Precision Medicine Approach May Improve Prostate Cancer Screening

Researchers from Stanford Medicine have developed a precision medicine approach that may help flag and separate normal variations in prostate-specific antigen (PSA) caused by genetic factors unrelated to cancer, which could improve the accuracy of prostate cancer screening.

The personalized approach, outlined in a study published last week in Nature Medicine, is an attempt to address the overdiagnosis and overtreatment of prostate cancer that can result from PSA screening.

PSA screening is a common test for prostate cancer, but the test often suggests the presence of cancer when there is none, the researchers explained. This has led the United States Preventive Services Task Force (USPSTF) to recommend against PSA-based screening in patients over the age of 70 and suggest that younger patients be given a choice about whether or not to undergo screening.

Elevated PSA levels can be a sign of prostate cancer, but other factors, such as an enlarged prostate, an infection, inflammation, or older age can also cause variations in PSA.

“Some men have higher PSA levels due to their genetics,” said John Witte, PhD, a professor of epidemiology and population health and of biomedical data sciences at Stanford and the senior author of the study, in the press release. “They don’t have cancer, but the higher PSA level leads to a cascade of unnecessary medical interventions like biopsy.”

Evaluations of the clinical utility of PSA screening have demonstrated that of patients with elevated PSA, less than one-third had prostate cancer. Research examining the prevalence of prostate cancer among those with normal PSA has also found that 15 percent of patients were later confirmed to have cancer after screening.

These findings suggest that there is ‘noise’ present in PSA screening that makes it less accurate at identifying prostate cancer, the research team indicated.

“To improve the signal, which is the variation in PSA levels caused by a prostate tumor, we subtract out the noise, which in this case comes from genetics,” said Linda Kachuri, PhD, an assistant professor of epidemiology and population health at Stanford and the lead author of the study.

To reduce this genetic noise, the researchers evaluated the genomes and PSA levels of 95,768 men of European ancestry without prostate cancer. This analysis revealed that 30 to 40 percent of PSA level variation was determined by genetic factors unrelated to cancer.

By separating these normal sources of noise within PSA screening, the research team aims to make PSA a more specific, accurate biomarker that can determine the presence of prostate cancer.

The researchers successfully identified 128 sites within the genome that can impact patient PSA levels, and they developed a polygenic score to help calculate PSA levels by taking variations at these sites into account.

To evaluate the score, the research team applied it to a dataset of just under 32,000 men without prostate cancer. Using these data, the score accurately predicted nearly ten percent of PSA variation.

However, the tool was significantly more effective among patients of European ancestry than their East Asian or African counterparts.

When the score was applied to a group of men with and without prostate cancer as confirmed via biopsy, the researchers determined that approximately 30 percent of the cohort could have been spared the biopsy.

Despite these successes, the tool does have limitations. Notably, the score would have missed roughly nine percent of positive biopsies. The researchers indicated that most of these were cases of slow-growing tumors that may not have required treatment, but the oversight highlights room for improvement in the score.

Since the tool was also developed with data from mostly men of European ancestry, the press release states that the research team will be collaborating on a larger study with the Million Veteran Program to incorporate more diverse patient data.