added 12.11.2013

The University of Michigan (UM) Prostate SPORE has an extensive track record of collaboration. Our history demonstrates many longstanding as well as new partnerships that have been established in the pursuit of translational research of the highest impact. Engagement in both horizontal and vertical collaborations have resulted from interactions with investigators across the UM campus, SPOREs at UM and other institutions, industry and other Universities. In addition to these productive collaborations, the Tissue/Informatics Core continues to be a critical resource supplying essential biologic resources and expertise to researchers across the country.

Recent and Ongoing Collaborations of the UM Prostate SPORE

A. Discovery, characterization, and translation of the ETS fusions in prostate cancer.

In 2005, using a bioinformatics approach to nominate oncogenes in a compendium of microarray data (, Prostate SPORE investigators Drs. Chinnaiyan, Tomlins, Pienta, Montie, Wei, Mehra, Rubin and others identified a recurrent fusion of the androgen-regulated gene TMPRSS2 to the ETS family of transcription factors in a majority of prostate cancers (approx. 50%). The fusions were first characterized from lethal metastatic prostate cancer samples procured from the Rapid Autopsy Program founded by Dr. Pienta, and supported by the UM Prostate SPORE Tissue/Informatics Core. The discovery of recurrent ETS fusions in prostate cancer likely represents one of the seminal collaborative discoveries emanating from the SPORE programs in general.

Since the initial discovery of the ETS gene fusions in 2005, our Prostate SPORE facilitated the transition of these biomarkers into clinical use. Several groups, including our own, have correlated presence of the TMPRSS2-ERG gene fusion with more aggressive disease in certain patient cohorts making it possible to molecularly type and sub-type prostate cancers based on the gene fusion partners and variants they harbor. Molecular sub-typing of prostate cancer biopsies using antibody based IHC or FISH analysis of the most common prostate cancer gene fusion product, ERG, is available in CLIA laboratories across the country and is generally used for challenging prostate biopsies or isolated precursor lesions. This was developed through a licensing agreement from the University of Michigan to Gen-Probe and Ventana/Roche based on our work.

As the TMPRSS2-ERG gene fusion is exquisitely specific to prostate cancer, we have worked diligently to translate this into a non-invasive urine test for prostate cancer. While the TMPRSS2-ERG gene fusion was discovered in the context of a Prostate SPORE project and infrastructure, we collaborated with the Early Detection Research Network (EDRN) Consortium to turn TMPRSS2-ERG into a urine biomarker of prostate cancer. Through this collaboration, we developed a new urine test that can help aid early detection of and treatment decisions about prostate cancer. The test supplements the standard prostate specific antigen (PSA) screening test. The multiplex test is designed to detect the gene fusion, TMPRSS2:ERG gene fusion in combination with another marker that was identified earlier, PCA3. The combination was more predictive of cancer than either marker alone. We employed this assay in nearly 1300 patients across 10 medical centers to demonstrate its performance to detect prostate cancer. In addition to detecting prostate cancer, quantitative levels of the TMPRSS2-ERG fusion transcript in urine are associated with aggressive features of prostate cancer including high Gleason grade and large tumor volume. In September of 2013, we were able to introduce the urine TMPRSS2-ERG + PCA3 test in our CLIA lab facilitated by collaboration with Gen-Probe/Hologic, this test is now available for patient testing.

The identification of TMPRSS:ERG fusions as a driver mutation for approximately 50% of prostate cancers has led to multiple efforts to target this pathway as a treatment for prostate cancer. Research in our laboratory revealed that PARP-1 is a downstream target of the ETS family and that inhibition of this pathway results in tumor inhibition in preclinical models. PARP1 function is critical to the pro-tumorigenic functions of the androgen receptor (AR): Specifically, new data strongly support the use of ABT-888, a PARP inhibitor, as a means to antagonize PARP1-dependent AR function in prostate cancer cells, especially in tumor models with intrinsically high AR activity or which have achieved castration resistance. These results led to the clinical trial, "A Randomized Gene Fusion Stratified Phase 2 Trial Of Abiraterone With Or Without ABT888 For Patients With Metastatic Castration-Resistant Prostate Cancer" with the primary objectives of 1) evaluating the role of ETS gene fusion as a predictive biomarker for response to hormone therapy (abiraterone) alone or hormone therapy plus PARP-1 targeted therapy (ABT-888) in patients with metastatic castration resistant prostate cancer and; 2) evaluating whether the addition of PARP targeted therapy is superior to hormone therapy alone based on ETS gene fusion status. This collaboration involves Drs. Hussain, Feng, Kunju, and Chinnaiyan among others from our Prostate SPORE.

B. Discovery and validation of genetic factors of prostate cancer including HOXB13

SPORE Project, "Defining Genetic Risk Factors for Brothers of Men with Prostate Cancer", led by Drs. Kathy Cooney and Julie Douglas sought to define genetic risk factors associated with prostate cancer, including the importance of BRCA-1 as a risk factor in prostate cancer.

Prostate cancer linkage to chromosome 17 markers was initially reported by the UM Prostate Cancer Genetics Project (PCGP) and was subsequently confirmed by additional collaborative reports. Using a targeted resequencing approach focusing on probands from 94 prostate cancer families, a recurrent mutation (G84E) in the homeobox transcription factor HOXB13 was identified. Through analysis of over 5000 DNA samples from men with prostate cancer diagnosed at UM or Johns Hopkins University, this mutation was shown to be associated with a significantly increased risk of prostate cancer (10-20 fold). The mutation is more common in young men with prostate cancer emphasizing the role of germline variation in early-onset disease. This collaboration was facilitated by Dr. Cooney with Dr. Isaacs from the Johns Hopkins SPORE. The relationship between HOXB13 G84E mutation and young age at cancer diagnosis was confirmed in two recent reports from Canada and Sweden. Notably, four additional rare HOXB13 mutations were identified in patients and cell lines.

C. Discovery, characterization and translation of EZH2 and other tissue biomarkers of prostate cancer

Our Prostate SPORE group has been engaged in the molecular profiling of prostate cancer to identify novel clinical biomarkers, candidate regulatory genes, and therapeutic targets. One of the key candidate genes that emerged from our tumor profiling studies was the Polycomb Group (PcG) protein EZH2. EZH2 is a central component of Polycomb Repressive Complex 2 (PRC2), possessing histone methyltransferase enzymatic activity. Using lethal metastatic prostate cancer samples from UM SPORE Rapid Autopsy Program, our group identified EZH2 as being over-expressed in metastatic prostate cancers and aggressive clinically localized prostate tumors. Subsequent studies by our group and others suggested that EZH2 overexpression was common in other aggressive sub-types of multiple solid tumors including breast, lung and bladder cancers.

The Prostate Cancer Biomedical Informatics Network (PCBIN, formerly Inter SPORE Prostate Biomarker Study (IPBS)), developed informatics resources that are interoperable and readily available to investigators within and outside of the consortium to search for prostate cancer samples and to discover potentially available materials and data from the participating PCBIN sites. Under the retrospective arm, a promising biomarker EZH2 discovered in the UM Prostate SPORE is being analyzed by several SPORE institutions. EZH2 as a biomarker of aggressive solid tumors has been licensed to Ventana/Roche for the development of a tissue biomarker for assessment of aggressiveness in solid tumors including prostate cancer.

D. Translation of an MDM2 small molecule inhibitor in prostate cancer and other cancers

Previous SPORE Project, "Preclinical Evaluation and Clinical Development of Potent Small-Molecule Inhibitors of the MDM2-p53 Interaction as a New Therapy for the Treatment of Human Prostate Cancer" led by Drs. Shaomeng Wang and David Smith, aimed to develop a potent MDM2 inhibitor to reactivate the tumor suppressor function of p53 as a new strategy for the treatment of advanced human prostate cancer. Dr. Wang's team designed a class of highly potent and orally active small-molecule inhibitors to target the MDM2-p53 interaction. Dr. Wang, in close collaboration with Ascenta Therapeutics, a small biotech company that licensed the MDM2 inhibitor technology from the University of Michigan, successfully secured Sanofi as the development partner for clinical development and commercialization. By actively working together with Sanofi and Ascenta, a clinical candidate was selected in December, 2010 and the IND enabling studies were completed in October 2011. A Phase I trial commenced in early 2012 and the first US site at MSKCC is active and enrolling patients with MDM2 amplification with p53 wild-type status. The newly synthesized MDM2 inhibitors are 100-times more potent than Nutlin-3, the gold standard of MDM2 inhibitors in the field.

E. Evaluating Anti-CCL2 as a Therapy for Metastatic Prostate Cancer

The SPORE Project, "Inhibition of CCL2 to Treat Metastatic Prostate Cancer" was co-led by Drs. Ken Pienta and Maha Hussain. The 5 year translational goal of this project was to further the understanding of systemic inhibition of CCL2 on prostate cancer growth and developing a novel treatment strategy of inhibiting macrophage infiltration into prostate tumors to inhibit growth. This project included "first in prostate" trials with an anti-CCL2 antibody.

UM Prostate SPORE investigators also collaborated with the Harvard Prostate SPORE to study the importance of CCL2 genotype on cancer risk and aggressiveness and investigate inherited variants and prostate cancer aggressiveness utilizing the Harvard patient cohort. An association study between six single nucleotide polymorphisms (SNP) in the CCL2 gene and prostate cancer clinicopathologic variables was conducted in a large hospital-based Caucasian patient cohort (N = 4,073). Genetic variation at CCL2 was associated with markers of disease aggressiveness. Three SNPs, each in strong linkage disequilibrium, was associated with a higher (>7) biopsy Gleason score: CCL2 -1811 A/G, -2835 A/C, and +3726 T/C (P = 0.01, 0.03, and 0.04, respectively). The CCL2 -1811 G allele was additionally associated with advanced pathologic stages in patients who underwent radical prostatectomy (P = 0.04). In haplotype analysis, we found that the frequency of a common haplotype, H5, was higher among patients with D'Amico good risk features (P (permutation)=0.04). These results supported the association of specific CCL2 variants on prostate cancer development and progression.

Other studies clearly demonstrated that CCL2 was a major chemoattractant for monocytes that trafficked to the tumor site that were then educated/differentiated into cancer promoting tumor-associated macrophages (TAMs). In collaboration with Dr. David Rowley at Baylor University, an expert in tumor-stroma interactions, we demonstrated the role of CCL2 in prostate tumor growth and metastasis through its regulatory role in mediating monocyte/ macrophage infiltration into the tumor microenvironment and stimulating phenotypic changes to TAMs. Dr. Rowley had demonstrated that human prostate cancer reactive stroma is composed of myofibroblasts that initiate during prostatic intraepithelial neoplasia and continually co-evolve with adjacent carcinoma during organ-confined progression. Our combined studies resulted in a funded U01 application utilizing the extensive set of human normal and prostate cancer samples at Baylor University and University of Michigan SPORE Tissue Banks, including samples obtained through the Rapid Autopsy Program and from mouse models of prostate cancer growth in primary prostate and bone.

Successful translation of these findings was accomplished through two clinical trials. In collaboration with the company Centocor and headed by Dr. Pienta, we conducted a multi-center trial testing the antibody to CCL2 provided by Centocor. Studying the CCL2 axis also revealed that inhibiting the CCL2/CCR2 axis also inhibited osteoclast formation. MLN1202, an antibody to CCR2, was utilized in a SWOG cooperative group Phase II clinical trial.

back to top

Bookmark and Share


How You Can Help

Give to cancer research