The success of the current paradigm of PSA screening and prostate cancer treatment, while contributing to a reduction in mortality, has also led to the problems of over diagnosis and over treatment. In an effort to reduce over treatment in men with """"""""low risk"""""""" tumors (low Gleason grade and low tumor volume), programs of active surveillance are increasingly being used to allow men to forgo or delay treatment, while being carefully monitored for the evidence of higher risk pathology. However, the ability to safely offer active surveillance as an alternative to immediate treatment depends on how well the biopsy can capture the phenotype of the tumor in the whole prostate. The need to predict the possibility of high grade tumor for a patient with a low grade biopsy has only recently emerged as an important clinical problem, coincident with increasing interest in active surveillance for carefully selected men. There are currently no biomarkers that were developed specifically to predict the risk of high-grade cancer in prostates of men who choose not to be treated based on a biopsy that shows a small amount of low grade cancer. We propose a multi-institutional study to develop biomarkers to predict high grade tumor in men with low grade biopsies. This proposal is based on an ongoing clinical and biomarker collaboration among the prostate cancer SPORE centers, the Inter-SPORE Prostate Biomarkers Study (IPBS). The IPBS is a prospective collaboration among the 11 SPOREs to evaluate prognostic biomarkers, and also to rigorously evaluate the influence of pre- analytic variability in tissue processing and handling on the performance of biomarker assays. IPBS investigators have a proven infrastructure for collection, distribution, and analysis of biomarkers in biopsy tissue, with linkage by a bioinformatics network. This study will also benefit from collaboration between the Early Detection Research Network (EDRN) and the prostate SPOREs to incorporate a validation cohort of patient samples accrued through the EDRN. Development of the biomarker panel is based on the hypothesis that emergence of high-grade prostate cancer is driven by increased genomic instability, global chromatin reorganization and adaptation of cancer cells to hypoxic conditions in the tumor microenvironment. Together these forces generate an enormous diversity in the tumor, which in some cells leads to the phenotype of Gleason grade (GG) 4. Further we hypothesize that markers of GG4 cancer exist in tumors in which the GG3 growth pattern occurs in conjunction with GG4, i.e. Gleason score (GS) 7, and that these markers may be detectable in biopsy specimens exhibiting only GG3. All of the candidate biomarkers were selected based on their consistency with these hypothesized mechanisms, and a number of the candidate biomarkers have also been strongly linked to high grade tumors. We will measure genomic instability either directly with FISH, or by analyzing the expression of genes and proteins that are involved in regulation of the DNA damage checkpoint, modification of chromatin, or that link hypoxia to genomic instability. We propose to develop and validate the biomarker panel to identify Gleason upgrading in the prostatectomy specimen through a stepwise process. In the first aim, tissue microarrays (TMAs) with samples from prostates with pure GS 6 (3+3), pure GS 3+4, and pure GS 4+3 will be evaluated with the candidate biomarkers (100 patients per GS category. This screen will identify the biomarkers most closely associated with a signature of high grade (GG 4) tumor. Biomarkers with strong associations in this aim will be tested on an independent set of specimens with biopsy GS 3+3 in Aim 2. Samples from 50 patients with biopsy GS 3+3 and also pure GS 3+3 in the prostatectomy will be compared to samples from 50 patients also with biopsy GS 3+3 but GS 7 (3+4 or 4+3) in the prostatectomy. The goal will be to evaluate whether the biomarkers measured on the biopsy can discriminate prostates with only GG 3 from those with GG 4 in conjunction with GG3. Biomarkers that pass this initial validation phase will undergo an independent validation using specimens from an EDRN cohort carefully established for the purpose of biomarker validation.
This aim will also evaluate whether measurements on biopsy specimens with only GS 3+3 can predict the the presence of GG 4 in the prostate. This hypothesis-driven study will be larger than previous studies evaluating a possible signature of high grade prostate tumor, and will involve samples representing pure tumor phenotypes. It makes use of a proven consortium of prostate SPORE investigators with an established infrastructure for specimen and biomarker evaluation, and a carefully developed EDRN reference set, combining two NCI-funded research entities with excellence in prostate cancer surgery, pathology and biomarker research. The collaborative experience of these consortia, and the access to appropriate tissues makes this project """"""""shovel ready."""""""" If successful this study could change the paradigm for treatment decisions in men with biopsy GG 3 and help to reduce over treatment of prostate cancer.

Public Health Relevance

Over treatment of prostate cancer (treatment that does not extend lifespan) has increased dramatically with the advent of PSA testing. Active surveillance (AS) is an approach to reduce over treatment by carefully monitoring men with low-risk prostate cancer and delaying treatment as long as the tumor continues to show low risk characteristics. A problem with the AS paradigm is that the determination of whether a cancer is low risk depends primarily on the Gleason grade as measured on the biopsy specimen. Although most men with only Gleason grade 3 on the biopsy have low risk prostate cancer suitable for AS, more than 25% of such men will be found to have higher grade (grade 4) tumor that was not captured by the biopsy, when they undergo prostatectomy. We propose to develop a panel of biomarkers that can be measured on biopsy specimens with Gleason grade 3 to predict whether higher grade tumor exists elsewhere in the prostate. Our hypothesis is that genomic instability, hypoxia in the tumor microenvironment, and chromatin modifications drive the development of high grade tumor. We have chosen biomarkers consistent with this hypothesis, and which also have been associated with Gleason grade, or with prostate cancer prognosis. We will develop a predictive model based on biomarkers that are most strongly associated with Gleason grade, and then validate the model in samples derived from cohorts at Johns Hopkins and from the Early Detection Research Network's Prostate Cancer Reference Set. If successful, this project may lead to methods that can increase the acceptability and safety of following patients with a program of AS, and reduce over treatment of prostate cancer.

National Institute of Health (NIH)
National Cancer Institute (NCI)
High Impact Research and Research Infrastructure Programs (RC2)
Project #
Application #
Study Section
Special Emphasis Panel (ZCA1-RPRB-J (O9))
Program Officer
Hruszkewycz, Andrew M
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Johns Hopkins University
Schools of Medicine
United States
Zip Code
Trock, Bruce J; Fedor, Helen; Gurel, Bora et al. (2016) PTEN loss and chromosome 8 alterations in Gleason grade 3 prostate cancer cores predicts the presence of un-sampled grade 4 tumor: implications for active surveillance. Mod Pathol 29:764-71
Inoue, Lurdes Y T; Trock, Bruce J; Partin, Alan W et al. (2014) Modeling grade progression in an active surveillance study. Stat Med 33:930-9