Skip to Content

Support MinnPost

MinnPost logo 7th Anniversary

MinnPost’s online auction is now live!
Register and start bidding today

This content is made possible by the generous sponsorship support of UCare.

PSA testing: Doctor explains benefits vs. risks

It’s been a week since the U.S. Preventive Services Task Force (USPSTF) issued its new draft recommendation regarding the prostate-specific antigen (PSA) blood test, which is widely used to screen for prostate cancer.

They recommended against men aged 50 and older routinely taking the test, saying the medical evidence overwhelmingly shows that the test's harms outweigh its benefits.

The pushback from some people in and out of the medical community was fast and furious, reminiscent of the often virulent public reaction that met the task force in 2009 when it issued its recommendations regarding routine mammograms. Two main groups of people have been most critical: urologists, who argue that the PSA test is their best tool for finding prostate cancer, and men who have been treated for prostate cancer. Many of them believe the test saved their lives.

Even Newt Gingrich got into the fracas, charging during a GOP presidential debate in New Hampshire last Tuesday that the new USPSTF recommendation will cause some men “to die unnecessarily.”

Dr. Timothy Wilt, a professor of medicine at the University of Minnesota and co-director of the Minnesota Agency for Healthcare Research & Quality (AHRQ) Evidence-based Practice Center, was one of the 16 members of the task force that issued the new PSA recommendation. I recently asked him about how men should interpret the recommendation, how the panel came to make it, and how he’s personally coping with the pushback. Here is an edited version of that conversation.

Dr. Timothy Wilt
va.gov
Dr. Timothy Wilt

MinnPost: The task force downgraded its previous “I” [insufficient evidence] recommendation for the PSA test to a “D.” What exactly does a “D” rating mean?

Timothy Wilt: A “D” recommendation is provided when there is moderate or high certainty that the benefits do not exceed the harms, or that the harms exceed the benefits. We then recommend against physicians providing the service.

Previously, the task force had made a “D” recommendation for men 75 years of age or older, but said there was insufficient evidence to accurately assess the benefits and harms of younger men. Now there is good evidence from five large randomized screening trials that prostate cancer screening with the PSA blood test does not prolong lives or reduced prostate cancer mortality, but is associated with considerable harms. Thus, the harms exceeded the benefits. We were moderately to highly certain of that, and thus the “D” rating against PSA screening.

MP: One of the points critics of your new recommendation have made is that there has been a 39 percent reduction in prostate cancer mortality between 1990, when the PSA was introduced, and 2007. How do you interpret that?

TW: There has been a reduction in prostate cancer mortality from its peak, which was a little after the PSA blood test was introduced. Prostate cancer mortality rates now are approximately equal to what they were prior to PSA screening. The problem with those numbers is that, although encouraging, they don’t conclusively demonstrate that PSA testing is the reason for that.

[To begin with], areas of this country that have the greatest amount of testing and treatment do not have lower mortality rates than areas of this country that have far lower testing and treatment. Furthermore, the rates of decline in this country are somewhat paralleled by other countries that have less PSA testing and less aggressive treatment. So there could be a lot of reasons why the cancer mortality rate as estimated has come down.

The gold standard for assessing clinical benefits is the randomized controlled trial. There have now been five large randomized trials [involving PSA testing] that have enrolled more than 300,000 men. That’s equal to about six Metrodomes filled with men between the ages of 50 and 74 followed for more than 10 years. About 18,000 men [in those studies] were diagnosed with prostate cancer, and, unfortunately, about 1,200 men died of prostate cancer. PSA testing did not reduce either prostate cancer deaths or overall mortality.

MP: And the testing actually did harm.

TW: The test itself may not be harmful. It’s the downstream problems. Those problems include a need for biopsies with an elevated PSA blood test. Those biopsies can cause pain, infection and bleeding.

Yet, while problematic, those really aren’t the main issues. [The primary issue] is the detection of cancers that otherwise would cause no problem in a man’s life. And those are very common. The vast majority of PSA-based cancers that are detected would never cause a problem during a man’s lifetime. But because it’s very difficult for us to tell which ones will cause problems from which ones won’t, and because the word cancer causes a lot of anxiety in patients and concern among physicians, almost 90 percent of men with PSA-detected cancers are treated. And treatments have real harms [including incontinence and erectile dysfunction] that are serious, frequent and occur early on and may persist.

MP: Why isn’t it feasible to separate the diagnostic process from the treatment decision-making process?

TW: That is one of the main issues. But keep in mind, PSA testing does not reduce mortality. Three hundred thousand men were enrolled in screening trials and followed for 10 years. Men randomized to get the PSA blood test did not have prostate cancer or all-cause mortality reduced compared to men who did not [have the test]. It did not do what it was supposed to do. And it has harms.

MP: Why is so hard for much of the public to accept this finding — as it was for them to accept the task force’s findings regarding mammogram screening?

 

TW: For a long period of time, [there’s been a belief that] early detection and early treatment must be beneficial. And for many conditions, it may be so. Screening for colorectal cancer is an example. Cervical cancer screening may also be an example. Diagnosis of high blood pressure and treatment of that — another excellent example.

However, it may not always be the case. Screening for ovarian cancer is not effective. There are blood tests you can use to detect ovarian cancer, but it does not reduce mortality.

In some situations the test picks up things very early and among many individuals who would never be bothered by [the disease]. On the other hand, for those individuals who would, unfortunately, have problems with their disease, the test may not be able to pick up those at a time when treatment is effective. Yet because a test picks up cancers that would never cause problems, and because people undergo treatment for those cancers, they attribute the good outcome to that diagnosis and treatment.

MP: And that’s the case with PSA testing. It doesn’t prolong lives or reduce deaths.

TW: We wish that wasn’t so. I, as a clinician, need to be able to give my patients accurate, honest information to try and improve their health. The best information I can give them now is that, unfortunately, the PSA blood test does not reduce prostate cancer mortality, it has harms, and those harms exceed the benefits. As their health care provider, I cannot recommend that test.

Having said that, if an individual says, “Doctor, I hear what you’re saying. I still want to do whatever I can to find prostate cancer,” the PSA blood test is probably the best test for finding it. It will double your chances of being diagnosed with prostate cancer. I would then order the test and try to set up some discussion prior to it with the patient about how we would react on the results.

MP: It’s reported that this task force recommendation was held up for two years? Is that accurate? Did the task force vote on this in 2009?

TW: Yes. I was a member of the task force at the time. We looked at the evidence at that time, voted on it, and had a similar recommendation [to the one issued this month]. However, we were also very aware that it was important to get the science right. We voted to conduct an additional evidence report, [one that] particularly revolved around the harms associated with diagnosis and treatment, as well as any possible new and extended follow-up information about the benefits from the randomized trials. A new evidence review was commissioned by the Oregon Health & Sciences University’s Evidence Based Practice Center. Those results were conducted over about a year’s period of time, presented to the task force in March of this year, and voted on once again with a similar recommendation — with sound science. From then on we worked on drafting the recommendation statement as you see it today.

MP: Did calling for that extra evidence review have anything to do with the controversy that surrounded the USPSTF’s 2009 mammogram recommendations?

TW: I think we’re all cognizant now of the role that task force recommendations play in clinicians’ and patients’ health care decisions. We also knew that this would be a very important recommendation, one that would not necessarily be widely or enthusiastically endorsed by many, and perhaps met with some skepticism. We felt it was also very, very important to get the science right on this.

MP: Is it difficult to be on the task force these days? Are you getting a lot of personal pushback?

TW:  It’s a wonderful opportunity to try and improve the lives of all Americans. It’s really the highest honor I’ve had both in terms of being a physician, a health policy person, and a scientist. I can tell you that the individuals on this task force take their jobs both as task force members and as primary care providers with the greatest integrity and the highest degree of responsibility.

We know that our recommendations are closely scrutinized. We are thankful that there’s actually so much interest in them, whether it’s positive or negative, because it means we can, hopefully, help make a difference — a difference in improving the lives of all Americans through developing evidence-based recommendations about disease prevention, detection and health promotion. That’s a wonderful opportunity, and a large responsibility that we take very seriously.

MP: What else should the public understand about this issue?

TW:  I just want your readers to know that this recommendation is based on sound science. It’s intended to help them and their doctors make accurate, well-informed decisions that can improve their lives.

It’s also important for them to know that not all things that we once believed to be true are found to be true by science. There’s a long history of that. The fact that medicine is willing to put its beliefs to a test is really what separates it out from other disciplines.

Get MinnPost's top stories in your inbox

Comments (2)

I am disturbed that a Task Force member should be citing the available studies as showing that PSA screening does not reduce overall mortality.

In the European study, PSA screening was estimated to reduce prostate-cancer-related mortality (which included deaths from prostate cancer surgery or biopsies) by 0.07%. In a sample size of about 80,000 each in the screening and control arms, there is no way such an impact on overall mortality could be statistically detected if overall mortality was 10%. Similarly, even in the meta-analyses that have been done, with a total sample of over 250,000, there is no way to detect plausible effects on overall mortality.

In fact, a power analysis suggests that in order to have an 80% chance of statistically detecting effects on overall mortality of PSA screening, we would have to have an experiment with over 4 million participants.

I also think that the Task Force's D grade, which states that "there is moderate or high certainty that the service has no net benefit or that the harms outweigh the benefits," is a somewhat skewed valuing of the evidence.

The European study, which is arguably the best study, although not perfect, and certainly one of the two largest and best-designed studies, did find a significant effect of PSA screening in reducing prostate cancer specific mortality. Based on the European study, researchers have estimated a "number needed to treat" to reduce one prostate cancer death of 48 after a mean follow-up of 9 years, and 18 after 12 years.

What these results imply is that treatment due to PSA screening reduces the risk of prostate cancer deaths by 2% after 9 years and 5% after 12 years. Men must weigh this possible reduced risk against a 30% to 50% or more increased risk of serious side-effects from the treatment.

Some men may judge the harms of treatment to outweigh the benefits. Other men may judge the benefits of treatment to outweigh the harms. However, I simply do not see how any reasonable weighting of the scientific evidence can say that there is even MODERATE certainty that the harms outweigh the benefits.

By the way, I am not a urologist. I am an economist who has had to recently face some of these decisions about screening and treatment. I have tried to use my training in econometrics to familiarize myself with the available statistical studies.

I personally think the evidence reviewed my the Task Force is far more consistent with a C rating, which says " The USPSTF recommends against routinely providing the service. There may be considerations that support providing the service in an individual patient."

I am all in favor of fully informing an individual patient that PSA screening can lead one on a slippery slope to biopsy and then treatment. If a man is reasonably sure that he would be unlikely to opt for definitive treatment after a positive biopsy, he should not do the screening. But if a man would like the information, because he thinks there IS a reasonable chance that he might take the risks of treatment to reduce his probability of prostate cancer death, then he should consider screening. This individual choice is far more consistent with a C grade than a D grade.

The USPSTF considered all available evidence from studies around the world. They commissioned an independent organization to review the studies and then evaluated those results. Their work has all the markings of unbiased objectivity anyone would desire.

We have a spirited conversation with Dr. Michael Lutz, MD and principal with the nation's third largest urology practice on the USPSTF's recommendations against routine screening on our recent episode of Graceful Aging. Please view it and use it to continue your study on this important health topic.
The show is at http://youtu.be/bUX7-iOmkdc.