- Email: [email protected]
Reasonable Delay of Surgical Treatment in Men with Localized Prostate Cancer - Impact on Prognosis?
European Urology
European Urology 47 (2005) 756–760
Reasonable Delay of Surgical Treatment in Men with Localized Prostate Cancer - Impact on Prognosis? Markus Graefen*, Jochen Walz, K.-H.F. Chun, Thorsten Schlomm, Alexander Haese, Hartwig Huland Department of Urology, University Hospital Hamburg-Eppendorf, Martinistr.52, 20246 Hamburg, Germany Accepted 10 February 2005 Available online 14 March 2005
Abstract Purpose: In many centers patients with clinically localized prostate cancer might be confronted with a delay in therapy due to not immediately available treatment capacity at that specific center. Furthermore, a growing amount of patients want to have a second or third opinion before they finally decide what therapeutic option to choose. We investigated whether a reasonable delay from diagnosis to definitive treatment impact recurrence free survival rates in men undergoing radical prostatectomy (RP) for localized prostate cancer. Material and methods: Preoperative data of 795 men treated for localized prostate cancer by RP between 1/1992 and 6/2000 were evaluated including pretreatment PSA, clinical stage and biopsy Gleason score. In addition, time from biopsy to the date of RP was obtained and investigated as a potential prognostic factor. The influence of the time gap between biopsy and surgery was statistically evaluated by univariate Cox regression analyses and KaplanMeier analyses; a multivariate Cox Modell was performed including all preoperative parameters. Relapse following RP was defined as a postoperative PSA level >0.1 ng/ml. Results: Mean followup of the patients was 33 months (1–116 months). Twenty-five percent of the patients failed during that time period. Mean time gap between diagnosis and treatment was 62 days (median 54 days) ranging from 5 to 518 days. Univariate Cox regression analysis showed no significant correlation (p = 0.062) of waiting time with recurrence rate. Multivariate Cox regression documented a highly significant association of PSA (p < 0.001), clinical stage (p = 0.001) and biopsy Gleason grade (p < 0.001) but not not for time to treatment (p = 0.841). In patients with high-grade cancer again no significant impact of treatment delay was found. Conclusion: Treatment delay in the investigated time span of a few months did not adversely affect recurrence free survival rates. Patients can be reassured that they can evalute treatment options without compromising efficacy due to a delay in treatment. # 2005 Elsevier B.V. All rights reserved. Keywords: Prostate cancer; Radical prostatectomy; Treatment delay
1. Introduction A patient diagnosed with localized prostate cancer seeking for therapy might be confronted with a delay in therapy due to limited capacities. Furthermore, in the face of a variety of proven effective therapeutic alternatives a growing number of patients will request a * Corresponding author. Tel.: +040 42803 3443; Fax: +040 42803 6837. E-mail address: [email protected] (M. Graefen).
second or third opinion before they finally decide what option they favour. However, only few data exist investigating to what degree a delay in therapy affects outcome in these men [1,2]. In general, progression of prostate cancer is shown to be slow with a tumor doubling time in early detected cancers of 2 to 4 years [3].Yet, it remains uncertain whether these data can be extrapolated to contemporarily diagnosed patients with prostate cancer as it remains unknown for how long the disease is already prevalent. In addition, a delay in
0302-2838/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2005.02.004
M. Graefen et al. / European Urology 47 (2005) 756–760
therapy might differently effect men harboring more advanced cancers compared to those with early disease. We therefore investigated to what degree the time between diagnosis and treatment of prostate cancer impact prognosis in men treated solely with radical prostatectomy.
757
To investigate the impact of time from diagnosis with prognosis in conjunction with traditional prognosticators a multivariate Cox regression model was calculated. Subgroup analysis were performed for those men with a short waiting time compared to those with a long time form diagnosis to treatment for all patients and for men with high grade cancer (biopsy Gleason score 7). All statistical tests were 2-sided, commercially available software (SPSS11) was used in all analyses.
2. Materials and methods 3. Results Data from 1298 patients who were treated with RP for clinically localized PCa between January 1992 and June 2000 at the University Hospital Hamburg-Eppendorf, Germany, were evaluated. Patients with neoadjuvant hormonal therapy, missing information on clinical stage, pretreatment PSA level, biopsy Gleason sum or date of diagnostic biopsy were excluded, resulting in a cohort of 795 consecutive patients available for statistical evaluation. The distribution of clinical features in our patient is shown in Table 1. In all men, serum was obtained for PSA testing before prostatic manipulation using the monoclonal immunoassay, Immulite DPC. [Diagnostic Products Corporation, Los Angeles, California.] All radical retropubic prostatectomy specimens were prospectively processed according to the Stanford protocol [4]. For clinical and pathological staging we used the second revision of the fourth edition of the TNM classification and for histological grading we used the Gleason system [5,6]. All patients were followed after radical prostatectomy every 6 months for the first 2 years and on a yearly base thereafter. All patients who recurred developed biochemical failure before clinical failure. Biochemical failure was defined as a PSA level of 0.1 ng/ml and rising. None of the patients received adjuvant therapy before proven failure. 2.1. Statistical methods To evaluate the association of preoperative features with biochemical failure univariate Cox regression models were calculated.
Mean follow-up of the patients was 33 months (1– 116 months). Twenty-five percent of the patients failed (defined as a PSA level >0,1 ng/ml) during that time period. Mean time gap between diagnosis and treatment was 62 days (median 54 days) ranging from 5 to 518 days (see Fig. 1). Univariate Cox regression analysis showed no significant correlation (p = 0.062) of waiting time with biochemical free survival. Multivariate Cox regression documented a highly significant association of PSA (p < 0.001), clinical stage (p = 0.001) and biopsy Gleason grade (p < 0.001) but not not for time to treatment (p = 0.841) (see Table 2). In a subgroup analyses comparing men with a waiting time less than 1 month (n = 111) to those who waited more than 4 months (n = 42) the univariate Cox regression model showed a significant correlation (p < 0,0001) between treatment delay and recurrence free survival. However, the rate ratio was below 1 (Rate ratio 0.498, 95% confidence intervall 0.344–0.720) documenting a protective factor of treatment delay. Looking only at the univariate analyses, patients who waited more than 4 months for treatment had only a
Table 1 Clinical and pathological characteristics of the 795 patients operated on between 1/1992 and 6/2000 at the University Hospital Hamburg Characteristics Clinical stage T1a/b T1c T2a T2b T2c T3a PSA (ng/ml) 0–4.0 4.1–10.0 10.1–20.0 >20 Biopsy Gleason sum 5 6 7 8–10
%
n=
0.3 49.7 17.0 21.7 6.5 4.3
2 395 135 172 52 34
9.8 47.6 26.7 14.8
78 378 212 118
5.8 53.8 36.7 3.7
46 428 292 29
Fig. 1. Distribution of delay of surgery (days after diagnostic biopsy) in 795 men undergoing radical prostatectomy.
758
M. Graefen et al. / European Urology 47 (2005) 756–760
Table 2 Univariate and multivariate analyses on association of delay of surgery and recurrence free survival after radical prostectomy
Univariate Cox regression model Days after diagnosis
p=
Rate ratio
95% confidence intervall
0.062
0.996
0.991–1.0
1.0 1.342 1.029 3.914
0.995–1.004 1.186–1.518 1.020–1.038 2.828–5.417
Multivariate Cox regression model Days after diagnosis 0.841 Clinical stage <0.001 Preoperative PSA level <0.001 Biopsy Gleason score <0.001
no significant difference recurrence rate in the KaplanMeier curves between men with early radical prostatectomy (<31 days after diagnosis) compared to those with delayed surgery (>70 days to RP) (Log rank test p = 0.8) (see Fig. 2). Due to a low number of patients who waited for more than 6 months (n = 12) or more than1 year for treatment (n = 4) no meaningful analyses was possible in our patient cohort to determine the effect of long term treatment delay in men with prostate cancer.
4. Discussion 50% likelihood of experiencing recurrence compared those who were treated in between 1 month after diagnosis. This result reflects the policy in our institution where we try to schedule men with more aggressive cancers earlier than those with early diagnosed disease. In 49.6% of patients with a delay of less than 1 month high-grade cancer was found in the diagnostic biopsy compared to 26.2% of those patients who waited for more than 4 months. To control for this bias a multivariate analyses was performed for this subset of patients which showed no independent prognostic effect of time to treatment (p = 0.345) when Gleason grade, pretreatment PSA and clinical stage were considered. In another subgroup analyses of men with highgrade cancer (n = 312) univariate Cox regression analysis showed no significant association between waiting time and progression free survival (p = 0.776). A Kaplan Meier Analysis of men with high grade cancer was performed to compare those individuals with a short vs. those with a long delay of surgery. There was
Fig. 2. Subanalysis of recurrence-free survival (RFS) in patients with aggressive cancer (biopsy Gleason score 7) divided by short (<30 days to surgery) vs. long (>70 days to surgery) delay of radical prostatectomy.
Patients recently diagnosed with localized prostate cancer have a variety of therapeutic options for treatment including radical prostatectomy, brachytherapy and external beam radiation. Therefore, a growing number of patients ask for a second or third opinion before they finally choose to undergo a specific intervention. This implies that a certain delay in time will occur until a treatment decision is finally made. Furthermore, many centers for prostate cancer treatment face a waiting list due to limited capacities leading to an additional delay for treatment. While the prognostic impact of pretreatment parameters on recurrence like biopsy Gleason score, clinical stage or serum PSA level is well established, the delay in treatment in men diagnosed with prostate cancer as a potential prognostic factor is not extensively investigated. The presented study based on data of 795 men could not find an impact of delay of surgery on recurrence free survival following radical prostatectomy in multivariate analyses. As the impact of surgical delay might be different in men harboring a high-grade cancer compared to those harboring a lowgrade cancer, we additionally investigated this subgroup. Again, Kaplan-Meier analyses revealed no significant difference in men diagnosed for high-grade cancer and early treament (<31 days to radical prostatectomy) and those with a a delay in surgery (>70 days). In another subgroup analyses we compared men with a waiting time less than 1 month to those who waited more than 4 months for surgery. The univariate Cox regression model showed a significant correlation (p < 0.0001) between treatment delay and recurrence free survival. However, the rate ratio was below 1 (Rate ratio 0.498, 95% confidence intervall 0.344–0.720) documenting a protecticve factor of treatment delay. Looking only at the univariate analyssis misleading as it suggests that patients who waited more than 4 months for treatment had only a 50% likelihood of experiencing recurrence compared those who were
M. Graefen et al. / European Urology 47 (2005) 756–760
treated in between 1 month after diagnosis. This result reflects the policy in our institution where we try to schedule men with more aggressive cancers earlier than those with early diagnosed disease. However, a multivariate analyses that controlled for this bias of voluntary treament scheduling was performed which showed no independent prognostic effect of time to treatment (p = 0.345) when Gleason grade, pretreatment PSA and clinical stage were considered. These findings are in line with a recent study published by Khan et al. [1] In their series based on 926 men who underwent radical prostatectomy a delay of up to several months from prostate cancer diagnosis did not impact long-term biochemical cancer control rates. Five and 10-year biochemical disease-free survival rates were similar in men operated on after 60 or less days (82% and 78%), 61 to 90 days (86% and 78%), 91 to 120 days (86% and 75%), and 120 to 150 days (86% and 82%) after diagnosis. However, the investigated patients cohort contains a remarkable number of early diagnosed patients reflecting the referal pattern of this specific institution. Therefore, it remained uncertain whether these findings can be extrapolated to centers treating a higher percentage of more advanced cancers like in our center. A preoperative PSA level >20 ng./ml. was found in 14.8% of our patients compared to only 5.9% of the patients presented by Khan et al. representing the difference in pretreatment patient selection. A delay of treatment in early detected cancers has probabely a less prominent effect on prognosis than a delay in aggressive highgrade cancers. It seems therefore important that the impact of treatment delay is investigated in heterogenous patient cohorts before a final statement can be made. However, even in the subgroup analysis of our patients in which we compared only patients with aggressive cancers stratified for early treatment versus delayed surgery no significant impact on prognosis could be found (Log rank test p = 0.8). An explanation might be, that the variance of days patients were awaiting surgery in our institution is fairly small (less than 30 days versus more than 70 days). Neither from our data nor from the literature it is possible to estimate the cut-off of days a patient can await treatment before prognosis is adversely affected due to a delay of treatment. In contrast to our findings Nam et al recently reported on an adverse effect of surgical delay beyond 3 months after surgery [2]. The results of their study based on 645 evaluable men demonstrated that those patients who underwent radical prostatectomy within 3 months of diagnosis experienced a significantly higher
759
10-year biochemical recurrence-free survival than those who underwent surgery 3 months or greater after diagnosis (74.6% vs. 61.3%, p = 0.05). The difference in patient selection and the fact that Nam et al reported on a multiple surgeon series were accounted for the differences of the findings shown by the abovementioned study of Khan et al from Baltimore. However, patient selection and number of surgeons are similar to our series and still the findings of Nam are controversy to our experience. A possible explanation might be the relatively short delay of the vast majority of our patients compared to the comparably long treatment delay patients experienced in the study by Nam et al. There are limitations to our study. The followup period was relatively short with a median followup of almost 3 years. However, several investigators have shown, that the majority of recurrences after radical prostatectomy occur in the first years [7]. Therefore, we are certain, that the impact of surgical delay will not differ when the followup period will be longer. Likewise, the preoperative PSA level, clinical stage, and biopsy Gleason grade in our study were significantly associated with recurrence considering the same followup period. Another limitation might be the clustering of surgical delay around 2 months after diagnosis. Even though the time span between diagnosis and treatment ranges from 5 to 518 days, 56.5% (n = 449) of patients were operated inbetween 40 to 80 days after diagnosis. This implies a limited variance of surgical delay which potentially limits its performance in a multivariate setting. Furthermore, the conclusion drawn from this study can only be, that a reasonable treatment delay of a few months does not adversely effect outcome in men treated with radical prostatectomy. This study can not answere the question regarding the effect of long term treatment delay in such patients. However, a surgical delay clustering around 2 months reflects daily practise and the investigated time span is therefore of relevance for a large number of patients.
5. Conclusions A delay of surgery in the investigated time span did not adversely effect prognosis of men undergoing radical prostatectomy for localized prostate cancer. These findings might reassure patients and their counselling physicians and support the patients need for evaluation of treatment options without compromising their effectiveness factored by treatment delay.
760
M. Graefen et al. / European Urology 47 (2005) 756–760
References [1] Nam RK, Jewett MA, Krahn MD, Robinette MA, Tsihlias J, Toi A, et al. Delay in surgical therapy for clinically localized prostate cancer and biochemical recurrence after radical prostatectomy. Can J Urol 2003;10(3):1891–8. [2] Khan MA, Mangold LA, Epstein JI, Boitnott JK, Walsh PC, Partin AW. Impact of surgical delay on long-term cancer control for clinically localized prostate cancer. J Urol 2004;172(5 Pt 1): 1835–9. [3] Scardino PT. The Gordon Wilson Lecture. Natural history and treatment of earlystage prostate cancer. Trans Am Clin Climatl Assoc 2000;201– 16.
[4] McNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribution of prostaic adenocarcinoma. Correlation with histologic pattern and direction of spread. Am J Surg Pathol 1988;12:897. [5] Beahrs OH, Henson DE, Hutter RV, et al. AJCC Cancer Staging Manual. 4th ed. Piladelphia, PA: Lippincott; 1992, 181–186. [6] Gleason DF. Histologic grading and clinical staging of prostatic carcinoma. In: Tannenbaum M, editor. Urologic pathology: The prostate. Philadelphia: Lea & Ferbiger; 1977. p. 171–97 chapter 9. [7] Hull GW, Rabbani F, Abbas F, Wheeler TM, Kattan MW, Scardino PT. Cancer control with radical prostatectomy alone in 1,000 consecutive patients. J Urol 2002;167(2 Pt 1):528–34.
European Urology 47 (2005) 756–760
Reasonable Delay of Surgical Treatment in Men with Localized Prostate Cancer - Impact on Prognosis? Markus Graefen*, Jochen Walz, K.-H.F. Chun, Thorsten Schlomm, Alexander Haese, Hartwig Huland Department of Urology, University Hospital Hamburg-Eppendorf, Martinistr.52, 20246 Hamburg, Germany Accepted 10 February 2005 Available online 14 March 2005
Abstract Purpose: In many centers patients with clinically localized prostate cancer might be confronted with a delay in therapy due to not immediately available treatment capacity at that specific center. Furthermore, a growing amount of patients want to have a second or third opinion before they finally decide what therapeutic option to choose. We investigated whether a reasonable delay from diagnosis to definitive treatment impact recurrence free survival rates in men undergoing radical prostatectomy (RP) for localized prostate cancer. Material and methods: Preoperative data of 795 men treated for localized prostate cancer by RP between 1/1992 and 6/2000 were evaluated including pretreatment PSA, clinical stage and biopsy Gleason score. In addition, time from biopsy to the date of RP was obtained and investigated as a potential prognostic factor. The influence of the time gap between biopsy and surgery was statistically evaluated by univariate Cox regression analyses and KaplanMeier analyses; a multivariate Cox Modell was performed including all preoperative parameters. Relapse following RP was defined as a postoperative PSA level >0.1 ng/ml. Results: Mean followup of the patients was 33 months (1–116 months). Twenty-five percent of the patients failed during that time period. Mean time gap between diagnosis and treatment was 62 days (median 54 days) ranging from 5 to 518 days. Univariate Cox regression analysis showed no significant correlation (p = 0.062) of waiting time with recurrence rate. Multivariate Cox regression documented a highly significant association of PSA (p < 0.001), clinical stage (p = 0.001) and biopsy Gleason grade (p < 0.001) but not not for time to treatment (p = 0.841). In patients with high-grade cancer again no significant impact of treatment delay was found. Conclusion: Treatment delay in the investigated time span of a few months did not adversely affect recurrence free survival rates. Patients can be reassured that they can evalute treatment options without compromising efficacy due to a delay in treatment. # 2005 Elsevier B.V. All rights reserved. Keywords: Prostate cancer; Radical prostatectomy; Treatment delay
1. Introduction A patient diagnosed with localized prostate cancer seeking for therapy might be confronted with a delay in therapy due to limited capacities. Furthermore, in the face of a variety of proven effective therapeutic alternatives a growing number of patients will request a * Corresponding author. Tel.: +040 42803 3443; Fax: +040 42803 6837. E-mail address: [email protected] (M. Graefen).
second or third opinion before they finally decide what option they favour. However, only few data exist investigating to what degree a delay in therapy affects outcome in these men [1,2]. In general, progression of prostate cancer is shown to be slow with a tumor doubling time in early detected cancers of 2 to 4 years [3].Yet, it remains uncertain whether these data can be extrapolated to contemporarily diagnosed patients with prostate cancer as it remains unknown for how long the disease is already prevalent. In addition, a delay in
0302-2838/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2005.02.004
M. Graefen et al. / European Urology 47 (2005) 756–760
therapy might differently effect men harboring more advanced cancers compared to those with early disease. We therefore investigated to what degree the time between diagnosis and treatment of prostate cancer impact prognosis in men treated solely with radical prostatectomy.
757
To investigate the impact of time from diagnosis with prognosis in conjunction with traditional prognosticators a multivariate Cox regression model was calculated. Subgroup analysis were performed for those men with a short waiting time compared to those with a long time form diagnosis to treatment for all patients and for men with high grade cancer (biopsy Gleason score 7). All statistical tests were 2-sided, commercially available software (SPSS11) was used in all analyses.
2. Materials and methods 3. Results Data from 1298 patients who were treated with RP for clinically localized PCa between January 1992 and June 2000 at the University Hospital Hamburg-Eppendorf, Germany, were evaluated. Patients with neoadjuvant hormonal therapy, missing information on clinical stage, pretreatment PSA level, biopsy Gleason sum or date of diagnostic biopsy were excluded, resulting in a cohort of 795 consecutive patients available for statistical evaluation. The distribution of clinical features in our patient is shown in Table 1. In all men, serum was obtained for PSA testing before prostatic manipulation using the monoclonal immunoassay, Immulite DPC. [Diagnostic Products Corporation, Los Angeles, California.] All radical retropubic prostatectomy specimens were prospectively processed according to the Stanford protocol [4]. For clinical and pathological staging we used the second revision of the fourth edition of the TNM classification and for histological grading we used the Gleason system [5,6]. All patients were followed after radical prostatectomy every 6 months for the first 2 years and on a yearly base thereafter. All patients who recurred developed biochemical failure before clinical failure. Biochemical failure was defined as a PSA level of 0.1 ng/ml and rising. None of the patients received adjuvant therapy before proven failure. 2.1. Statistical methods To evaluate the association of preoperative features with biochemical failure univariate Cox regression models were calculated.
Mean follow-up of the patients was 33 months (1– 116 months). Twenty-five percent of the patients failed (defined as a PSA level >0,1 ng/ml) during that time period. Mean time gap between diagnosis and treatment was 62 days (median 54 days) ranging from 5 to 518 days (see Fig. 1). Univariate Cox regression analysis showed no significant correlation (p = 0.062) of waiting time with biochemical free survival. Multivariate Cox regression documented a highly significant association of PSA (p < 0.001), clinical stage (p = 0.001) and biopsy Gleason grade (p < 0.001) but not not for time to treatment (p = 0.841) (see Table 2). In a subgroup analyses comparing men with a waiting time less than 1 month (n = 111) to those who waited more than 4 months (n = 42) the univariate Cox regression model showed a significant correlation (p < 0,0001) between treatment delay and recurrence free survival. However, the rate ratio was below 1 (Rate ratio 0.498, 95% confidence intervall 0.344–0.720) documenting a protective factor of treatment delay. Looking only at the univariate analyses, patients who waited more than 4 months for treatment had only a
Table 1 Clinical and pathological characteristics of the 795 patients operated on between 1/1992 and 6/2000 at the University Hospital Hamburg Characteristics Clinical stage T1a/b T1c T2a T2b T2c T3a PSA (ng/ml) 0–4.0 4.1–10.0 10.1–20.0 >20 Biopsy Gleason sum 5 6 7 8–10
%
n=
0.3 49.7 17.0 21.7 6.5 4.3
2 395 135 172 52 34
9.8 47.6 26.7 14.8
78 378 212 118
5.8 53.8 36.7 3.7
46 428 292 29
Fig. 1. Distribution of delay of surgery (days after diagnostic biopsy) in 795 men undergoing radical prostatectomy.
758
M. Graefen et al. / European Urology 47 (2005) 756–760
Table 2 Univariate and multivariate analyses on association of delay of surgery and recurrence free survival after radical prostectomy
Univariate Cox regression model Days after diagnosis
p=
Rate ratio
95% confidence intervall
0.062
0.996
0.991–1.0
1.0 1.342 1.029 3.914
0.995–1.004 1.186–1.518 1.020–1.038 2.828–5.417
Multivariate Cox regression model Days after diagnosis 0.841 Clinical stage <0.001 Preoperative PSA level <0.001 Biopsy Gleason score <0.001
no significant difference recurrence rate in the KaplanMeier curves between men with early radical prostatectomy (<31 days after diagnosis) compared to those with delayed surgery (>70 days to RP) (Log rank test p = 0.8) (see Fig. 2). Due to a low number of patients who waited for more than 6 months (n = 12) or more than1 year for treatment (n = 4) no meaningful analyses was possible in our patient cohort to determine the effect of long term treatment delay in men with prostate cancer.
4. Discussion 50% likelihood of experiencing recurrence compared those who were treated in between 1 month after diagnosis. This result reflects the policy in our institution where we try to schedule men with more aggressive cancers earlier than those with early diagnosed disease. In 49.6% of patients with a delay of less than 1 month high-grade cancer was found in the diagnostic biopsy compared to 26.2% of those patients who waited for more than 4 months. To control for this bias a multivariate analyses was performed for this subset of patients which showed no independent prognostic effect of time to treatment (p = 0.345) when Gleason grade, pretreatment PSA and clinical stage were considered. In another subgroup analyses of men with highgrade cancer (n = 312) univariate Cox regression analysis showed no significant association between waiting time and progression free survival (p = 0.776). A Kaplan Meier Analysis of men with high grade cancer was performed to compare those individuals with a short vs. those with a long delay of surgery. There was
Fig. 2. Subanalysis of recurrence-free survival (RFS) in patients with aggressive cancer (biopsy Gleason score 7) divided by short (<30 days to surgery) vs. long (>70 days to surgery) delay of radical prostatectomy.
Patients recently diagnosed with localized prostate cancer have a variety of therapeutic options for treatment including radical prostatectomy, brachytherapy and external beam radiation. Therefore, a growing number of patients ask for a second or third opinion before they finally choose to undergo a specific intervention. This implies that a certain delay in time will occur until a treatment decision is finally made. Furthermore, many centers for prostate cancer treatment face a waiting list due to limited capacities leading to an additional delay for treatment. While the prognostic impact of pretreatment parameters on recurrence like biopsy Gleason score, clinical stage or serum PSA level is well established, the delay in treatment in men diagnosed with prostate cancer as a potential prognostic factor is not extensively investigated. The presented study based on data of 795 men could not find an impact of delay of surgery on recurrence free survival following radical prostatectomy in multivariate analyses. As the impact of surgical delay might be different in men harboring a high-grade cancer compared to those harboring a lowgrade cancer, we additionally investigated this subgroup. Again, Kaplan-Meier analyses revealed no significant difference in men diagnosed for high-grade cancer and early treament (<31 days to radical prostatectomy) and those with a a delay in surgery (>70 days). In another subgroup analyses we compared men with a waiting time less than 1 month to those who waited more than 4 months for surgery. The univariate Cox regression model showed a significant correlation (p < 0.0001) between treatment delay and recurrence free survival. However, the rate ratio was below 1 (Rate ratio 0.498, 95% confidence intervall 0.344–0.720) documenting a protecticve factor of treatment delay. Looking only at the univariate analyssis misleading as it suggests that patients who waited more than 4 months for treatment had only a 50% likelihood of experiencing recurrence compared those who were
M. Graefen et al. / European Urology 47 (2005) 756–760
treated in between 1 month after diagnosis. This result reflects the policy in our institution where we try to schedule men with more aggressive cancers earlier than those with early diagnosed disease. However, a multivariate analyses that controlled for this bias of voluntary treament scheduling was performed which showed no independent prognostic effect of time to treatment (p = 0.345) when Gleason grade, pretreatment PSA and clinical stage were considered. These findings are in line with a recent study published by Khan et al. [1] In their series based on 926 men who underwent radical prostatectomy a delay of up to several months from prostate cancer diagnosis did not impact long-term biochemical cancer control rates. Five and 10-year biochemical disease-free survival rates were similar in men operated on after 60 or less days (82% and 78%), 61 to 90 days (86% and 78%), 91 to 120 days (86% and 75%), and 120 to 150 days (86% and 82%) after diagnosis. However, the investigated patients cohort contains a remarkable number of early diagnosed patients reflecting the referal pattern of this specific institution. Therefore, it remained uncertain whether these findings can be extrapolated to centers treating a higher percentage of more advanced cancers like in our center. A preoperative PSA level >20 ng./ml. was found in 14.8% of our patients compared to only 5.9% of the patients presented by Khan et al. representing the difference in pretreatment patient selection. A delay of treatment in early detected cancers has probabely a less prominent effect on prognosis than a delay in aggressive highgrade cancers. It seems therefore important that the impact of treatment delay is investigated in heterogenous patient cohorts before a final statement can be made. However, even in the subgroup analysis of our patients in which we compared only patients with aggressive cancers stratified for early treatment versus delayed surgery no significant impact on prognosis could be found (Log rank test p = 0.8). An explanation might be, that the variance of days patients were awaiting surgery in our institution is fairly small (less than 30 days versus more than 70 days). Neither from our data nor from the literature it is possible to estimate the cut-off of days a patient can await treatment before prognosis is adversely affected due to a delay of treatment. In contrast to our findings Nam et al recently reported on an adverse effect of surgical delay beyond 3 months after surgery [2]. The results of their study based on 645 evaluable men demonstrated that those patients who underwent radical prostatectomy within 3 months of diagnosis experienced a significantly higher
759
10-year biochemical recurrence-free survival than those who underwent surgery 3 months or greater after diagnosis (74.6% vs. 61.3%, p = 0.05). The difference in patient selection and the fact that Nam et al reported on a multiple surgeon series were accounted for the differences of the findings shown by the abovementioned study of Khan et al from Baltimore. However, patient selection and number of surgeons are similar to our series and still the findings of Nam are controversy to our experience. A possible explanation might be the relatively short delay of the vast majority of our patients compared to the comparably long treatment delay patients experienced in the study by Nam et al. There are limitations to our study. The followup period was relatively short with a median followup of almost 3 years. However, several investigators have shown, that the majority of recurrences after radical prostatectomy occur in the first years [7]. Therefore, we are certain, that the impact of surgical delay will not differ when the followup period will be longer. Likewise, the preoperative PSA level, clinical stage, and biopsy Gleason grade in our study were significantly associated with recurrence considering the same followup period. Another limitation might be the clustering of surgical delay around 2 months after diagnosis. Even though the time span between diagnosis and treatment ranges from 5 to 518 days, 56.5% (n = 449) of patients were operated inbetween 40 to 80 days after diagnosis. This implies a limited variance of surgical delay which potentially limits its performance in a multivariate setting. Furthermore, the conclusion drawn from this study can only be, that a reasonable treatment delay of a few months does not adversely effect outcome in men treated with radical prostatectomy. This study can not answere the question regarding the effect of long term treatment delay in such patients. However, a surgical delay clustering around 2 months reflects daily practise and the investigated time span is therefore of relevance for a large number of patients.
5. Conclusions A delay of surgery in the investigated time span did not adversely effect prognosis of men undergoing radical prostatectomy for localized prostate cancer. These findings might reassure patients and their counselling physicians and support the patients need for evaluation of treatment options without compromising their effectiveness factored by treatment delay.
760
M. Graefen et al. / European Urology 47 (2005) 756–760
References [1] Nam RK, Jewett MA, Krahn MD, Robinette MA, Tsihlias J, Toi A, et al. Delay in surgical therapy for clinically localized prostate cancer and biochemical recurrence after radical prostatectomy. Can J Urol 2003;10(3):1891–8. [2] Khan MA, Mangold LA, Epstein JI, Boitnott JK, Walsh PC, Partin AW. Impact of surgical delay on long-term cancer control for clinically localized prostate cancer. J Urol 2004;172(5 Pt 1): 1835–9. [3] Scardino PT. The Gordon Wilson Lecture. Natural history and treatment of earlystage prostate cancer. Trans Am Clin Climatl Assoc 2000;201– 16.
[4] McNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribution of prostaic adenocarcinoma. Correlation with histologic pattern and direction of spread. Am J Surg Pathol 1988;12:897. [5] Beahrs OH, Henson DE, Hutter RV, et al. AJCC Cancer Staging Manual. 4th ed. Piladelphia, PA: Lippincott; 1992, 181–186. [6] Gleason DF. Histologic grading and clinical staging of prostatic carcinoma. In: Tannenbaum M, editor. Urologic pathology: The prostate. Philadelphia: Lea & Ferbiger; 1977. p. 171–97 chapter 9. [7] Hull GW, Rabbani F, Abbas F, Wheeler TM, Kattan MW, Scardino PT. Cancer control with radical prostatectomy alone in 1,000 consecutive patients. J Urol 2002;167(2 Pt 1):528–34.