Surgical management after active surveillance for low-risk prostate cancer: pathological outcomes compared with men undergoing immediate treatment
Marc A. Dall’Era, Department of Urology, University of California, Davis, 4860 Y Street, Sacramento, CA 95817, USA.
Study Type – Therapy (case control)
Level of Evidence 3b
What's known on the subject? and What does the study add?
The risks of delayed radical prostatectomy for men who progress on active surveillance are largely unknown. Two series have reported that prostatectomy after active surveillance has similar results to immediate therapy. Our data add to this growing body of evidence that appropriately selected men with prostate cancer can undergo active surveillance with delayed prostatectomy without added risk of missing an opportunity for cure as the majority of tumours remain organ confined.
• To compare the pathological outcomes of men undergoing radical prostatectomy (RP) after a period of active surveillance (AS) with those of a similar risk group undergoing immediate surgery.
PATIENTS AND METHODS
• We identified men through our institutional database who underwent RP within 6 months of diagnosis or after a period of AS. The primary outcome of the present study was Gleason upgrade to ≥7 after prostatectomy.
• Pathological stage and positive surgical margin rate were assessed as secondary outcomes. Binomial logistic regression models were used to determine associations of treatment subgroups with pathological upgrade, upstage and positive margins.
• Thirty-three men with initially low-risk cancer features underwent RP after a median (range) of 18 (7–76) months of AS. A total of 278 men with low-risk disease features underwent immediate RP within 6 months of diagnosis. Rates of Gleason upgrading to ≥7, pathological category pT3 and positive surgical margins did not differ significantly from the immediate RP group.
• On multivariate analysis of low-risk patients, adjusting for baseline pathological features, treatment group (AS followed by prostatectomy vs immediate prostatectomy) was not associated with Gleason upgrading (odds ratio, OR, 0.35; 95% CI, 0.12–1.04), non-organ-confined disease (OR, 1.67; 95% CI, 0.32–8.65) or positive surgical margins at prostatectomy (OR, 0.95; 95% CI, 0.16–5.76).
• The present analysis did not show an association between RP after a period of AS and adverse pathological features for men with low-risk disease.
University of California, San Francisco.
Active surveillance (AS) is a viable treatment option for carefully selected men with low-risk prostate cancer as defined by grade, stage, volume and serum PSA. Recent studies describe oncological outcomes of close surveillance with delayed, selective intervention for men with progressive disease [1,2]. Median follow-up from these reports is relatively short and the long-term risks of AS remain unknown. Until the results of larger trials comparing AS with immediate treatment for men with very low-risk prostate cancer are available, more readily available surrogate markers of risk or outcome must be relied on to determine the best candidates for this approach. An update to the Swedish randomized trial comparing radical prostatectomy (RP) with traditional watchful waiting in primarily unscreened men diagnosed with prostate cancer showed a survival benefit with RP for some men . A critical aspect is to determine which men will benefit the most from prostatectomy with the possibility of decrements in health-related quality of life and to determine when this treatment should be delivered. It should also be realized that not all men will likely benefit from immediate radical intervention. The present study aimed to describe the rates of high-risk pathological findings of men undergoing prostatectomy after a period of AS.
Natural history data show that tumour grade is the most important predictor of disease-specific mortality for men being followed with prostate cancer [4,5]. RP series also show that biochemical recurrence and disease-specific survival are highly dependent on the stage of disease at presentation . Accurate clinical staging and grade assessment based on physical examination, prostate biopsy and available imaging modalities are paramount to risk assessment for AS. Surgical series of men who would have been considered candidates for AS yet chose RP show a substantial risk of clinical under-grading and staging with current modalities [7,8]. These risks appear to be minimized by selecting men for AS with low-grade and low- tumour volume characteristics based on total number of biopsy cores with cancer and the percentage of cores involved with tumour . It remains unknown how the pathological outcomes change over time for men on AS who are carefully followed and whether some patients will miss the benefits of RP by delayed therapy. The present study aimed to describe the pathological findings of men undergoing RP after a period of AS for prostate cancer and compare them with a group of similar men undergoing immediate surgery. Correlating clinical characteristics with pathological outcomes after RP for men initially treated with AS can provide insight into potential predictors of adverse disease features for men choosing AS for management of their prostate cancer.
PATIENTS AND METHODS
Patients who selected AS as the primary management option after the diagnosis of prostate cancer were identified through the University of California, San Francisco (UCSF) Urologic Oncology Database after Institutional Review Board approval. Men from this group who then selected RP as a secondary treatment comprised the surveillance cohort for the present study. Inclusion criteria were no therapy received before diagnosis/presentation at our institution, primary therapy designated as AS and no active treatment (surgery, external beam radiation, brachytherapy or androgen ablation therapy) received within 6 months after diagnosis. Patients were selectively offered AS if they met the low-risk diagnostic criteria: PSA < 10 ng/mL, Gleason sum ≤6, absence of Gleason grade 4 or 5, cancer involvement of ≤33% of biopsy cores and ≤50% of any single core, clinical stage T1/T2. In addition to these men, some patients who did not meet these criteria were placed on AS for other medical reasons or personal choice. The surveillance regimen consisted of office visits with digital rectal examination, serial PSA measurements (usually at 3-month intervals) and TRUS at 6–12-month intervals. Twelve core biopsies were routinely performed under ultrasound guidance. Additional cores were taken at the discretion of the operator for visible abnormalities. Repeat prostate biopsies were recommended at 12–24-month intervals. All biopsy and RP specimens were reviewed by a UCSF genitourinary pathologist.
Clinical and pathological characteristics of the cohort of AS patients who subsequently underwent surgery (AS + RP) were compared with men who had immediate surgery. Initially, four risk-treatment groups were studied: low-risk AS + RP in men meeting all criteria for AS; higher-risk AS + RP (i.e. men selecting AS who did not meet all criteria); low-risk immediate RP (i.e. men meeting all of our AS criteria); and higher-risk immediate RP. The pathological outcomes for AS + RP and immediate RP patients with low-risk disease only (i.e. meeting all of our criteria for AS) were then analyzed.
The primary outcome of our analysis was Gleason upgrade to at least 7 after RP compared to the most recent prostate biopsy. Pathological Gleason sum was compared with the Gleason sum from the most recent biopsy before RP in the AS + RP group and with diagnostic Gleason grade in the immediate RP group. Pathological stage category pT3 or greater and positive surgical margin rate were assessed as secondary outcomes.
Chi-squared was used to compare categorical variables, and t-tests and analysis of variance were used to compare continuous variables between groups. Logistic regression models were used to determine associations of treatment subgroups with pathological upgrade, upstage and positive margins. Multivariate models were adjusted for age, PSA at diagnosis and tumour volume at pathology. Preliminary correlations between covariates were carried out using Pearson’s R and chi-squared. All variables were included in the models a priori based on clinical relevance. P < 0.05 was considered statistically significant. All statistical analysis was performed using SAS, version 9.1, for Windows (SAS Institute, Cary, NC, USA).
There were 1902 prostate cancer patients diagnosed at UCSF in 1996–2008 and managed with either surgery or AS who consented to allow their data to be used for research at UCSF. Of those, 570 (30%) met the low-risk criteria for AS eligibility, whereas the remaining 1332 had a higher clinical risk at diagnosis. Within the low-risk group, 337 patients underwent primary RP (278 within 6 months of diagnosis and 59 at 6–18 months after diagnosis) and 233 (41%) selected AS. Some 33 AS patients eventually underwent RP, 32 received other active treatments and 168 continued on surveillance. Among men with higher clinical risk characteristics, 1169 patients were in the primary RP group (1067 within 6 months and 102 at 6–18 months after diagnosis) and 163 opted for AS, with 30 ultimately having surgery, 31 having other treatments and 102 continuing on surveillance. Low-risk AS patients were recommended for intervention most commonly in response to Gleason upgrading on surveillance biopsy (23 patients) or patient desire.
The analysis focused on 1408 patients who had immediate RP within 6 months after diagnosis and AS patients who subsequently had RP (AS + RP). At diagnosis, the median (range) age was 59 (37–79) years and median (range) PSA was 5.8 (0.02–100+) ng/ml. The median (range) number of biopsy cores taken at diagnosis was 10 (1–60). Some 25% of diagnostic biopsies were performed at UCSF and 75% were performed at outside institutions and verified by a UCSF pathologist. The median (range) time on surveillance was 18 (7–72) months for AS + RP patients and median (range) months to immediate RP was 3 (<1–6) months. Median (range) follow-up after surgery was 12 (<1–60) months for AS + RP patients and 27 (<1–162) months for immediate RP patients.
Age and clinical characteristics are shown in Table 1 for the four risk-treatment groups: 33 low-risk AS + RP, 278 low-risk immediate RP, 30 higher-risk AS + RP and 1067 higher-risk immediate RP patients. Surgical pathology outcomes by risk-treatment groups are shown in Table 2. Among the low-risk patients, pre-surgery Gleason grade was higher in the AS + RP group because most of these patients elected for surgical intervention as a result of higher-grade disease being detected on surveillance biopsy. When analyzing these data by disease characteristics at diagnosis, 61% of initially low-risk men were upgraded at the time of surgery. Considering that several men were appropriately diagnosed with more aggressive disease during surveillance, and thus were appropriately treated with curative intent, the present study aimed to determine rates of clinical under-grading despite efforts at close surveillance. When analyzing the data using Gleason scores obtained from the last recorded biopsy, rates of upgrading to ≥7, staging of pT3 and positive surgical margins did not differ significantly between the AS + RP and immediate RP groups. Some 30% of men undergoing AS + RP were upgraded after surgery compared to their most recent biopsies and 35% of immediate RP patients were upgraded (P= 0.57), the majority of whom were upgraded from Gleason 3 + 3 to Gleason 3 + 4. One low-risk AS + RP patient was upgraded from 3 + 3 to 4 + 3 and five were downgraded (from 3 + 4 to 3 + 3l one from 8 to 4 + 3) after surgery. Of 28 low-risk AS + RP men with at least one surveillance biopsy after diagnostic biopsy, six (21%) were upgraded at RP compared to four of five men who did not undergo a surveillance re-biopsy.
Table 1. Baseline cohort characteristics
|Charecteristic, mean ± SD*|| || || || || || |
| Age||58.5 ± 6.77||59.3 ± 7.62||0.55||59.0 ± 6.84||62.4 ± 5.10||<0.01|
| PSA||5.5 ± 1.87||5.0 ± 2.06||0.18||8.1 ± 7.50||7.5 ± 3.93||0.65|
| % Free PSA||13.7 ± 5.77||21.2 ± 8.17||<0.01||12.4 ± 7.73||17.1 ± 4.63||0.15|
| PSA density||0.2 ± 0.72||0.1 ± 0.09||0.55||0.3 ± 0.23||0.2 ± 0.30||0.62|
| Prostate volume||37.6 ± 20.31||40.2 ± 20.28||0.51||32.8 ± 16.59||43.1 ± 27.55||<0.01|
| Cores taken||11.9 ± 6.49||14.2 ± 9.03||0.08||9.6 ± 6.04||10.0 ± 8.49||0.73|
| Months to RP||3.2 ± 1.25||25.9 ± 16.18||<0.01||3.1 ± 1.28||23.7 ± 15.41||<0.01|
|Chi-squared, n (%)**|| || || || || || |
| T-stage|| || || || || || |
| cT1||137 (49)||21 (64)||0.12||454 (43)||13 (43)||0.94|
| cT2||141 (51)||12 (36)|| ||609 (57)||17 (57)|| |
| cT3||–||–|| ||4 (<1)||0 (0)|| |
| Gleason|| || || || || || |
| 6||278 (100)||33 (100)||–||412 (39)||26 (87)||<0.01|
| 7 (3 + 4)||–||–|| ||336 (31)||3 (10)|| |
| 7 (4 + 3)||–||–|| ||8 (1)||0 (0)|| |
| 8–10||–||–|| || 311 (29)||1 (3)|| |
| Cores taken|| || || || || || |
| <6||22 (8)||0 (0)||0.02||237 (22)||10 (33)||0.08|
| 6–11||123 (44)||9 (27)|| ||461 (43)||7 (23)|| |
| 12–17||133 (48)||24 (73)|| ||369 (35)||13 (43)|| |
| Cores positive|| || || || || || |
| <33%||278 (100)||33 (100)|| ||343 (32)||14 (47)||0.24|
| 33–66%||–||–|| ||380 (36)||8 (27)|| |
| >66%||–||–|| ||344 (32)||8 (27)|| |
| Single core positive|| || || || || || |
| ≤10%||106 (48)||16 (50)||0.83||134 (15)||11 (39)||<0.01|
| >10%||115 (52)||16 (50)|| ||745 (85)||17 (61)|| |
| Missing||57||1|| ||188||2|| |
Table 2. Pathological outcomes after radical prostatectomy
|Gleason upgrade ≥7|
| Yes|| 98||35||10||30||.||260||24|| 8||27||.|
|Upstage to pT3|
| Yes|| 28||10|| 7||21||.||320||30|| 11||37||.|
| Yes|| 33||12|| 6||18||.||214||20|| 5||17||.|
In total, 26 (77%) low-risk AS + RP and 250 (90%) immediate RP patients had organ-confined disease (pT2) at the time of RP (P < 0.01). Of the men with non-organ-confined disease, all men in the AS + RP group had pT3a disease, whereas 18 (6%) and 10 (4%) of men in the immediate RP group with non-organ-confined disease had pT3a and pT3b, respectively. Low-risk AS + RP men with organ-confined disease had lower rates of pathological upgrading (P < 0.01). At the time of RP, six (18%) of AS + RP and 33 (12%) of immediate RP patients had positive surgical margins (P= 0.30).
Of the men who did not meet the low-risk criteria for AS, 11 (36%) of AS + RP and 323 (31%) of immediate RP patients had pT3 category disease, with one AS + RP and 113 immediate RP having seminal vesicle involvement. Some 49% of higher-risk immediate RP patients underwent lymphadenectomy and, of those, 39 (7%) had lymph node metastases. No positive nodes were found in the three (9%) low-risk AS + RP, 10 (33%) high-risk AS + RP or 35 (13%) low-risk immediate RP men who had a lymph node dissection.
To investigate the effect of time on AS, the outcomes for AS + RP patients were stratified based on whether they had surgery within 18 months of surveillance or after 18 months (Table 3). No difference was noted between the groups for the outcomes analyzed, with the majority of tumours remaining organ-confined even well after 18 months of AS. The results were similar when low and higher-risk men were analyzed separately. Of 167 immediate RP and 15 AS + RP with at least 4 years of follow-up biochemical disease-free survival after surgery was 98% and 100%, respectively.
Table 3. Pathological outcomes for active surveillance plus radical prostatectomy (AS + RP) men stratified by time on active surveillance before surgery
|Gleason upgrade ≥7|| || || || || |
| Yes|| 9||29|| 9||28||.|
|Upstage to pT3|| || || || || |
| Yes|| 7||23|| 11||34||.|
|Positive margins|| || || || || |
| Yes|| 5||16|| 6|| 9 ||.|
On multivariate analysis of low-risk patients, adjusting for baseline demographic and pathological features, treatment group (AS + RP vs immediate RP) was not associated with Gleason upgrading (odds ratio, OR, 0.35; 95% CI, 0.12–1.04), upstaging to non-organ-confined disease (OR, 1.67; 95% CI, 0.32–8.65) or positive surgical margins at prostatectomy (OR, 0.95; 95% CI, 0.16–5.76) (Table 4). Gleason upgrade and upstaging to non-organ-confined disease were correlated.
Table 4. Binomial logistic regression results for low-risk patients only
|Univariate|| || || |
| AS then RP vs Immediate RP (Referent)||0.80 (0.37–1.75)||2.40 (0.96–6.04)||1.00 (0.98–1.03)|
|Multivariate|| || || |
| AS then RP vs Immediate RP (Referent)||0.35 (0.12–1.04)||1.67 (0.32–8.65)||0.95 (0.16–5.76)|
| Months from diagnosis or last repeat biopsy to RP (continuous)||1.17 (1.03–1.33)*||1.04 (0.98–1.09)||1.03 (0.97–1.08)|
| Age at diagnosis (continuous)||1.01 (0.97–1.06)||1.06 (0.99–1.13)||0.99 (0.93–1.05)|
| PSA at diagnosis (continuous)||1.08 (0.93–1.24)||1.21 (0.95–1.54)||1.08 (0.87–1.35)|
| Tumour volume at pathology (continuous)||0.96 (0.89–1.03)||1.04 (0.96–1.12)||1.08 (1.02–1.16)*|
| Gleason upgrade at path: Yes vs No (referent)||–||2.86 (1.18–6.93)*||1.76 (0.77–4.00)|
| Upstage to pT3: Yes vs No (referent)||2.69 (1.11–6.50)*||–||2.48 (0.85–7.23)|
| Positive margins: Yes vs No (referent)||1.74 (0.75–4.03)||2.62 (0.90–7.62)||–|
The present analysis did not show an association between treatment group and adverse pathological features at the time of prostatectomy for men with low-risk disease at presentation. After a median period of 18 months of AS, 79% of men had organ-confined disease after prostatectomy and only one patient had primary Gleason pattern 4 disease. In a similar analysis, Warlick et al.  reported that 23% of 38 patients had what was considered ‘non-curable’ prostate cancer with RP after a median of 26 months of surveillance, which was statistically similar to men undergoing immediate prostatectomy. Noncurable prostate cancer was defined as pathological features associated with a <75% chance of remaining disease-free for 10 years after surgery. A more recent analysis of this data set of men undergoing RP after AS reported that 65% had organ-confined disease, with 53% having a Gleason sum of 6 . Some 35% of tumours showed evidence of extraprostatic extension at the time of surgery. These data, similar to our findings, suggest that appropriately selected men with prostate cancer can undergo AS with delayed prostatectomy without the added risk of missing an opportunity for cure because the majority of tumours remain organ-confined. Klotz reported on 29 men undergoing delayed RP after AS and found non-organ-confined disease in 12 men and lymph node metastases in one man . Of men with a rapid PSA doubling time of <3 years, seven of 18 (38%) had positive surgical margins . It is important to note that this series includes some men with higher PSA values and Gleason scores of 7. A smaller series of men reported similar findings after delayed prostatectomy .
The risk of clinical under-grading is of great concern for men with features of low-risk disease because tumour grade is well recognized as one of the best predictors of outcome. AS protocols commonly employ a strategy of repeat surveillance biopsy with the aim of limiting under-grading at the same time as detecting progressive disease early [13–15]. It was previously shown that finding higher-grade disease on surveillance biopsy is the greatest driver of treatment in this series of men . In the present study, 30% of men overall were upgraded at the time of surgery and men with at least one surveillance biopsy had a lower rate of under-grading then men not undergoing surveillance biopsies (chi-squared: P < 0.01). It is interesting to note that 20 of 33 AS men (61%) with low-risk disease features at diagnosis from this cohort ultimately had Gleason 7 disease at prostatectomy, ten of which were diagnosed before surgery on surveillance biopsy. This rate is considerably higher than what would be expected from a group of low-risk men undergoing primary surgery, confirming how the surveillance process selected for a higher-risk group of individuals for radical therapy. Although, the true pathology of untreated men continuing surveillance is not known, these findings suggest that carefully monitoring PSA and employing a strategy of surveillance prostate biopsies can identify men with higher-risk disease. It is not known whether this change in Gleason score represents a true grade progression or simply sampling error, although current evidence suggest the latter [10,17]. Multiple retrospective surgical series well describe the rates of change of Gleason grade from biopsy to immediate RP in men meeting one of several criteria for AS [18,19]. Griffin et al.  described the pathological outcomes for 292 men with Gleason 6, low-volume disease, defined as no greater than two positive cores on biopsy, who underwent immediate prostatectomy after diagnosis. In that group of men, 27% had Gleason 7 or higher stage disease, 16% had T3 disease and 10% had positive surgical margins . Those results compare favourably with the data from the series reported in the present study, especially for the men meeting all of the criteria for low-risk disease who underwent surgery a median of 18 months after diagnosis. These data also show how current methods for risk stratification still miss a substantial number of men with higher-risk features. They emphasize the important need for close surveillance of men choosing this approach with the aim of managing their prostate cancer. Performing an immediate second biopsy within 3 months of diagnosis may serve to reduce the risk of clinical under-grading and can identify higher-risk men who should be recommended for more aggressive treatement. Rates of extraprostatic disease were higher among patients with a higher clinical risk at diagnosis; 36% of AS + RP and 31% of immediate RP patients who did not meet all criteria for AS had pT3 disease. One man who had been on AS with higher-risk features had seminal vesicle involvement at prostatectomy. Men from this series who did not meet all criteria for AS had either high PSA values, a high number of positive biopsy cores or high tumour involvement of each positive core, suggesting that these parameters may act as surrogates for more aggressive disease. These data show the importance of carefully selecting men with low-risk disease features, including estimates of tumour volume based on needle core biopsy, characteristic for AS.
We acknowledge that there several limitations to the present study, which comprises a retrospective analysis of men undergoing RP after AS for prostate cancer at UCSF. Not all patients were uniformly surveyed. Group sample sizes were uneven; data were analyzed from only 33 low-risk AS + RP patients compared to 278 immediate RP patients, reflected by relatively wide CIs for the outcomes analyzed. The limited sample size in the present study limits the ability to detect small differences in outcomes between the groups. Detailed data on decision-making of AS patients to pursue active treatment and the effect on surgical pathology were not part of the present study. Although all AS + RP men delayed RP for at least 6 months after diagnosis, not all men underwent repeat prostate needle biopsy. Variability in pathologist interpretation may affect the results obtained as well, although all pathology was reviewed at our institution in an attempt to limit this. Although the specimens were not all reviewed by the same individual, this central review should limit this bias. It is acknowledged that the main pathological outcomes of the present study are simply surrogates for the more clinically relevant outcomes of disease-specific and overall survival. These pathological features, however, have been integrated into a predictive model after RP and, until randomized trials describe the true results of AS for low-risk prostate cancer in a contemporary cohort, these data will help identify candidates for this approach with the lowest risk of progression . The pathological findings at the time of RP can provide important insight for predicting prognosis and risks of prostate cancer mortality after primary treatment.
The results obtained in the present study describe the pathological outcomes for men undergoing RP after a period of AS with delayed surgical intervention. These data suggest that men on AS with higher-risk disease features can be identified by close surveillance and offered timely curative intervention. Early re-biopsy may serve to reduce the risk of under-diagnosis of more aggressive disease.
CONFLICT OF INTEREST
Peter R. Carroll is an investigator for NIH and DOD and provides research support for Abbott.