New Evidence Shows that Large-Scale Prostate Cancer Screening Saves Lives

SAN FRANCISCO—Two new studies presented at the 2010 annual meeting of the American Urological Association (AUA) suggest that large-scale prostate cancer screenings can indeed save lives.

Researchers in Innsbruck, Austria, evaluated data from the state of Tyrol within Austria, where an early detection and treatment program has been in place for more than 20 years. They found that these programs have been associated with a reduction in mortality in areas where effective treatment is available to all men.

The Tyrol Project was started in 1988, offering free prostate-specific antigen (PSA) testing to men aged 45 to 75years beginning in 1993. Approximately 120,000 men were screened. In men with organ-confined lesions, prostatectomy was recommended (2153 radical prostatectomies were performed); 86.3% of patients with T1 or T2 disease were treated with low-morbidity radical prostatectomy; 8.7% received brachytherapy; and 8.7% received radiotherapy. After 1 year, 95.1% of men were continent, and potency was preserved in 78.9% of men younger than 65 years of age.

The researchers found that since 1996, there was a significant reduction in mortality from prostate cancer in Tyrol. In the years 2003-2008, prostate cancer mortality rates decreased by 48% (2003), 55% (2004), 52% (2005), 49% (2006),

41% (2007), and 64% (2008). In other states within Austria, there were also declines in prostate cancer mortality, but the percentages in those areas were only approximately 30%. The investigators concluded that when screening and treat ment are available and free, prostate cancer mortality is decreased by population-wide screening efforts.

NNS and NNT to save one life with PSA screening

One way of decreasing prostate cancer deaths is through screening with PSA serum testing. However, the tradeoff between reducing prostate cancer deaths through screening and possible overdiagnosis and over-treatment is the subject of continuing intense debate.

In 2009, prospective, randomized clinical trials of prostate cancer screening reported disparate results, with the Prostate, Lung, Colorectal, Ovarian Cancer (PLCO) trial finding no mortality benefit and the European Randomized Study of Screening for Prostate Cancer (ERSPC) showing a 20% mortality benefit (30% in men actually screened). The ERSPC researchers estimated, however, that at a median follow-up of 9 years, 1410 men would need to be screened (number needed to screen [NNS]) and 48 treated (number needed to treat [NNT]) to avoid one prostate cancer death. The most frequently quoted and troubling statistic to clinicians and patients alike is the estimate that 48 men need to be treated to prevent one prostate cancer death, which is very high compared with an NNT of 10 for breast cancer screening.

Alternative explanations for a high NNT could be that screening overdetects a large proportion of indolent cancers or that the limited follow-up of the ERSPC population overestimated the true NNS and NNT. Using extrapolated data from the ERSPC, a multicenter team of researchers set out to discover the true NNS for prostate cancer and the true NNT to save one life as well as to assess the effect of follow-up times on these calculations.

Based on published ERSPC data, researchers from Chicago and Baltimore estimated the cumulative hazard ratios and NNS/NNT out to 12 years of follow-up. At year 10, the model yielded an NNS of 837 and NNT of 29, similar to the ERSPC report. However, by year 12, the NNS decreased to 503 and the NNT was 18. Noting that the NNS and NNT to save one life are directly affected by the length of follow-up, the researchers concluded that more than 10 years of follow-up may be necessary to truly show the value of population-based prostate cancer screening. A prominent feature of prostate cancer screening is that the benefits take a long time to achieve and the true magnitude of overdiagnosis and overtreatment remain largely unquantified.

“The Tyrol study shows the benefits of freely available PSA testing and the importance of effective treatment once cancer is found,” said AUA spokesman Christopher Amling, MD, a professor of urology at Oregon Health & Science University, Portland. “Altho ugh the ERSPC screening study showed a significant mortality reduction with PSA screening, it also showed that with early follow-up, a relatively large number of men need to be screened and treated to prevent one prostate cancer death. By extrapolation of data from the ERSPC trial, the researchers from Chicago and Baltimore were able to demonstrate that with longer follow-up, the NNS and NNT are significantly lower, suggesting that the value of PSA-based screening may be greater than this study suggests.”

Amling said the AUA believes that early detection of and risk assessment for prostate cancer should be offered to asymptomatic men 40 years of age or older who have a life expectancy of at least 10 years.