PSA screening and the overdiagnosis of prostate cancer

by Charles Bankhead, Staff Writer, MedPage Today

Only one of every 20 prostate cancer diagnoses leads to a benefit that would not have been realized without PSA screening, an analysis of 20-year trends suggests.

medpage-today An estimated 1 million excess diagnoses have accrued since 1986, and the incidence of prostate cancer remains well above levels that existed prior to widespread PSA screening, according to H. Gilbert Welch, MD, of Dartmouth Medical School in Hanover, N.H., and Peter C. Albertsen, MD, of the University of Connecticut in Farmington.

Particularly dramatic is the sevenfold greater incidence of prostate cancer in younger men compared with the pre-PSA era, Welch and Albertsen reported online in the Journal of the National Cancer Institute.

“Given the considerable time that has passed since PSA screening began, most of this excess incidence must represent overdiagnosis,” the authors concluded.

“It is important that we begin to explicitly communicate to men who are considering screening the relative magnitude of the number of deaths averted to the number overdiagnosed,” they added.

“Overdiagnosed patients cannot benefit from treatment because their disease is not destined to progress to cause symptoms or death.”

The findings added another chapter to a long-standing debate about the risks and benefits of widespread PSA screening for prostate cancer.

Little doubt has remained about the impact of PSA screening on diagnosis of prostate cancer, Welch and Albertsen said. However, considerable uncertainty has surrounded the effect of PSA screening on prostate cancer death.

To provide another perspective to the debate, the authors queried the NCI Surveillance, Epidemiology, and End Results database regarding age-specific incidence of prostate cancer and initial course of therapy.

Using U.S. census data, they estimated the excess or deficit in prostate cancer diagnoses and treatment since 1986, the year before PSA screening was introduced.

They found that prostate cancer incidence increased steadily until 1992 and then stabilized at levels considerably higher than those of 1986.

From 1986 to 2005 prostate cancer incidence increased by 26%, from 119 to 150.5 per 100,000.

The overall trend, however, obscured distinct age-specific trends, the authors noted.

Analysis by decade of age revealed an inverse association between age and the change in relative risk (RR) of diagnosis from 1986 to 2005:

* ≥80: 1,146.5 to 637.4 per 100,000, RR 0.56
* 70 to 79: 819.2 to 896.8 per 100,000, RR 1.09
* 60 to 69: 349.4 to 666.9 per 100,000, RR 1.91
* 50 to 59: 58.4 to 212.7 per 100,000, RR 3.64
* <50: 1.3 to 9.4 per 100,000, RR 7.23 Confidence intervals for all of the differences fell within the limits of statistical significance. Combining all age groups, Welch and Albertsen estimated that an additional 1,305,600 men have been diagnosed with prostate cancer since 1986, and an additional 1,004,800 men have been definitively treated. "Using the most optimistic assumption about the benefit of this additional diagnosis and treatment -- namely, that the entire decline in prostate cancer mortality observed during this period is attributable to screening -- we estimate that approximately 56,500 prostate cancer deaths have been averted and that approximately 23 men had to be diagnosed and approximately 18 treated for each man experiencing the presumed benefit," they said. Noting that prostate cancer treatment has risks, Welch and Albertsen concluded that more than 1 million American men have been "needlessly exposed" to the risks, the financial implications, and the anxiety associated with being a cancer patient. In an accompanying editorial, Otis Brawley, MD, chief medical officer of the American Cancer Society, said the reasons behind a 40% decline in prostate cancer mortality in the U.S. since 1993 remain unclear. He noted that prostate cancer mortality also has declined in some countries that do not have widespread PSA screening. In a reference to the ongoing national debate over healthcare reform, Brawley said the "irrational tendency to adopt treatments and technologies without adequate assessment is a form of 'medical gluttony' and a major reason that U.S. per capita healthcare costs are the highest in the world. We do not get what we pay for; our life expectancy is 29th among developed countries." Brawley concluded that some of the confusion surrounding prostate cancer screening can be avoided "if we all clearly label what we know, as what we know; what we do not know, as what we do not know; and what we believe, as what we believe ... [and] not confuse what is believed with what is known." The article drew strong reactions on both sides of the debate. Patrick Walsh, MD, of Johns Hopkins, said the authors ignored the fact that prostate cancer mortality declined by 28% from 1994 to 2009 and evidence that definitive surgery for prostate cancer has reduced 10-year mortality by as much as 50%. "The problem with the article is simple," said Walsh. "In 1986 there was an enormous reservoir of men in the population with advanced, asymptomatic, incurable prostate cancer, so that when they were diagnosed with cancer, treatment had no effect on their outcome. "Then, beginning in the mid- to late-1990s, we finally were seeing men who were curable and who were going to live long enough to be cured, but if they had not been treated, it would have taken them 15 to 20 years to die of the disease, a time that has not yet passed." Edward Gelmann, MD, of Columbia University in New York City, said PSA testing's limitation is that it does not help distinguish between men who have cancer that should be treated and those who do not. "What is really needed are markers that distinguish between indolent and more active forms of cancer," said Gelmann. Mark Soloway, MD, of the University of Miami, suggested that increased use of active surveillance could give men "the opportunity to avoid the morbidity of treatment ... and take the small chance of progression if they delay or avoid treatment." Visit for more prostate cancer news.

✓ Join 150,000+ subscribers
✓ Get KevinMD's most popular stories