When discussing the place for surveillance in the treatment of localized prostate cancer, several questions have to be answered. First of all, we must ask, does prostate cancer constitutes a major health problem? Is the mortality from prostate cancer acceptable or is it worthwhile to decrease the mortality? The answer may seem obvious but we have to face the fact that the mean age at death in patients dying from prostate cancer exceeds the mean life expectancy of the male population.<sup>6 Aus G., Hugosson J., Norlén L. Long-term survival and mortality in prostate cancer treated with noncurative intent J Urol 1995 ;  154 : 460  [cross-ref]

Cliccare qui per andare alla sezione Riferimenti</sup> Only one fourth of prostate cancer deaths occur in males who have not reached the age where half of the male population is dead.<sup>13 Cancer Registry of National Board of Health and Welfare   Cancer Incidence in Sweden 1993 Stockholm: Fritzes (1994). 

Cliccare qui per andare alla sezione Riferimenti</sup> To determine whether surveillance is a good option, we therefore also have to consider the suffering and costs associated with the disease course in conservatively managed prostate cancer. If we still believe that preventing death in prostate cancer is worthwhile, new questions arise. Because no curative treatment seems available for advanced disease, we must rely on early diagnosis. Prostate cancer, however, is a peculiar disease in that approximately 50% of men older than 50 years have microscopic foci of carcinoma in the prostate (i.e., latent carcinoma).<sup>12 Breslow N., Chan C.W., Dhom G. , e al. Latent carcinoma of prostate at autopsy in seven areas Int J Cancer 1977 ;  20 : 680  [cross-ref]

Cliccare qui per andare alla sezione Riferimenti, 50 Sakr W.A., Haas G.P., Cassin B.F. , e al. The frequency of carcinoma and intraepithelial neoplasia of the prostate in young male patients J Urol 1993 ;  150 : 379

Cliccare qui per andare alla sezione Riferimenti</sup> This is not a very well-understood condition and the distinction between significant and nonsignificant carcinomas is usually made only by calculating the size of the neoplasm.<sup>15 Catalona W.J., Smith D.S., Ratliff T.L. , e al. Measurement of prostate-specific antigen in serum as a acreening test for prostate cancer N Engl J Med 1991 ;  324 : 1156  [cross-ref]

Cliccare qui per andare alla sezione Riferimenti, 52 Stamey T.A., Kabalin J.N., McNeal J.E. Prostate specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate. II. Radical prostatectomy treated patients J Urol 1989 ;  141 : 1076

Cliccare qui per andare alla sezione Riferimenti</sup> Even though fatal tumors at some time must have started as microscopic foci, we do not know how often such foci progress to larger tumors. Clinical cancer is supposed by most urologists to represent progressive disease, but exactly how often these tumors progress and the time from a localized stage to death is under continuous debate.<sup>19 Chodak G.W., Thistedt R.A., Gerber G.S. , e al. Results of conservative management of clinically localized prostate cancer N Engl J Med 1994 ;  330 : 242  [cross-ref]

Cliccare qui per andare alla sezione Riferimenti, 20 Epstein J.I., Paull G., Eggleston J.C. , e al. Prognosis of untreated stage A1 prostatic carcinoma: A study of 94 cases with extended followup J Urol 1986 ;  136 : 837

Cliccare qui per andare alla sezione Riferimenti, 35 Johansson J.E., Adami H.O., Andersson S.O. , e al. High 10-year survival rate in patients with early, untreated prostatic cancer JAMA 1992 ;  267 : 2191

Cliccare qui per andare alla sezione Riferimenti, 43 Lerner S.P., Seale-Hawkins C., Carlton C.E. , e al. The risk of dying of prostate cancer in patients with clinically localized disease J Urol 1991 ;  146 : 1040

Cliccare qui per andare alla sezione Riferimenti, 45 Menon M., Parulkar B.G., Bakre S. Should we treat localized prostate cancer? An opinion Urology 1995 ;  46 : 607  [cross-ref]

Cliccare qui per andare alla sezione Riferimenti, 46 Moskovitz B., Nitecki S., Richter-Levin D. , e al. Cancer of the prostate: Is there a need for agressive treatment? Urol Int 1987 ;  42 : 49  [cross-ref]

Cliccare qui per andare alla sezione Riferimenti</sup>

Further problems arise when one tries to analyze the effects of curative treatment. Because no results from large-scale randomized studies between surveillance and radical treatment are available, we are reduced to comparing the outcome in nonrandomized series. When comparing data between treated and untreated patients, much selection bias is introduced because patients offered these two options usually differ a lot.<sup>3 Adolfsson J., Steineck G., Whitmore W.F. Recent results of management of palpable clinically localized prostate cancer Cancer 1993 ;  72 : 310  [cross-ref]

Cliccare qui per andare alla sezione Riferimenti</sup> Because prostate cancer is slow-growing and affects the male population at an age when a lot of competing mortality is present, many men do not live long enough to develop progressive prostate cancer. No treatment may seem optimal in elderly men and men with limited life expectancy. But do we have the tools today to forecast how long the remaining life expectancy will be? How valid is such a calculation in the individual patient? At what risk for prostate cancer death is no treatment justified? How do these men perceive receiving no treatment? How are men affected psychologically by knowing that they have a cancer not treated because they have a limited life expectancy and are supposed to die from something else?

High age at diagnosis and a long natural course of the disease not only makes the individual patient's fate hard to predict, it also makes special demands of statistical methods used if comparisons between patients scheduled for different treatments are to be possible. A variety of methods have been used to describe the long-term outcome in prostate cancer. How do these methods correlate to each other and to what extent does the choice of method affect the impression of the results? Patients diagnosed today are typically diagnosed by repeated prostate-specific antigen (PSA) screening,<sup>9 Benson M.C., Whang I.S., Olsson C.A. , e al. The use of prostate specific antigen density to enhance the predictive value of intermediate levels of serum prostate specific antigen J Urol 1992 ;  147 : 817

Cliccare qui per andare alla sezione Riferimenti, 15 Catalona W.J., Smith D.S., Ratliff T.L. , e al. Measurement of prostate-specific antigen in serum as a acreening test for prostate cancer N Engl J Med 1991 ;  324 : 1156  [cross-ref]

Cliccare qui per andare alla sezione Riferimenti</sup> in contrast to patients in most series with deferred treatment, who were diagnosed either by infrequent digital rectal examinations or due to micturition symptoms.<sup>48 Otnes B., Harvei S., Fosså S.D. The burden of prostate cancer from diagnosis until death Br J Urol 1995 ;  76 : 587

Cliccare qui per andare alla sezione Riferimenti</sup> Is comparison between these two populations at all possible, and if so can we calculate the biases introduced?