Indisputably, early detection and treatment of prostate cancer are more effective than ever before with a consistent decline in mortality rates in many Western countries (2). However, this statement does not accomodate the fact that many men continue to die with prostate cancer and that the consequences of this tumour and its treatments may contribute significantly to the demise of many of these men, both directly and indirectly. Albertsen et al (207, 208) estimated that, for conservative treatment, men with moderately and poorly differentiated tumours lost approximately 4-5 years and 6-8 years of life, respectively, compared with those with well-differentiated tumours whose life-expectancies were little different to age-matched controls (207, 208).

This problem of identifying patients with the more aggressive cancers is relevant to men of all ages, consistent with the findings of Parker et al (2001) (209) who concluded in their meta-analysis that age is not a significant prognostic factor in contemporary clinical practice (209). Using the SEER database, Lu-Yao and Yao (1997) (210) found that the disease-specific 5 year survival for poorly differentiated tumours was 63-69% (210). Albertsen et al (2005) (211), reporting on 767 men diagnosed between 1971 and 1984 and treated without curative intent, found that, for low-grade prostate cancers (Gleason 2-4), men have a minimal risk of dying from prostate cancer during 20 years of follow-up. Those with Gleason scores of 5 or 6 have an intermediate risk of prostate cancer being the cause for their demise but patients with high-grade prostate cancers (Gleason 8-10) have a high probability of dying from prostate cancer within 10 years of diagnosis (211).

The zeal to diagnose and treat must be tempered by realisation that neither detection nor therapeutic intervention may be in the best interests of many men. Establishing the right balance between the benefits and disadvantages of diagnosis and treatment is problematical and, consistent with the variation in the natural history of prostate cancer, it is not difficult for the unwanted effects of intervention to be worse than the disease being treated. Of necessity, it is important to recognise that the risks of competing mortalities are very different for men of different ages. Based on Queensland data, Baade et al (2005) estimated that about 60% of men diagnosed in their 50s could be expected to die a premature death (ie, before reaching 80 years of age) from prostate cancer compared with corresponding rates of 50% and 38% for men diagnosed at 60 and 70 years. Thus, for a 50 year-old man, a diagnosis of prostate cancer is more likely to result in a premature death from prostate cancer than for men diagnosed in their 70s (212). Hence, whether or not to diagnose the condition should be a carefully considered decision.

While prostate cancer is the most common male malignancy in the developed world and the second most common cause of cancer deaths, uncertainties remain about management practices at several points in the illness continuum. For example, owing to a lack of definitive data confirming that widespread screening for prostate cancer will reduce the death rate from this disease, population-based screening for prostate cancer in asymptomatic men is not currently recommended in most countries . Rather, it is suggested that men should be able to access PSA testing as long as they are fully informed of the pros and cons of testing.

For those diagnosed with localised prostate cancer, further decisions present with three possible treatment options at a minimum: watchful waiting, radiation therapy, or RRP - extending more recently to include in some settings brachytherapy and laparoscopic and robotic surgery . Men who are diagnosed with advanced disease will also face difficult treatment decisions such as when to commence treatment and what method of hormonal ablation to select, each with various quality of life ‘trade-offs’, to accept . In the setting in which no one treatment approach is clearly superior with regards to cancer cure and where quality of life outcomes differ markedly, the quality of patients’ decision making about medical treatments is critical. As a result, strategies to assist in meaningfully considering prostate cancer treatment options, and the risks and benefits of these options in order to achieve high quality patient decisions, are essential .

The approach that is considered to be optimal for achieving high quality patient decisions is shared decision making . Shared decision making is defined as a process carried out between a patient and his health care professional where both parties share information and the patient understands the risks and benefits of each treatment option, participates in the decision to the extent that he desires and makes a decision consistent with his preferences and values, or defers the decision to another time. Shared decision making may not be easy to achieve for all patients . For example, although many patients with cancer indicate a preference for sharing decision making with their clinicians, some, in the case of prostate cancer between 8% to 58% of men, prefer a passive decision making role where clinicians make treatment decisions on their behalf . However, clinicians still need to understand patients’ preferences to ensure that they are making quality decisions on behalf of their patients. As well, there is often a gap between the clinical ideal of shared decision making and actual clinical practice where decision complexity and time constraints may make this approach difficult for both parties to achieve . There are, however, defined strategies and decision aids that can facilitate this process .

Many groups advocate an informed decision-making process as an evidence-based approach and necessary precursor to screening for early prostate cancer . Others have suggested that informed decision-making on this health topic is also necessary as a medico-legal risk management strategy . While some researchers have suggested a set of information that needs to be communicated to men about this health decision , there are few explicit guidelines on this subject . Problematically, patients and clinicians do not agree on core content . It has been advised that, for any screening test, patients need to understand the purpose of the test, the likelihood of false-negatives and false-positives, the uncertainties and risks associated with testing, significant medical, social or financial implications of testing and any possible sequelae and follow up care plans (www.ipdas.ohri.ca).

Such information needs to be communicated to patients in a logical and balanced sequence in order to promote better understanding and increased decisional control by men. One approach that has been widely tested in primary care in Australia is the use of six decision steps (see Table 1). Each decision step logically follows to prompt the clinician to overview important health information, with tailoring suggested in Step 1 to ensure the discussion is consistent with the patient’s concerns. For example, for a man with a significant family history of prostate cancer, this factor is likely to be central to the patient discussion . Men who experience uncomplicated lower urinary tract symptoms (LUTS) often worry about prostate cancer, so addressing this concern first may be priority . In this regard, resources for patients that explain about male reproductive health problems such as urinary symptoms and sexual dysfunction are available at www.andrologyaustralia.org. As well, National Health and Medical Research Council guidelines are available about the management of LUTS (http://www.health.gov.au/nhmrc/publications/synopses/cp42syn.htm).

From this point, checking to ensure the patient has a basic understanding of both the prostate and possible tests is needed and, given many men may be unaware of the location and function of the prostate gland, an anatomical diagram may be a useful teaching tool here. Next, a consideration of individual risk with regard to both the incidence and mortality of prostate cancer is needed. Communicating health risks effectively is a challenge in the provision of effective decision support. In general people find probabilities hard to understand, often estimate their level of risk incorrectly, and tend not to weigh up pros and cons in a systematic way when deciding about treatments . As well, population based statistics provide data about populations, not individuals, so risk communication needs to acknowledge this as a limitation and, where possible, refer to age-based risk estimates and relevant individual factors such as family history ).

There are a number of communication strategies that have been suggested to help patients understand risk. These include

However, a quality health decision goes beyond the simple transfer of information and includes consideration and incorporation of each patient’s values and personal preferences . Thus, Step 5 in Box 1 prompts the clinician to discuss each man’s individual preferences. A number of strategies can be used to do this, most commonly the use of a pros and cons exercise in which patients are encouraged to explicitly consider the factors that matter most to them personally in this decision, and the direction and leaning of their preferences either for or against each possible option. One approach to support this process for this health topic is the inclusion of a values table within a decision card (see Table 1). A decision aid that incorporates both the six decision steps and this values clarification exercise can be found on http://www.ncci.org.au/services/PSA_decision_card.pdf .

Decision aids are also effective in supporting patients to make informed choices. With regards to PSA testing, patient focussed decision aids have been found to be effective in increasing men’s knowledge about PSA testing and decreasing decision-related distress , with a variable effect on actual testing behaviour.

A range of aids is freely available from the web ( www.prostatehealth.org.au; www.cdc.gov/cancer/prostate; www.cancerbacup.org.uk).

Cancer helplines also often provide such information, for example, The Cancer Council Australia Cancer Helpline on 13 11 20; the UK helpline on 0808 800 1234; the USA Cancer Helpline on 1800 227 2345.

Traditionally, palpation of the prostate by digital rectal examination (DRE) was the manner by which a diagnosis of prostate cancer was suspected. In historical series, up to 50% of palpable masses were attributable to prostate cancer (243, 244). Although DRE by itself is a poor method for diagnosing this malignancy (245, 246), it does still have an important diagnostic role as 25% of tumours are detected in men with normal PSA levels (247). Unfortunately, when a prostate cancer is diagnosed based on a palpable tumour, the risk of the patient already harbouring metastatic or locally advanced malignancy is considerable (248-250). However, a PSA-based prostate cancer detection strategy which omits DRE runs the low risk of missing some curable cancers (251).

PSA testing has revolutionized the detection of prostate cancer and monitoring of its treatment. However, its application in early detection is contentious with attitudes to population-based screening for prostate cancer in asymptomatic men varying considerably. (252, 253)

In the US, PSA screening is currently recommended for all men older than 50 years and advocated to begin at 45 years in those with first-degree relatives with prostate cancer and African-American men because of their higher risk of contracting this tumour.( 247, 254) Population screening is not currently recommended in most Western countries (213) although case selection is widely practised with peer organisations endorsing the importance of informed patient participation in decision-making for diagnostic testing, as outlined above.

Deficiencies in the use of PSA in the diagnosis of prostate cancer have led to research into examining the cellular contents of the prostate more directly. Bostwick Laboratories provide a test that assays for PCA3/DD3 RNA ( http://www.bostwicklaboratories.com/) from prostatic cells in urine immediately following DRE (168, 169). Tinzl et al (2004) (169) reported that detection of the non-coding PCA/DD3 RNA, which is highly overexpressed in most prostate cancers (167), provided sensitivities of 73%, 84% and 84% with specificities of 61%, 80% and 70% for serum PSA values of <4 ng/ml, 4-10 ng/ml and >10 ng/ml, respectively, in the detection of prostate cancer (169).

Fradet et al (2004) (170) reported on a multi-centre evaluation of this marker in a paper titled ‘uPM3, a new molecular urine test for the detection of prostate cancer’. 443 of 517 (86%) samples provided by patients undergoing prostatic biopsies at 5 centres were assessable by this molecular assay. The overall uPM3 sensitivity and specificity were 66% and 89%, respectively. In men with a PSA level <4 ng/ml, the sensitivity was 74% and specificity 91%; for PSA levels 4-10 ng/ml, the sensitivity was 58% and specificity 91%; for PSA levels >10 ng/ml, the sensitivity and specificity was 79% and 80%, respectively. The overall accuracy was 81% compared with 43% and 47% for total PSA at a cutoff of 2.5 and 4.0 ng/ml, respectively.

Once the possibility of a prostatic cancer is raised, whether by rectal examination, PSA parameters, or a combination of both, the second part of the contemporary two-step early-diagnostic approach, TRUS-guided prostate biopsies, is usually performed. TRUS imaging permits spatial positioning of spring-loaded biopsy needles to provide a methodical approach for obtaining tissue cores for standard histopathology. With few exceptions, TRUS imaging by itself is non-diagnostic as only gross changes register as an abnormal appearance on the monitor. The number of biopsy cores taken is important with the chance of missing a cancer by standard sextant biopsy estimated to be approximately 25% (281) so that, more recently, the numbers of cores recommended are at least 8 and preferably a minimum of 10. In addition, it is advocated that biopsies should be directed laterally and that they should include the anterior horns of the peripheral zone (263, 282-287). Many urologists routinely take 12 biopsy cores now to minimise the likelihood of missing cancer.

Guidelines established by the American Urologic Association recommend prostate needle biopsies for any man with a PSA value greater than 4.0 ng/ml, or an abnormal prostate on digital rectal examination (247).

The issue of repeat biopsies was addressed by Djavan et al (2001) particularly in relation to when it is reasonable to stop repeating the biopsies. Cancer-detection rates in 1051 men biopsied were 22%, 10%, 5% and 4% with 1-4 TRUS biopsy sessions with 58%, 60.9%, 86.3% and 100%, respectively, having organ-confined disease. Recently, Yanke et al (2005) extended experience with the Kattan Nomogram to predict the likelihood of a positive finding at a subsequent biopsy session. Predictor variables studied in the nomogram were patient age, family history of prostate cancer, prostate specific antigen slope, months from initial negative biopsy session, months from previous negative biopsy session, cumulative number of negative cores previously taken and previously detected high grade PIN or atypical small acinar proliferation. The authors evaluated a total of 356 repeat biopsy procedures for 230 patients. The mean number of total cores per patient was 17.9 with 78 men having biopsies positive for cancer. The area under the ROC curve was 0.71, which was greater than any single risk factor (288).

Routine practice involves peri-operative antibiotic prophylaxis with a pre-procedural enema and not proceeding if any faeces at all is present in the rectum, as determined by DRE. Since TRUS biopsies are unpleasant and uncomfortable, many urologists use anaesthesia (local or general) as a routine. Minor morbidity is common with this procedure with well over 50% of patients experiencing at least one complication. Fortunately the dreaded complication of life-threatening sepsis is uncommon, generally <1%, even though rates of bacteraemia vary greatly; blood in the urine, ejaculate and faeces are not infrequent sequelae with some men having difficulty voiding immediately following the procedure. (289-291).