Draft screening guidelines spark debate

Possible drawbacks and long-term effects of new draft guidelines for prostate cancer screening


  • Carrie Printz

Editor's note: This is Part 2 of a 2-part series. Part 1 published in the April 15 issue of “CancerScope.”

Many critics have found flaws with the recent draft guidelines for prostate cancer screening released last fall by the US Preventive Services Task Force (USPSTF). The proposed guidelines, discussed in more detail in Part 1 of this series, were based on findings from 2 large screening trials: the US-based Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial and the European Randomized Study of Screening for Prostate Cancer (ERSPC). The recent guidelines issued a “D” recommendation for prostate-specific antigen (PSA) screening for men younger than age 75 years, indicating that the benefits do not outweigh the harms.1

Among the critics of the guidelines is William Catalona, MD, director of the clinical prostate cancer program at Northwestern University in Chicago, Illinois, and a pioneer of PSA screening. Specifically, Dr. Catalona believes the USPSTF's assessment of harms is inaccurate. “They used very outdated studies from 20 years ago to assess the risk of prostatectomy and radiation therapy,” he says. “They overestimated the risk of screening and treatment. For example, they said that no study has ever reported on anxiety associated with screening, but the Göteborg study [part of the ERSPC trial] found that only 6% of patients had high anxiety associated with biopsies.”

Indeed, the USPSTF stated in its study discussion that the evidence on treatment-related harms was most applicable to open retropubic radical prostatectomy and external-beam radiation therapy, although details of techniques and specific regimens were frequently lacking. Also, the task force stated that it found little evidence comparing newer techniques for prostatectomy, such as nerve-sparing approaches that use robotic-assisted or free-hand laparoscopy, with watchful waiting. Still, the task force added that it did not find a pattern suggesting that more recent studies reported different risk estimates than older studies.

Michael LeFevre, MD, MSPH, co-vice chair of USPSTF, says, “We updated our review with the most recent literature based on large, population-based studies. If people feel they have significant information on this—that the harms have declined—they should publish it.”

One of the really bad things is that [the USPSTF guideline] shuts down the discussion and shuts down the decision making. To deny patients the opportunity to make an informed decision puts them in harm's way. And to do this based on flawed science is really unconscionable. —William Catalona, MD

Illustration 1.

According to Dr. Catalona, the true extent of overdiagnosis is unknown, because if a patient is diagnosed, undergoes a radical prostatectomy, and is cured, researchers have no way of knowing whether he was overdiagnosed. Many factors can be taken into account to avoid overdiagnosis, he says, including a patient's age, race, family history, PSA velocity, and more; after diagnosis is made, physicians also can look at Gleason grade and percentage of cancer in the biopsy.

“The PCLO's protocol is not how things are done today,” he says. “If I had a 50-year-old black patient with 2 brothers [who had] aggressive prostate cancer and a 2.5 ng/mL PSA, I wouldn't wait until it got to a 4 to biopsy him. On the other hand, if I had an 80-year-old patient with a PSA around 5 that hadn't changed much in the past 10 years, I wouldn't biopsy him.”

He adds that these factors do not come out in large, randomized trials and that the trials are no longer relevant to how the field has progressed in understanding overdiagnosis and overtreatment. “I think the pendulum has swung too far,” he says. “Patients with life-threatening prostate cancer look at me and wonder if they're being overdiagnosed.”

On the other side, he says, is the risk of underdiagnosis. Although recent studies and a National Institutes of Health consensus panel have recommended active surveillance for men with low-grade prostate cancer, Dr. Catalona says there are still risks involved. There is the risk of a sampling error in which initial biopsies do not show prostate cancer (or if they do, they do not indicate how aggressive it is). As a result, some patients are later diagnosed with aggressive disease. Studies show that about 35% to 50% of men fall off active surveillance and must be treated, he says.

Task Force Make-Up

Another criticism that Dr. Catalona and others have made of the USPSTF is that it did not have a member who is an expert in prostate cancer, but Otis Brawley, MD, chief medical officer of the American Cancer Society (ACS), says that this make-up of the task force is actually one of their strengths. “The task force is very much like a grand jury,” he says. “You get a group of smart people who are not schooled in that particular area, so they have no emotional or financial conflicts, and you bring in experts in that area to testify. You also hire PhD students—usually from a school of public health—to do an extensive literature review of the studies and their quality. Then you come up with recommendations.”

The Institute of Medicine recently issued recommendations on review groups, suggesting that the panels should not consist of anyone who treats the disease they are reviewing because this could lead to emotional or financial conflicts. It also suggests the panels provide comment periods for the public to respond before final guidelines are issued. The ACS has changed its guideline review process accordingly, says Dr. Brawley.

Dr. Catalona and others who support screening also point out that the US rates of metastatic cancer and prostate cancer mortality have declined significantly—declines of greater than 75% and 40%, respectively—since PSA testing first began. Yet, Drs. Brawley and LeFevre say the link is not entirely clear.

The prostate cancer mortality rate also has declined in European countries, such as Germany and Great Britain, that do not allow screenings except in clinical trials, says Dr. Brawley. Dr. LeFevre argues that it would have taken longer to see a decline in prostate cancer mortality if it were directly related to screening: the substantial decline occurred between 1992 and 2002, and an actual decline related to screening would not have been apparent until 2002, given what researchers now know about the timeline between screening and death from the disease.

“We don't know what caused the decline,” Dr. LeFevre says. “Changes in treatment and monitoring during those 10 years is our best guess.”

Impact of the Guidelines

As the debate over the data and studies continues, physicians also have differing opinions as to the ultimate impact of the USPSTF's guidelines.

If the task force does not revise the “D” recommendation, Dr. Catalona thinks that Medicare, Medicaid, and private insurers may not pay for prostate cancer screenings. He believes that would only serve to worsen the racial and economic disparities that exist in cancer care. “One of the really bad things is that it shuts down the discussion and shuts down the decision-making,” he says. “To deny patients the opportunity to make an informed decision puts them in harm's way. And to do this based on flawed science is really unconscionable.”

The US rate of prostate cancer mortality has declined significantly since PSA testing first began. But, the rate also has declined in European countries that do not allow screenings except in clinical trials.

Dr. Brawley hopes the guidelines will serve to focus attention on the controversial nature of PSA screenings and lead more men to have conversations with their physicians. “We've been very ineffective with our ‘informed decision-making’ campaign in this country,” he says. “Most men don't know prostate cancer screening is a procedure where there are huge risks, and 90% of men diagnosed with prostate cancer in the US get treated.”

Additional studies on the issue are unlikely because they take too long, are too costly, and are too prone to manipulation, says Dr. Brawley, adding, “We recommend that when science doesn't know the answer, we don't make it up.”

Dr. LeFevre also hopes the draft (and final) guidelines will lead to more discussion on the issue. “I think more primary care physicians will feel compelled to tell their patients this isn't just something you check off on a lab sheet,” he says. Over time, he hopes research will provide more answers as to which patients need to be treated and which do not, and that science will lead to better treatments that cause less harm. On that point, at least, all parties are likely to agree.