Editorials

Controversy in managing patients with prostate cancer

Banish dogma, get more data 

Life is uncertain, and never more so than when a serious illness like prostate cancer strikes and a decision must be made about how to proceed. Ideally, the clinician would find (or remember) the relevant research, interpret the findings as they apply to the patient at hand, estimate prognosis, and discuss treatment options objectively and with compassion and support. Even then, life remains uncertain; for any particular patient, no matter how good the evidence and precise the probability estimates, there can be no guarantee.

This irreducible uncertainty is confronted routinely by doctors and patients and is rarely a source of clinical controversy. Trouble begins when experts reach different conclusions from the same piece of evidence. The poorer the evidence, the more discretionary the interpretation, and the more controversial the conclusion. When available evidence is totally inadequate to inform decisions that must be made, then clinical controversy may border on chaos. Savage and others have shown that in Britain and elsewhere controversy and chaos reign in the management of prostate cancer.^1-3

For early prostate cancer, there have been no controlled trials good enough to show whether survival is increased by active intervention with radical prostatectomy, radiation, or hormonal therapy. For men with early disease, any benefit is likely to be realised long after treatment and the immediate side effects of incontinence and impotence have been well documented.⁴ In light of uncertain, delayed survival benefits and known, immediate harm, it is no wonder that clinicians' recommendations vary.

In the survey conducted by Savage and colleagues among 274 British urologists, nine out of 10 favoured active intervention for men aged under 70 years with poorly differentiated early prostate cancerfive recommended radiation, three radical prostatectomy, and one immediate hormone treatment. For men aged over 70, three out of 10 urologists would recommend some form of active treatment, most often radiation. The authors did not ask about moderately differentiated disease, by far the most common type of prostate cancer now being detected and for which prognostic uncertainty is greatest. For patients with well differentiated early disease, who probably have a good prognosis regardless of treatment, four out of 10 British urologists would still recommend radical surgery and three would recommend radiation for those aged under 70. For men over 70, seven out of 10 would favour observational management.

The inclination toward active management for younger men reported by a majority of urologists is at odds with their views about screening. Only a quarter thought that early detection of prostate cancer conferred a survival advantage. The survey also found divergent recommendations for treating locally advanced and metastatic prostate cancer and relapsed prostate cancer. The recent publication of the Medical Research Council trial showing improved outcomes for patients with early prostate cancer treated with androgen deprivation may (or may not) increase consensus.⁵

The controversy and chaos are not limited to Britain. Similar variability in treatment choices have been described in the Nordic countries.² In the United States there is stronger consensus and greater enthusiasm for surgery among urologists, who, in recent years, have performed more than 100 000 radical prostatectomies annually.³ But the enthusiasm is not universal. Rates of radical prostatectomy vary widely among American states.³ American radiation oncologists generally favour radiation treatment. With no evidence for long term effectiveness, brachytherapy and cryotherapy now compete with external beam radiation and radical prostatectomy as the preferred treatment for early cancer.

What can be done to bring reason and order to the management of prostate cancer? Savage and colleagues recommend establishing standards of practice.¹ But standards or guidelines can inform decisions only when the evidence on which they are based is adequate. For the foreseeable future, recommendations for managing prostate cancer, especially clinically localised disease, will rely more on dogma than data.

This is the conclusion reached by the Prostate Cancer Clinical Guidelines Panel of the American Urological Association.⁶ Rather than offer recommendations for patients with different clinical characteristics, the panel concluded that treatment alternatives should be presented as options, each with its advantages and disadvantages. Its only recommended standard was that patients with newly diagnosed cancer should be informed of all commonly accepted treatments.⁶ Programmes to support communication of options, leading to treatment choices that reflect the preferences and attitudes towards risk of the individual patients who will live with the consequences, have been shown to be feasible in busy urology practices.⁷ Decisions would be supported with access to the best available information, with candour about what existing evidence does not allow us to know, and with compassion for the patient facing an uncertain future. Well supported decisions would lead some men to opt for possible future survival benefits, with one or another active treatment depending on their own assessment of the impact of side effects on quality of life. Some might not be willing to accept any compromise in quality of life and choose expectant management. But for others, perhaps many, the benefits and harms would balance, such that the best choice might well be participation in a randomised trial. Trials of treatment for localised prostate cancer are under way.⁸ But they will involve a mere fraction of the men who are eligible and, we believe, a mere fraction of the men who would choose participation if well informed.⁹ Undoubtedly, we could reduce our collective ignorance more quickly if we redoubled efforts to bring men who found themselves at "effective equipoise" into controlled trials.¹⁰ We could also learn from registries, or preference trials, of men whose preferences and attitudes led them to a clear choice among standard or evolving treatments.¹¹

The constructive professional response to the controversy and chaos in the management of patients with prostate cancer is not to develop standards of practice that include treatment recommendations unsupported by currently available evidence. The men we care for now and in the future will be better served if the new standard is to promote patient choice with compassion and care, and then learn from their experiences.