261417 Use of multi-gene panels involving single nucleotide polymorphisms (SNPs) for prostate cancer risk assessment

Sunday, October 28, 2012

Julian Little, PhD , Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, ON, Canada
Brenda Wilson, MBChB, MSc , Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, ON, Canada
Ron Carter, PhD , Department of Pathology and Molecular Medicine, McMaster University, Hamilton, ON, Canada
Eva Tomiak, MD , Department of Genetics, Children's Hospital of Eastern Ontario, Ottawa, ON, Canada
Joseph Beyene, PhD , Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, Canada
Kathryn Walker, MSc , Evidence-Based Practice Center, McMaster University, Hamilton, ON, Canada
Pasqualina Santaguida, PhD , Evidence-Based Practice Center, McMaster University, Hamilton, ON, Canada
Parminder Raina, PhD , Evidence-Based Practice Center, McMaster University, Hamilton, ON, Canada
Background: Prostate cancer presents many challenges, in terms of risk prediction, screening, and distinguishing latent from aggressive disease. Genome-wide association studies have identified many single nucleotide polymorphisms (SNPs) which appear to be associated with disease risk and prognosis. We conducted a systematic review to identify, synthesize and appraise the literature on the analytic validity, clinical validity, and clinical utility of using panels of SNPs in the prediction, detection, and management of prostate cancer.

Methods: We searched MEDLINEŽ, Cochrane CENTRAL, Cochrane Database of Systematic Reviews, and EMBASE from inception to October 11, 2011, using combinations of controlled vocabulary and text words. We also searched grey literature. Eligibility criteria were developed separately for the three areas of interest. We abstracted data on the area under the receiver-operator characteristic curve, sensitivity, specificity, and other relevant metrics.

Results: From 1,998 unique citations, 14 were retained after title and abstract and full text screening. All addressed risk stratification (clinical validity), were based on case control designs, and none were conducted in routine clinical settings. They evaluated 15 individual panels (2-35 SNPs), and none performed more than modestly in predicting presence of prostate cancer or distinguishing aggressive from latent disease. The overall risk of bias was at least moderate.

Conclusions: The evidence on currently available SNP panels does not permit meaningful assessment of analytic validity. The limited evidence on clinical validity is insufficient to conclude that the panels assessed would perform adequately as risk stratification, screening or prognostic tests. No evidence is available on clinical utility.

Learning Areas:
Chronic disease management and prevention
Clinical medicine applied in public health
Epidemiology
Provision of health care to the public
Public health or related research

Learning Objectives:
1. Evaluate the evidence for the use of genomic panels in assessing prostate cancer risk

Keywords: Cancer, Cancer Screening

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am the principal investigator on an AHRQ-funded systematic review of the potential value of SNP-panels in the early detection or prediction of risk for prostate cancer. This is part of my research program in human genome epidemiology, which is focused on gene characterization and the evaluation of the clinical validity and clinical utility of genomic-related health application stages that are needed in the translation of genetic discoveries to health care and public health practice,
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.

Back to: 2039.0: Translational Research in Genomics Across the Lifespan