Informed consent for prostate-specific antigen-based screening—European view

Present uncertainties and need for informed consent 

To be acceptable as a general health care policy, screening will have to be shown to lead to a decrease in prostate cancer mortality in a randomized study. Additionally, information will have to be produced that will allow judgment on the quality of life with and without screening and treatment. Knowledge about these two issues will be the main parameters to be taken into account in a subsequent risk-benefit analysis.

The level of evidence available at this time has recently been categorized (categories I to III) by an international panel.1 Category I evidence based on randomized studies is not available but will be within a few years. Category II evidence for the effectiveness of screening comes from decreasing prostate cancer mortality in the United States and other areas of intense application of screening tests2 and the so-called “Innsbruck study,”3 which showed in 1999, a reduction in the mortality rate with respect to expected mortality of 33% after years of extensive screening and treatment. Unfortunately, no randomized studies of radical prostatectomy or radiotherapy against watchful waiting have been done. On September 12, 2002 the results of the SPG4 study comparing radical prostatectomy (RP) to watchful waiting were published.4 The data show a significant advantage of 50% in terms of prostate cancer mortality in advantage of radical prostatectomy. Closest to a randomized study is the careful evaluation of 59,876 patients recruited to the Surveillance, Epidemiology, and End Results (SEER) database of the U.S. National Cancer Institute by Lu-Yao and Yao.5 The patients are almost equally distributed among radical prostatectomy, radiotherapy, and watchful waiting. Outcomes in terms of overall survival and relative survival are equivalent for grade 1 and 2 (well and moderately differentiated) disease. Large differences in favor of active treatment, however, emerge in the intention-to-treat and treatment-received analyses of those men who had poorly differentiated localized disease. Obviously, the presence of selection bias precludes statistical analysis.

Although potential benefits of screening become more likely through observational data, solid information on the potential harms of screening remains scant. Essink-Bot et al.6 found in a study based on the European Randomized Study of Screening for Prostate Cancer (ERSPC), section Rotterdam, that the effect of screening itself in terms of anxiety and physical harm is limited and of short duration. Of the 600 invited participants, discomfort and pain were experienced by 52%, 39%, and 98% after rectal examination, transrectal ultrasonography, and biopsy, respectively; 4% of the men used pain medication and 4% to 6% experienced temporary interference with functioning. A 1-year follow-up study comparing quality-of-life parameters between screening and control7 compared the impact of radical prostatectomy and radiotherapy on health-related quality of life and specific urologic symptoms. Although adverse effects relating to urination and sexual functioning are significant with both forms of treatment, the acceptance by the study participants specifically after radical prostatectomy was remarkably high.

The greatest potential harm of screening comes from the possibility of overdiagnosis and overtreatment. McGregor et al.8 related the prevalence of prostate cancer with screening to the risk of disease-related death and found a risk ratio of 6.25 (ie, 16 of 100 screen-detected cancers are estimated to cause death). Zappa et al.9 defined overdiagnosis as the diagnosis of cases that otherwise would not be diagnosed during the patient’s lifetime. Depending on age, they projected overdiagnosis rates of 50% to 100%. Schröder et al.10 presenting data on the ERSPC, section Rotterdam, correlated rates of cancer detection with underlying regional mortality and found a ratio of diagnosis to mortality in the control group of 2.5 and in the screening group of 14.6. Additional estimates of overdiagnosis will emerge from the modeling of observational data and data from ongoing randomized studies. The risk of overdiagnosis is substantial and needs to be carefully considered in informing potential candidates requesting early detection measures.

Present policies worldwide 

The report by Albertsen et al.1 contains a section provided by Perrin, who has established an inventory of screening policies worldwide. A careful search resulted in only one country (Luxembourg) in which screening has formally been introduced. He found that in 12 of 28 evaluated countries, early detection measures are either available or encouraged. In 20 of these 28 countries, the PSA test is reimbursed. Obviously, reimbursement does not necessarily relate to screening. There is wide agreement that screening cannot be refused to well-informed men who request PSA-based testing.11, 12 Recently (July 4, 2001), the Secretary of Health of the United Kingdom issued the following statement: “The national screening committee has been considering the strength of feeling about testing for prostate cancer among men who have had prostate cancer diagnosed, as well considering the evidence about screening. It recognised that many men are asking general practitioners for a PSA test and that the current policy [does not give] clear guidance to general practitioners about how to respond. …” The policy statement then states that the PSA test should be made available to men who are well informed about the risks and benefits and that information in the form of leaflets, electronic information sites, and brochures should be made available. If a patient then wishes to have a PSA test, the National Health Service is obliged to provide the testing, biopsy, and treatment, as necessary.

The big question that remains is what is the content of the information referred to in this policy statement. Clearly, the development and validation of such information lags behind.

What should be the content of informed consent to screening? 

This question should be related to the wishes of potential participants. Men at risk should be in a position to determine the priorities with respect to their interest in the risks and benefits of screening. Items offered to potential participants should, however, include the following:

The necessary information can be developed in the form of leaflets, brochures, videos, or carefully designed and described conversational strategies. The few questionnaires that have been designed and applied in validation samples include the one by Wolf et al.14, 15 Wolf et al. showed in a cohort of 56 uninformed and 48 informed patients that the category “low interest in screening” increased with information and the category “high interest in screening” was cut to less than half after applying the validated information.

Conclusions 

No conclusive answers are possible at this time concerning mortality reduction by screening. In most European countries, the issue of prostate cancer screening is under heavy debate. Few countries have established clear policies. The European Committee of Cancer Screening has found that the evidence does not support the introduction of population-based screening at the European level at this time. However, in line with other policy statements, screening cannot be refused to well-informed men.16

The key issue is that validated information material is not yet available. The production of such materials is a top priority that should have preceded the issue of national guidelines and policies such as the one issued in the United Kingdom in 2001. The best way of dealing with this problem at the European level would be through a European or worldwide task force charged with the development and validation of appropriate information material. National efforts should be included in this process. Until such materials become available, the burden of information lies with the physician or urologist who is confronted with the request for PSA-based testing. The investment of time is considerable and cannot be provided for in practice situations in most countries. Clearly, PSA testing without information or the inclusion of PSA testing in routine examinations such as employment healthcare checks and insurance checks is unacceptable.