Skip to Content
Translational Research Program (TRP)
Contact CIP
Show menu
Search this site
Last Updated: 10/10/18

Dana-Farber/Harvard Cancer Center (DF/HCC)

SPORE in Prostate Cancer

Current PD/PIs: Massimo Loda and Steven Balk
Former PD/PI: Philip W. Kantoff
Period of Performance: 07/01/2007-06/30/2012

Principal Investigator Contact Information

Massimo Loda, MD
Professor of Pathology, Harvard Medical School
Principal Investigator, Pathology DFCI
Senior Pathologist, Brigham and Women's Hospital
Dana-Farber Cancer Institute
Medical Oncology
450 Brookline Ave. DA1537
Boston, MA 02215
Tel: (617) 632-4001
Fax: (617) 632-4005
Email: massimo_loda@dfci.harvard.edu

Steven Balk, MD, PhD
Professor of Medicine
Beth Israel Deaconess Medical Center
CLS Building Room 443
3 Blackfan Circle
Boston, MA 02215
Tel: (617) 735-2065
Fax: (617) 735-2050
Email: sbalk@bidmc.harvard.edu

Project 1: Biguanides for the treatment of prostate cancer (Bubley/Vander Heiden)

Several observations point towards the use of biguanides such as metformin and phenformin for cancer treatment or prevention. Some of the initial interest stems from population studies that have shown that diabetic patients treated with metformin have a dramatic decrease in risk of cancer and cancer mortality compared to diabetic patients treated with insulin or sulfonylureas (1). Metformin use was associated with a 44% risk reduction in prostate cancer (PCa) among Caucasian men (2). The addition of metformin also chemotherapy might make this more effective. For example, in a retrospective study of >2,500 breast cancer patients treated with neoadjuvant chemotherapy, patients taking metformin had significantly higher pathologic complete responses than patients not taking metformin (3).

Our understanding of how to incorporate metformin into clinical trials is complicated by the fact that the mechanism by which metformin acts is not completely understood. Both direct and indirect effects of metformin could contribute to a beneficial effect on prevention or treatment. The indirect effects of metformin have been ascribed to its effects on plasma glucose levels (4,5). Metformin affects transcription of gluconeogenesis genes in the liver and increases glucose uptake in skeletal muscle. This has the effect of decreasing circulating insulin and IGF-1 levels, and thereby decreasing signaling through insulin/IGF-1 receptors. Indirect effects are consistent with the observation that hyperinsulinemic states such as those associated with obesity have been associated with poorer outcomes and higher risk (6). By a direct mechanism, metformin may inhibit cell proliferation and/or apoptosis independent of its effects on circulating insulin levels. Metformin may exert a growth-inhibitory effect on epithelial cell cancers through the activation of AMPK and other downstream signaling relevant to cell growth and survival (7-9). Cancer cell line experiments have shown that metformin activates AMPK, inhibits mTOR, and decreases global protein synthesis.(7) Evidence that AMP kinase affects carcinogenesis is that this enzyme is activated by LKB1, a tumor suppressor gene lost in some cases of lung and squamous cell cancer (10). The concentration of metformin needed to achieve results in in vitro and in animal model experiments (11-14) is at least ten fold higher than what is achieved with typical diabetic dosing (15). This makes it more likely that many of the preclinical findings stem from metformin having a direct effect. Nonetheless there has been a great excitement from pre-clinical findings. For example recently Hirsch et al demonstrated that metformin treatment inhibited transformation and selectively killed cancer stem cells from 4 genetically distinct breast cancer lines (11).

The projects below are all beginning within the DF/HCC. They will, in their totality, be able to answer many of the questions related to the effect of biguanides in PCa treatment and in prevention, and their probable mechanism of action.

The specific aims of this project are to:

  • Determine if metformin has a direct versus indirect effect:
  • Determine a strategy for maximizing the potential effects of biguanides alone and in combination with docetaxel using PCa animal models
  • Perform a Phase 1b trial of phenformin in combination with docetaxel in patients with CRPC
  • Use the resources from the Physicians Health Study to assess the effect of metformin treatment on PCa outcome, and correlate this with effects in signaling pathways in tumors using IHC methodology.

Project 2: Genetic and Clinical Characterization of the 8q24 Prostate Cancer Risk Locus (Freedman/Pomerantz)

Inherited factors for prostate cancer prostate cancer risk have proven difficult to identify. The results of linkage analysis, the workhorse of gene finding methods for Mendelian disorders, have been notoriously difficult to validate. Within the past year, however, two independent studies have demonstrated that a region on 8q24 confers a strong risk of developing sporadic prostate cancer in men of African and European ancestries. 8q24 is the first bona fide genetic risk locus that is responsible for an appreciable fraction of sporadic prostate cancer cases in the general population, particularly, in African-American men. Our hypotheses are that a genetic variant within 8q24 initiates prostate cancer and impacts the clinical features of the disease.

The discovery of this novel locus presents the opportunity to gain deeper insight into prostate cancer biology. Although the region has been identified, the causal allele and the gene that it influences remain to be determined. Since 8q is one of the most frequent regions of somatic amplification in prostate cancer, we will also have the opportunity to explore connections between the germline and somatic genomes. Understanding the genetic and molecular pathways driving prostate cancer can ultimately lead to improved diagnostic and therapeutic capabilities. From a clinical perspective, patients can be stratified into carriers and non-carriers of this risk allele to explore its impact on clinical variables, such as presentation and disease outcomes. Our goals are to elucidate the genetic pathogenesis and the clinical impact of the 8q24 variant allele on prostate cancer.

The specific aims of this project are to:

  • To identify the causal germline allele through fine mapping of the 8q24 region
  • To characterize inherited variation at 8q24 and somatic molecular associations at 8q24
  • To determine the effect of the germline 8q24 variant on clinical presentation and outcomes in prostate cancer patients

Project 3: TMPRSS2-ERG and SPINK1 in Lethal Prostate Cancer (Mucci/Loda)

Abstract – Not available.

The specific aims of this project are to:

  • Characterizing the phenotype of TMPRSS2:ERG and SPINK1 prostate cancer.
  • Genetic susceptibility, TMPRSS2:ERG and SPINK1, and lethal prostate cancer.
  • Somatic alterations, TMPRSS2:ERG and SPINK1, andlethal prostate cancer.

Project 4: Modulating Transcription Factor Targets Via Chemical Genomics (Golub/Taplin)

The molecular basis of prostate cancer is becoming increasingly elucidated as the genetic abnormalities underlying the disease are identified. Most prominent among these is the androgen receptor (AR), which for decades has been known to be essential for the survival of prostate cancer cells. However, because transcription factors (including AR) are generally considered ‘undruggable’, pharmacological approaches to modulating AR activity have been limited, focusing on blocking ligand binding to AR, and hormonally diminishing the body’s production of androgens. While tumors initially respond to such androgen deprivation therapy, recurrence inevitably occurs. New approaches to modulating AR activity are therefore needed. Similarly, the recently identified mutations in the ETS family of transcription factors observed in as many as 70% of prostate cancers call for a novel approach to pharmacologically inhibiting the activity of these mutant oncoproteins.

During the prior funding cycle, we developed a novel, gene expression-based chemical screening method (GE-HTS), and applied it to the discovery of AR-modulating small molecules. Most prominent among the hits that emerged from that screen were a group of structurally unique natural products that we demonstrated were functioning as novel HSP90 inhibitors.

The specific aims of this project are to:

  • Test the hypothesis that HSP90 inhibition will abrogate AR function and result in clinical responses in patients with advanced prostate cancer.
  • Develop a signature of TMPRSS2/ERG activity.
  • Screen small molecule libraries to identify compounds capable of modulating the TMPRSS2/ERG signature.
  • Validate the hits emerging from the TMPRSS2/ERG GE-HTS screen

Project 5: The Androgen Receptor in Castration Resistant Prostate Cancer (Balk/Brown)

The vast majority of prostate cancers are androgen dependent, and androgen deprivation therapy (ADT) remains the standard treatment for non-organ-confined prostate cancer. Unfortunately, patients treated with ADT invariably relapse with progressive systemic prostate cancer that has been termed hormone-refractory or androgen-independent prostate cancer (AICaP). Significantly, most cases of AICaP are associated with high levels of androgen receptor (AR) mRNA and protein, as well as renewed expression of androgen regulated genes (such as prostate specific antigen (PSA), TMPRSS2, and the recently identified TMPRSS2/ETS fusion genes), suggesting that AR and AR-regulated genes remain critical for tumor growth in AICaP. However, the molecular events mediating the apparent reactivation of AR transcriptional activity in AICaP, and the critical downstream AR-regulated genes, remain to be determined. We recently reported our results using Affymetrix oligonucleotide microarrays to compare gene expression in a series of laser capture microdissected primary prostate cancer samples versus AICaP bone marrow metastases. These studies demonstrated a high level of expression of AR mRNA in AICaP together with renewed expression of multiple strongly AR-regulated genes, confirming substantial reactivation of AR-mediated transcription despite castrate androgen levels. With respect to potential mechanisms for this AR reactivation, we found that AICaP samples had marked increases in enzymes that mediate androgen metabolism, suggesting that enhanced synthesis of the ligands testosterone and dihydrotestosterone (DHT) by AICaP is one potential mechanism contributing to relapse after castration.

The specific aims of this project are to:

  • Complete the current trial of ketoconozole, hydrocortisone, and dutasteride, and initiate new phase II clinical trials of compounds that suppress AR expression or function in CRPC.
  • Combine gene expression data from CRPC cell lines and xenografts with AR ChIP-on-chip from these cell lines to identify critical AR cooperating factors and regulated genes that may be therapeutic targets in CRPC.
  • Validate AR regulated genes as therapeutic targets in CRPC

Project 6: Interrogating the EZH2 pathway to develop therapies and prognostic biomarkers for prostate cancer (Cichowski/Sweeney)

The lifetime risk of developing prostate cancer is 1 in 6 1. If detected early, prostate cancer is curable; however, therapy for advanced disease while being able to prolong overall survival and improve quality of life, it is not curative. Consequently, while there has been decline in the death rate from prostate cancer, there were still an estimated 28,660 deaths due to prostate cancer in the United States in 2008.2. The current treatment for advanced or recurrent prostate cancer is surgical or medical castration in the first instance with about 90% of men responding with resolution of symptoms. However, nearly all patients ultimately relapse and develop metastatic castrate resistant disease 3. Therefore there is a significant need to better understand both the mechanisms that promote metastatic progression and resistance to therapy so we can develop more effective therapies.

While many genetic and transcriptional changes associated with metastatic prostate cancer have been have been reported, there is still a paucity of functional insight into the events that actually drive metastasis. One gene that has been proposed to play an important role in metastasis is EZH2, which encodes an enzyme that regulates the epigenetic silencing of specific target genes (reviewed in 4). EZH2 is amplified and/or over-expressed in advanced prostate cancers and was identified as the most differentially up-regulated gene in metastatic tumors 5. As such EZH2 has been hypothesized to promote prostate cancer metastasis; however, a direct role for EZH2 in driving metastatic disease has never been demonstrated. In addition while numerous EZH2 target genes have been identified, there is still little functional insight into how the epigenetic silencing of any of these genes might play a causal role in tumor progression.

We have recently developed a tractable murine model for metastatic prostate cancer and for the first time have shown that EZH2 plays a causal role in driving prostate cancer metastasis in vivo 6. Moreover, using two different experimental systems we identified a critical EZH2 target gene, known as DAB2IP, that when suppressed, also promotes metastasis. Genomic and immunohistochemical analysis of human prostate tumors further demonstrated that this novel oncogene-tumor suppressor pathway is deregulated in human cancer. Moreover, we found that this pathway critically regulates NF-kB and that NF-kB activation is essential for driving metastasis in both model systems. Importantly, this work represents the first study that has ever functionally identified a gene or pathway that drives metastatic prostate cancer.

In this application we propose to utilize the model systems that we have developed to 1) define the importance of EZH2 as a potential therapeutic target. Specifically, we will determine if its acute inhibition suppresses established primary prostate tumors, metastases or both. We also aim to 2) systematically identify other EZH2 targets in prostate cancer progression and metastasis. These studies will not only provide mechanistic insight into prostate cancer progression, but may also identify critical molecular predictors of metastatic disease and may ultimately also serve as biomarkers to predict patient outcome and guide drug development with ultimate goal of improving therapeutic efficacy. Finally, because we have shown that NF-kB is activation is essential for driving metastasis we will utilize human tumor arrays to determine whether NF-kB activation and/or DAB2IP loss can accurately predict progression and outcome. Thus we aim to harness the combined power of genetic, genomic, and functional in vivo modeling approaches to identify therapeutic targets, drivers, and predictive biomarkers of metastatic prostate cancer.

The specific aims of this project are to:

  • Use mouse models to investigate the importance of EZH2 as a potential therapeutic target in established primary prostate tumors and metastases
  • Identify additional critical EZH2 target genes that function in prostate cancer progression
  • Determine whether NF-?B activation, and/or loss of DAB2IP, predict outcome in a large cohort of individuals with prostate cancer with the ultimate goal of establishing the most accurate, predictive markers of progression.

Core 1: Administration, Planning and Evaluation (Kantoff/Bubley/Cantley/D’Amico/Sanda/Stampfer)

The purpose of the Administration, Evaluation, and Planning Core is to ensure the coordination of the Dana-Farber/Harvard Cancer Center (DF/HCC) Prostate Cancer SPORE components and to provide oversight and leadership of the scientific, administrative, and fiscal aspects of the SPORE.

The DF/HCC Prostate Cancer SPORE is a large organization, with research performed in six institutions. Despite its size, its strength is that the SPORE and the performance institutions are part of an existing cancer research organization, the DF/HCC. We take advantage of the Dana-Farber Cancer Institute’s (DFCI) large and efficient administrative staff and its central Research Administration office. This office is also responsible for administering the DF/HCC. The DF/HCC Prostate Cancer SPORE utilizes the DF/HCC’s operating procedures for expenditure reporting oversight. DF/HCC provides both fiscal and administrative oversight to its operations, and the DF/HCC Prostate Cancer SPORE uses the same structure.

Within the DF/HCC Prostate Cancer SPORE, there are several layers of oversight and evaluation. Dr. Kantoff, as SPORE Director, monitors the progress of the Projects and Cores and oversees the Career Development Program (along with Drs. Sanda and D’Amico), the Developmental Projects Program (along with Drs. Cantley and Stampfer), and all other proposed activities. Within the DF/HCC, Dr. Kantoff has the authority and resources to ensure the success of this SPORE. Our Governance Committee, made up of senior members of the DF/HCC Prostate Cancer Program, meet monthly to provide immediate decision-making. We have two strong Co-PIs. Dr. Kantoff will be joined by Lewis Cantley, PhD, Professor of Systems Biology and Chief, Division of Signal Transduction at the Beth Israel Deaconess Medical Center, to assume the responsibility as Co-PI of the SPORE. Dr. Cantley, a member of the National Academy since 2001 and the PI of a PO1 on signal transduction in prostate cancer, brings outstanding basic science expertise to our Program and SPORE. Meir Stampfer, MD, Dr.P.H., Professor of Nutrition and Epidemiology, Chair, Department of Epidemiology, Harvard School of Public Health, will also serve as Co-PI. Given Dr. Stampfer’s interest and expertise in prostate cancer and the centrality of population science in this SPORE application, his leadership is welcome. We have a strong Internal Advisory Board, comprised of prominent members of the Harvard Medical School community and representing the participating institutions and major cancer research disciplines. Our External Advisory Board will meet in the Boston area annually during the five-year funding cycle. The membership and functions of these committees are discussed in detail in the description of this Core.

The Administration, Evaluation, and Planning Core allows for the provision of stimulating intellectual activities, organization of venues for planning future research through seminars and retreats, and the oversight of research and spending. This Core also provides the tools to work with institutions inside and outside of Harvard University to leverage the considerable power of the SPORE in order to raise more research funding for prostate cancer.

The Specific Aims of this Core are to:

  • Monitor research progress and plan for the future
  • Foster collaborative research within the SPORE and between SPOREs
  • Integrate the Prostate SPORE into the DF/HCC structure
  • Provide necessary resources and fiscal oversight
  • Promote rapid dissemination of significant research findings

Core 2: Biostatistics and Computational Biology Core (Regan/Tamayo)

Biostatistics and Computational Biology Core provides design and analytic support to SPORE Projects, Developmental Projects, Projects of the Career Development Awardees, and other SPORE Cores.

The specific aims of this Core are:

  • Provide biostatistical and computational biology expertise for the planning and design, conduct, analysis, and reporting of laboratory, genomic, animal, translational, clinical (including associated correlative studies), and epidemiological studies for SPORE projects, Developmental Projects, Projects of the Career Development Awardees, and other SPORE Cores.
  • Provide consultation on data collection, storage and quality assurance, statistics, and computational biology software and programs and coordination of laboratory results with parameters and outcomes from clinical studies or clinical/translational research databases.
  • Provide short-term biostatistics and computational biology consulting to the entire group of SPORE researchers.

Core 3: Tissue and Pathology Core (Loda)

The DF/HCC Prostate Cancer SPORE Tissue and Pathology Core has provided and will continue to provide collaborating investigators multiple pathology services including histology, immunohistochemistry, in situ hybridization, fluorescent in situ hybridization (FISH), computer-assisted image analysis, laser capture microdissection, and the generation of and access to tissue microarrays (TMAs).

Our overarching goal is to create a seamless informatics link between the extensive existing tissue resources among the Dana-Farber/Harvard Cancer Center (DF/HCC) hospitals and collaborators. Thus, one of the major goals of this Core is to maintain and grow an existing tissue and blood resource (henceforth referred to as “biobank”), linked to clinical outcome data, behind a secure data management system that will be available to DF/HCC SPORE investigators as well as SPORE investigators at other institutions. In this regard there is and will continue to be close collaboration between the Tissue and Pathology Core and the Biostatistical and Computational Biology Core. Dr. Loda is cognizant that first and foremost we are responsible for the optimal clinical care and protection of our patients. Only after ensuring that sufficient information has been acquired to accurately grade and stage a patient’s tumor, do we register his samples into our biobank. The protection of patient confidentiality is guarded throughout the entire process, from collection to use in research projects. We are also greatly encouraged by the patient advocates involved in the SPORE program, who remind us of the importance of using these annotated samples for translational research to reduce suffering from CaP. We consider this our mission.

  • Guiding Principles (or Specific Aims): There are some guiding principles, which are critical to the success of this Core and allow us a means of presenting the various facets of our Tissue and Pathology Core. We believe these principles are consistent with NCI guidelines for the collection and storage of biospecimens as recently published in the Federal Register. Successful translational research requires a wide range of well-annotated human analytes, mouse models, xenografts, and cell lines.
  • Patient protection and regulatory issues are the top priority of any biobank.
  • Access to samples and rules of usage (governance) need to be determined prior to performing experiments.
  • Data generated from each sample increase that sample’s value.
  • The biobank should facilitate research of qualified investigators regardless of their affiliation to the DF/HCC Prostate Cancer SPORE (i.e., outstanding research is encouraged from all corners of the world).
  • Pre-analytic variability of samples should be monitored and standard operating procedures (SOPs) should be employed when known.
  • Synergy with pre-existing programs leads to economy of effort.
  • Given limited resources, the biobank should focus first on the five SPORE Projects and the Developmental Projects and Projects of the Career Development Awardees.
  • Bioinformatics enhances the use of data generated from biobank samples.
  • High quality pathology interpretation and expertise in methodology is critical to working with heterogeneous prostate cancer samples to ensure optimal use of samples.

Career Development Program (CDP)

The investigators assembled in the DF/HCC Prostate Cancer SPORE have a substantial record in mentorship and development of junior faculty working in the prostate cancer field (detailed below). The goal of the Career Development Award Program of our SPORE is to build upon this record and continue a formal process for the identification, selection, funding, and mentoring of individuals pursuing careers in the study of the basic, translational, and clinical aspects of prostate cancer.

The CDP aims to attract and develop the next generation of translational investigators in prostate cancer.

The specific aims of the CDP are:

  • Solicitation of CDP Award Applications
  • Evaluation of Applicants and Selection of Awardees
  • CDP Evaluation and Review
  • Mentorship for Career Development Program Awardees and Career Progress

Developmental Research Program (DRP)

The Developmental Research Program of the DF/HCC Prostate Cancer Program is intended to attract established investigators to the field of prostate cancer in order to develop new ideas of investigation which may change the field or be developed into full projects in the SPORE

The objectives of the Developmental Projects Program are to ensure a continual renewal of high-quality scientific endeavors in the DF/HCC Prostate Cancer SPORE and to fund efforts that will complement or enhance the overall quality of the DF/HCC Prostate Cancer SPORE. In general, the Developmental Projects Program will fund established investigators.

The specific aims of the DRP are to:

  • Solicit applications and/or identify novel prostate cancer research projects
  • Evaluate these projects for funding
  • Fund innovative developmental projects
  • Re-evaluate projects for possible transition into full project status
  • Evaluate the success of the program