Screening asymptomatic men for prostate cancer: A comparison of international guidelines on prostate-specific antigen testing

Objective To summarise and compare the key recommendations on prostate-specific antigen (PSA)-based screening for prostate cancer, and so highlight where more evidence is required to facilitate consistent recommendations. Methods The Medline database and websites of 18 national screening organisations and professional associations were searched between January 2010 and November 2020 to identify screening guidelines published in English, considering recent clinical trials. Results Population-based PSA testing of asymptomatic men is not widely recommended. Guidelines emphasize shared patient-clinician decision making. For ‘average-risk’ men choosing to be screened, the recommended age varies from 50–55 to 70 years, alongside consideration of life expectancy (ranging from 7–15 years). Screening intervals, when specified, are biennial (most common), annual, or determined from baseline PSA. The earliest age for screening high-risk men (frequently defined as of African descent or with a family history of prostate cancer) is 40 years, but recommendations often defer to clinical judgement. Conclusions Population screening of asymptomatic men is not widely recommended. Instead, balancing the potential harms and benefits of PSA testing is endorsed. Variation between guidelines stems from differing interpretations of key trials and could lead to clinician-dependent screening views. The development of clinical decision aids and international consensus on guidelines may help reduce national and international variation on how men are counselled.


Introduction
Prostate cancer (PCa) is the second commonest malignancy and the fifth leading cause of cancer death in men worldwide. The early detection and prompt treatment of potentially lifethreatening PCa is key to any policy aimed at reducing mortality. However, screening asymptomatic men using the serum prostate-specific antigen (PSA) test is a contentious public health issue. Although the PSA test itself is useful in monitoring PCa progression, it lacks specificity in the detection of clinically important PCa in asymptomatic men. 1 The identification of indolent cancers introduces the potential for harm, either from complications arising from unnecessary investigations such as infection after biopsy, 2 or from treatment complications such as incontinence or impotence after surgery or radiotherapy ('over-treatment'). 3 False-negative PSA tests may offer false reassurance, leading to delayed presentation with more advanced disease and a poorer prognosis. 1 Three randomized controlled trials (RCTs) investigated the effect of PSA-based screening on  1 There is, therefore, uncertainty about the magnitude of the effects of PSA-based screening on PCa mortality and whether any potential benefits outweigh the harms from over-detection or overtreatment, leading to different guidelines on PSA screening worldwide. This paper summarises and compares the key recommendations on PSA-based screening for PCa for both average-risk (someone with no identifiable PCa risk factors) and high-risk (those of African descent or with a family history of PCa) men, identifying areas of disagreement. Our overarching objective was to highlight areas where more evidence is required to facilitate consistent PSA screening recommendations.

Methods
Medline was searched using the terms listed in Appendix 1 (online supplementary material). Records were exported to be screened in Rayyan. 6 We conducted a grey literature search of websites associated with national screening and professional organisations. References to unidentified guidelines were used to uncover any that were missed (snowball sampling).
Two assessors, among five reviewers overall [MdLR, TG, SJ, AJ, SW], screened each record against inclusion crietria and differences resolved by consensus, as illustrated in a PRISMA flow diagram (Figure 1, Appendix 2 in the online supplementary material). Information was abstracted on the main recommendations for different target populations, stratified by age, risk status and screening intervals.

Results
Published recommendations on PSA testing for asymptomatic men were identified for nine countries or regions, and 18 national or professional organisations, as shown in Table 1. The references and comprehensive recommendations are available in Appendix 3 (online supplementary material). Table 1 summarises the key recommendations for men at 'average and high risk' of PCa. There is general agreement against recommending 'population-based', 'systematic' or 'routine screening' for prostate cancer. The Japanese Urological Society (JUA) is the only organisation to specifically recommend a 'population-based screening' programme in men over 50 years.
Guidelines converge on the importance of shared decision making between men and their clinicians, but many lack clarity on who should lead the discussion. The UK's Prostate Cancer Risk Management Programme (PCRMP) recommends that clinicians should not proactively raise the issue with average-risk men and should exercise "clinical judgement" in the case of high-risk men under 50 years. For average-risk men, most guidelines recommend 50 or 55 years as an appropriate age to start the discussion and to cease around 70 years. The exceptions are the National Comprehensive Cancer Network (NCCN), which recommends that screening discussions occur between the ages of 45 and 75 years, and the National Cancer Control Programme (NCCP), which recommends screening discussions for men younger than 50 years.
Most guidelines recommend a life expectancy of at least 10 years when considering screening, except the Prostate Cancer Foundation of Australia and Cancer Council Australia (PCFACCA) who recommend 7 years for average-risk men. How to assess life expectancy objectively is unclear.
There is less guidance on testing frequency. Ten of the organisations addressed recommend screening biennially, whilst others suggest annual and some recommendations are based on a baseline PSA level.
Six guidelines provide 'no specific recommendations' for those considered high risk but some of these suggest it would be reasonable to have a shared discussion. The critera for 'high risk' vary (most specifiy family history and being of Black race or African-American descent). A lower screening discussion age of 40 or 45 years, if specified, is recommended.

Discussion
There is widespread rejection of population-based screening using the PSA test and an emphasis on providing opportunities for shared decision making. This shift towards more patientcentred care is arguably needed in this setting where the potential for harm is significant. 2 The ambiguity about whether clinicians introduce the discussion could lead to subjective screening decisions and impact the number of individuals counselled. 7 If the man is responsible, influences of their education level and health literacy may lead to health inequities. 8 Certain factors, such as African descent and family history, increase an individual's PCa risk. Yet only 5% of participants enrolled in the PLCO trial were Black men and the ERSPC trial did not analyse this demographic, despite Black men being twice as likely to die from PCa as Caucasian men. 9 Some organisations do not provide specific recommendations for high-risk men leading to poor clarity and subjectivity. Non-uniformity in medical funding and access to healthcare may play a role in magnifying disparities particularly amongst high-risk men.
Research in PSA screening could consider both increasing the recruitment of high-risk men into clinical trials, and focus on the clinician's role in raising PCa screening. The generation of decision aids may illustrate and aid existing evidence as well as improve knowledge and reduce decision conflict amongst patients and clinicians. 10 These advances could engender more cohesion among recommendations, with a shift towards more evidence-based policy.