose of the Administrative Core to provide continuous leadership and general administrative support for all SPORE-related activities. Dr. Christopher Logothetis, Director, and Dr. Timothy Thompson, Co-Director, provide support for all SPORE research projects, cores, and programs and ensure compliance with all governmental regulations, policies, and requirements. The Administrative Core coordinates data quality control and quality assurance in conjunction with the Biostatistics and Bioinformatics and Specimen Cores and the Internal and External Advisory Boards. The Administrative Core convenes all meetings of the SPORE, including seminars and lectures presented by scientific leaders in basic, translational, and clinical prostate cancer research brought to the SPORE by the directors. With the program directors, the Administrative Core monitors the selection of projects and trainees for the Developmental Research and Career Development Programs, respectively. Dr. Logothetis chairs the Executive Committee, composed of the co-principal investigator and senior scientific leaders with extensive SPORE experience. To address the ongoing needs of the minority and medically underserved communities in Houston and Harris County, Texas, Dr. Curtis Pettaway (collaborator) has developed the Prostate Outreach Program integrated into SPORE research through the Administrative Core. Special contributor Dr. Randall Millikan has developed the MD Anderson Prostate Research database (Research Repository), which serves the increasing data and bioinformatics needs of SPORE investigators, in collaboration with the Biostatistics and Bioinformatics and Specimen Cores. The Patient Advocacy Program brings patient representation to the SPORE and increases awareness and education of prostate cancer research, treatment, and prevention in the community.
Core B: Biostatistics and Bioinformatics Core
Kim-Anh Do, Ph.D., Director
Kevin Coombes, Ph.D., Co-Director
The research proposed by The University of Texas MD Anderson Cancer Center SPORE in Prostate Cancer encompasses a broad range of activities, including studies in cell lines, animal models, and clinical trials. These studies will generate many different types of data, including clinical, epidemiological, biochemical, immunohistochemical, pharmacokinetics, genotype, and immunologic. The Biostatistics and Bioinformatics Core provides comprehensive biostatistics and bioinformatics expertise to ensure the statistical integrity and to optimize data analysis of the studies by the SPORE. It will incorporate sound experimental design principles within each project that will enhance interpretability of study results, will carry out data analyses using appropriate statistical methodology, and will contribute to interpretation of results through written reports and frequent interaction with project investigators. Members of the Core will participate in monthly SPORE meetings with all project investigators, ensuring that proper consideration is taken of biostatistics and data management issues during all phases of SPORE experiments. The Biostatistics and Bioinformatics Core will further provide an integrated data management system to facilitate communication between all projects and cores, which will be customized to meet the needs of the Prostate Cancer SPORE. This process includes prospective data collection, data quality control, data security, and patient confidentiality. Thus, from inception to reporting, translational experiments will benefit from SPORE resources that will be used to augment existing MD Anderson Cancer Center biostatistics resources.
To provide valid statistical designs of laboratory research, clinical trials, and translational experiments arising from the ongoing research of the SPORE.
To develop and conduct the innovative statistical modeling, simulations, and data analyses needed by the projects, developmental projects, and other cores to achieve their specific aims.
To ensure that the results of all projects are based on well-designed experiments and are appropriately interpreted and to assist in the preparation of manuscripts describing these results.
To develop integrated computational libraries and tools for producing documented, reproducible statistical analyses and to make these tools available to all SPORE participants.
Core C: Specimen Core
Patricia Troncoso, M.D., Director
Randall Millikan, M.D., Co-Director
The Specimen Core will provide SPORE investigators with well-characterized biologic specimens, including tissues, blood, and tissue derivatives that are essential for achieving the aims of the projects. The Specimen Core has a large repository of paraffin blocks, frozen samples, plasma, serum, and bone marrow that spans the entire spectrum of prostate cancer. The repository includes primary tumors and metastases from therapy-resistant tumors and tumors derived from radical prostatectomy specimens from patients given novel preoperative therapies as part of studies conducted at The University of Texas MD Anderson Cancer Center and other, multi-institutional efforts. This material, supplemented in select cases with matching biopsy specimens, plasma, serum, and bone marrow aspirates collected before therapy, will provide SPORE investigators with optimal tissue samples with which to address the proposed basic and translational research tissue requirements of the projects. We have also established biologic models (cell lines and xenografts) relevant to the projects. Tissue requests and approval by the Tissue Acquisition and Distribution Committee will be handled electronically. Tissue derivatives, including tissue microarrays, cRNA, and DNA, will optimize the use of limited samples and enhance collaboration among investigators at MD Anderson and other institutions. Structured information derived from standardized, high-throughput assays of differential gene expression, both of protein (immunohistochemistry) and RNA (oligonucleotide arrays and multiplexed PCR), will be available to individual SPORE investigators to facilitate modular and gene network analysis in specific clinical contexts. A unique feature of the Specimen Core will be the ability to link comprehensive clinical and pathologic information to the morphologic and molecular characterization of selected pathways. This linkage will be accomplished in cooperation with the Biostatistics and Bioinformatics Core by using a Web-based portal. By using this Web-based system to access and query SPORE data, we will pool laboratory resources, facilitate the translational research proposed in the projects, and accelerate successful achievement of the proposed aims.
Collect, process, annotate, characterize, store, and distribute human biospecimens related to prostate cancer.
Create well-characterized and quality-controlled tissue derivatives for translational research (including xenografts, cell lines, tissue microarrays, protein lysates, RNA, and DNA) and conduct selected studies, including immunohistochemical, reverse-transcription polymerase chain reaction (RT-PCR), and reverse-phase protein array (RPPA) evaluations.
Provide investigators with expertise to optimally select and use biospecimen resources and analytical techniques for tissue-based studies.
Provide longitudinal clinical annotation for biospecimens, consisting of structured data on comorbidity, family history, diagnosis, treatment, and outcome.
Provide an integrated information system that supports a secure archive (for both clinical data and datasets from translational studies) along with a corresponding application for browsing and querying data related to biospecimen repositories, clinical annotation, and translational experiments.