- Email: [email protected]
Surveillance and Deferred Treatment for Localized Prostate Cancer. Population Based Study in the National Prostate Cancer Register of Sweden
Surveillance and Deferred Treatment for Localized Prostate Cancer. Population Based Study in the National Prostate Cancer Register of Sweden Pär Stattin,*,† Erik Holmberg, Ola Bratt,‡ Jan Adolfsson,§ Jan-Erik Johansson and Jonas Hugosson on behalf of the National Prostate Cancer Register From the Department of Surgical and Perioperative Sciences, Urology and Andrology, Umeå University, Umea (PS), Oncological Centre (EH) and Department of Urology (JH), Sahlgrenska University Hospital, Gothenburg, Department of Urology, University Hospital of Lund, Lund (OB), Oncological Centre, CLINTEC, Karolinska Institute, Stockholm (JA), and Department of Urology, University Hospital of Örebro, Örebro (JEJ), Sweden
Purpose: To what extent active surveillance and deferred treatment for localized risk prostate cancer are used is unclear. We assessed the use of surveillance and of deferred treatment in a population based, nationwide cohort in Sweden. Materials and Methods: In the National Prostate Cancer Register of Sweden, with a 98% coverage vs the compulsory Swedish Cancer Registry, we identified 8,304 incident cases of prostate cancer in 1997 to 2002 with age younger than 70 years, clinical local stage T1 or 2, N0 or Nx, M0 or Mx and serum prostate specific antigen less than 20 ng/ml. Data were extracted from medical charts for 7,782 of these men (94%) at a median of 4 years after diagnosis. Results: Primary treatment was surveillance for 2,065 men (26%), radical prostatectomy for 3,722 (48%), radiotherapy for 1,632 (21%) and hormonal treatment for 363 (5%). Men on surveillance had lower local tumor stage, grade and prostate specific antigen, and were older than those who received active primary treatment (p ⬍0.001). After a median surveillance of 4 years 711 men (34%) on surveillance had received deferred treatment, which was radical prostatectomy for 279 (39%), radiotherapy for 212 (30%) and hormonal treatment for 220 (30%). Conclusions: Surveillance was a common treatment for patients younger than 70 years with localized prostate cancer in Sweden in 1997 to 2002, 26% of men with localized prostate cancer started surveillance and after a median followup of 4 years, 66% of these men remained on surveillance. Key Words: prostatic neoplasms, data collection, registries, therapeutics
idespread testing of serum PSA has drastically increased the number of nonpalpable, well or moderately differentiated prostate cancers, ie T1c tumors with Gleason score 6 or less and small tumor volume in core biopsies.1 As a consequence many men who undergo curative therapy today have a low risk of tumor progression and there is a growing concern of over treatment, in particular of elderly men.2,3 Therefore, active surveillance as a treatment option for low risk prostate cancer has become increasingly attractive.4,5 The aim of surveillance is to avoid side effects of active treatment by treating only men who experience
W
disease progression during followup. However, relatively little is known to what extent surveillance is used and what proportion of men on surveillance subsequently receives deferred treatment. Some previous studies, notably 2 large register based studies in the United States3,6 plus a number of modestly sized, mostly single institution series, have reported on the use of surveillance, and some have also reported the use of deferred treatment.7–15 In this study we assessed the use of surveillance for localized prostate cancer and assessed the use of deferred treatment in a large, population based, nationwide cohort in Sweden. MATERIALS AND METHODS
Submitted for publication May 14, 2008. Supported by grants from the Swedish Cancer Foundation and Västerbotten County Council. Study received approval from the Research ethical committee of Gothenburg University. * Correspondence: Department of Surgical and Perioperative Sciences, Umea University, Umea, Västerbotten S-901 85 Sweden (telephone: ⫹46-90-785 22 91; FAX: ⫹46-90-125396; e-mail: [email protected]). † Financial interest and/or other relationship with Onkologi i Sverige, Orion Pharma and GSK. ‡ Financial interest and/or other relationship with Active Biotech, Ferring, AstraZeneca and Orion Pharma. § Financial interest and/or other relationship with The Scandinavian Journal of Urology and Nephrology.
See Editorial on page 2299.
0022-5347/08/1806-2423/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
The Swedish Cancer Registry and the NPCR In Sweden registration in the Cancer Registry of all cancer cases is mandatory, regulated by law, and the capture rate has been reported to be virtually complete at approximately 98% for solid tumors in patients younger than 75 years.16,17 Currently 98% of all incident prostate cancer cases in the Cancer Registry are also registered in the NPCR which contains data on TNM stage, tumor differentiation, serum PSA at the date of diagnosis and primary treatment within 6 months from the date of diagnosis. Expectancy was one of the treatment options that could be actively indicated in the registration.1,18 In NPCR clinical local tumor stage T2 is
2423
Vol. 180, 2423-2430, December 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.08.044
2424
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER
8304
7960
Cases fulfilling inclusion criteria
Eligible cases for study
No informed consent 344 cases
2065
7782 Cases included in study cohort
Cases on continued surveillance
Cases on surveillance
Primary treatment 3722 RP 1632 RT 363 HT
Data cleaning excluded cases 46 other primary treatments 49 unknown primary treatment 18 missing data for PSA 52 missing data for tumour grade 13 missing data for tumour stage
1354
Terminated surveillance and deferred treatment 279 RP 212 RT 220 HT
FIG. 1. Flow chart for primary treatment and deferred treatment in NPCR of Sweden in men younger than 70 years with clinical local stage T1 or T2 prostate cancer, no signs of metastatic disease (Nx or N0 and M0 or Mx) and serum PSA less than 20 ng/ml.
medical records by cancer registry nurses under supervision by a urologist in each region. The retrospectively extracted data from the charts included date of last followup, date of termination of surveillance which was defined as the date of start of deferred treatment, reason for termination of surveillance and type of deferred treatment. As this was a retrospective observational study, followup and initiation of deferred treatment was made at the discretion of the physician and the patient, and there was no protocol for followup or criteria for termination of surveillance. The study was approved by the Research ethical committee of Gothenburg University and consent was obtained from all study subjects by the use of an opt-out protocol.
reported without any further subclassification into T2a, b or c. We identified all men with an incident prostate cancer diagnosis in NPCR between 1997 and 2002 in 5 of 6 regions in Sweden and in 1 region for men diagnosed between 1998 and 2002, who were 70 years or younger at diagnosis, with clinical local stage T1a, b, c or T2, without signs of lymph node metastasis (Nx or N0) or bone metastasis (Mx or M0) and with serum PSA 20 ng/ml or less. We defined low risk cases as stage T1a, b or c, with tumor differentiation Gleason score 2-6 or WHO I–II and serum PSA less than 10 ng/ml. Intermediate risk cases were defined as T2 tumors, or tumor differentiation of Gleason score 7, or PSA greater than 10 ng/ml. High risk tumors were defined as tumors with Gleason score 8-10 or WHO III, irrespective of PSA or local stage, ie T1a, b, c or T2.
Statistical Methods Chi-square tests were used to test for difference in distribution of surveillance vs active treatment and vs termination of surveillance according to tumor characteristics and age. A nonparametric test for trend across ordered groups was used
Data Extraction in Followup Study Data on observations made more than 6 months after the date of diagnosis of prostate cancer were extracted from
TABLE 1. Primary treatment in the NPCR Surveillance No. pts No. T stage (%):† T1a T1b T1c T2 No. Gleason score (%):†,‡ 2–6 7 8–10 Missing No. WHO grade (%): I II III No. ng/ml PSA (%):† 0–4 4–10 10–20 Median ng/ml PSA (25–75 percentile) No. pt age (%):† Younger than 60 60–64 65–70 Mean pt age (SD)
2,065
RP
RT
3,722
HT*
1,632
All
363
7,782
337 97 1,032 599
(81) (49) (27) (18)
57 51 1,930 1,684
(14) (25) (51) (49)
17 34 686 895
(4) (17) (18) (26)
3 15 113 232
(⬍1) (8) (3) (7)
414 197 3,761 3,410
(100) (100) (100) (100)
1,444 100 19 502
(29) (8) (6) (40)
2,424 716 150 432
(49) (57) (46) (34)
925 343 110 254
(19) (27) (34) (20)
131 104 48 80
(3) (8) (15) (6)
4,924 1,263 327 1,268
(100) (100) (100) (100)
375 110 17
(51) (24) (20)
218 177 37
(30) (39) (43)
111 123 20
(15) (27) (23)
26 41 13
(4) (9) (15)
730 451 87
(100) (100) (100)
429 (46) 1,066 (25) 570 (22) 6.7 (4.1–10.0)
367 (40) 2,217 (52) 1,138 (44) 7.5 (5.0–11.0)
117 (13) 869 (20) 646 (25) 8.8 (6.0–12.0)
17 (2) 150 (3) 196 (8) 10.0 (7.7–14.0)
930 (100) 4,302 (100) 2,550 (100) 7.8 (5.0–11.0)
301 525 1,239 64.7
1,314 1,196 1,212 61.4
349 469 814 63.5
31 79 253 65.8
1,995 2,269 3,518 62.9
(15) (23) (35) (4.6)
(66) (53) (34) (5.3)
(17) (21) (23) (4.9)
(2) (3) (7) (4.2)
* Included monotherapy with per oral antiandrogen blockade in 83 men, castration therapy in 212 and hormonal treatment not specified in 68. † pdiff ⬍0.001 between surveillance vs active primary treatment. ‡ Gleason score and WHO combined GS 2–6 ⫹ WHO I, II, GS 7, and GS 8–10 ⫹ WHO III.
(100) (100) (100) (5.2)
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER to test the association between primary surveillance (and continued surveillance) and ordinal variables, ie risk category and age category. All analyses were done using the statistical package Stata 10™.
RESULTS Figure 1 is a flow chart of the identification and selection of the study cohort. For 7,784 men included in the study cohort primary treatment was surveillance for 2,065 (26%), RP for 3,722 (48%), RT for 1,632 (21%) and hormonal therapy for 363 (5%). There was a linear decrease during the study period of the proportion of men that selected surveillance (ptrend ⬍0.001). In 1997, 42% of men
2425
selected surveillance whereas 20% selected surveillance in 2002. Primary Treatment vs Tumor Characteristics and Patient Age Men who started surveillance had lower tumor stage, tumor grade, PSA and were older than men who underwent active treatment (pdiff ⬍0.001) (table 1). Of 593 men with stage T1a or b tumors 434 (73%) were put on surveillance and 159 (27%) received curative therapy. Among men with palpable tumors (T2) 599 (18%) underwent surveillance and 2,579 (76%) received curative therapy. Of 6,105 men with well or moderately differentiated tumors on biopsy 1,929 men (32%) were put on surveillance whereas the corresponding number
FIG. 2. Primary treatment according to risk category and age category in NPCR of Sweden for 7,782 men younger than 70 years diagnosed between 1997 and 2002 with T1 or T2 tumor, Nx or N0, Mx or M0 and serum PSA less than 20 ng/ml. Low risk category— clinical local stage T1a, b or c, and Gleason score 2– 6 or WHO I–II and serum PSA less than 10 ng/ml. Intermediate risk category—stage T2, Gleason score 7 or serum PSA greater than 10 ng/ml. High risk category—Gleason score 8 –10, WHO III, any T stage, any serum PSA. Trend test for surveillance vs active primary treatment according to age in low risk p ⬍0.001, intermediate risk p ⬍0.001 and high risk groups p ⬍0.01. Trend test for surveillance vs active primary treatment according to risk groups p ⬍0.001 in all 3 age groups.
2426
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER Termination of Surveillance vs Tumor Characteristics and Age After a median followup of 4 years 711 of 2,065 men (34%) on surveillance had terminated surveillance. The proportion of men terminating surveillance during followup is depicted in figure 3. Surveillance was terminated because of patient choice in 24%, PSA progression in 49%, other signs of progression in 14%, other reasons in 11% and the reason could not be assessed in 3%. For these 711 men deferred treatment was RP for 279 (39%), RT for 212 (29%) and hormonal treatment for 220 (30%) (table 2). The termination of surveillance was less strongly related to tumor characteristics and age than the choice of primary treatment but was still significantly associated with stage, PSA and age at diagnosis. The distribution of the deferred treatment according to risk and age is depicted in figure 4.
FIG. 3. Proportion of men terminating surveillance in NPCR of Sweden among 2,015 men younger than 70 years with T1 or T2 tumor, Nx or N0, M0 or Mx and PSA less than 20 ng/ml. Shaded area shows 95% CI for estimate. Of 2,065 cases on surveillance 50 were excluded from Kaplan-Meier analysis. Cause for exclusion was date of diagnosis after date of end of surveillance (9 cases), no date for last control (18) and surveillance terminated without date of end of surveillance (23).
DISCUSSION In this nationwide, retrospective register based study in Sweden of men younger than 70 years diagnosed in 1997 to 2002 with localized prostate cancer, 26% were initially treated with surveillance. Two-thirds of men on surveillance remained untreated after a median followup of 4 years. Older men and men with lower risk tumors were significantly more likely to opt for surveillance. There are some previous, mostly smaller reports on the use of surveillance as initial treatment for localized prostate cancer, with the proportion of men on surveillance ranging from 7% to 45%. In a recent report based on cases in the Connecticut Tumor Registry diagnosed from 1990 to 1992, surveillance had been elected for 114 of 1,618 men (7%) younger than 75 years with localized prostate cancer.14 In the CaPSURE™ study 366 of 3,976 men (9%) with low risk tumors selected surveillance.19 In the Netherlands in the
of men with poorly differentiated tumors was 36 of 414 (9%). Among men younger than 60 years at diagnosis 15% selected surveillance and among men 60 to 70 years old the corresponding figure was 35%. The distribution of primary treatment for low, intermediate and high risk tumors among men younger than 60 years, 60 to 65 and 66 to 70 years old is depicted in figure 2. Surveillance was significantly associated with age and risk group, and was selected for 55% of men 66 to 70 years old with a low risk tumor vs 8% of men younger than 60 years with an intermediate risk tumor.
TABLE 2. Deferred treatment in NPCR after termination of surveillance in 2,065 men Surveillance Continued No. pts No. T stage (%): T1a T1b T1c T2 No. Gleason score (%): 2–6 7 8–10 Missing No. WHO grade (%): I II III No. ng/ml PSA (%): 0–4 4–10 10–20 Median ng/ml PSA (25–75 percentile) No. pt age (%): Younger than 60 60–64 65–70 Mean pt age (SD)
1,354
Surveillance Terminated
Surveillance Terminated 711
RP
RT
279
HT*
212
pdiff†
220
⬍0.001
304 76 621 353
(90) (78) (60) (59)
33 21 411 246
(10) (22) (40) (41)
12 8 199 60
(36) (38) (48) (24)
9 2 124 77
(27) (10) (30) (31)
12 11 88 109
(36) (52) (21) (44)
960 67 9 318
(67) (67) (47) (66)
484 33 10 184
(33) (33) (53) (37)
224 6 1 48
(46) (18) (10) (26)
143 15 3 51
(30) (45) (30) (28)
117 12 6 85
(24) (36) (60) (46)
252 54 12
(67) (49) (71)
123 56 5
(33) (51) (29)
38 10
(31) (18)
37 13 1
(30) (23) (20)
48 33 4
(39) (59) (80)
0.63‡
—
349 (81) 671 (63) 334 (59) 6.0 (3.9–9.8)
80 (19) 395 (37) 236 (41) 7.5 (5.0–11.0)
50 (61) 178 (45) 51 (22) 6.1 (4.4–9.0)
22 (28) 120 (30) 70 (30) 7.1 (5.0–11.0)
8 (10) 97 (25) 115 (49) 10.0 (7.2–15.0)
180 324 850 65.0
121 201 389 64.2
77 110 92 62.1
37 64 111 63.9
7 27 186 67.3
(60) (62) (69) (4.5)
(40) (38) (32) (4.7)
(64) (55) (24) (4.5)
(31) (32) (29) (4.7)
(6) (13) (48) (3.1)
* Included monotherapy with per oral antiandrogen blockade in 51 men, castration therapy in 116 and hormonal treatment not specified in 53. † pdiff between continued surveillance vs terminated surveillance. ‡ Gleason score and WHO combined GS 2–6 ⫹ WHO I, II, GS 7, and GS 8–10 ⫹ WHO III.
⬍0.001
⬍0.01
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER
2427
FIG. 4. Continued surveillance and terminated surveillance with start of deferred active treatment in NPCR of Sweden according to risk category and age for 2,065 men younger than 70 years diagnosed between 1997 and 2002 with T1 or T2 tumor, Nx or N0, Mx or M0 and serum PSA less than 20 ng/ml with primary surveillance. Low risk category— clinical local stage T1a, b, c, and Gleason score 2– 6 or WHO I–II and serum PSA less than 10 ng/ml. Intermediate risk category—stage T2, Gleason score 7 or serum PSA greater than 10 ng/ml. High risk category—Gleason score 8 –10, any T stage, any serum PSA. High risk categories for men 65 years or younger were not demonstrated as there were only 9 men in these groups. Only 2 men younger than 60 years in high risk category both received hormonal treatment as deferred treatment. Of 7 men 60 to 65 years old in high risk category 1 received RT, 1 received hormonal treatment and 5 remained on surveillance. Test for trend for surveillance vs active primary treatment according to age in low risk p ⬍0.01 and intermediate risk p ⬍0.12. Trend test for surveillance vs active primary treatment according to risk groups p ⬍0.001 in all 3 age groups in low and intermediate risk categories.
Rotterdam section of the ERSPC, 278 of 1,772 men (16%) with screen detected prostate cancer selected surveillance, whereas in the Gothenburg section of ERSPC in Sweden 270 of 660 men (41%) chose surveillance.20 In a study in the Geneva Cancer Registry, Switzerland 378 of 844 men (45%) with localized prostate cancer in all ages were treated expectantly.10 Finally, 2 large United States studies based on data from the SEER database reported on the frequency in much larger cohorts.3,6 One study reported that 10,868 of 24,405 men (65%) with low risk tumors (defined as Gleason score 2– 4 or Gleason score 2–7 for men older than 70 years) selected no curative treatment.3 Another SEER study on Medicare patients reported that 12,608 of 44,630 men (37%)
aged 65 to 80 years and diagnosed with organ confined, well or intermediately differentiated prostate cancer between 1991 and 1999 selected no curative treatment.6 Importantly, surveillance and androgen deprivation therapy could not be distinguished from one another in the SEER registry. Furthermore, these studies did not specifically report local tumor stage or serum PSA at diagnosis. Thus, there are large differences in the proportion of men who select surveillance as treatment for localized prostate cancer, and the proportion seems to be strongly related to patient age and tumor characteristics as demonstrated in our study. Treatment strategy was also strongly related to geographic region with different treatment policies.
2428
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER
After a median followup of 4 years the proportion of men who had terminated surveillance in our study was 711 of 2,065 (34%). This proportion is similar to that found after similar followup times in the Rotterdam ERSPC cohort in which 19 of 64 (30%) men subsequently received active treatment,11 and in the Gothenburg ERSPC cohort where the number of men converting to active therapy was 104 of 270 (39%).20 The corresponding number in a Canadian single institution series study was 101 of 299 men (35%),8 and in a similar study from the United States 103 of 407 men (25%).9 Two small single institution series have reported that approximately 10% of the patients converted to active therapy after shorter followup time,12,13 and 1 United Kingdom study reported that 238 of 326 men (73%) remained on surveillance after 2 years.15 Active surveillance has recently received increasing attention as a treatment strategy for low and intermediate risk prostate cancer. Our study showed that surveillance was a relatively common treatment option in Sweden in 1997 to 2002 for men younger than 70 years with localized prostate cancer. More recent data from NPCR show that the percentage of men diagnosed with prostate cancer, fulfilling the same inclusion criteria as in our study and opting for surveillance is quite stable in Sweden at 24% in 2003, 21% in 2004, 22% in 2005 and 26% in 2006. Our study shows that there is a large number of men with low and intermediate risk tumors who would be eligible for inclusion in prospective studies on active surveillance that are necessary to elucidate the long-term outcome of this treatment strategy. To the best of our knowledge our study is the first report based on a nationwide, population based study cohort on the number of men with localized prostate cancer put on surveillance, and the number of men who terminated surveillance within the first 4 years and for whom data were available for tumor characteristics as well as primary and deferred treatment. There was an ongoing screening study in Gothenburg during the study period but in the rest of Sweden no formal screening program was in operation. The strengths of this study include the population based design which included men from all types of medical facilities in the entire nation. Approximately 90% of all men in Sweden younger than 70 years with an incident localized prostate cancer in 1997 (1998 for 1 region) to 2002 were included in our study. This assumption is based on the 98% capture rate of the Swedish Cancer Registry of all prostate cancer cases, the 98% capture rate of the NPCR vs the Cancer Registry and the 94% coverage of this followup study within the NPCR. There are some weaknesses in our study design. Active surveillance and watchful waiting were not distinguished as separate entities in the NPCR during the study period and the selection of deferred treatment suggests that the surveillance group was a mix of these 2 strategies. Some men on surveillance underwent deferred curative treatment, whereas some men subsequently received hormonal therapy. There was no information available in our cohort on tumor extent on core biopsies, PSA density or PSA doubling time, and there was no predefined protocol according to which the patients were followed and no criteria for termination of surveillance. We do not have data on how many patients underwent repeat biopsies, nor do we have data on comorbidity among men.
CONCLUSIONS This population based nationwide study showed that 26% of Swedish men younger than 70 years diagnosed with localized prostate cancer in 1997 to 2002 were treated with surveillance. Older men and men with lower risk tumors underwent surveillance significantly more often than younger men and men with higher risk tumors. Two-thirds of men on surveillance remained untreated after a median followup of 4 years. ACKNOWLEDGMENTS NPCR steering group: Pär Stattin, Anders Widmark, Magnus Hellström, Magnus Törnblom, Jan Adolfsson, Anna BillAxelson, Jan-Erik Johanssson, Christer Ahlstrand, Bill Pettersson, Jonas Hugosson, Jan-Erik Damber, Ola Bratt, Göran Ahlgren and Roy Ehrnström.
Abbreviations and Acronyms CaPSURE ⫽ ERSPC ⫽ GS HT NPCR PSA RP RT SEER
⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽
Cancer of the Prostate Strategic Urologic Research Endeavor European Randomized Study of Screening for Prostate Cancer Gleason score hormonal treatment National Prostate Cancer Register prostate specific antigen radical retropubic prostatectomy radiotherapy Surveillance, Epidemiology and End Results
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8. 9.
Adolfsson J, Garmo H, Varenhorst E, Ahlgren G, Ahlstrand C, Andrén O et al: Clinical characteristics and primary treatment of prostate cancer in Sweden between 1996 and 2005. Scand J Urol Nephrol 2007; 41: 456. Barry MJ: Early detection and aggressive treatment of prostate cancer: groping in the dark. J Gen Intern Med 2000; 15: 749. Miller DC, Gruber SB, Hollenbeck BK, Montie JE and Wei JT: Incidence of initial local therapy among men with lowerrisk prostate cancer in the United States. J Natl Cancer Inst 2006; 98: 1134. Parker C: Active surveillance: towards a new paradigm in the management of early prostate cancer. Lancet Oncol 2004; 5: 101. Klotz L: Active surveillance with selective delayed intervention using PSA doubling time for good risk prostate cancer. Eur Urol 2005; 47: 16. Wong YN, Mitra N, Hudes G, Localio R, Schwartz JS, Wan F et al: Survival associated with treatment vs observation of localized prostate cancer in elderly men. JAMA 2006; 296: 2683. Harlan SR, Cooperberg MR, Elkin EP, Lubeck DP, Meng MV, Mehta SS et al: Time trends and characteristics of men choosing watchful waiting for initial treatment of localized prostate cancer: results from CaPSURE. J Urol 2003; 170: 1804. Klotz L: Active surveillance for prostate cancer: for whom? J Clin Oncol 2005; 23: 8165. Carter HB, Kettermann A, Warlick C, Metter EJ, Landis P, Walsh PC et al: Expectant management of prostate cancer
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER with curative intent: an update of the Johns Hopkins experience. J Urol 2007; 178: 2359. 10. Merglen A, Schmidlin F, Fioretta G, Verkooijen HM, Rapiti E, Zanetti R et al: Short- and long-term mortality with localized prostate cancer. Arch Intern Med 2007; 167: 1944. 11. Roemeling S, Roobol MJ, de Vries SH, Wolters T, Gosselaar C, van Leenders GJ et al: Active surveillance for prostate cancers detected in three subsequent rounds of a screening trial: characteristics, PSA doubling times, and outcome. Eur Urol 2007; 51: 1244. 12. Hardie C, Parker C, Norman A, Eeles R, Horwich A, Huddart R et al: Early outcomes of active surveillance for localized prostate cancer. BJU Int 2005; 95: 956. 13. Soloway MS, Soloway CT, Williams S, Ayyathurai R, Kava B and Manoharan M: Active surveillance; a reasonable management alternative for patients with prostate cancer: the Miami experience. BJU Int 2008; 101: 165. 14. Albertsen PC, Hanley JA, Penson DF, Barrows G and Fine J: 13-Year outcomes following treatment for clinically localized prostate cancer in a population based cohort. J Urol 2007; 177: 932. 15. van As NJ, Norman AR, Thomas K, Khoo VS, Thompson A, Huddart RA et al: Predicting the probability of deferred radical treatment for localised prostate cancer managed by active surveillance. Eur Urol 2008; Epub ahead of print. 16. Mattsson B and Wallgren A: Completeness of the Swedish Cancer Register. Non-notified cancer cases recorded on death certificates in 1978. Acta Radiol Oncol 1984; 23: 305. 17. Barlow L, Westergren K, Holmberg L and Talback M: The completeness of the Swedish Cancer Register—a sample survey for year 1998. Acta Oncol 2008; 3: 1. 18. The National Prostate Cancer Register (NPCR) in Sweden 2002–2006. Regional Oncologic Center: Uppsala 2008. Available at http://www.roc.se/prostata/rapport/rapport_ 02_06.pdf. 19. Cooperberg MR, Broering JM, Kantoff PW and Carroll PR: Contemporary trends in low risk prostate cancer: risk assessment and treatment. J Urol 2007; 178: S14. 20. Khatami A, Aus G, Damber JE, Lilja H, Lodding P and Hugosson J: PSA doubling time predicts the outcome after active surveillance in screening-detected prostate cancer: results from the European randomized study of screening for prostate cancer, Sweden section. Int J Cancer 2007; 120: 170.
with selective deferred therapy, and compare outcomes with immediate treatments. We must refine selection criteria for surveillance and standardize trigger points for initiating therapy, etc. Reaching these goals will entail substantial investment in long-term clinical trials. However, implementation of surveillance strategies may depend more on the decision making process than the available information (as evidenced by the fact that Europeans and Americans have the same information now and yet make different management decisions). Observational, population based studies such as the NPCR study and the establishment of risk assessment tools and decision analysis tools may inform efforts to bring surveillance into the mainstream in the United States. Daniel A. Barocas Department of Urologic Surgery Vanderbilt University Medical Center Nashville, Tennessee 1.
Barocas DA, Cowan JE, Smith JA Jr and Carroll PR: What percentage of patients with newly diagnosed carcinoma of the prostate are candidates for surveillance? An analysis of the CaPSURETM database. J Urol 2008; 180: 1330.
This population based study provides insight into the Swedish patterns of care in recent years. Of all men younger than 70 years diagnosed with localized disease and PSA less than 20 ng/ml, surveillance was chosen by as many as 26%. At 4 years the actuarial freedom from treatment for these men was 63% with PSA trends the main indication for intervention. It will be interesting to learn how practice has evolved more recently in the light of the Scandinavian Prostate Cancer Group-4 trial data showing that PSA kinetics in an individual are a poor predictor of the risk of lethal prostate cancer.1 The contrast with practice in the United States is obvious. The Swedish approach provides research opportunities that are not readily available in the United States. Prospective studies in men on surveillance could address the role of repeat biopsies and the impact of tertiary prevention strategies. In addition, with long-term followup the Swedish NPCR will provide important data on the natural history of the disease.
EDITORIAL COMMENTS Stattin et al and the NPCR of Sweden are to be commended on their insightful analysis. The capture rate (98% of solid tumors in the nation of Sweden) of this registry makes it a powerful resource for this type of observational study. One of the most interesting facets of this research is the geographic variation in the management of localized prostate cancer. The authors found that 32% of low and intermediate risk patients chose initial surveillance. In contrast, a recent study of the CaPSURE database in the United States showed that only 6% of low and intermediate risk patients chose surveillance and in a highly selected subset of low risk patients the rate was still just 9%.1 This variation in practice patterns warrants further investigation and studies of this type are an excellent point of departure. There are many obstacles one can posit to the widespread use of surveillance in the United States. We must establish the safety and efficacy of initial surveillance
2429
Chris Parker Institute of Cancer Research Royal Marsden Hospital Sutton, Surrey, United Kingdom 1.
Fall K, Garmo H, Andrén O, Bill-Axelson A, Adolfsson J, Adami HO et al: Prostate-specific antigen levels as a predictor of lethal prostate cancer. J Natl Cancer Inst 2007; 99: 526.
REPLY BY AUTHORS A quarter of men in Sweden with localized prostate cancer diagnosed in 1997 to 2002 did not receive active primary treatment and two-thirds remained on surveillance after a median followup of 4 years. For whom is this an appropriate treatment strategy? Long-term outcomes of this treatment strategy as well as active treatments can in the near future be assessed in this nationwide, population based cohort by linkage to the Swedish Cause of Death Register, which will
2430
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER
provide an important benchmark of outcomes obtained in daily clinical practice. We continue a tradition of Scandinavian registries which have changed treatment guidelines for rectal cancer, inguinal hernia and hip replacement.1– 4 1.
2.
Påhlman L, Bohe M, Cedermark B, Dahlberg M, Lindmark G, Sjödahl R et al: The Swedish Rectal Cancer Registry. Br J Surg 2007; 94: 1285. Haapaniemi S and Nordin P: The Swedish Hernia Register. In: Recurrent Hernia: Prevention and Treatment. Edited by V
3.
4.
Schumpelick and RJ Fitzgibbons. Berlin: Springer-Verlag 2007. Kehlet H, Bay-Nielsen M and Danish Hernia Database Collaboration: Nationwide quality improvement of groin hernia repair from the Danish Hernia Database of 87,840 patients from 1998 to 2005. Hernia 2008; 12: 1. Malchau H, Garellick G, Eisler T, Kärrholm J and Herberts P: Presidential guest address: the Swedish Hip Registry: increasing the sensitivity by patient outcome data. Clin Orthop Relat Res 2005; 441: 19.
Purpose: To what extent active surveillance and deferred treatment for localized risk prostate cancer are used is unclear. We assessed the use of surveillance and of deferred treatment in a population based, nationwide cohort in Sweden. Materials and Methods: In the National Prostate Cancer Register of Sweden, with a 98% coverage vs the compulsory Swedish Cancer Registry, we identified 8,304 incident cases of prostate cancer in 1997 to 2002 with age younger than 70 years, clinical local stage T1 or 2, N0 or Nx, M0 or Mx and serum prostate specific antigen less than 20 ng/ml. Data were extracted from medical charts for 7,782 of these men (94%) at a median of 4 years after diagnosis. Results: Primary treatment was surveillance for 2,065 men (26%), radical prostatectomy for 3,722 (48%), radiotherapy for 1,632 (21%) and hormonal treatment for 363 (5%). Men on surveillance had lower local tumor stage, grade and prostate specific antigen, and were older than those who received active primary treatment (p ⬍0.001). After a median surveillance of 4 years 711 men (34%) on surveillance had received deferred treatment, which was radical prostatectomy for 279 (39%), radiotherapy for 212 (30%) and hormonal treatment for 220 (30%). Conclusions: Surveillance was a common treatment for patients younger than 70 years with localized prostate cancer in Sweden in 1997 to 2002, 26% of men with localized prostate cancer started surveillance and after a median followup of 4 years, 66% of these men remained on surveillance. Key Words: prostatic neoplasms, data collection, registries, therapeutics
idespread testing of serum PSA has drastically increased the number of nonpalpable, well or moderately differentiated prostate cancers, ie T1c tumors with Gleason score 6 or less and small tumor volume in core biopsies.1 As a consequence many men who undergo curative therapy today have a low risk of tumor progression and there is a growing concern of over treatment, in particular of elderly men.2,3 Therefore, active surveillance as a treatment option for low risk prostate cancer has become increasingly attractive.4,5 The aim of surveillance is to avoid side effects of active treatment by treating only men who experience
W
disease progression during followup. However, relatively little is known to what extent surveillance is used and what proportion of men on surveillance subsequently receives deferred treatment. Some previous studies, notably 2 large register based studies in the United States3,6 plus a number of modestly sized, mostly single institution series, have reported on the use of surveillance, and some have also reported the use of deferred treatment.7–15 In this study we assessed the use of surveillance for localized prostate cancer and assessed the use of deferred treatment in a large, population based, nationwide cohort in Sweden. MATERIALS AND METHODS
Submitted for publication May 14, 2008. Supported by grants from the Swedish Cancer Foundation and Västerbotten County Council. Study received approval from the Research ethical committee of Gothenburg University. * Correspondence: Department of Surgical and Perioperative Sciences, Umea University, Umea, Västerbotten S-901 85 Sweden (telephone: ⫹46-90-785 22 91; FAX: ⫹46-90-125396; e-mail: [email protected]). † Financial interest and/or other relationship with Onkologi i Sverige, Orion Pharma and GSK. ‡ Financial interest and/or other relationship with Active Biotech, Ferring, AstraZeneca and Orion Pharma. § Financial interest and/or other relationship with The Scandinavian Journal of Urology and Nephrology.
See Editorial on page 2299.
0022-5347/08/1806-2423/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
The Swedish Cancer Registry and the NPCR In Sweden registration in the Cancer Registry of all cancer cases is mandatory, regulated by law, and the capture rate has been reported to be virtually complete at approximately 98% for solid tumors in patients younger than 75 years.16,17 Currently 98% of all incident prostate cancer cases in the Cancer Registry are also registered in the NPCR which contains data on TNM stage, tumor differentiation, serum PSA at the date of diagnosis and primary treatment within 6 months from the date of diagnosis. Expectancy was one of the treatment options that could be actively indicated in the registration.1,18 In NPCR clinical local tumor stage T2 is
2423
Vol. 180, 2423-2430, December 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.08.044
2424
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER
8304
7960
Cases fulfilling inclusion criteria
Eligible cases for study
No informed consent 344 cases
2065
7782 Cases included in study cohort
Cases on continued surveillance
Cases on surveillance
Primary treatment 3722 RP 1632 RT 363 HT
Data cleaning excluded cases 46 other primary treatments 49 unknown primary treatment 18 missing data for PSA 52 missing data for tumour grade 13 missing data for tumour stage
1354
Terminated surveillance and deferred treatment 279 RP 212 RT 220 HT
FIG. 1. Flow chart for primary treatment and deferred treatment in NPCR of Sweden in men younger than 70 years with clinical local stage T1 or T2 prostate cancer, no signs of metastatic disease (Nx or N0 and M0 or Mx) and serum PSA less than 20 ng/ml.
medical records by cancer registry nurses under supervision by a urologist in each region. The retrospectively extracted data from the charts included date of last followup, date of termination of surveillance which was defined as the date of start of deferred treatment, reason for termination of surveillance and type of deferred treatment. As this was a retrospective observational study, followup and initiation of deferred treatment was made at the discretion of the physician and the patient, and there was no protocol for followup or criteria for termination of surveillance. The study was approved by the Research ethical committee of Gothenburg University and consent was obtained from all study subjects by the use of an opt-out protocol.
reported without any further subclassification into T2a, b or c. We identified all men with an incident prostate cancer diagnosis in NPCR between 1997 and 2002 in 5 of 6 regions in Sweden and in 1 region for men diagnosed between 1998 and 2002, who were 70 years or younger at diagnosis, with clinical local stage T1a, b, c or T2, without signs of lymph node metastasis (Nx or N0) or bone metastasis (Mx or M0) and with serum PSA 20 ng/ml or less. We defined low risk cases as stage T1a, b or c, with tumor differentiation Gleason score 2-6 or WHO I–II and serum PSA less than 10 ng/ml. Intermediate risk cases were defined as T2 tumors, or tumor differentiation of Gleason score 7, or PSA greater than 10 ng/ml. High risk tumors were defined as tumors with Gleason score 8-10 or WHO III, irrespective of PSA or local stage, ie T1a, b, c or T2.
Statistical Methods Chi-square tests were used to test for difference in distribution of surveillance vs active treatment and vs termination of surveillance according to tumor characteristics and age. A nonparametric test for trend across ordered groups was used
Data Extraction in Followup Study Data on observations made more than 6 months after the date of diagnosis of prostate cancer were extracted from
TABLE 1. Primary treatment in the NPCR Surveillance No. pts No. T stage (%):† T1a T1b T1c T2 No. Gleason score (%):†,‡ 2–6 7 8–10 Missing No. WHO grade (%): I II III No. ng/ml PSA (%):† 0–4 4–10 10–20 Median ng/ml PSA (25–75 percentile) No. pt age (%):† Younger than 60 60–64 65–70 Mean pt age (SD)
2,065
RP
RT
3,722
HT*
1,632
All
363
7,782
337 97 1,032 599
(81) (49) (27) (18)
57 51 1,930 1,684
(14) (25) (51) (49)
17 34 686 895
(4) (17) (18) (26)
3 15 113 232
(⬍1) (8) (3) (7)
414 197 3,761 3,410
(100) (100) (100) (100)
1,444 100 19 502
(29) (8) (6) (40)
2,424 716 150 432
(49) (57) (46) (34)
925 343 110 254
(19) (27) (34) (20)
131 104 48 80
(3) (8) (15) (6)
4,924 1,263 327 1,268
(100) (100) (100) (100)
375 110 17
(51) (24) (20)
218 177 37
(30) (39) (43)
111 123 20
(15) (27) (23)
26 41 13
(4) (9) (15)
730 451 87
(100) (100) (100)
429 (46) 1,066 (25) 570 (22) 6.7 (4.1–10.0)
367 (40) 2,217 (52) 1,138 (44) 7.5 (5.0–11.0)
117 (13) 869 (20) 646 (25) 8.8 (6.0–12.0)
17 (2) 150 (3) 196 (8) 10.0 (7.7–14.0)
930 (100) 4,302 (100) 2,550 (100) 7.8 (5.0–11.0)
301 525 1,239 64.7
1,314 1,196 1,212 61.4
349 469 814 63.5
31 79 253 65.8
1,995 2,269 3,518 62.9
(15) (23) (35) (4.6)
(66) (53) (34) (5.3)
(17) (21) (23) (4.9)
(2) (3) (7) (4.2)
* Included monotherapy with per oral antiandrogen blockade in 83 men, castration therapy in 212 and hormonal treatment not specified in 68. † pdiff ⬍0.001 between surveillance vs active primary treatment. ‡ Gleason score and WHO combined GS 2–6 ⫹ WHO I, II, GS 7, and GS 8–10 ⫹ WHO III.
(100) (100) (100) (5.2)
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER to test the association between primary surveillance (and continued surveillance) and ordinal variables, ie risk category and age category. All analyses were done using the statistical package Stata 10™.
RESULTS Figure 1 is a flow chart of the identification and selection of the study cohort. For 7,784 men included in the study cohort primary treatment was surveillance for 2,065 (26%), RP for 3,722 (48%), RT for 1,632 (21%) and hormonal therapy for 363 (5%). There was a linear decrease during the study period of the proportion of men that selected surveillance (ptrend ⬍0.001). In 1997, 42% of men
2425
selected surveillance whereas 20% selected surveillance in 2002. Primary Treatment vs Tumor Characteristics and Patient Age Men who started surveillance had lower tumor stage, tumor grade, PSA and were older than men who underwent active treatment (pdiff ⬍0.001) (table 1). Of 593 men with stage T1a or b tumors 434 (73%) were put on surveillance and 159 (27%) received curative therapy. Among men with palpable tumors (T2) 599 (18%) underwent surveillance and 2,579 (76%) received curative therapy. Of 6,105 men with well or moderately differentiated tumors on biopsy 1,929 men (32%) were put on surveillance whereas the corresponding number
FIG. 2. Primary treatment according to risk category and age category in NPCR of Sweden for 7,782 men younger than 70 years diagnosed between 1997 and 2002 with T1 or T2 tumor, Nx or N0, Mx or M0 and serum PSA less than 20 ng/ml. Low risk category— clinical local stage T1a, b or c, and Gleason score 2– 6 or WHO I–II and serum PSA less than 10 ng/ml. Intermediate risk category—stage T2, Gleason score 7 or serum PSA greater than 10 ng/ml. High risk category—Gleason score 8 –10, WHO III, any T stage, any serum PSA. Trend test for surveillance vs active primary treatment according to age in low risk p ⬍0.001, intermediate risk p ⬍0.001 and high risk groups p ⬍0.01. Trend test for surveillance vs active primary treatment according to risk groups p ⬍0.001 in all 3 age groups.
2426
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER Termination of Surveillance vs Tumor Characteristics and Age After a median followup of 4 years 711 of 2,065 men (34%) on surveillance had terminated surveillance. The proportion of men terminating surveillance during followup is depicted in figure 3. Surveillance was terminated because of patient choice in 24%, PSA progression in 49%, other signs of progression in 14%, other reasons in 11% and the reason could not be assessed in 3%. For these 711 men deferred treatment was RP for 279 (39%), RT for 212 (29%) and hormonal treatment for 220 (30%) (table 2). The termination of surveillance was less strongly related to tumor characteristics and age than the choice of primary treatment but was still significantly associated with stage, PSA and age at diagnosis. The distribution of the deferred treatment according to risk and age is depicted in figure 4.
FIG. 3. Proportion of men terminating surveillance in NPCR of Sweden among 2,015 men younger than 70 years with T1 or T2 tumor, Nx or N0, M0 or Mx and PSA less than 20 ng/ml. Shaded area shows 95% CI for estimate. Of 2,065 cases on surveillance 50 were excluded from Kaplan-Meier analysis. Cause for exclusion was date of diagnosis after date of end of surveillance (9 cases), no date for last control (18) and surveillance terminated without date of end of surveillance (23).
DISCUSSION In this nationwide, retrospective register based study in Sweden of men younger than 70 years diagnosed in 1997 to 2002 with localized prostate cancer, 26% were initially treated with surveillance. Two-thirds of men on surveillance remained untreated after a median followup of 4 years. Older men and men with lower risk tumors were significantly more likely to opt for surveillance. There are some previous, mostly smaller reports on the use of surveillance as initial treatment for localized prostate cancer, with the proportion of men on surveillance ranging from 7% to 45%. In a recent report based on cases in the Connecticut Tumor Registry diagnosed from 1990 to 1992, surveillance had been elected for 114 of 1,618 men (7%) younger than 75 years with localized prostate cancer.14 In the CaPSURE™ study 366 of 3,976 men (9%) with low risk tumors selected surveillance.19 In the Netherlands in the
of men with poorly differentiated tumors was 36 of 414 (9%). Among men younger than 60 years at diagnosis 15% selected surveillance and among men 60 to 70 years old the corresponding figure was 35%. The distribution of primary treatment for low, intermediate and high risk tumors among men younger than 60 years, 60 to 65 and 66 to 70 years old is depicted in figure 2. Surveillance was significantly associated with age and risk group, and was selected for 55% of men 66 to 70 years old with a low risk tumor vs 8% of men younger than 60 years with an intermediate risk tumor.
TABLE 2. Deferred treatment in NPCR after termination of surveillance in 2,065 men Surveillance Continued No. pts No. T stage (%): T1a T1b T1c T2 No. Gleason score (%): 2–6 7 8–10 Missing No. WHO grade (%): I II III No. ng/ml PSA (%): 0–4 4–10 10–20 Median ng/ml PSA (25–75 percentile) No. pt age (%): Younger than 60 60–64 65–70 Mean pt age (SD)
1,354
Surveillance Terminated
Surveillance Terminated 711
RP
RT
279
HT*
212
pdiff†
220
⬍0.001
304 76 621 353
(90) (78) (60) (59)
33 21 411 246
(10) (22) (40) (41)
12 8 199 60
(36) (38) (48) (24)
9 2 124 77
(27) (10) (30) (31)
12 11 88 109
(36) (52) (21) (44)
960 67 9 318
(67) (67) (47) (66)
484 33 10 184
(33) (33) (53) (37)
224 6 1 48
(46) (18) (10) (26)
143 15 3 51
(30) (45) (30) (28)
117 12 6 85
(24) (36) (60) (46)
252 54 12
(67) (49) (71)
123 56 5
(33) (51) (29)
38 10
(31) (18)
37 13 1
(30) (23) (20)
48 33 4
(39) (59) (80)
0.63‡
—
349 (81) 671 (63) 334 (59) 6.0 (3.9–9.8)
80 (19) 395 (37) 236 (41) 7.5 (5.0–11.0)
50 (61) 178 (45) 51 (22) 6.1 (4.4–9.0)
22 (28) 120 (30) 70 (30) 7.1 (5.0–11.0)
8 (10) 97 (25) 115 (49) 10.0 (7.2–15.0)
180 324 850 65.0
121 201 389 64.2
77 110 92 62.1
37 64 111 63.9
7 27 186 67.3
(60) (62) (69) (4.5)
(40) (38) (32) (4.7)
(64) (55) (24) (4.5)
(31) (32) (29) (4.7)
(6) (13) (48) (3.1)
* Included monotherapy with per oral antiandrogen blockade in 51 men, castration therapy in 116 and hormonal treatment not specified in 53. † pdiff between continued surveillance vs terminated surveillance. ‡ Gleason score and WHO combined GS 2–6 ⫹ WHO I, II, GS 7, and GS 8–10 ⫹ WHO III.
⬍0.001
⬍0.01
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER
2427
FIG. 4. Continued surveillance and terminated surveillance with start of deferred active treatment in NPCR of Sweden according to risk category and age for 2,065 men younger than 70 years diagnosed between 1997 and 2002 with T1 or T2 tumor, Nx or N0, Mx or M0 and serum PSA less than 20 ng/ml with primary surveillance. Low risk category— clinical local stage T1a, b, c, and Gleason score 2– 6 or WHO I–II and serum PSA less than 10 ng/ml. Intermediate risk category—stage T2, Gleason score 7 or serum PSA greater than 10 ng/ml. High risk category—Gleason score 8 –10, any T stage, any serum PSA. High risk categories for men 65 years or younger were not demonstrated as there were only 9 men in these groups. Only 2 men younger than 60 years in high risk category both received hormonal treatment as deferred treatment. Of 7 men 60 to 65 years old in high risk category 1 received RT, 1 received hormonal treatment and 5 remained on surveillance. Test for trend for surveillance vs active primary treatment according to age in low risk p ⬍0.01 and intermediate risk p ⬍0.12. Trend test for surveillance vs active primary treatment according to risk groups p ⬍0.001 in all 3 age groups in low and intermediate risk categories.
Rotterdam section of the ERSPC, 278 of 1,772 men (16%) with screen detected prostate cancer selected surveillance, whereas in the Gothenburg section of ERSPC in Sweden 270 of 660 men (41%) chose surveillance.20 In a study in the Geneva Cancer Registry, Switzerland 378 of 844 men (45%) with localized prostate cancer in all ages were treated expectantly.10 Finally, 2 large United States studies based on data from the SEER database reported on the frequency in much larger cohorts.3,6 One study reported that 10,868 of 24,405 men (65%) with low risk tumors (defined as Gleason score 2– 4 or Gleason score 2–7 for men older than 70 years) selected no curative treatment.3 Another SEER study on Medicare patients reported that 12,608 of 44,630 men (37%)
aged 65 to 80 years and diagnosed with organ confined, well or intermediately differentiated prostate cancer between 1991 and 1999 selected no curative treatment.6 Importantly, surveillance and androgen deprivation therapy could not be distinguished from one another in the SEER registry. Furthermore, these studies did not specifically report local tumor stage or serum PSA at diagnosis. Thus, there are large differences in the proportion of men who select surveillance as treatment for localized prostate cancer, and the proportion seems to be strongly related to patient age and tumor characteristics as demonstrated in our study. Treatment strategy was also strongly related to geographic region with different treatment policies.
2428
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER
After a median followup of 4 years the proportion of men who had terminated surveillance in our study was 711 of 2,065 (34%). This proportion is similar to that found after similar followup times in the Rotterdam ERSPC cohort in which 19 of 64 (30%) men subsequently received active treatment,11 and in the Gothenburg ERSPC cohort where the number of men converting to active therapy was 104 of 270 (39%).20 The corresponding number in a Canadian single institution series study was 101 of 299 men (35%),8 and in a similar study from the United States 103 of 407 men (25%).9 Two small single institution series have reported that approximately 10% of the patients converted to active therapy after shorter followup time,12,13 and 1 United Kingdom study reported that 238 of 326 men (73%) remained on surveillance after 2 years.15 Active surveillance has recently received increasing attention as a treatment strategy for low and intermediate risk prostate cancer. Our study showed that surveillance was a relatively common treatment option in Sweden in 1997 to 2002 for men younger than 70 years with localized prostate cancer. More recent data from NPCR show that the percentage of men diagnosed with prostate cancer, fulfilling the same inclusion criteria as in our study and opting for surveillance is quite stable in Sweden at 24% in 2003, 21% in 2004, 22% in 2005 and 26% in 2006. Our study shows that there is a large number of men with low and intermediate risk tumors who would be eligible for inclusion in prospective studies on active surveillance that are necessary to elucidate the long-term outcome of this treatment strategy. To the best of our knowledge our study is the first report based on a nationwide, population based study cohort on the number of men with localized prostate cancer put on surveillance, and the number of men who terminated surveillance within the first 4 years and for whom data were available for tumor characteristics as well as primary and deferred treatment. There was an ongoing screening study in Gothenburg during the study period but in the rest of Sweden no formal screening program was in operation. The strengths of this study include the population based design which included men from all types of medical facilities in the entire nation. Approximately 90% of all men in Sweden younger than 70 years with an incident localized prostate cancer in 1997 (1998 for 1 region) to 2002 were included in our study. This assumption is based on the 98% capture rate of the Swedish Cancer Registry of all prostate cancer cases, the 98% capture rate of the NPCR vs the Cancer Registry and the 94% coverage of this followup study within the NPCR. There are some weaknesses in our study design. Active surveillance and watchful waiting were not distinguished as separate entities in the NPCR during the study period and the selection of deferred treatment suggests that the surveillance group was a mix of these 2 strategies. Some men on surveillance underwent deferred curative treatment, whereas some men subsequently received hormonal therapy. There was no information available in our cohort on tumor extent on core biopsies, PSA density or PSA doubling time, and there was no predefined protocol according to which the patients were followed and no criteria for termination of surveillance. We do not have data on how many patients underwent repeat biopsies, nor do we have data on comorbidity among men.
CONCLUSIONS This population based nationwide study showed that 26% of Swedish men younger than 70 years diagnosed with localized prostate cancer in 1997 to 2002 were treated with surveillance. Older men and men with lower risk tumors underwent surveillance significantly more often than younger men and men with higher risk tumors. Two-thirds of men on surveillance remained untreated after a median followup of 4 years. ACKNOWLEDGMENTS NPCR steering group: Pär Stattin, Anders Widmark, Magnus Hellström, Magnus Törnblom, Jan Adolfsson, Anna BillAxelson, Jan-Erik Johanssson, Christer Ahlstrand, Bill Pettersson, Jonas Hugosson, Jan-Erik Damber, Ola Bratt, Göran Ahlgren and Roy Ehrnström.
Abbreviations and Acronyms CaPSURE ⫽ ERSPC ⫽ GS HT NPCR PSA RP RT SEER
⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽
Cancer of the Prostate Strategic Urologic Research Endeavor European Randomized Study of Screening for Prostate Cancer Gleason score hormonal treatment National Prostate Cancer Register prostate specific antigen radical retropubic prostatectomy radiotherapy Surveillance, Epidemiology and End Results
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8. 9.
Adolfsson J, Garmo H, Varenhorst E, Ahlgren G, Ahlstrand C, Andrén O et al: Clinical characteristics and primary treatment of prostate cancer in Sweden between 1996 and 2005. Scand J Urol Nephrol 2007; 41: 456. Barry MJ: Early detection and aggressive treatment of prostate cancer: groping in the dark. J Gen Intern Med 2000; 15: 749. Miller DC, Gruber SB, Hollenbeck BK, Montie JE and Wei JT: Incidence of initial local therapy among men with lowerrisk prostate cancer in the United States. J Natl Cancer Inst 2006; 98: 1134. Parker C: Active surveillance: towards a new paradigm in the management of early prostate cancer. Lancet Oncol 2004; 5: 101. Klotz L: Active surveillance with selective delayed intervention using PSA doubling time for good risk prostate cancer. Eur Urol 2005; 47: 16. Wong YN, Mitra N, Hudes G, Localio R, Schwartz JS, Wan F et al: Survival associated with treatment vs observation of localized prostate cancer in elderly men. JAMA 2006; 296: 2683. Harlan SR, Cooperberg MR, Elkin EP, Lubeck DP, Meng MV, Mehta SS et al: Time trends and characteristics of men choosing watchful waiting for initial treatment of localized prostate cancer: results from CaPSURE. J Urol 2003; 170: 1804. Klotz L: Active surveillance for prostate cancer: for whom? J Clin Oncol 2005; 23: 8165. Carter HB, Kettermann A, Warlick C, Metter EJ, Landis P, Walsh PC et al: Expectant management of prostate cancer
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER with curative intent: an update of the Johns Hopkins experience. J Urol 2007; 178: 2359. 10. Merglen A, Schmidlin F, Fioretta G, Verkooijen HM, Rapiti E, Zanetti R et al: Short- and long-term mortality with localized prostate cancer. Arch Intern Med 2007; 167: 1944. 11. Roemeling S, Roobol MJ, de Vries SH, Wolters T, Gosselaar C, van Leenders GJ et al: Active surveillance for prostate cancers detected in three subsequent rounds of a screening trial: characteristics, PSA doubling times, and outcome. Eur Urol 2007; 51: 1244. 12. Hardie C, Parker C, Norman A, Eeles R, Horwich A, Huddart R et al: Early outcomes of active surveillance for localized prostate cancer. BJU Int 2005; 95: 956. 13. Soloway MS, Soloway CT, Williams S, Ayyathurai R, Kava B and Manoharan M: Active surveillance; a reasonable management alternative for patients with prostate cancer: the Miami experience. BJU Int 2008; 101: 165. 14. Albertsen PC, Hanley JA, Penson DF, Barrows G and Fine J: 13-Year outcomes following treatment for clinically localized prostate cancer in a population based cohort. J Urol 2007; 177: 932. 15. van As NJ, Norman AR, Thomas K, Khoo VS, Thompson A, Huddart RA et al: Predicting the probability of deferred radical treatment for localised prostate cancer managed by active surveillance. Eur Urol 2008; Epub ahead of print. 16. Mattsson B and Wallgren A: Completeness of the Swedish Cancer Register. Non-notified cancer cases recorded on death certificates in 1978. Acta Radiol Oncol 1984; 23: 305. 17. Barlow L, Westergren K, Holmberg L and Talback M: The completeness of the Swedish Cancer Register—a sample survey for year 1998. Acta Oncol 2008; 3: 1. 18. The National Prostate Cancer Register (NPCR) in Sweden 2002–2006. Regional Oncologic Center: Uppsala 2008. Available at http://www.roc.se/prostata/rapport/rapport_ 02_06.pdf. 19. Cooperberg MR, Broering JM, Kantoff PW and Carroll PR: Contemporary trends in low risk prostate cancer: risk assessment and treatment. J Urol 2007; 178: S14. 20. Khatami A, Aus G, Damber JE, Lilja H, Lodding P and Hugosson J: PSA doubling time predicts the outcome after active surveillance in screening-detected prostate cancer: results from the European randomized study of screening for prostate cancer, Sweden section. Int J Cancer 2007; 120: 170.
with selective deferred therapy, and compare outcomes with immediate treatments. We must refine selection criteria for surveillance and standardize trigger points for initiating therapy, etc. Reaching these goals will entail substantial investment in long-term clinical trials. However, implementation of surveillance strategies may depend more on the decision making process than the available information (as evidenced by the fact that Europeans and Americans have the same information now and yet make different management decisions). Observational, population based studies such as the NPCR study and the establishment of risk assessment tools and decision analysis tools may inform efforts to bring surveillance into the mainstream in the United States. Daniel A. Barocas Department of Urologic Surgery Vanderbilt University Medical Center Nashville, Tennessee 1.
Barocas DA, Cowan JE, Smith JA Jr and Carroll PR: What percentage of patients with newly diagnosed carcinoma of the prostate are candidates for surveillance? An analysis of the CaPSURETM database. J Urol 2008; 180: 1330.
This population based study provides insight into the Swedish patterns of care in recent years. Of all men younger than 70 years diagnosed with localized disease and PSA less than 20 ng/ml, surveillance was chosen by as many as 26%. At 4 years the actuarial freedom from treatment for these men was 63% with PSA trends the main indication for intervention. It will be interesting to learn how practice has evolved more recently in the light of the Scandinavian Prostate Cancer Group-4 trial data showing that PSA kinetics in an individual are a poor predictor of the risk of lethal prostate cancer.1 The contrast with practice in the United States is obvious. The Swedish approach provides research opportunities that are not readily available in the United States. Prospective studies in men on surveillance could address the role of repeat biopsies and the impact of tertiary prevention strategies. In addition, with long-term followup the Swedish NPCR will provide important data on the natural history of the disease.
EDITORIAL COMMENTS Stattin et al and the NPCR of Sweden are to be commended on their insightful analysis. The capture rate (98% of solid tumors in the nation of Sweden) of this registry makes it a powerful resource for this type of observational study. One of the most interesting facets of this research is the geographic variation in the management of localized prostate cancer. The authors found that 32% of low and intermediate risk patients chose initial surveillance. In contrast, a recent study of the CaPSURE database in the United States showed that only 6% of low and intermediate risk patients chose surveillance and in a highly selected subset of low risk patients the rate was still just 9%.1 This variation in practice patterns warrants further investigation and studies of this type are an excellent point of departure. There are many obstacles one can posit to the widespread use of surveillance in the United States. We must establish the safety and efficacy of initial surveillance
2429
Chris Parker Institute of Cancer Research Royal Marsden Hospital Sutton, Surrey, United Kingdom 1.
Fall K, Garmo H, Andrén O, Bill-Axelson A, Adolfsson J, Adami HO et al: Prostate-specific antigen levels as a predictor of lethal prostate cancer. J Natl Cancer Inst 2007; 99: 526.
REPLY BY AUTHORS A quarter of men in Sweden with localized prostate cancer diagnosed in 1997 to 2002 did not receive active primary treatment and two-thirds remained on surveillance after a median followup of 4 years. For whom is this an appropriate treatment strategy? Long-term outcomes of this treatment strategy as well as active treatments can in the near future be assessed in this nationwide, population based cohort by linkage to the Swedish Cause of Death Register, which will
2430
SURVEILLANCE AND DEFERRED TREATMENT FOR LOCALIZED PROSTATE CANCER
provide an important benchmark of outcomes obtained in daily clinical practice. We continue a tradition of Scandinavian registries which have changed treatment guidelines for rectal cancer, inguinal hernia and hip replacement.1– 4 1.
2.
Påhlman L, Bohe M, Cedermark B, Dahlberg M, Lindmark G, Sjödahl R et al: The Swedish Rectal Cancer Registry. Br J Surg 2007; 94: 1285. Haapaniemi S and Nordin P: The Swedish Hernia Register. In: Recurrent Hernia: Prevention and Treatment. Edited by V
3.
4.
Schumpelick and RJ Fitzgibbons. Berlin: Springer-Verlag 2007. Kehlet H, Bay-Nielsen M and Danish Hernia Database Collaboration: Nationwide quality improvement of groin hernia repair from the Danish Hernia Database of 87,840 patients from 1998 to 2005. Hernia 2008; 12: 1. Malchau H, Garellick G, Eisler T, Kärrholm J and Herberts P: Presidential guest address: the Swedish Hip Registry: increasing the sensitivity by patient outcome data. Clin Orthop Relat Res 2005; 441: 19.