Prostate cancer has the highest incidence rates among men with African ancestry; it’s the second cause of cancer death and is related to a plethora of health disparities. Polygenic risk scores (PRS) have the potential to improve screening of some cancers, as they provide information about an individual’s genetic risk of a disease. The Darst Group, from the Fred Hutch Public Health Sciences Division, along with colleagues from the University of Southern California, including co-first author Dr. Fei Chen and last author, Dr. Christopher Haiman, and from the Veteran Affairs Million Veteran Program, utilized PRS as a measure of genetic susceptibility to prostate cancer in a recent study. The goal of the study was to validate the ability of a multi-ancestry PRS, which they developed in a previous publication, to predict prostate cancer risk in men with different ancestries. Their findings were published recently in eLife.
The study included a total of 31,925 cases and 490,507 controls from European, African, and Hispanic populations from the Veteran Affairs Million Veteran Program and other diverse replication studies. A multi-ancestry PRS was calculated as a weighted sum of the number of prostate cancer risk alleles carried, weighted by variant-specific conditional weights, based on Darst’s previous paper. The association between the PRS and prostate cancer risk was analyzed in each replication study separately and then meta-analyzed across studies using a fixed-effects inverse-variance weighted meta-analysis. Two of the replication studies were utilized to evaluate age-specific effects of PRS on prostate cancer risk. Absolute risk of prostate cancer was calculated for a range of ages for each PRS category using the association results along with incidence and mortality rates from the Surveillance, Epidemiology, and End Results (SEER) Program. Their findings validated the ability of the multi-ancestry PRS to identify men at increased risk of developing prostate cancer across diverse populations, suggesting the potential of the PRS to help men to make informed decisions regarding prostate-specific antigen (PSA) screening.
Dr. Darst, the co-first author, summarized the importance of the results, “These findings validate the predictive ability of the PRS in men from diverse populations and underscore the potential for the PRS to impact prostate cancer screening practices. In particular, the PRS could be used to identify men at highest risk of prostate cancer who may most benefit from receiving screening. Likewise, very few prostate cancer cases were found to have low PRS and reducing screening in men with low PRS could drastically reduce the over-diagnosis of prostate cancer. Although the PRS was highly predictive of disease risk across diverse populations, the performance of the PRS was lower in men of African ancestry compared to men of European ancestry; it will be crucial for future genetic studies to include more African ancestry men in order to improve the performance of the PRS.” Younger men exhibited a stronger association between PRS and prostate cancer risk, younger men may benefit most from genetic testing, which could inform the age at which PSA testing is initiated and frequency of testing thereafter. “Further, while our findings verify the validity of the PRS, it will be important to investigate the clinical utility of incorporating this type of information into screening, particularly among Black men, who face the greatest incidence and mortality rates in the US and the state of Washington”, concluded Darst.
This research was supported by the National Cancer Institute, Prostate Cancer Foundation, Achievement Rewards for College Scientists Foundation, and the Million Veteran Program.
Fred Hutch/University of Washington/Seattle Children's Cancer Consortium member Burcu F. Darst contributed to this work.
Chen F, Darst BF, Madduri RK, Rodriguez AA, Sheng X, Rentsch CT, Andrews C, Tang W, Kibel AS, Plym A, Cho K, et al. 2022. Validation of a multi-ancestry polygenic risk score and age-specific risks of prostate cancer: A meta-analysis within diverse populations. Elife. 11:e78304.