The effect of assessing genetic risk of prostate cancer on the use of PSA tests in primary care: A cluster randomized controlled trial.

Paul Pinsky,Jacob Fredsøe,Michael Væth,Peter Vedsted,Pia Kirkegaard,Flemming Bro,Adrian Edwards,Karina D Sørensen,Torben F Ørntoft,Jan Koetsenruyter

doi:10.1371/journal.pmed.1003033

Abstract

BackgroundAssessing genetic lifetime risk for prostate cancer has been proposed as a means of risk stratification to identify those for whom prostate-specific antigen (PSA) testing is likely to be most valuable. This project aimed to test the effect of introducing a genetic test for lifetime risk of prostate cancer in general practice on future PSA testing.Methods and findingsWe performed a cluster randomized controlled trial with randomization at the level of general practices (73 in each of two arms) in the Central Region (Region Midtjylland) of Denmark. In intervention practices, men were offered a genetic test (based on genotyping of 33 risk-associated single nucleotide polymorphisms) in addition to the standard PSA test that informed them about lifetime genetic risk of prostate cancer and distinguished between “normal” and “high” risk. The primary outcome was the proportion of men having a repeated PSA test within 2 years. A multilevel logistic regression model was used to test the association.After applying the exclusion criteria, 3,558 men were recruited in intervention practices, with 1,235 (34.7%) receiving the genetic test, and 4,242 men were recruited in control practices. Men with high genetic risk had a higher propensity for repeated PSA testing within 2 years than men with normal genetic risk (odds ratio [OR] = 8.94, p < 0.01). The study was conducted in routine practice and had some selection bias, which is evidenced by the relatively large proportion of younger and higher income participants taking the genetic test.ConclusionsProviding general practitioners (GPs) with access to a genetic test to assess lifetime risk of prostate cancer did not reduce the overall number of future PSA tests. However, among men who had a genetic test, knowledge of genetic risk significantly influenced future PSA testing.Trial registrationThis study is registered with ClinicalTrials.gov, number NCT01739062.

Highlights

Prostate cancer is the most common cancer among men in Europe, affecting about 190,000 men for the first time every year and causing about 80,000 deaths [1,2]
Providing general practitioners (GPs) with access to a genetic test to assess lifetime risk of prostate cancer did not reduce the overall number of future prostate-specific antigen (PSA) tests
By introducing a genetic lifetime risk test for prostate cancer in general practice, we aimed to investigate its possible effect on future PSA testing, with the hypothesis that the number of PSA tests would increase in the small group of men with “high” risk and could be reduced in the large group of men with “normal” risk [17]

Summary

Introduction

Prostate cancer is the most common cancer among men in Europe, affecting about 190,000 men for the first time every year and causing about 80,000 deaths [1,2]. A commonly used method for early detection of prostate cancer is the prostate-specific antigen (PSA) test, this method has limited accuracy [3]. Assessing genetic lifetime risk for prostate cancer has been proposed as a means of risk stratification to identify those for whom prostate-specific antigen (PSA) testing is likely to be most valuable. This project aimed to test the effect of introducing a genetic test for lifetime risk of prostate cancer in general practice on future PSA testing

Objectives

Methods

Results

Conclusion