A study from Nature Medicine has significantly refined the use of prostate-specific antigen (PSA) screening for prostate cancer using an individual’s genome to fine-tune their diagnosis.
PSA under scrutiny
Prostate cancer is the second most prevalent form of cancer in men and makes up 15.1% of all male cancer diagnoses. To screen for prostate cancer, clinicians typically measure the levels of PSA. Prostate cells usually produce PSA in small amounts, but the levels noticeably increase when something goes wrong with the prostate.
However, PSA screening has continued to give clinicians a headache. It has been estimated that fewer than 1 in 3 men with high PSA levels actually have prostate cancer. This means that most men identified as at risk from this test undergo an unnecessary biopsy. As a result, PSA screening is seen as a woefully inaccurate method to diagnose prostate cancer.
Genomics to the rescue
The PSA screen can still provide valuable information regarding prostate health and so this research, led by scientists at Stanford Medicine, hoped to refine PSA diagnoses using genomics.
Every individual will have innate differences in PSA level, completely unrelated to prostate health. By sequencing germline DNA (from saliva, blood etc.), the researchers could screen for genetic variants that influence PSA levels. This means the naturally high PSAers and low PSAers can be identified before the PSA levels are measured in a clinical setting, refining the process.
PSA redeemed
Using 128 genomic sites linked to PSA levels, and evaluating the resulting PSA polygenic risk score, the researchers were able to improve the detection of aggressive forms of prostate cancer using individual-adjusted PSA levels. When applied to a large dataset of men with and without prostate cancer, its diagnosis could have spared 31% of men from unnecessary biopsy.
There are a few caveats, namely, by using mostly men of European ancestry, the benefit of genomics was only noticeable for this group. However, the researchers are keen to address this limitation and lead researcher, Linda Kachuri, stated that they “ideally, we want to come up with a single score that works well for everybody, across the spectrum of ancestry.”
