Should We Screen for Prostate Cancer?

Cancer screening is like fencing in an animal: the goal is to catch the cancer/animal in order to control it. The complicating factor is that cancers turn out to be highly variable in their behavior. Imagine the slow growing ones as turtles. Fencing in a turtle will create the illusion that you’ve controlled it (“the cancer is cured!”), but in fact that turtle was never going to go anywhere in the first place.

My father passed away of heart disease a few years ago. He was 77 years old. Though he’d been active and appeared to be in good health, he’d had a history of high cholesterol and his own father had died of heart disease in his 60s, so it wasn’t a total surprise. What does this have to do with prostate cancer screening? I’ll explain below.

For years, many physicians, including most urologists, have encouraged men to undergo PSA testing as a screen for prostate cancer. Prostate cancer is the most common cancer in men, and the third leading cause of cancer death in men. So it seems logical to many people that we ought to screen for early cancer in order to reduce this death rate. The problem is, most clinical research on this topic has not been able to demonstrate any significant survival benefit to PSA screening. Why not?

To explain, you have to go back to the rationale behind cancer screening. The idea is that if you can catch a cancer early, it’s more likely to be curable. This sounds reasonable. Is it true? Well, it depends.

Gilbert Welch is a primary care doctor in Vermont, and the author of several books on cancer screening. His latest is Less Medicine, More Health: 7 Assumptions that Drive Too Much Medical Care. He explains cancer screening by comparing it to fencing in an animal: the goal is to catch the cancer/animal in order to control it. The complicating factor is that cancers turn out to be highly variable in their behavior. Some are very slow growing, and Welch compares them to turtles. Fencing in a turtle will create the illusion that you’ve controlled it (“the cancer is cured!”), but in fact that turtle was never going to go anywhere in the first place. The majority of prostate cancers are turtles. Although many men do die of prostate cancer (such as my maternal grandfather), many more men die with prostate cancer, either diagnosed or undiagnosed (i.e., there happens to be an area of cancer in their prostate gland at the time that they die of some other cause). Other cancers are highly aggressive, analogous to birds, for whom fences are useless because they can just fly over the top. The only cancers for which screening is useful are the ones of intermediate speed, such as most cervical and colon cancers. These are the rabbits.

Now back to my dad. In the decade prior to his death, he’d seen a urologist for some urinary issues. The urologist happened to order a PSA test during that visit. The result came back a bit elevated, so the doctor performed a prostate biopsy. This, gratefully, came back negative. But what if it had come back positive? Most patients with prostate cancer end up treating it with either surgery or radiation. These procedures have significant rates of long-term complications; roughly between 10 and 30 percent of patients end up with urinary incontinence or impotence or both. In fact, even the biopsies used for the initial diagnosis carry a non-trivial complication rate; roughly 0.5 percent of patients end up with a serious infection. A patient has to accept all these risks in exchange for the theoretical benefit of extending his life through cancer treatment. But if the cancer in question is a turtle or even a rabbit that happens to be on the slow side, and the patient is fated to die of something else in the near future, then the patient may be better off not knowing about the cancer at all.

My dad got lucky: his PSA test didn’t discover a cancer, and his prostate biopsy didn’t result in infection. But in retrospect, he would have been better off not having the PSA test at all. For more information on prostate cancer screening visit the American Cancer Society or for test selection and interpretation, visit ARUP’s Consult site.

Brian Jackson, MD, MS, Vice President and Chief Medical Informatics Officer, ARUP