USPSTF Makes PSA Screening Recommendations Without Urology Representative
The US Preventive Services Task Force (USPSTF), with no urology representative, reviewed what it considered the relevant literature and concluded that available evidence is insufficient to assess the balance between potential benefits and harms of using PSA to screen men less than 75 years old for prostate cancer and has recommended against screening men over 75 years old (even those at high-risk).
A report (Moul et al) found that 78% of men surveyed at a screening clinic disagreed with this recommendation. This study also presented evidence that older patients generally have more aggressive disease and worse outcomes. Although there is a selection bias in studying only men attending a screening clinic and those treated with surgery, they raise a valid question about the USPSTF recommendation.
Physicians striving to prevent prostate cancer deaths screen their patients because of the increasing evidence that it reduces metastases and death from prostate cancer. In the PSA era, advanced cancer at diagnosis in the US has decreased by over 75%, and cancer death rates have fallen by over 40% from 1993 to 2006, more than for any cancer in men or women.
Statistical models suggest that 45-70% of this decrease is due to screening. Globally, prostate cancer mortality has decreased in countries where screening is practiced and has remained stable or increased where it is not. If screening detected only harmless cancers, treating them could not produce such striking benefits.
The recently reported randomized screening trials from Europe (ERSPC) and the U.S (PLCO) that did not include men over 75, nevertheless, have added to the controversy. The ERSPC found that screening reduced prostate cancer deaths by at least 20% (27% in men who were actually screened) with a 41% reduction in metastatic disease, while PLCO did not find a difference.
Flaws in PLCO Study
The PLCO trial was destined to fail due to extensive PSA testing before study entry (removing many patients with life-threatening prostate cancer from the study population), extensive screening of controls during the study, use of an outdated PSA cutoff (4 ng/ml), failure of those in the screening arm with abnormal results to undergo prompt biopsy, and premature reporting of mortality results at a median follow-up of 5-6 years for cancer patients (at which point death rates in the European trial had not yet begun to diverge). In the ESRPC, the mortality rates continue to diverge with ongoing follow-up; whereas, the PLCO results are unlikely to change due to the aforementioned flaws.
The ERSPC estimated that 1400 men would need to be screened and 48 treated to prevent 1 prostate cancer death, while only 25 men would need to be treated to prevent 1 case with metastases. However, these estimates are strongly affected by the difference in death rates in the two arms of the study, and, thus, are overestimates because of screening in the “no-screening” arm of the study and the incomplete follow-up to date. It is likely that these estimates will decrease.
Because prostate cancer arises silently and becomes incurable before causing symptoms, the only way to detect it early is through screening. Eliminating screening would also eliminate the possibility for early diagnosis and curative treatment. Instead, PSA-based parameters (i.e., density, velocity and free PSA) should be used to decrease unnecessary biopsies and selectively identify men with aggressive tumors.
With respect to the 2008 USPSTF guideline, not all 75-year-old men are the same. Rather than discontinuing screening based solely upon chronological age, the decision to screen in this population should take into account the absolute PSA level and PSA trends over time, as well as general health status.
The USPSTF’s misguided recommendation (and the resulting media coverage) could give reluctant men of any age an excuse to postpone or forgo screening. This recommendation, if followed, will undoubtedly result in delays in potentially lifesaving treatment, with possible unnecessary morbidity and mortality.