Some elderly men get prostate cancer tests despite limited benefit
{{#rendered}} {{/rendered}}
About one in five elderly U.S. men get invasive tests to look for prostate tumors even though they may not live long enough to benefit from treatment, a recent study suggests.
The U.S. Preventive Services Task Force (USPSTF), a government-backed panel of independent physicians, in 2008 recommended against routine prostate cancer tests for men at least 75 years old or with a limited life expectancy.
The USPSTF cited concerns that widespread screening often caught harmless tumors that didn't need treatment and led to unnecessary procedures with side effects like impotence and incontinence.
{{#rendered}} {{/rendered}}
But when researchers reviewed data on 3,035 men over 65 who received prostate biopsies from urologists in Michigan in 2013 and 2014, they found 18 percent of these patients had a life expectancy of less than a decade.
Some men with limited life expectancy may have needed biopsies because they had painful symptoms, said senior study author Dr. Khurshid Ghani of the University of Michigan in Ann Arbor.
Or, Ghani said, they may have had elevated blood levels of a substance called prostate-specific antigen (PSA). But PSA levels can't reliably distinguish between dangerous tumors and tumors that are growing so slowly that they don't need treatment, he added.
{{#rendered}} {{/rendered}}
"Men with a limited life expectancy and a minimally elevated PSA, with no symptoms, and a normal prostate on clinical examination are unlikely to benefit from a biopsy," Ghani added by email.
At least half of the men in the study were about 74 years old, and they were typically overweight. The men with limited life expectancy tended to be older and have more medical problems.
Across 42 locations in Michigan that participated in the study, the proportion of biopsies done on men with limited life expectancy ranged from about 4 percent to 39 percent.
{{#rendered}} {{/rendered}}
Overall, 547 men with limited life expectancy received prostate biopsies, the researchers report in the journal European Urology. These men were a bit older and, compared to men with longer to live, they had significantly higher PSA levels as well as abnormal results from rectal exams, the study found.
Biopsy results led to a prostate cancer diagnosis for 69 percent of men with limited life expectancy, and within this group 74 percent received treatment - but fewer than half had aggressive tumors.
One limitation of the study is that some tools for estimating life expectancy can be cumbersome, and many doctors may not use them to determine whether patients needed biopsies, the authors note.
{{#rendered}} {{/rendered}}
Even so, older men also tend to have higher PSA levels and are more likely than younger men to have enlarged prostates for reasons that are benign or due to slow-growing tumors, said Dr. Victoria Tang of the University of California, San Francisco, and the San Francisco VA Medical Center.
"Men with limited life expectancy should not be getting a prostate biopsy if a diagnosis and treatment of prostate cancer will not improve the quality or length of life," Tang, who wasn't involved in the study, said by email.
For these men, it makes sense for doctors to be selective about ordering a biopsy, said Dr. Quoc-Dien Trinh, a urologist at Brigham and Women's Hospital and Harvard Medical School in Boston who wasn't involved in the study.
{{#rendered}} {{/rendered}}
"For example, I am not interested in diagnosing low-risk prostate cancer in a man who is 85 with three recent heart attacks," Trinh said by email.
In a perfect world, doctors would be able to accurately pinpoint exactly how long it might take for tumors to have an effect on patients' lives and objectively compare this to the competing risk of death from other causes, said Dr. Alexander Kutikov, a researcher at Fox Chase Cancer Center in Philadelphia who wasn't involved in the study.
"In reality, we are still in the Wild West era with regard to our abilities to accurately assess risks surrounding each disease," Kutikov said by email. "Indeed, predictive models for cancer prognosis and life expectancy are far from perfect, while effects of many treatments are not wholly known."