Dr. Laurence Klotz, Professor of Surgery, University of Toronto and Chief, Division of Urology, Sunnybrook and Women’s College Health Sciences Centre, Toronto, Ontario
Prostate cancer (PC) screening based on prostate biopsy for men with levels of serum prostate-specific antigen (PSA) above an empirical level, or abnormal digital rectal examination (DRE), results in diagnosing many men with prostate cancer for whom the disease does not pose a threat to their life. Welch has recently calculated that there are 2.74 million U.S. men who are 50-70 with a PSA > 2.5. If all American men in this age group had a PSA, and a PSA > 2.5 is used as an indication for biopsy, 775,000 cases would be diagnosed this year in the U.S. alone. This is 543,000 more than the 232,000 cases diagnosed in 2005, and 25 times more than the 30,350 men expected to die of PC per year in the U.S.1
Several autopsy studies of men dying of other causes have documented the high prevalence of histologic prostate cancer.2 A large proportion of this histological, or ‘latent’ prostate cancer is never destined to progress or affect the lifespan of the patient.
Since the introduction of PSA screening, the lifetime risk of being diagnosed with prostate cancer has almost doubled from around 10%, in the pre-PSA era, to 17%.3 This means that many cases of localized prostate cancer are over-treated, since some patients not destined to experience prostate cancer death or morbidity will be subjected to radical therapy.4
Cancer aggressiveness can be predicted to some degree using existing clinical parameters. The ones mostly widely used are tumor grade, or Gleason score; PSA; and tumor stage. Favorable-risk prostate cancer is characterized as a Gleason 6 or less, a PSA 10 or less, and T1c-T2a disease.5 As a result of stage migration due to PSA screening, the proportion of newly diagnosed patients who fall into the ‘favorable-risk’ category has increased, and now constitutes 50-60% of patients. While patients with these characteristics have a much more favorable natural history and progression rate than those with a higher Gleason grade or PSA, some of them still progress to advanced, incurable prostate cancer and death.
An update of a large group of patients in Connecticut treated with watchful waiting has recently reported the results of a 20-year follow-up.6 The study data confirms the powerful predictive value of Gleason score. In that pre-PSA screening cohort, 23% of untreated Gleason 6 patients died of prostate cancer within 20 years. For Gleason 7 prostate cancer, about 65% of untreated men died of prostate cancer within 20 years. In addition, the author recently subjected the original slides to re-analysis using contemporary Gleason scoring.7 This demonstrated clearly that there has been a shift in grade interpretation over the last 20 years (as reported in the August 2006 issue of Insights). Many Gleason 6 cancers diagnosed 20 years ago would be called Gleason 7 today. Thus it is likely that the Connecticut results represent a ‘worst case’ scenario for the expected mortality from untreated Gleason 6 cancer. This means that the prostate cancer mortality of untreated, non-screen-detected, contemporary Gleason 6 cancer may be as low as 10% at 20 years.
Autopsy studies have demonstrated that prostate cancer typically begins in the third or fourth decade of life.1 This means that, in most patients, there is a period of slow subclinical tumor progression that lasts approximately 30 years, followed by a period of clinical progression (potentially to metastatic disease and death) lasting about 15 years. The implication is that most patients have a long window of curability. This is particularly true for patients with favorable risk, low volume disease.
One approach to achieving prediction of tumor aggressiveness is to use this window of curability to identify patients at higher risk for progression based on a rapid PSA doubling time (PSADT) and/or a histologic progression over time.
Numerous cohorts have reported the results of a watchful waiting approach where there was no treatment until there was progression to metastatic or locally advanced disease, at which point androgen ablation therapy was implemented.8-17 These older studies consistently describe non-progression in many patients. However, the results are difficult to apply in the current era for two reasons: (1) because the cohorts described are from the pre-PSA era, and constitute patients with more extensive disease at the time of diagnosis, and (2) because patients were not offered the opportunity for selective definitive therapy at an earlier stage when their disease was still potentially curable. In the era of PSA monitoring, patients who are treated conservatively receive periodic PSA tests. This raises the tantalizing prospect that treatment of favorable prostate cancer could be deferred indefinitely in the majority; and that effective delayed therapy need only be offered to the patient subset in whom PSA progresses rapidly or the tumor grade increases.18-19
The Prostate Cancer Prevention Trial (PCPT), a 19,000-member trial comparing the effectiveness of Proscar® and a placebo in preventing prostate cancer, incorporated a strategy of routine systematic biopsies of the prostate, regardless of PSA level. Twenty four percent of patients in the placebo arm were diagnosed with prostate cancer over a seven year period, even though their PSAs were still in the normal range.20 This high proportion means that, in sharp contrast to accepted wisdom, a routine prostate biopsy will result in the detection of latent micro-foci of disease in many men.
The lifetime risk of dying from prostate cancer remains less than 3%.3 As the lifetime risk of being diagnosed approaches the known rate of histological (mostly insignificant) prostate cancer, there is a greater risk of over-treatment. At least two studies have attempted to model the rate of diagnosing clinically insignificant disease, suggesting that it ranges from 30% to 84%.4-5 The current incidence-to-mortality ratio of about 7:1 suggests that the higher (84%) figure is more likely. Factors contributing to this are the increasing use of PSA screening and more extensive biopsy strategies that employ eight to 13 cores.21 Additionally, biopsies are often repeated over and over until a cancer diagnosis is made. More biopsies mean the diagnosis of more prostate cancer and of more clinically insignificant disease (as well as more clinically important disease).
A large series of patients from Johns Hopkins treated with radical prostatectomy22 showed that a median period of 16 years elapsed from surgery until death in patients dying of prostate cancer following disease recurrence. Many watchful waiting studies, most of which accrued patients from the pre-PSA era, also demonstrate that disease-related mortality in populations of prostate cancer patients only becomes substantial after 10 years.23-34 Low-grade prostate cancer in particular is associated with low progression rates and high survival rates in the intermediate to long term. This is also supported by the Albertson’s Connecticut data.6
Moreover, the estimated lead-time between diagnosis based on PSA, and diagnosis based on clinical factors such as a palpable nodule in the prostate as was resported in the Connecticut series has also been estimated to be around 10 years by many authors.35,36 Thus, many patients currently diagnosed by PSA screening with favorable prognostic factors are diagnosed considerably earlier in the development of the disease than was the average patient in this unscreened population in the older watchful waiting studies. Therefore, these screened patients are likely to have prostate cancer with an even longer and more benign natural history. And since patients on active surveillance who become re-classified as higher risk over time still have the opportunity for radical intervention, it seems obvious that the expected prostate cancer mortality in this group is likely to be exceptionally low.
Because the prediction of clinically insignificant disease is problematic and inaccurate, an alternative strategy has been developed that allows patient entry into an expectant management protocol with rigorous monitoring and the option of curative salvage therapy, should signs of progression develop. This is referred to as active surveillance.18-19
Klotz and Choo were the first to report on a prospective active surveillance protocol incorporating selective delayed intervention for the subset with rapid PSA progression or grade progression on repeat biopsy.37,38 The eligibility criteria for this included patients with T1c or T2a prostate cancer who had a Gleason of 6 and a PSA of 10. For patients over age 70, these were relaxed to include Gleason <= 7 (3+4) and/or PSA <= 15. The current cohort comprises 331 patients. The median age was 70 years with an age range of 49 to 84 years. 80% of the patients had a Gleason score of 6 or less, and the same proportion had a PSA < 10 ng/ml (median 6.5 ng/ml). With a median follow-up of 72 months, 101 patients (34%) came off active surveillance, while 198 have remained on surveillance. Of patients discontinuing surveillance, the reason was rapid biochemical progression (PSA rise) in 15%, clinical progression in 3%, histologic progression (new or enlarging nodule in the prostate gland) in 4%, and patient preference (deciding to treat even though nothing had changed) in 12%.With a median follow up of seven years (range 2-11 years), the overall survival was 85% and the disease-specific survival was 99%.
Only three out of the 331 patients had died of prostate cancer at the time of writing this review. All three patients had PSADTs of < 2 years, and death occurred 3.0, 5 .1, and 5.2 years after diagnosis. All three exhibited the same pattern of clinical progression: initial favorable prognostic factors, but subsequently a rapid rise in PSA which led to treatment in 6, 9, and 11 months after the initial diagnosis; after treatment, all three had a progressive rise in PSA and clinically apparent bone metastases within a year after treatment, leading to androgen deprivation therapy. All three patients died within three years of the initiation of ADT. This very rapid progression after diagnosis suggests that these patients had occult metastases outside the prostate at the time of their initial disease presentation, and that their outcome would not have been altered by earlier treatment. Even in the Swedish trial,39 there were almost no ‘saves’ before five years had elapsed.
The median PSADT for all of the men in our study, calculated by logarithmic regression, was seven years. Twenty-two percent of the patients had a PSADT of less than three years, whereas 42% had a PSADT of over 10 years that suggests an indolent course of disease in these patients.
At the time of repeat biopsy, the Gleason score remained stable in 92% of patients; only 8% demonstrated a significant rise in Gleason score, classed as an increase of equal or greater than 2. It is not known whether this represents true grade progression or initial under-sampling; however, it is consistent with other similar series reported by other researchers, demonstrating a 4% rate of grade progression over 2-3 years.40 In our group, 29 patients (10% of the cohort) had a radical prostatectomy as a result of a short PSA doubling time or grade progression. Of these patients, all had an initial Gleason score of 5-6, a PSA < 10 ng/ml, and a tumor stage pT1-2 at study entry. The final pathology was stage pT2 in 18 patients (64%%), pT3a in 11, T3c in 1, and N+ in 1. Among the 18 patients with a PSA DT < 3 years (18 patients), only seven had positive margins. This suggests that even among the worst subset of the cohort, i.e. those reclassified as higher risk over time, the majority remained curable despite having delayed therapy.
Who Benefits from Treatment?
The recent landmark trial from Sweden demonstrated, for the first time, that radical prostatectomy improves survival.39 In that study, the treatment of about 600 patients was randomized between radical prostatectomy and watchful waiting. The study showed a 5% absolute survival benefit at 10 years, and a 50% reduction in prostate cancer mortality with surgery. However, this cohort was a group with many patients who had intermediate-to high-risk disease that was much worse than the proposed candidates for active surveillance. In this study only 5% were diagnosed based on PSA screening, and the median PSA was 12.8. The volume of disease in these patients represented a pre-stage migration cohort. (Even in this group, however, the number needed to treat to prevent each prostate cancer death was 19). The distribution of disease volume and grade is higher than the expected distribution in a contemporary screened population, where a substantial proportion of newly diagnosed patients have small volume low-grade disease.
The Swedish study should not be interpreted to mean that all patients with localized prostate cancer should be treated radically. Many studies emphasize that the patients with Gleason 4-5 pattern disease are at the greatest risk for death from prostate cancer. In the Swedish study, the mortality improvement began to appear at five years. It would be most unusual for a patient with low grade, low volume disease to die within 5-7 years of diagnosis. (In the Toronto surveillance cohort, this is 1% of patients.37) This means that the majority of the benefit seen in the Swedish trial likely represented mortality reduction in the high-risk group.
We have used this data and the Connecticut watchful waiting data to estimate, for each prostate cancer death averted at 20 years, the number of patients with favorable risk prostate cancer that would have to be treated at the time of diagnosis. The number-needed-to-treat (NNT) for each death avoided at 10 years in the Swedish trial was 20. It is likely that with additional (i.e. 20 year) follow up, the survival benefit in the Swedish trial, now 10 years, will increase. This is likely to be balanced by the lead-time inherent in PSA screening.
Thus, in a screened patient with intermediate grade and PSA similar to the Swedish cohort, the NNT at 20 years is estimated to also be around 20. The Albertsen data6 indicate that the mortality for intermediate-risk disease was about 2.5 times greater at 20 years than it was for favorable risk disease. This number may be an under-estimate if the shift in contemporary Gleason scoring is factored in. Thus, compared to no treatment, about 50 favorable-risk patients need to be treated for each death that will be prevented by surgery. However, if one offers selective delayed intervention to those patients who progress, it can be conservatively estimated that at least 50% can be salvaged. The conclusion is that about 80-100 radical prostatectomies would be required for each prostate cancer death averted in favorable risk disease. Correcting the Connecticut data for grade migration, as referred to earlier, would increase this even further.
Finally, how much benefit does that one patient whose prostate cancer death is averted by all those radical treatments achieve? Experience from 2000 patients at Johns Hopkins suggests that the prostate cancer deaths averted would have occurred on average 16 years after diagnosis, meaning that the number of life years saved in each of these 1 in 100 averted deaths is modest. Unfortunately, no one lives forever. The rare individual who benefits from surgery (who is typically 60 years old on average lives to be 82, so his life would be prolonged an average of five years by having his prostate cancer death averted.7 If each prostate cancer death averted adds five years to that individual’s life, each radical prostatectomy would add 0.6 months of life (60 months per 100 operations). This is of dubious merit.
Since some apparently low risk patients may re-classify as high risk over time, patients should be followed carefully and treated if they show evidence of rapid PSA progression or Gleason grade progression on repeat biopsy.
In young, healthy patients on surveillance, the optimal PSADT threshold for radical intervention should be around three years. In our series at Sunnybrook Health Services Centre, patients with a PSADT of three years or less constituted 22% of the cohort. The decision to use this cut point for intervention remains empirical and speculative. However, the selection of this cut point is supported by findings reported by others. For example, 20-25% of patients with a three-year doubling time represents a rough approximation of the proportion of good risk patients ‘at risk’ for disease progression.38 For patients with a PSA in the 6-10 range, it also approximates an annual rise of 2 ng/ml, an adverse predictor of outcome as described by D’Amico.41
The psychological effects of living for many years with untreated cancer are a potential concern. Does the cumulative effect, year after year, of knowing one is living with untreated cancer lead to depression or other adverse effects? The best data on this comes from a companion study to the Holmberg randomized trial of surgery vs. watchful waiting in Sweden. It found absolutely no significant psychological difference between the two groups after five years. Worry, anxiety, depression, all were equal between the two arms.29 While surveillance may be stressful for some men, the reality is that most patients with prostate cancer, whether treated or not, are concerned about the risk of progression. Anxiety about PSA recurrence is common among both treated and untreated patients. It is hoped that with education patients will begin to understand the very indolent natural history of most good-risk prostate cancers and, with the realization that the disease is not life-threatening, may avoid much of this anxiety.
Our follow-up strategy for managing patients with active surveillance and selective delayed intervention is described in Table 1.
Conservative management has been resisted in many constituencies due to concern about the inaccuracies of clinical staging and grading. The advent of widespread PSA screening has the positive effect of identifying patients with life threatening prostate cancer at a time when they are more curable, and the negative effect of identifying many patients with non-life threatening cancer who are susceptible to over-treatment. In a population subjected to regular screening, the latter group is far more prevalent. PSA testing will result in hundreds of thousands of patients needlessly subjected to the side effects of therapy. A rational approach to therapy is to offer aggressive treatment to the intermediate and high-risk group, and little or no treatment to he low risk group.
However, some apparently favorable-risk patients harbor more aggressive disease. In these patients, there are benefits of curative treatment. A policy of close monitoring with selective intervention for those whose cancers exhibit characteristics of higher risk disease over time is an appealing way to deal with this. Intervention is offered for a PSADT greater than 3 years (depending on patient age, co-morbidity, etc.), or grade progression to a predominant Gleason 4 pattern. This approach is currently the focus of several clinical trials, and preliminary analysis of these has demonstrated that it is feasible. Most patients who understand the basis for this approach will remain on long-term surveillance. If patients are selected properly (i.e. good-risk and low-volume disease) and are followed carefully to enable early intervention if there is evidence of progression, it is likely that the majority of men with indolent disease will not suffer from clinical disease progression or PC death, and the minority with aggressive PC will still be amenable to cure. Using two different approaches, we estimate that if all such patients were offered radical prostatectomy compared to this strategy, the number-needed-to-treat would be approximately 100 for each patient who avoids a PC death. Thus, the proportion of patients who die of PC is not likely to be significantly different from the proportion dying in spite of aggressive treatment of all good risk patients at the time of diagnosis. This approach is currently being evaluated in a large-scale phase 3 study.
1. Welch, Prostate-specific antigen levels in the United States: implications of various definitions for abnormal. JNCI 2005:97:1132-7.
2. Sakr, W.A., Haas, G.P., Cassin, B.F. et al.: The frequency of carcinoma and intraepithelial neoplasia of the prostate in young male patients. J Urol, 150: 379, 1993.
3. Jemal, A., Tiwari, R. C., Murray, T. et al.: Cancer Statistics, 2004.CA Cancer J Clin, 54: 8, 2004.
4. Etzioni, R., Penson, D. F., Legler, J. M. et al.: Overdiagnosis due to prostate-specific antigen screening: lessons from U.S. prostate cancer incidence trends. J Natl Cancer Inst, 94: 981, 2002.
5. D’Amico AV, Whittington R, Malkowicz SB, et al, Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA.; 280(11): 969-74, 1998.
6. Albertsen P, Hanley JA, Fine J. 20-Year Outcomes Following Conservative Management of Clinically Localized Prostate Cancer JAMA.2005; 293: 2095-2101.
7. Albertsen PC, Hanley JA, Barrows GH, Penswon DF, Kowalczyk PD, Sanders MM, Fine J. Prostate Cancer and the Will Rogers Phenomenon Natl Cancer Inst 2005; 97: 1248-1253.
8. Hanash, K., DC, U., EN, C.: Carcinoma of the prostate: a 15 year follow up. J Urol, 107: 450, 1972.
9. Johansson, J., L ,H., Johansson: Fifteen-year survival in prostate cancer. A prospective, population-based study in Sweden. JAMA, 277: 467, 1997.
10. Lerner, S., C, S.-H., CE, C.: The risk of dying of prostate cancer in patients with clinically localized disease. J Urol, 146: 1040, 1991.
11. Handley, R., TW, C., D, T.: Deferred treatment for prostate cancer. Br I Urol, 62: 249, 1988.
12. Adolfsson, J., J, C., T, L.: Deferred treatment in clinically localized prostate carcinoma. Brit J Urol, 69: 183, 1992.
13. Waaler G and Stenwig AE (1993) Prognosis of localised prostatic cancer managed by “Watch and Wait” policy. Br J Urol, 72: 214-219.
14. Whitmore, W., IA,W., IM, T.: Expectant management of localized prostatic cancer. Cancer, 67: 1091, 1991.
15. George, N.: Natural history of localised prostatic cancer managed by conservative therapy alone. Lancet, 1: 494, 1988.
16. Aus, G., Hugosson, I., Norlen, L.: Long-term survival and mortality in prostate cancer treated with noncurative intent. J Urol, 154: 460, 1995.
17. Sandblom, G., M, D.,Varenhorst: Long-term survival in a Swedish population-based cohort of men with prostate cancer.Urology, 56: 442, 2000.
18. Parker, C.:Active surveillance: towards a new paradigm in the management of early prostate cancer. Lancet Oncol, 5: 101, 2004.
19. Parker, C.: Active surveillance: an individualized approach to early prostate cancer. BJU Int, 92: 2, 2003.
20. Thompson, I.M.,Goodman,P.J., Tangen, C.M.et al.: The influence of finasteride on the development of prostate cancer. N Engl J Med, 349: 215, 2003.
21. Presti, J. C., Jr.: Prostate biopsy: how many cores are enough? Urol Oncol, 21: 135, 2003.
22. Pound, C., AW, P., MA, E. et al.: Natural history of progression after PSA elevation following radical prostatectomy. JAMA, 281: 1591, 1999.
23. Stamey, T.A., Freiha, F.S., McNeal, J.E. et al.:Localized prostate cancer. Relationship of tumor volume to clinical significance for treatment of prostate cancer.Cancer, 71: 933, 1993.
24. Epstein, J. I., Walsh, P. C., Carmichael, M.et al.: Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. Jama, 271: 368, 1994.
25. Epstein, J. I., Pizov, G., Walsh, P. C.: Correlation of pathologic findings with progression after radical retropubic prostatectomy. Cancer, 71: 3582, 1993.
26. Epstein, J.I., Carmichael, M., Partin, A.W.et al.: Is tumor volume an independent predictor of progression following radical prostatectomy? A multivariate analysis of 185 clinical stage B adenocarcinomas of the prostate with 5 years of followup. J Urol, 149: 1478, 1993.
27. Epstein, J. I., Chan, D.W., Sokoll, L. J. et al.: Nonpalpable stage T1c prostate cancer: prediction of insignificant disease using free/total prostate specific antigen levels and needle biopsy findings. J Urol, 160: 2407, 1998.
28. Goto, Y., Ohori, M., Arakawa, A. et al.: Distinguishing clinically important from unimportant prostate cancers before treatment: value of systematic biopsies. J Urol, 156: 1059, 1996.
29. Kattan, M.W., Eastham, J.A., Wheeler, T.M. et al.: Counseling men with prostate cancer: a nomogram for predicting the presence of small, moderately differentiated, confined tumors. J Urol, 170: 1792, 2003.
30. Augustin, H., Hammerer, P. G., Graefen, M. et al.: Insignificant prostate cancer in radical prostatectomy specimen: time trends and preoperative prediction. Eur Urol, 43: 455, 2003.
31. Noguchi, M., Stamey, T.A., McNeal, J.E. et al.: Relationship between systematic biopsies and histological features of 222 radical prostatectomy specimens: lack of prediction of tumor significance for men with nonpalpable prostate cancer. J Urol, 166: 104, 2001.
32. Irwin MB, Trapasso JG. Identification of insignificant prostate cancers: analysis of preoperative parameters. Urology. 1994 Dec; 44(6): 862-7; discussion 867-8.
33. Cupp MR, Bostwick DG, Myers RP, Oesterling JE. The volume of prostate cancer in the biopsy specimen cannot reliably predict the quantity of cancer in the radical prostatectomy specimen on an individual basis. J Urol. 1995 May; 153(5): 1543-8.
34. Anast JW, Andriole GL, Bismar TA, Yan Y, Humphrey PA. Relating biopsy and clinical variables to radical prostatectomy findings: can insignificant and advanced prostate cancer be predicted in a screening population? Urology. 2004 Sep; 64(3): 544-50.
35. Draisma G, Schroeder FH et al Lead times and over detection due to PSA screening: estimates from the European Randomized Study of Screening for Prostate Cancer. J Natl Cancer Inst. 2003 Jun 18; 95(12): 868-78.
36. Tornblom M et al.Lead time associated with screening for prostate cancer. Int J Cancer. 2004 Jan 1; 108(1): 122-9.
37. Choo, R., Klotz, L., Danjoux, C. et al.: Feasibility study: watchful waiting for localized low to intermediate grade prostate carcinoma with selective delayed intervention based on prostate specific antigen, histological and/or clinical progression. J Urol, 167: 1664, 2002.
38. Choo, R., DeBoer, G., Klotz, L. et al.: PSA doubling time of prostate carcinoma managed with watchful observation alone. International Journal of Radiation Oncology, Biology, Physics, 50: 615, 2001.
39. Bill-Axelson A, Holmberg L. Radical prostatectomy versus watchful waiting (update). N Engl J Med.2005 May 12; 352(19): 1977-84.
40. Epstein, J. I., Walsh, P. C., Carter, H.B.: Dedifferentiation of prostate cancer grade with time in men followed expectantly for stage T1c disease. J Urol, 166: 1688, 2001.
41. D’Amico AV, Chen MH, Roehl KA, Catalona WJ. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med. 2004 Jul 8; 351(2):