BJUI. Effect of delaying surgery on radical prostatectomy outcomes: a contemporary analysis

BJUI BJU INTERNATIONAL Effect of delaying surgery on radical prostatectomy outcomes: a contemporary analysis Ruslan Korets, Catherine M. Seager, Max S. Pitman, Gregory W. Hruby, Mitchell C. Benson and James M. McKiernan Department of Urology, Columbia University College of Physicians and Surgeons, New York, NY, USA Accepted for publication 27 July 2011 Study Type Therapy (case series) Level of Evidence 4 OBJECTIVE To examine the effect of time from last positive biopsy to surgery on clinical outcomes in men with localized prostate cancer undergoing radical prostatectomy (RP). PATIENTS AND METHODS We conducted a retrospective review of 2739 men who underwent RP between 1990 and 2009 at our institution. Clinical and pathological features were compared between men undergoing RP 60, 61 90 and > 90 days from the time of prostate biopsy. A Cox proportional hazards model was used to analyse the association between clinical features and surgical delay with biochemical progression. Biochemical recurrence (BCR)-free rates were assessed using the Kaplan Meier method. What s known on the subject? and What does the study add? For patients electing surgical treatment, the question of the effect of surgical delay on clinical outcomes in prostate cancer is controversial. In this study we examined the effect of delay from diagnosis to surgery on outcomes in men with localized prostate cancer and found no association between time to surgery and risk of biochemical recurrence, even for patients with longer delays and high-risk disease. Men with localized prostate cancer can be reassured that reasonable delays in treatment will not influence disease outcomes. RESULTS Of the 1568 men meeting the inclusion criteria, 1098 (70%), 303 (19.3%) and 167 (10.7%) had a delay of 60, 61 90 and >90 days, respectively, between biopsy and RP. A delay of > 60 days was not associated with adverse pathological findings at surgery. The 5-year survival rate was similar among the three groups (78 85%, P = 0.11). In a multivariate Cox model, men with higher PSA levels, clinical stages, Gleason sums, and those of African-American race were all at higher risk for developing BCR. A delay to surgery of > 60 days was not associated with worse biochemical outcomes in a univariate and multivariate model. CONCLUSIONS A delay of > 60 days is not associated with adverse pathological outcomes in men with localized prostate cancer, nor does it correlate with worse BCR-free survival. Patients can be assured that delaying treatment while considering therapeutic options will not adversely affect their outcomes. KEYWORDS time factors, prostatic neoplasm/surgery, prostatic neoplasm/pathology, disease progression INTRODUCTION Patients diagnosed with prostate cancer must navigate an increasingly complicated web of decisions about treatment, including if and when radical prostatectomy (RP) is appropriate. While patient anxiety may often lead to a short interval between diagnosis and surgery, other logistic factors such as health insurance coverage, medical or personal problems, and surgeon or operating room availability may also cause delays in surgery of variable length [1 ]. As localized prostate cancer is generally accepted to have an indolent course, such delays tend to be tolerated by physicians and patients alike; however, the question of whether treatment delays negatively impact patient outcomes is controversial, and the oncology literature varies with respect to different cancers. Delaying cystectomy is well known to adversely impact survival among patients with bladder cancer [2 ] and early treatment may improve survival for head and neck and lung cancers as well [3,4 ]. Treatment delays do not appear to affect survival for either breast or colon cancer [5,6 ], but the current literature provides conflicting information regarding delayed RP for prostate cancer [1,7 11 ]. Pretreatment prognostic factors including serum PSA, Gleason score, AJCC staging, and 2011 BJU INTERNATIONAL 110, 211 216 doi:10.1111/j.1464-410x.2011.10666.x 211

KORETS ET AL. preoperative risk models have been used to predict response to therapeutic intervention and patient outcomes. It is therefore not surprising that these same clinical markers have also been used by physicians, less formally, to help determine the urgency of treatment. In the present study, we examine the effect of delay from diagnosis to surgery on outcomes in men with localized prostate cancer and analyse the impact of the delay across well described patient risk categories. PATIENTS AND METHODS STUDY POPULATION Using the Institutional Review Boardapproved Columbia University Urologic Oncology Database, we conducted a retrospective review of all patients treated with RP by five high-volume surgeons between 1990 and 2010. This query identified 2739 patients. Time to surgery (TTS) was defined as the period between the most recent positive prostate biopsy and RP. Of these patients, we excluded patients who had received neoadjuvant androgen deprivation ( n = 41), adjuvant ( n = 59) androgen deprivation or immediate adjuvant external beam radiation ( n = 45), who had unknown biopsy dates ( n = 615), and who had < 12 months of follow-up after RP ( n = 411). Records of patients with surgical delays of > 180 days were reviewed to ensure that none of the patients included in the analysis had been placed on active surveillance with delayed curative intervention protocols. Biochemical recurrence (BCR) was defined as a single serum PSA level of 0.2 ng/ml at least 21 days after RP [12 ]. STATISTICAL ANALYSIS Men were divided into three groups based on surgical delay that has been previously reported in the literature: Group 1 ( 60 days), Group 2 (61 90 days), Group 3 ( 91 days) [8,10 ]. The clinicopathological characteristics were compared among the groups using ANOVA for continuous variables and the chi-squared for categorical variables. Age at diagnosis and logarithmically transformed preoperative PSA levels were examined as continuous variables. Patient race/ethnicity (Caucasian vs. African-American vs. Hispanic vs. other), year of surgery (1990 1994 vs. 1995 1999 TABLE 1 Clinical and pathological characteristics Group 1: 60 days Group 2: 61 90 days Group 3: >90 daysp No. of patients (%) 1098 (70) 303 (19.3) 167 (10.7) Race/Ethnicity, n (%) 0.284 White 767 (69.9) 228 (75.3) 124 (74.3) Black 118 (10.7) 21 (6.9) 14 (8.4) Hispanic 87 (7.9) 22 (7.3) 8 (4.8) Other 126 (11.5) 32 (10.5) 21 (12.5) Median age (IQR) 60 (55, 66) 61 (56, 66) 61 (57, 66) 0.164 Year of surgery, n (%) 0.004 1990 1994 63 (5.8) 37 (12.2) 23 (13.8) 1995 1999 317 (28.9) 109 (36.0) 44 (26.4) 2000 2004 354 (32.2) 79 (26.0) 53 (31.7) 2005 2009 364 (33.1) 78 (25.8) 47 (28.1) Mean preoperative PSA (IQR) 5.7 (4.3, 8.0) 6.2 (4.7, 8.7) 6.2 (4.5, 8.2) 0.073 Clinical stage, n (%) 0.107 ct1 680 (61.9) 186 (61.4) 112 (67.1) ct2 410 (37.4) 112 (36.9) 55 (32.9) ct3 8 (0.7) 5 (1.7) 0 (0) Biopsy Gleason sum, n (%) 0.012 < 7 544 (50.0) 159 (53.0) 101 (61.2) 7 412 (37.8) 112 (37.3) 57 (34.6) > 7 133 (12.2) 29 (9.7) 7 (4.2) D Amico Risk-Group, n (%) 0.005 Low 351 (34.1) 109 (37.1) 73 (47.7) Intermediate 554 (53.8) 158 (53.7) 72 (47.1) High 125 (12.1) 27 (9.2) 8 (5.2) Surgical approach 0.024 Open 884 (80.5) 264 (87.1) 140 (83.8) Robot-assisted 214 (19.5) 39 (12.9) 27 (16.2) Median follow-up (IQR) 57 (27, 86) 65 (38, 108) 64 (37, 99) vs. 2000 2004 vs. 2005 2009), clinical stage (T1 vs. T2 vs. T3), Gleason sum ( < 7 vs. = 7 vs. >7), D Amico risk group [13 ] and surgical approach were examined as categorical variables. The odds ratios (ORs) of adverse pathological findings at surgery (Gleason sum upgrade, positive surgical margins, extracapsular extension, seminal vesicle invasion and positive node status) were calculated for all TTS groups using logistical regression. Linear regression modelling was used to determine the effect of TTS as a continuous variable on BCR. Cox proportional hazards model was used to analyse the effect of age, race/ethnicity, PSA, clinical stage, biopsy Gleason sum, year of surgery, surgical approach and TTS on risk of BCR. BCR-free survival was compared across the TTS groups using the Kaplan Meier method and the log-rank test. For all tests, two-sided P values < 0.05 were considered to indicate statistical significance. All analyses were conducted using S TATA version 11.0 (StataCorp, College Station, TX, USA). RESULTS In all, 1568 patients met the inclusion criteria for the study. The median age of patients at the time of diagnostic biopsy was 60 years (interquartile range [IQR ] : 56, 66). The median (range) time to RP among the entire study cohort was 45 (36 847) days (IQR: 42, 65). A total of 1098 patients (70.0%) had surgery within 60 days, 303 patients (19.3%) waited 61 90 days and 167 patients (10.7%) delayed surgery for >90 days. Forty-one patients (2.6%) experienced a delay of >180 days. Table 1 shows the clinical and pathological characteristics of the three study groups, stratified by TTS. Men with lower biopsy Gleason grades and low-risk D Amico stratification experienced 212 2011 BJU INTERNATIONAL

DELAYED SURGERY AND RADICAL PROSTATECTOMY OUTCOMES TABLE 2 Odds ratio of adverse pathological findings at surgery 60 days 61 90 days (95% CI) > 90 days (95% CI) P Gleason sum upgrade 1.00 (ref) 1.11 (0.83 1.48) 1.08 (0.78 1.44) 0.48, 0.96 Positive surgical margin 1.00 (ref) 1.13 (0.85 1.51) 1.06 (0.66 1.41) 0.38, 0.86 Extracapsular extension 1.00 (ref) 1.03 (0.78 1.35) 0.95 (0.69 1.04) 0.84, 0.07 Seminal vesicle invasion 1.00 (ref) 0.91 (0.69 1.20) 0.91 (0.62 1.11) 0.49, 0.45 Positive lymph nodes 1.00 (ref) 0.87 (0.56 1.33) 0.99 (0.59 1.68) 0.53, 0.68 TABLE 3 Cox proportional hazards model of risk of BCR Univariate Multivariate HR CI P HR CI P Age 1.03 1.01 1.06 0.002 1.02 1.00 1.04 0.03 Race/Ethnicity White ref ref Black 2.24 1.57 3.21 0.001 1.93 1.31 2.83 0.001 Hispanic 1.92 1.19 3.11 0.007 1.59 0.97 2.63 0.07 Other 1.61 1.12 2.30 0.009 1.3 0.89 1.90 0.17 Preoperative PSA 1.95 1.61 2.36 <0.001 1.53 1.26 1.85 <0.001 Clinical stage ct1 ref ref ct2 1.2 1.18 2.20 0.01 1.41 1.01 1.93 0.04 ct3 2.59 1.03 6.34 0.04 3.59 1.37 7.37 0.009 Biopsy Gleason < 7 ref ref 7 2.2 1.65 2.94 <0.001 1.95 1.45 2.62 <0.001 > 7 5.3 3.89 7.36 <0.001 4.42 3.15 6.21 <0.001 Year of surgery 1990 1994 ref 1995 1999 0.75 0.51 1.09 0.13 2000 2004 0.72 0.48 1.08 0.11 2005 2009 1.39 0.92 2.1 0.12 D Amico Risk-Group Low ref ref Intermediate 2.78 2.00 3.84 <0.001 1.89 1.35 2.78 <0.001 High 4.71 3.17 7.00 <0.001 1.93 1.18 3.01 0.007 Minimally invasive surgery No ref Yes 1.47 0.91 2.35 0.11 TTS Group 60 days ref ref 61 90 days 1.19 0.89 1.58 0.24 1.26 0.94 1.70 0.12 >90 days 1.27 0.75 1.4 0.13 1.13 0.73 1.31 0.43 longer delays than men from higher-risk groups or those with more advanced Gleason sum findings. Compared with men treated in the 1990s, men treated during the last 10 years of our study period were less likely to experience delays > 60 days ( P = 0.004). Furthermore, men electing to have robot-assisted prostatectomy experienced shorter delays to surgery than men undergoing open RP ( P = 0.02). The median follow-up time after surgery was 64 months (IQR: 30, 93). During this time 275/1568 (17.5%) men experienced BCR. The ORs of adverse findings at prostatectomy across the TTS groups are shown in Table 2. No significant association was observed between TTS and adverse pathological findings including: Gleason sum upgrade, positive surgical margins, extracapsular extension, seminal vesicle invasion and positive lymph nodes. To examine if treatment delay led to a higher risk of disease progression, we used a linear regression model to analyse TTS as a continuous variable. In this model, TTS was not associated with BCR (hazard ratio [ HR ] : 1.02, 95% CI: 0.98 1.04; P = 0.29). On univariate analysis older age, non-caucasian race, higher preoperative PSA level, advanced clinical stage and biopsy Gleason grade were all associated with increased risk of BCR (Table 3 ). A significant association was not found between year of surgery, surgical approach or TTS group and BCR. In a multivariate model comprised of all variables that showed a significant association with BCR on univariate analysis, older age (HR: 1.02), higher PSA (HR: 1.53), intermediate and high D Amico risk groups (HR: 1.89, 1.93), clinical stage >ct1 (HR: 1.41, 3.59), biopsy Gleason sum >6 (HR: 1.95, 4.42), and African-American race (HR: 1.93) were all significantly associated with BCR. Delay to prostatectomy of >60 days was not associated with higher rates of clinical progression. When examining only a group of men at high risk of BCR by D Amico criteria, no differences in recurrence rates were observed between the three TTS groups according to the log-rank test ( P = 0.479). Kaplan Meier analysis showed a 5-year BCR-free survival of 80% in men who underwent RP < 60 days after last positive biopsy. The results were similar for men who underwent RP within 61 90 days, and >90 days of last positive biopsy (78% and 85%, respectively; P = 0.11 [Fig. 1 ] ). DISCUSSION Patients initially diagnosed with prostate cancer face an ever-growing number of decisions about how best to proceed with their treatment. In this setting, logistical factors such as insurance coverage, personal conflicts and surgeon and operating room availability, as well as patients desire to explore multiple treatment options, are likely to cause delays in therapy. Choosing an appropriate treatment for prostate cancer has become even more complicated in recent years, as the dramatic increase in prevalence of PSA screening has led to a 2011 BJU INTERNATIONAL 213

KORETS ET AL. stage migration toward lower-risk disease among newly diagnosed patients [14 ]. This dynamic has allowed many patients to elect to have delayed treatment or active surveillance protocols in an effort to prevent overtreatment and the morbidities associated with aggressive therapeutic interventions [15,16 ]. Nevertheless, the characteristics of early lesions are not well understood, and the safety of active surveillance protocols has not been completely assured. The optimum management for these early-stage lesions of uncertain biological potential, therefore, remains controversial. Biochemical Recurrence-Free Survival 1.00 0.75 0.50 0.25 0.00 0 50 Time, months Log-rank: P = 0.11 100 <60 days 61 90 days >90 days FIG. 1. Kaplan Meier estimates of BCR rates, stratified by time from last positive biopsy to surgery. Given this uncertainty, attempting to understand the prognostic significance of delays in surgical treatment becomes essential to providing patients with the best chance of a cure without unnecessarily compromising quality of life. The benefits of early definitive treatment for prostate cancer have been suggested by reports of improved stage- and grade-specific progression-free survival rates after RP in the PSA era [ 17 ] ; however, the impact of delayed time to all treatment varies in the oncology literature. Early treatment appears to confer a survival advantage for bladder, head and neck and lung cancers [2 4 ], but delays in treatment are not associated with a decline in overall survival rates for either breast or colon cancer [5,6 ]. Meanwhile, the effect of delaying radiation therapy has been shown to affect outcomes in prostate cancer patients with high-risk but not intermediate- or low-risk disease [18 ]. For patients choosing surgical treatment, the question of the effect of surgical delay on clinical outcomes in prostate cancer remains uncertain, with studies to date showing inconsistent results [1,7 11 ]. Based on a study population of 645 Canadian patients, Nam et al. [7 ] found an insignificant increase in risk of BCR among patients who had a delay of >3 months between diagnostic biopsy and RP (OR 1.58, P = 0.09). The authors also found improved 10-year recurrence-free survival rates for patients who underwent early surgery vs those who had delayed intervention (74.6% vs 61.3%, P = 0.05). Similarly, Freedland et al. [8 ] explored the clinical outcomes of 895 patients in a low-risk group (PSA < 10 and Gleason score 6). While the authors found no association between surgical delay and negative pathological outcomes (high grade disease, positive surgical margins or extraprostatic extension), longer delays were associated with an increased risk of BCR (log rank P = 0.01). Controlling for other risk factors, surgical delay was associated with a significantly higher risk of BCR (RR 2.73, P = 0.002) for patients who waited >180 days before surgery. There have also been a number of studies that detected no such relationship between delayed surgical intervention and prostate cancer outcomes. Van den Bergh et al. [1 ] examined 227 men with low-risk disease who underwent either immediate RP (mean [SD ] time to RP 0.5 [ 0.2 ] years) or expectant management followed by RP (mean [ SD ] time to RP 2.6 [ 2.0 ] years), and found no significant difference in risk of negative pathological features (extracapsular extension, positive margins or tumour size) between the two groups. The authors did find a decreased rate of BCR among the immediate RP cohort (9% at 10 years, vs. 35% for delayed RP), but these results were not significant (log-rank P = 0.689). In an earlier series, Vickers et al. [19 ] reported on the effect of treatment delay on BCR for 3149 patients with clinically localized prostate cancer treated within 1 year of diagnosis [19 ]. In a multivariate model controlling for stage, Gleason score and PSA, the authors found that a delay of 1 year had no significant effect on risk of BCR at 3, 5, 8 or 10 years of follow-up ( P > 0.2 for all). Boorjian et al. [10 ] also reported on this same series of 3149 patients, and found that time to RP, treated as either a continuous or categorical variable (divided at 90 days), had no significant effect on risk of subsequent BCR. Notably, time to RP was still not an independent predictor of BCR after stratifying by preoperative risk, such that delays of > 90 days did not adversely affect outcome for patients with a nomogrampredicted risk of BCR of 40% ( P = 0.548) [9 ]. In two other large series, Khan et al. [ 10 ] showed no association between time to RP and risk of BCR for surgical delays of 151 days, and Graefen et al. [11 ] also reported no association between time to surgery (continuous variable) and BCR, even for patients with high-grade disease. In the present study, we found no association between TTS and risk of BCR, even for patients with longer delays or higher-risk disease. These data suggest that nearly all patients with localized prostate cancer can be reassured that wait times of reasonable length will not affect their chances of cure after definitive treatment. Furthermore, our results support the continued use of deferred treatment or active surveillance protocols for appropriately selected patients with low-risk disease [20 ]. While our findings are consistent with a number of previous studies [1,9 11 ], the reason for variation among reports in the literature is not immediately clear. Common thinking would suggest that the indolent nature of progression for localized prostate cancer is such that surgical delays which may seem quite long (up to 150 days in some reports [10 ] ) are actually relatively short compared with the disease course, and thus do not allow for clinical progression to occur before definitive treatment. 214 2011 BJU INTERNATIONAL

DELAYED SURGERY AND RADICAL PROSTATECTOMY OUTCOMES Unfortunately this logic cannot explain why some studies have reported worse outcomes among patients who delayed treatment, particularly since these studies [7,8 ] used similar time cut-off points (for defining early vs. delayed treatment) to studies that found no association between surgical delay and adverse outcomes [1,9 11 ]. Although the studies have attempted to control for potential confounders such as age, clinical or tumour stage and Gleason score between groups, all reports to date have been retrospective, so it is possible that unmeasurable factors affecting time to RP, such as surgeons clinical judgment, may have biased the results of these studies. As such, further study in prospective trials would be needed to assess more definitively the effects of treatment delay. It is well known, however, that some patients with clinically localized prostate cancer, particularly those with intermediateor high-risk disease, will eventually progress to metastatic disease if left untreated [21 ]. We must assume, therefore, that for these patients, there is a point after which it is no longer safe to delay definitive therapy. This reasoning is particularly relevant to African-American men, who are known to be at greater risk of dying from prostate cancer [22 ]. An interesting finding in the present study is that, on multivariate analysis, African-American race continued to be an independent predictor of BCR while controlling for all other covariates. This finding is consistent with earlier results from our centre, which showed that African-American men are significantly more likely than Caucasian or Hispanic men to have an adverse change in their nomogrampredicted risk of BCR after RP [23 ]. These results suggest that apart from standard measures of preoperative risk such as clinical stage, Gleason score, and PSA, there may be inherent differences in tumour biology that portend worse outcomes for African-American patients. The biology of localized prostate cancer also presents a potential research area that would help to address questions of the effect of treatment delay. Early tumour biology is not completely understood, but patients with longer TTS may have higher rates of micrometastases as a result of delays in the removal of the primary tumour. One study detected tumour cells in bone marrow using reverse transcriptase polymerase chain reaction and immunocytochemistry in 56% of patients with organ-confined disease [24 ]. Micrometastases may be a result of, or accompanied by, molecular changes that are, at least theoretically, more likely to accumulate with delay in removal of the primary lesion. Although they are still being defined, multiple gene alterations have been associated with both pathological features of high-risk prostate cancer and also with disease-specific mortality [25,26 ]. The present study was not designed to detect either micrometastases or molecular changes caused by delays in definitive surgical treatment, but further research may help to incorporate these factors into the clinical management of men with prostate cancer. The present study has several limitations that warrant a brief discussion. First, as it is a retrospective study, the findings could be confounded by selection bias, as well as by variables that were not included in the analysis. Additionally, as our institution is a tertiary academic centre, a significant number of patients that are treated have had their biopsies performed elsewhere, with the slides re-reviewed by our institution s uropathologist. The exact date of the biopsy could not be verified in 611 (22%) patients, and as a result they were excluded from the analysis. Given that detection of death from prostate cancer requires long follow-up times, this study used a number of surrogate endpoints, including pathological findings, a post-rp nomogram and BCR, to predict negative clinical outcomes associated with disease. These have all been previously validated as surrogate outcomes in prostate cancer [27 29 ], but we must assume that they are not perfect predictors of diseasespecific mortality. Despite these limitations, the present study represents one of the largest series to date examining the effect of delayed RP. As such, it supports the existing evidence that delays in surgical treatment have no adverse effects on long-term cancer control for men with localized prostate cancer. In conclusion, in this study of 1568 men with prostate cancer treated with RP, we found no association between delayed TTS and risk of either BCR or adverse pathological outcome. These data suggests that men with localized prostate cancer can be reassured that reasonable delays in treatment will not influence disease outcomes. CONFLICT OF INTEREST None declared. REFERENCES 1 van den Bergh RC, Steyerberg EW, Khatami A et al. Is delayed radical prostatectomy in men with low-risk screen-detected prostate cancer associated with a higher risk of unfavorable outcomes? Cancer 2010 ; 116 : 1281 90 2 Kulkarni GS, Urbach DR, Austin PC, Fleshner NE, Laupacis A. Longer wait times increase overall mortality in patients with bladder cancer. J Urol 2009 ; 182 : 1318 24 3 Robertson C, Robertson AG, Hendry JH et al. Similar decreases in local tumor control are calculated for treatment protraction and for interruptions in the radiotherapy of carcinoma of the larynx in four centers. Int J Radiat Oncol Biol Phys 1998 ; 40 : 319 29 4 O Rourke N, Edwards R. Lung cancer treatment waiting times and tumour growth. Clin Oncol (R Coll Radiol) 2000 ; 12 : 141 4 5 Sainsbury R, Johnston C, Haward B. Effect on survival of delays in referral of patients with breast-cancer symptoms: a retrospective analysis. Lancet 1999 ; 353 : 1132 5 6 Roncoroni L, Pietra N, Violi V, Sarli L, Choua O, Peracchia A. Delay in the diagnosis and outcome of colorectal cancer: a prospective study. Eur J Surg Oncol 1999 ; 25 : 173 8 7 Nam RK, Jewett MA, Krahn MD et al. Delay in surgical therapy for clinically localized prostate cancer and biochemical recurrence after radical prostatectomy. Can J Urol 2003 ; 10 : 1891 8 8 Freedland SJ, Kane CJ, Amling CL, Aronson WJ, Presti JC Jr, Terris MK. Delay of radical prostatectomy and risk of biochemical progression in men with low risk prostate cancer. J Urol 2006 ; 175 : 1298 302 ; discussion 302 3 9 Boorjian SA, Bianco FJ Jr, Scardino PT, Eastham JA. Does the time from biopsy to surgery affect biochemical recurrence 2011 BJU INTERNATIONAL 215

KORETS ET AL. after radical prostatectomy? BJU Int 2005 ; 96 : 773 6 10 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 : 1835 9 11 Graefen M, Walz J, Chun KH, Schlomm T, Haese A, Huland H. Reasonable delay of surgical treatment in men with localized prostate cancer impact on prognosis? Eur Urol 2005 ; 47 : 756 60 12 Freedland SJ, Sutter ME, Dorey F, Aronson WJ. Defining the ideal cutpoint for determining PSA recurrence after radical prostatectomy. Prostatespecific antigen. Urology 2003 ; 61 : 365 9 13 D Amico AV, Whittington R, Malkowicz SB et al. Biochemical outcome after radical prostatectomy or external beam radiation therapy for patients with clinically localized prostate carcinoma in the prostate specific antigen era. Cancer 2002 ; 95 : 281 6 14 Littrup PJ. Prostate cancer screening. Appropriate choices? Investigators of the American Cancer Society National Prostate Cancer Detection Project. Cancer 1994 ; 74 : 2016 22 15 Litwin MS, Gore JL, Kwan L et al. Quality of life after surgery, external beam irradiation, or brachytherapy for early-stage prostate cancer. Cancer 2007 ; 109 : 2239 47 16 Sanda MG, Dunn RL, Michalski J et al. Quality of life and satisfaction with outcome among prostate-cancer survivors. N Engl J Med 2008 ; 358 : 1250 61 17 Desireddi NV, Roehl KA, Loeb S et al. Improved stage and grade-specific progression-free survival rates after radical prostatectomy in the PSA era. Urology 2007 ; 70 : 950 5 18 Nguyen PL, Whittington R, Koo S et al. The impact of a delay in initiating radiation therapy on prostate-specific antigen outcome for patients with clinically localized prostate carcinoma. Cancer 2005 ; 103 : 2053 9 19 Vickers AJ, Bianco FJ Jr, Boorjian S, Scardino PT, Eastham JA. Does a delay between diagnosis and radical prostatectomy increase the risk of disease recurrence? Cancer 2006 ; 106 : 576 80 20 Shappley WV 3rd, Kenfield SA, Kasperzyk JL et al. Prospective study of determinants and outcomes of deferred treatment or watchful waiting among men with prostate cancer in a nationwide cohort. J Clin Oncol 2009 ; 27 : 4980 5 21 Albertsen PC, Hanley JA, Fine J. 20-year outcomes following conservative management of clinically localized prostate cancer. JAMA 2005 ; 293 : 2095 101 22 Jayadevappa R, Chhatre S, Johnson JC, Malkowicz SB. Variation in quality of care among older men with localized prostate cancer. Cancer 2010 ; [Epub ahead of print] 23 Laudano MA, Badani KK, McCann TR et al. Significant change in predicted risk of biochemical recurrence after radical prostatectomy more common in black than in white men. Urology 2009 ; 74 : 660 4 24 Melchior SW, Corey E, Ellis WJ et al. Early tumor cell dissemination in patients with clinically localized carcinoma of the prostate. Clin Cancer Res 1997 ; 3 : 249 56 25 Pressinotti NC, Klocker H, Schafer G et al. Differential expression of apoptotic genes PDIA3 and MAP3K5 distinguishes between low- and high-risk prostate cancer. Mol Cancer 2009 ; 8 : 130 42 26 Reid AH, Attard G, Ambroisine L et al. Molecular characterisation of ERG, ETV1 and PTEN gene loci identifies patients at low and high risk of death from prostate cancer. Br J Cancer 2010 ; 102 : 678 84 27 Stephenson AJ, Scardino PT, Eastham JA et al. Postoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Clin Oncol 2005 ; 23 : 7005 12 28 Freedland SJ, Humphreys EB, Mangold LA et al. Risk of prostate cancer-specific mortality following biochemical recurrence after radical prostatectomy. JAMA 2005 ; 294 : 433 9 29 Cheng L, Davidson DD, Lin H, Koch MO. Percentage of Gleason pattern 4 and 5 predicts survival after radical prostatectomy. Cancer 2007 ; 110 : 1967 72 Correspondence: James M. McKiernan, Department of Urology, Columbia University College of Physicians and Surgeons, 161 Fort Washington Ave, Herbert Irving Pavilion 11th Floor, New York, NY 10032, USA. e-mail: jmm23@columbia.edu Abbreviations : RP, radical prostatectomy ; BCR, biochemical recurrence ; TTS, time to surgery ; OR, odds ratio ; IQR, interquartile range ; HR, hazard ratio. 216 2011 BJU INTERNATIONAL