Dr. Vince on evaluation of prostate cancer polygenic risk scores

In this video, Randy A. Vince Jr, MD, MS, discusses the European Urology study “Assessing the Clinical Utility of Published Prostate Cancer Polygenic Risk Scores in a Large Biobank Data Set.” Vince is an assistant professor of urologic oncology at Case Western Reserve University and University Hospitals Urology Institute, as well as the director of Minority Men's Health at University Hospitals Cutler Center for Men in Cleveland, Ohio.

Transcription:

Please describe the background for this study.

Just for general background, polygenic risk scores incorporate what we call SNPs, which are obtained from what we call GWAS, which is a genome-wide association study. By looking at these SNPs, we're trying to look at the hereditability of prostate cancer. There have been a number of GWAS studies that have developed polygenic risk scores for prostate cancer. And what they found is that there is a really strong association for most polygenic risk scores with the incidence of prostate cancer or the risk of developing prostate cancer. However, what we know is that it is much bigger than just whether or not you develop prostate cancer. Because there's a whole debate among a lot of people in terms of, should we be diagnosing people with low-risk prostate cancer, that's pretty indolent? The reason why some of the guidelines changed in the past to recommend against prostate cancer screening was because we were overdiagnosing low-risk disease and overtreating people with low-risk disease. So now, we know there's a move to try to shift from just diagnosing anyone with prostate cancer, to purely being able to focus in on those men with clinically significant disease. That way, they can move on to biopsies and things like that and potential treatment. What we wanted to do with this was to look at 1) Out of the polygenic risk scores that have been published, how do they perform on an external data set with the association for incidence of prostate cancer and then for those top-performing polygenic risk scores, how do they associate with the aggressiveness of prostate cancer; in other words, does an increase in polygenic risk score correlate with increasing disease aggressiveness?

Considering 16 PRSs were evaluated, can you elaborate on the specific criteria used to identify the top-performing ones for further analysis of aggressive prostate cancer?

In terms of the 16 polygenic risk scores that were selected in general, our external data set from the University of Michigan was predominantly men of European ancestry. And so we wanted to use those polygenic risk scores that would develop on men of exclusively European ancestry. That way, it will be a strong correlation. And then the first step was, again, to look at the association with the incidence of prostate cancer for an increasing polygenic risk score. And what we found is that across the board, all 16 scores did relatively well with the association between an elevated polygenic risk score and prostate cancer incidence. However, what we did was those top 3 performing - so when we look at what we call an odds ratio, those risk scores that have the strongest association with prostate cancer incidence, we then took those polygenic risk score constructs and then applied it to evaluating disease aggressiveness. And so that's how we selected those top 3 polygenic risk scores to determine what to evaluate the association between disease aggressiveness with the increasing score.

This transcription was edited for clarity.