Skip to Main Content

An official website of the United States government

Principal Investigator
Name
Nora Pashayan
Degrees
BSc MSc MSt MD PhD FFPH
Institution
University College London
Position Title
Tenured Senior Clinical Lecturer
Email
About this CDAS Project
Study
PLCO (Learn more about this study)
Project ID
PLCO-85
Initial CDAS Request Approval
May 20, 2014
Title
The implications of polygenic risk-stratified screening for prostate cancer on overdiagnosis and cancer-specific mortality
Summary
The USPSTF recommends against PSA based screening for prostate cancer on the grounds that the expected harms of screening (mainly overdiagnosis and subsequent overtreatment) outweigh the potential benefits. Overdiagnosis is defined as detection by screening of tumours that would not have presented clinically in a person’s lifetime in the absence of screening.

To date, genome-wide association studies (GWAS) have identified 76 prostate cancer susceptibility loci, which explain approximately 30% of the familial risk of prostate cancer. Polygenic risk profile based on these loci could be used for population stratification and targeted screening. A risk-stratified screening strategy for prostate cancer with eligibility for screening based on an absolute risk that is dependent on age and polygenic risk-profile improves the efficiency of the screening programme. However, there is yet no empirical evidence that incorporating polygenic profiling into a screening program will reduce the burden of overdiagnosis and reduce prostate cancer-specific mortality.

We will explore whether polygenic risk-stratified screening for prostate cancer would improve the benefit to harm ratio of screening.

We propose Markov Chain multi-state modelling in case-cohort study design to estimate the transition intensities (instantaneous rate) between the different states (normal, pre-clinical screen detectable, clinical, and overdiagnosed).
We will derive polygenic risk score based on the known 76 prostate cancer susceptibility loci on 2252 men included in the CGEMS prostate cancer GWAS. The genotype data will be linked to the phenotype data (attendance of screening, outcome of screening, cancer characteristics, participant characteristics, and vital status). We will derive risk-stratified estimates of rate of overdiagnosis. Using the optimal risk score cut point, we will estimate the relative reduction in cancer death among the intervention and control arm stratified by risk categories.

PLCO as a randomised screening trial with data on outcome of several rounds of screening and interval cancers will provide the ideal opportunistic data to estimate the benefits and harms of polygenic risk-stratified screening for prostate cancer. The findings would inform screening policy.
Aims

1. To model the natural history of prostate cancer in the preclinical phase
2. To assess the effect of absolute risk dependent on age and polygenic risk score and absolute risk dependent on age, polygenic risk, family history and PSA level on disease progression parameters
3. To derive model-based estimate of the proportion of screen-detected cancers likely to be overdiagnosed
4. For different absolute risk cut points, to estimate the relative reduction in mortality
5. From the above, to devise an optimal screening strategy for use of the polygenic risk profile in selecting a population for early detection, and to quantify the benefits and harms of such a strategy in terms of prostate cancer deaths prevented and reduction in overdiagnosed cases

Collaborators

Prof Paul Pharoah, University of Cambridge
Prof Stephen Duffy, Queen Mary University of London
Dr Ardo van den Hout, University College London