In the news

Combine genetic variants with prostate-specific antigen (PSA) testing to improve the performance of prostate cancer screening

A genome-wide association study was performed to locate genetic variants associated with PSA levels, based on data from 15,757 Icelandic and 454 British men not diagnosed with prostate cancer. This was followed by an assessment of the improvement in predictive accuracy of PSA testing when combined with either these genetic variants or all prostate cancer susceptibility variants in 415 Icelandic and 1,291 British men with information on biopsy-outcome (i.e. biopsy positive or negative). 

Variants at six loci were found to be associated with PSA levels, two of which were not also associated with prostate cancer risk. By applying the combined genetic effect of all 23 known PSA and prostate cancer susceptibility variants on the commonly used PSA threshold for biopsy, the test was slightly better able to discriminate between outcomes: the area under the ROC curve (AUC, which varies from 0.5-1.0, where a value of 1.0 indicates a perfect performance) improved from 0.704 to 0.732 in Icelandic men and 0.571 to 0.636 in British men. Applying this genetic correction, 6-7% of Icelandic men would be reclassified with respect to whether they should undergo biopsy.

The authors propose that a personalised PSA cut-off value, based on genotype, should be used when deciding to perform a prostate biopsy.

Although there is substantial international variation, PSA screening for prostate cancer is currently not offered systematically in the UK because it is unclear whether the benefits outweigh the harms due to the large number of false positive results (see previous news). This study is an important step towards trying to improve the balance, by using genetic factors in addition to PSA testing to determine whether a man is likely to have prostate cancer and should go on to have further testing. Although the addition of genetic variants associated with prostate cancer did improve the predictive accuracy of PSA testing, the improvement was marginal and the advice for the vast majority of men would be unchanged. As the accompanying perspective points out, “although this study presents an auspicious beginning, on the basis of the AUCs and reclassification analysis, it does not appear that SNP-adjusted PSA values are primed for translation into clinical practice without additional follow-up studies”. Further large population studies are needed to investigate whether using individual genetic factors, in combination with age and other risk factors, to modulate the PSA threshold for biopsy would substantially improve outcomes.