Prostate cancer screenings: a second opinion
I probably have prostate cancer.
There’s no need to feel sorry for me -- so do about half the men my age (I’m in my mid-50s). We doctors have learned this from microscopic examinations of the prostates of men who are autopsied following an accidental death. And the older men get, the more likely it is that they have prostate cancer. Autopsies of men in their 70s have found that about 80% of them had the disease.
I almost certainly won’t die from prostate cancer, however. The lifetime risk of prostate cancer death for American males is only about 3%. So, although the prevalence of the cancer may sound alarming, 97% of men will die from something else.
These two observations have forced doctors to rethink what it means to have this cancer. Some have envisioned the problem to be like an iceberg. In the past, we only saw the part of the iceberg above the waterline -- the cancers that caused disease and death. With early detection, we can see below the waterline -- and there are a lot more cancers there. Many of these will never cause problems. They would have been better off undiagnosed.
But doctors can’t tell who is better off undiagnosed. We can’t reliably distinguish between prostate cancers that will never cause symptoms and those that are deadly. So we tend to treat everyone. The bulk of men who are treated won’t benefit from it, because there is nothing to fix. But many of them will be harmed. Treatment causes significant side effects in about 30% of those treated, most commonly a decline in sexual function, leaking urine and/or rectal irritation.
That’s why prostate cancer screening is such a challenging issue. Yes, it may save some men’s lives, but it will harm many others along the way.
Two weeks ago, we learned more. The results of two large, randomized trials of prostate cancer screening were published. The studies represented an enormous research effort: almost 20 years of work, involving more than a quarter of a million men and many millions of dollars.
Yet there is still some uncertainty whether screening saves any lives. The European study said yes; the U.S. study said no. That in itself tells you something: If there is a benefit, it is undoubtedly small. In contrast, researchers in the 1960s were able to convincingly demonstrate the benefit of treating very high blood pressure by studying about 150 men over a two-year period. Why were they able to do this with so few men so quickly? Because the benefit was huge.
I believe there probably is a benefit to prostate cancer screening. But it is accompanied by a substantial human cost. Let’s assume the European study is right. Its data give us some idea of the magnitude of the trade-off: For every man who avoids a prostate cancer death, about 50 are treated needlessly (some of my colleagues might say the number is closer to 30, others might say it’s closer to 100).
Being 50 times more likely to be diagnosed and treated needlessly than being the one man who avoids a prostate cancer death doesn’t strike me as a good gamble. To the extent I have control over my cause of death, avoiding a prostate cancer death isn’t my top priority (I’m more concerned about a lingering cognitive decline in a long-term care facility.) And death is not the only outcome that matters to me. I place considerable value on not being medicalized and suffering the side effects of treatment any more than I need to.
But it doesn’t matter what I think about the trade-off. What matters is what you think.
American men have been engaged in prostate cancer screening for almost two decades with relatively little effort given to communicating the trade-off between the benefit and the potential harm of unnecessary treatment. The time has come to make that trade-off clear. There are a lot of bad arguments out there for screening. They include:
* Doctors who tell you they don’t want to go back to the era when all their prostate cancer patients had advanced disease. It is true that the typical prostate cancer patient in the past had advanced disease. But we now know that the primary reason these patients now seem so rare is that they are being diluted by the many new prostate cancer patients who would have never been diagnosed in the past -- the majority of whom had cancers that weren’t destined to progress.
* Media messages that highlight the tremendous improvements in survival. It is true that over the last 50 years, the five-year survival for prostate cancer has increased more dramatically than any other cancer (from less than 50% to almost 100%). But we now know that these numbers too are largely an artifact of over-diagnosis -- diagnosing a lot of men with prostate cancer who were never destined to die from the disease.
* Friends, family, acquaintances or celebrities who “owe their life” to screening. There are now a lot of men who appear to be in this group. But once you understand the problem of over-diagnosis, you recognize an alternative explanation: They never needed treatment in the first place. Some have labeled this the popularity paradox of screening: The more over-diagnosis screening causes, the more people who feel they owe it their lives and the more popular screening becomes.
There is no imperative to be screened, or not screened, for prostate cancer. The only imperative is that men be informed about the consequences of either choice.
H. Gilbert Welch is a professor of medicine at the Dartmouth Institute of Health Policy and Clinical Practice. He is the author of “Should I Be Tested for Cancer? Maybe Not and Here’s Why.”