The controversy surrounding PSA screening is one of the most heated in oncology. The potential benefits include prevention of prostate cancer morbidity and mortality, but the men potentially harmed through overdiagnosis and overtreatment outnumber those who benefit. In their thorough review of the evidence on this topic, the authors outline the observational studies as well as the randomized controlled trials that have attempted to determine whether screening for prostate cancer with prostate specific antigen (PSA) testing with or without digital rectal exam (DRE) confers a mortality benefit. Two important studies demonstrated a mortality benefit: the European Randomised Study of Screening for Prostate Cancer (ERSPC) and the overlapping Göteborg, Sweden, trial.[1,2] Investigators on the ERSPC trial estimated that 48 men would need to be screened and treated (NNT) to prevent one prostate cancer death after a median of 9 years of follow-up; in the Göteborg trial, the NNT was 12 after 14 years. The other large randomized controlled trial, the Prostate, Lung, Colorectal, Ovarian (PLCO) cancer screening trial, showed a non-significant increase in prostate cancer–specific mortality associated with screening.[3] As the authors highlight, many explanations have been offered to account for the discrepancy in results between the two studies, including duration of follow-up, prescreening and overtreatment of men in the PLCO trial, and differences in the treatment of men diagnosed in the screened vs the control arm in the ERSPC study.
Recently, updated results of the Norrköping screening study have added fuel to the debate.[4] After 20 years of follow-up, the risk ratio for prostate cancer–specific death in the screening arm was 1.16 (95% confidence interval [CI], 0.78 to 1.73), a result very similar to that reported in the PLCO trial after 10 years (rate ratio, 1.11; 95% CI, 0.83 to 1.50). The Norrköping study randomly assigned 1494 men to screening every 3 years, first with DRE alone, then with added PSA testing with a biopsy threshold of 4 ng/mL in 1993; the control arm consisted of 7532 men not invited to screen. A significant stage and grade migration was associated with screening, and 25% of screened men diagnosed with prostate cancer underwent radical prostatectomy or radiation therapy, compared with 14% in the control arm. However, fewer men were diagnosed with prostate cancer in the screening arm of the Norrköping study than in the screening arms of the ERSPC and PLCO trials. The strengths of the Norrköping study include its low contamination rates, high compliance, and long follow-up—but the negative results of this study do not preclude a small benefit from screening. Based on the best available evidence at present—the ERSPC and the PLCO results—screening with PSA testing is associated with at best a modest improvement in mortality. It may be that with longer follow-up, the ERSPC and PLCO will provide a more definitive answer.[5]
The controversy regarding the benefits of PSA screening will continue until that time—or until the advent of a better screening tool. However, the harms caused by prostate cancer screening are far less controversial and are immediate. As many as 75% of men over the age of 50 years have had a PSA test in the United States, including 25% of men > 85 years of age.[6,7] This screening comes at a significant cost both to the individual and to the health care system.
The harms to the individual from PSA screening are well recognized. As the authors point out, false-positive PSA test results are found in up to 12.5% of screened men and can be associated with significant psychological distress[8,9]; adverse effects of biopsy can include pain, urosepsis, and acute urinary retention. Adverse effects associated with treatment can include erectile dysfunction, urinary irritative symptoms or incontinence, and bowel dysfunction in the case of the radiation modalities, among others. The majority of men will experience at least one adverse effect of treatment, and these adverse effects are longstanding and can be life-altering.[10]
However, the adverse effects of screening are not limited to the patient. Prostate cancer diagnosis and treatment were estimated to cost $1.3 billion in the United States in 2000, an increase of 30% over 1994 expenditures.[11] A recent cost-effectiveness analysis of PSA screening estimated that the cost of diagnosis and treatment is over $5,227,306 per patient to prevent one prostate cancer death—and this estimate assumes the ERSPC benefit seen after a median of 9 years is maintained with longer follow-up.[12] The introduction and rapid dissemination of new and expensive technologies and techniques, including intensity-modulated radiation therapy, robotic-assisted laparascopic radical prostatectomy, and proton beam therapy, may be expected to raise these costs even further.[13,14]
And for many men, treatment is unnecessary. Approximately 70% of men diagnosed with prostate cancer by screening have low-risk disease (stage < T2a, PSA level < 10 ng/mL, Gleason score < 3+3), which is associated with < 6% risk of prostate cancer–specific death at 15 years.[1,3,15,16] Over 90% of these men will be treated for their disease at diagnosis,[17] although it is estimated that up to 60% of men may not require therapy, even over the long term.[18] Currently available screening tests are not able to identify which individual patients require treatment and which men have disease that is truly clinically insignificant. However, in the face of the enormous financial and quality-of-life costs of prostate cancer treatment, attention is now turning to the question of how to limit the harms that result from PSA screening—that is, how to mitigate the conversion of overdiagnosis into overtreatment.
Active surveillance has emerged in recent years as a potential alternative to initial treatment. In this management strategy, men with low-risk prostate cancer are followed closely rather than treated at diagnosis; they are treated with curative intent if their disease exhibits more aggressive behavior. This approach differs significantly from watchful waiting, in which men—usually older men with comorbidities—are followed, with only palliative treatment offered if symptoms of their cancer develop. Although prospective studies of active surveillance have had only limited follow-up, in the most mature of these, 30% of the men involved had been treated, and the prostate cancer–specific survival was 97.2% at 10 years.[19] Results of several randomized controlled trials will become available over the next few years, but in the interim the optimal criteria defining eligibility, surveillance, and disease progression are being debated. However, this approach has been associated with improved quality-adjusted life expectancy compared to initial treatment, and two studies have suggested that it is significantly less expensive than initial treatment.[10,20,21] This spring, results from the Prostate Cancer Intervention Versus Observation Trial (PIVOT) comparing radical prostatectomy to watchful waiting will be presented; these will provide further insight into the consequences of observation vs initial treatment.
The authors of this thoughtful review question whether it is possible to change practice even in the face of evidence that PSA screening may do more harm than good. The controversy surrounding whether and how to screen has now expanded to include whether or not to treat the large numbers of men who will be diagnosed by screening. While the debate over the efficacy of prostate cancer screening rages and we await more mature data from the randomized controlled trials described above, it will continue to fall to the physician to convey not only the uncertainty surrounding screening using PSA testing, but also the uncertainty surrounding the optimal treatment strategy for newly-diagnosed disease.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
