Study
PLCO
(Learn more about this study)
Project ID
2009-0504
Initial CDAS Request Approval
Jun 19, 2009
Title
Phase III validation using PLCO specimens of a consensus panel of ovarian cancer biomarkers developed by SPORE/EDRN investigators: Study 4: Validation using PLCO serial specimens of PLCO Ovarian Cance
Summary
E.1. Abstract: We propose to validate and potentially to refine a multimodal screening strategy designed to be cost-effective in post-menopausal women. We have proposed to use this strategy in high-risk women in the Novel Markers Trial, a randomized controlled trial (RCT) proposed in the competing renewal application for the Pacific Ovarian Cancer Research Consortium (POCRC) Ovarian SPORE. The strategy uses CA125 and HE4, interpreted using the Parametric Empirical Bayes (PEB) longitudinal algorithm, to select women for imaging (transvaginal ultrasound). To implement this strategy, women will be tested annually for rising CA125 and/or HE4. If either marker is above a PEB rule threshold corresponding to 95% specificity the woman will be referred for imaging. If imaging is positive, or if both markers are above their 95% specificity thresholds, the woman will be referred for surgical consult. We propose here to use the serial samples from the PLCO trial to estimate the performance characteristics of this approach (Aim 1), and to explore protocol variations to determine whether or not adding Mesothelin and/or MMP7 to the panel improves performance (Aim 2). Our goal is to be as certain as possible of the optimal screening protocol before we implement a prospective trial. In the Novel Markers Trial as in the proposed work we will include assays performed on an automated platform so that the data generated by the proposed study can be used with other data to support an FDA application should that be warranted in the future. Project Co-Leaders: Nicole Urban, ScD, Martin McIntosh, PhD, Charles Drescher, MD. Alterations in global (genome-wide) DNA methylation levels are thought to promote carcinogenesis by weakening chromosomal stability and altering normal gene expression patterns. Changes in the epigenome are believed to be early events in cancer and recently have been used as early biomarkers of disease. Genome-wide cytosine hypomethylation occurs primarily in repetitive sequences of DNA that have no obvious impact on gene expression; however, they can also occur in oncogenes that would normally be methylated and silenced. In a case-control study of renal cancer conducted in Central and Eastern Europe, we found that levels of LINE-1 methylation in DNA from peripheral blood lymphocytes were significantly higher in cases compared to controls. Furthermore, higher levels of LINE-1 methylation acted as a strong risk factor for renal cancer, with the association appearing to be more pronounced among current smokers. However, because these samples were collected post-diagnosis, it remains unclear whether the observed differences occurred prior to, or as a result of carcinogenesis. To elucidate this issue, in the proposed study, we will examine the relationship between global DNA methylation (LINE-1) and renal cell cancer risk using DNA from prospectively collected blood samples obtained in the PLCO Screening Trial. Using prospectively collected genomic DNA from the PLCO Screening Trial, this study will provide an invaluable opportunity to examine the role of global DNA methylation in predicting the risk of renal cancer. This study will allow us to follow-up leads generated in the Central and Eastern European Renal Cancer Case-Control study but with pre-diagnostic samples. Results from the proposed study will greatly expand our knowledge on the role of DNA methylation and subsequent renal cancer risk and will help us determine whether global methylation changes are associated with cancer susceptibility, or alternatively as a result of disease.
Aims
E.2. Specific aims of the proposal: Stored serum samples from the PLCO trial were previously requested by our group; our joint proposal with Dan Cramer entitled "Phase III validation using PLCO specimens of a consensus panel of ovarian cancer biomarkers" developed by SPORE/EDRN investigators was approved in 2005. Analyses have been conducted as described in the document "PLCO Ovarian Cancer Early Detection Biomarker Studies, Data Release Strategy & Timeline, June 13, 2008." Having identified the most informative biomarkers using the unblinded half of the proximal samples from PLCO trial as well as serial samples from the Carotene and Retinol Efficacy Trial (CARET), we will use serial samples from the PLCO to validate, refine and further validate a screening algorithm. Our aims are as follows: 1) Using the unblinded half of the PLCO serial sample data, a) Validate 2-marker and 4-marker algorithms developed using CARET data, and b) Refine the 2-marker and 4-marker algorithms using the unblinded half of the PLCO serial sample data; 2) Using the blinded half of the PLCO serial sample data, validate the refined 2-marker and 4-marker algorithms. We base our power calculations on analyses to test the hypothesis that at comparable specificity the Parametric Empirical Bayes (PEB) longitudinal algorithm rule using the 4-marker panel will on average detect cancer earlier than CA125 alone as a first-line screening test for pre-clinical ovarian cancer. We propose to study the following specific aims in 242 renal cell cancer cases and 484 controls from a nested case-control study of renal cell cancer within the PLCO prospective cohort: <!--[if !supportLists]-->1. <!--[endif]-->To evaluate the association between global methylation levels in long interspersed nucleotide elements (LINE-1) in genomic DNA and risk of renal cell cancer, and replicate findings from the Central and Eastern European Renal Cancer Study to determine whether the observed differences are also apparent prior to cancer development. <!--[if !supportLists]-->2. <!--[endif]-->To explore whether the association between methylation levels and risk of renal cell cancer is modified by factors associated with oxidative stress, such as smoking and dietary variables obtained from PLCO questionnaire data. A secondary aim of the proposed study is to identify contributing factors to variation in genome-wide methylation among controls.
Collaborators
Chow, Wong-Ho (NCI, DCEG)
Hofmann, Jonathan (NCI, DCEG)
Martin McIntosh (Fred Hutchinson Cancer Research Center)
Nicole Urban (Fred Hutchinson Cancer Research Center)
Purdue, Mark (NCI, DCEG)
Rothman, Nathaniel (NCI, DCEG)
Related Publications
