Because no adequate randomized trials have compared active treatment modalities for localized prostate cancer, the authors analyzed risk-adjusted, cancer-specific mortality outcomes among men who underwent radical prostatectomy, men who received external-beam radiation therapy, and men who received primary androgen-deprivation therapy.
The Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) registry comprises men from 40 urologic practice sites who are followed prospectively under uniform protocols, regardless of treatment. In the current study, 7538 men with localized disease were analyzed. Prostate cancer risk was assessed using the Kattan preoperative nomogram and the Cancer of the Prostate Risk Assessment (CAPRA) score, both well validated instruments that are calculated from clinical data at the time of diagnosis. A parametric survival model was constructed to compare outcomes across treatments adjusting for risk and age.
In total, 266 men died of prostate cancer during follow-up. Adjusting for age and risk, the hazard ratio for cancer-specific mortality relative to prostatectomy was 2.21 (95% confidence interval [CI], 1.50-3.24) for radiation therapy and 3.22 (95% CI, 2.16-4.81) for androgen deprivation. Absolute differences between prostatectomy and radiation therapy were small for men at low risk but increased substantially for men at intermediate and high risk. These results were robust to a variety of different analytic techniques, including competing risks regression analysis, adjustment by CAPRA score rather than Kattan score, and examination of overall survival as the endpoint.
For the more than 192,000 men expected to be diagnosed with prostate cancer annually,1 decision-making with respect to the type and timing of treatment is complex: Prostate cancer is surpassed only by lung cancer in its mortality burden among men in the United States,1 yet the natural history of the disease frequently is indolent even among men who go untreated,2 and all available active treatments can be associated with significant adverse effects.3 To our knowledge, no contemporary studies that randomized patients across primary treatments have been reported. Indeed, a systematic review recently commissioned by the Agency for Healthcare Research and Quality concluded that insufficient high-quality evidence exists to support any given treatment modality over another.4
The American Urological Association's clinical practice guideline for localized prostate cancer states that alternatives offered to patients should include active surveillance, radical prostatectomy, external-beam radiation therapy, and brachytherapy but draws no conclusions regarding the relative efficacy of these alternatives.5 Primary androgen-deprivation monotherapy for localized disease is not endorsed by the guideline given inadequate evidence regarding outcomes; nonetheless, it is commonly used in practice.6, 7
Given the often prolonged course of prostate cancer even among most men who ultimately have lethal disease,8 studies with short-term to intermediate-term follow-up may report outcomes only in terms of recurrence-free survival based on prostate-specific antigen (PSA)-based definitions. Because many disparate definitions of biochemical recurrence have been proposed,9 comparing outcomes across modalities using PSA endpoints is problematic. Clinical endpoints—in particular, prostate cancer-specific mortality (CSM) and all-cause mortality (ACM)—do not vary across treatments and ultimately are more relevant to patients. However, analyses at these endpoints require long-term follow-up.
To ascertain risk-adjusted comparative effectiveness of primary treatment approaches for prostate cancer, we conducted an analysis comparing CSM and ACM outcomes after prostatectomy, external-beam radiation therapy, and primary androgen deprivation in a well defined, multicenter, prospective cohort of patients with prostate cancer.
MATERIALS AND METHODS
Data were abstracted from the Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE), a national disease registry that accrues men with biopsy-proven prostate adenocarcinoma who receive treatment at any of 40 (primarily community-based) urology practices across the United States. Participating urologists recruit men consecutively at diagnosis and report initial and follow-up clinical data, including staging tests and treatments. Comorbidities are recorded at baseline and in follow-up, and comorbidity scoring is based on the Charlson index.10 The registry was initiated in 1995. Between 1995 and 1998, accrual was both prospective and retrospective; after 1998, all accrual has been prospective. Patients provide written, informed consent under local and central institutional review board supervision.
Patients are treated according to their clinicians' usual practices and are followed until death or withdrawal from the study. Clinicians report mortality events, and copies of state death certificates are obtained. CSM is determined if prostate cancer is listed as a primary, secondary, or tertiary cause of death on the death certificate and if no other malignancy is listed as a higher order cause. Perioperative mortality and death from complications of radiation and/or androgen deprivation count toward ACM but not CSM. If the patient has been lost to follow-up or the certificate is not available, then the National Death Index is queried to identify the date and cause of death. Additional details regarding CaPSURE's methodology have been reported previously.11, 12
In total, 13,805 men had enrolled in CaPSURE as of July 2008. Of these, 8982 men with localized disease (clinical stage ≤T3aN0) underwent prostatectomy, received external-beam radiation, or received primary androgen deprivation; and had at least 6 months of follow-up recorded. The 1444 men who were missing data needed to calculate both risk instruments described below were excluded. Thus, 7538 men comprised the analytic dataset. Years of treatment ranged from 1987 to 2007: Twenty-six percent received treatment before 1997, 10% received treatment before 1995, and 1% received treatment before 1991.
Demographic and clinical characteristics of the patients in each treatment group were compared using analyses of variance or chi-square tests as appropriate for continuous and categorical variables. To ensure that the analyses did not depend on a specific risk-adjustment approach, prostate cancer risk was assessed using 2 well validated pretreatment instruments. The first instrument was the original nomogram published by Kattan et al, which uses a score from 0 to 100 calculated from the PSA level, Gleason grade, and clinical tumor (T) classification to estimate the likelihood of recurrence-free survival after radical prostatectomy.13-16 For the current analysis, risk was expressed as the Kattan score subtracted from 100 (100-Kattan score), and higher numbers indicated a greater risk of disease.
The second instrument was the University of California, San Francisco Cancer of the Prostate Risk Assessment (CAPRA), a score with a range from 0 to 10 calculated from the PSA level, biopsy Gleason grade, clinical T stage, age at diagnosis, and percentage of positive biopsy cores.16-19 The CAPRA score predicts pathologic stage and biochemical recurrence-free survival, and each 2-point increase in score indicates roughly a doubling of the risk of recurrence. Most recently, it was demonstrated that the score predicts metastasis, CSM, and ACM across multiple primary treatments.20
Kaplan-Meier time-to-event curves were generated,21 and outcomes were compared using the log-rank test. Then, Weibull parametric survival models were constructed to compare outcomes, adjusting for age and case mix using either the Kattan score or the CAPRA score. The primary endpoint was CSM, and ACM was assessed as a secondary endpoint. For each endpoint, the hazard ratio (HR) with 95% confidence interval (CI) was calculated for radiation and androgen deprivation compared with prostatectomy. The model was used to predict CSM at 10 years at various levels of risk. For the CSM analyses, patients who died of other causes were censored at the time of death. In a sensitivity analysis, the CSM analyses also were conducting using competing risks regression.22 Tests for interaction between risk and treatment also were performed.
Adjustment for neoadjuvant androgen deprivation did not alter the statistical significance of any variable in the model and had minimal impact on the parameter estimates; therefore, this variable was not included in the final model. The model also was tested excluding the 136 men who received adjuvant radiation therapy after prostatectomy, both with and without the inclusion of adjuvant radiation as an additional predictor variable in the model. To limit the analysis to men who received radiation treatment under relatively contemporary standards, we performed a subset analysis that was limited to men who received treatment after 1998. Finally, although it has been demonstrated that the models are accurate in predicting CSM across multiple treatments,20 it is possible that neither the Kattan score nor the CAPRA score adequately reflect differences in risk across patients. Therefore, as an additional test, we reassessed the model with Kattan scores artificially increased progressively by 5-point increments for patients who underwent radical prostatectomy to estimate the degree of unmeasured confounding beyond measured risk that would need to be assumed to nullify the results. All statistical tests were 2-sided, and analyses were performed using the Stata software package (version 11; Stata Corporation, College Station, Tex).
In total, 1293 men died (17.2%), including 226 who died from prostate cancer (3%). Sociodemographic and clinical factors of the patients in each primary treatment group are summarized in Table 1. All comparisons between treatment groups for clinical and sociodemographic factors were statistically significant (P<.001). Patients who underwent prostatectomy were younger, more frequently Caucasian, and had less comorbidity and lower risk disease features than patients in the other groups. There were 3 perioperative deaths (0.06% of prostatectomy cases). Overall, 49.7% of patients who received radiation received neoadjuvant and/or adjuvant hormone therapy, including 33.7% of those who had CAPRA scores from 0 to 2, 50.6% of those who had CAPRA scores from 3 to 5, and 67.6% of those who had CAPRA scores from 6 to 10. In addition, 6.7% of patients who underwent prostatectomy received neoadjuvant therapy, including 4.5% of those who had CAPRA scores from 0 to 2, 7.7% of those who had CAPRA scores from 3 to 5, and 19.3% of those who had CAPRA scores from 6 to 10. The mean ± standard deviation (SD) duration of therapy was 7.9 ± 3.1 months.
Table 1. Sociodemographic and Clinical Factors for Patients in Each Primary Treatment Group
RP indicates radical prostatectomy; EBRT, external-beam radiotherapy; PADT, primary androgen-deprivation therapy; CAPRA, Cancer of the Prostate Risk Assessment; PSA, prostate-specific antigen; PPB, percentage of biopsy cores positive.
Median [25%/75% quartile]
No. of comorbidities
No. of patients (%)
The mean ± SD and median times to death were 6.8 ± 4.0 years and 6.4 years, respectively, and the mean ± SD and median follow-up times among those who survived were 4.2 ± 3.3 years and 3.9 years, respectively. The median follow-up was similar across treatments (3.9 years, 4.5 years, and 3.6 years for prostatectomy, radiation, and androgen deprivation, respectively) and across risk groups (3.6 years, 4.1 years, and 4.0 years for CAPRA scores of 0-2, 3-5, and 6-10, respectively). Unadjusted time-to-event curves for CSM are presented in Figure 1. The differences in outcomes across treatments were statistically significant (log-rank test; P<.001). Relative to prostatectomy, the unadjusted HRs for CSM were 2.46 (95% CI, 1.8-3.43) for radiation and 4.36 (95% CI, 3.21-5.93) for androgen deprivation.
The results of the primary risk-adjusted analysis are presented in Table 2. Adjusting for age and case mix using the Kattan score, the HRs for CSM relative to prostatectomy for radiation and androgen deprivation were 2.21 (95% CI, 1.50-3.24) and 3.22 (95% CI, 2.16-4.81), respectively. The HR for CSM for androgen deprivation relative to radiation was 1.45 (95% CI, 1.02-2.07). Adjusting for the CAPRA score rather than the Kattan score yielded somewhat lower but similar HRs relative to prostatectomy: 1.63 (95% CI, 1.09-2.45) for radiation, 2.65 (95% CI, 1.75-4.01) for androgen deprivation, and 1.62 (95% CI, 1.11-2.36) for androgen deprivation relative to radiation. The use of competing risks regression likewise yielded similar results: relative to prostatectomy, the HRs were 2.00 (95% CI, 1.33-3.01) and 2.56 (95% CI, 1.62-4.03) for radiation and androgen deprivation, respectively; relative to radiation, the HR was 1.27 (95% CI, 0.88-1.84) for androgen deprivation.
Table 2. Results of Survival Analysis for Predicting Prostate Cancer-Specific Mortality
Excluding 136 men who received adjuvant radiation therapy after prostatectomy had no effect on the results of the model whether or not radiation was included as a predictor in the model. In interaction analyses, there was no evidence that the difference between prostatectomy and radiation depended on the baseline risk (P = .20). There was suggestion that improved outcome with radiation compared with androgen deprivation increased for patients with higher risk disease (P = .07); however, because this difference did not meet statistical significance, survival differences were modeled assuming constant relative risk among treatments across different levels of risk.
Table 3 presents the results for ACM: Adjusting for age, Kattan score, and comorbidity, the HR relative to prostatectomy was 1.58 (95% CI, 1.32-1.89) for radiation and 2.25 (95% CI, 1.86-2.72) for androgen deprivation. Relative to radiation, the HR for ACM for androgen deprivation was 1.43 (95% CI, 1.21-1.69). Virtually identical results were produced with adjustment for the CAPRA score rather than the Kattan score. Figure 2 and Table 4 present predicted 10-year CSM by 100-Kattan and CAPRA scores, respectively, for each treatment. Predicted CSM increased consistently with rising CAPRA scores, from 1.5% to 32.8% for prostatectomy, from 2.5% to 48.7% for radiation, and from 4.0% to 66.3% for androgen deprivation.
Table 3. Results of Survival Analysis for Predicting All-Cause Mortality
Table 4. Predicted 10-Year Cancer-Specific Mortality According to Cancer of the Prostate Risk Assessment Scores for Each Primary Treatment Group
HR [95% CI]
HR indicates hazard ratio; CI, confidence interval; CAPRA, Cancer of the Prostate Risk Assessment; RP, radical prostatectomy; EBRT, external-beam radiotherapy; PADT, primary androgen-deprivation therapy.
In restricting the analysis to those who were treated after 1998, the number of CSM events fell to 67 among 5143 patients at risk. The HRs for CSM relative to prostatectomy in this subset were 2.7 (95% CI, 1.2-6.2) for radiation therapy and 6.5 (95% CI, 3.1-13.5) for androgen deprivation. In our sensitivity analysis for unmeasured confounding, in calculating the model with Kattan scores artificially increased for patients who underwent prostatectomy, the difference between patients who underwent prostatectomy and patients who received radiation remained statistically significant until the Kattan scores were increased by 20 points for prostatectomy patients, and it did not change direction until the scores were increased by >30 points (Table 5).
Table 5. Hazard Ratios With 95% Confidence Intervals for Cancer-Specific Survival in the External-Beam Radiotherapy and Primary Androgen-Deprivation Therapy Groups Relative to the Radical Prostatectomy Group Controlling for Age and Kattan Scorea
The Kattan score for each patient who underwent prostatectomy was increased artificially by 0 to 35 points.
Uncertainty regarding the optimal management of localized prostate cancer has produced wide and excessive local and regional variation in the use of various interventions.23-25 In general, with increasing risk, men are less likely to undergo prostatectomy, more likely to receive radiation, and much more likely to receive androgen-deprivation monotherapy.26 In particular, over time, the use of androgen deprivation has increased for high-risk men.6, 26 Although several large centers recently reported outcomes after prostatectomy in high-risk patients that compared favorably with outcomes from earlier series,27 to date, there are no indications that these findings have had an impact on community practice.
These trends have not been evidence-driven; indeed, given the existing dearth of high-quality comparative data, the Institute of Medicine recently included treatment for localized prostate cancer among the 25 most important topics for comparative-effectiveness research.28 To our knowledge, only 3 randomized trials have been published comparing major primary management approaches. Bill-Axelson et al reported a survival benefit for prostatectomy over watchful waiting, with a 35% relative reduction in risk of CSM at 10 years.29 An earlier, smaller randomized study likewise reported longer overall survival for patients who underwent prostatectomy compared with patients who received watchful waiting.30 Another recent trial randomized patients with clinical T3N0M0 disease to receive flutamide with or without radiation therapy. The results from the study indicated a strong benefit for the combination treatment arm,31 although flutamide monotherapy generally would be considered inadequate therapy by contemporary standards, particularly for locally advanced disease.
Randomized trials in localized prostate cancer face challenges related to high costs associated with long follow-up and patient and/or clinician biases a priori in favor of any given approach or another. The Surgical Prostatectomy Versus Interstitial Radiation Intervention Trial (SPIRIT) intended to randomize men to radical prostatectomy versus brachytherapy. Despite a 90-minute patient education session that was intended to facilitate accrual, only 56 patients accrued at 31 centers over 2 years, and the study was closed early.32 The Prostate Cancer Intervention Versus Observation Trial (PIVOT) screened 13,022 men at 52 sites over 7 years to identify 5023 eligible men, of whom 731 men (14.5%) agreed to be randomized between surgery and observation. Initial results from that trial are expected later this year.33 The Prostate Testing for Cancer and Treatment (ProtecT) study is the only ongoing randomized trial that includes multiple active treatment arms—prostatectomy, external-beam radiation, and watchful waiting. It has had greater success attributed to a complex intervention aimed at increasing the acceptance of randomization.34 However, results from that study will require years to reach maturity.
Meanwhile, important insights into outcomes have been gained from research based on large data sources, such as the Surveillance, Epidemiology and End Results (SEER) Program and Medicare.35, 36 However, these analyses are limited by relatively scant clinical information in the datasets—for example, absent PSA, Gleason, and treatment details. Therefore, prospective disease registries provide an important source of evidence for comparative-effectiveness research analyses.37 We performed such an analysis in CaPSURE, a large, national, community-based registry of men who were followed prospectively and uniformly from diagnosis regardless of treatment selection.
The current analysis produced evidence of significant differences in CSM and ACM across primary treatments when controlling for age, disease risk, and comorbidity. Especially striking was the progressive increase in differences across treatments with increasing risk (Fig. 2, Table 4). Mortality at 10 years was uncommon among men who had low-risk disease regardless of treatment; whereas, among the men with higher risk disease (in contrast to observed treatment trends26), men who underwent prostatectomy were much less likely to die than men who received external-beam radiation, and men in both local therapy groups had better survival than men who received androgen deprivation alone.
Several caveats should be considered with regard to the current analyses. CaPSURE practice sites are not a random sample of the US population. However, they represent a range of practice locations, sizes, and treatment patterns and do approximate the community prostate cancer patient experience in the United States.12 CaPSURE patients who reach mortality endpoints are more likely to have been diagnosed earlier, usually with a sextant biopsy; thus, their likelihood of clinical under staging is greater than would be expected for contemporary patients who undergo extended-template biopsies. Therefore, the mortality predictions from our analysis may be higher than might be expected in contemporary practice. It is possible that improvements in technique and outcomes among patients who receive radiation have been more pronounced over the past decade than those for patients who undergo surgery; however, we observed that, if anything, the differences in survival were greater when the analysis was restricted to a more contemporary cohort.
CaPSURE does not include consistent data on radiation dose and technique or on tertiary Gleason scores. There were insufficient events to control adequately for type and timing of salvage therapies, which are quite variable—reflecting inconsistent community practices in the face of little evidence-based guidance—and have been discussed in detail previously.38 In a recent report from a large academic cohort comparing prostatectomy with radiation under relatively uniform protocols, adjustment for salvage therapy had no impact on the outcomes of the analysis.39
Higher doses of radiation have been associated with a 12% improvement in recurrence-free survival,40 but it has not been demonstrated that they improve the likelihood of CSM or ACM.4 Likewise, it has not been established that variations in technique, such as intensity modulation, improve mortality. Variation in radiation practice seems unlikely to explain more than a fraction of the results of this analysis. Indeed, the academic series noted above included only radiation patients who received ≥81 gray. The results were concordant with the current study, with an approximately 3-fold reduction in case mix-adjusted rates of metastasis and prostate CSM in the surgery group.39 CaPSURE does include a large cohort of patients who received brachytherapy and active surveillance/watchful waiting. However, those patients generally were diagnosed in the more recent years of the registry, and their follow-up is not yet sufficiently mature to assess mortality.
Overall, 51% of the patients who received external-beam radiation in the current analysis received neoadjuvant and/or adjuvant androgen-deprivation therapy, a proportion similar to the 56% reported in the recent academic series noted above.39 In CaPSURE, the likelihood of receiving neoadjuvant therapy together with external-beam radiation for high-risk disease has increased steadily over time.6 Adjustment for neoadjuvant androgen deprivation with radiation therapy did not modify the results, probably because the use of neoadjuvant therapy in CaPSURE is closely associated with disease risk: because higher risk patients are much more likely to receive neoadjuvant therapy,6 the impact of neoadjuvant therapy is reflected in the risk adjustment, and, in a model adjusted for risk, the likelihood of neoadjuvant therapy is not an independent predictor. The mean duration of therapy was longer than that reported in the academic series (7.9 months in the current cohort vs 3 to 6 months in the academic cohort).39 A recent analysis of duration of neoadjuvant therapy indicated a relatively small difference (<5%) in cancer-specific survival for patients who received longer term therapy and an overall survival difference only among those with high-grade disease.41 Therefore, the longer duration of therapy among higher risk radiation patients in CaPSURE might be expected to improve outcomes.
To address the possibility that our results were affected by differences in death rates from causes other than prostate cancer, we used competing risks regression, which produced minimal changes in the findings. The attribution of CSM may not be accurate in all patients, particularly those ascertained from the National Death Index; however, the findings were observed both for CSM and for ACM and were robust to different considerations of risk and in several other sensitivity analyses. Finally, it is possible that other unmeasured confounding might explain some part of results. The Charlson score, for example, may not adequately reflect differences in comorbidity that may drive treatment decision-making. (A subset of CaPSURE patients have completed a more comprehensive comorbidity evaluation,42 but that group was too small for the current analysis.)
To evaluate the possibility that the limitations discussed—or other sources of unmeasured confounding—may explain the results, we artificially raised the Kattan scores for radical prostatectomy patients and observed that, in the risk-adjusted model, the benefit for surgery over radiation persisted until the scores for prostatectomy patients were increased by at least 20 points. In other words, the nomogram would need to systematically underestimate radiation patients' risk of progression relative to surgical patients' risk by 20 absolute percentage points; thus, a patient receiving radiation, for example, with a Gleason score of 3 + 3, PSA 4.0 ng/mL, and T1c tumor would have to have the same true risk as a surgical patient with a Gleason score of 3 + 4, PSA 9.0 ng/mL, and T2a tumor. The accuracy of the prediction model for predicting CSM is 80% across multiple treatments,20 and we cannot identify unmeasured confounders that would be expected to have such a large impact on risk-adjusted outcomes. The magnitude of the differences between treatments might be expected to vary with additional adjustment, but a qualitative change in the findings seems very unlikely. An additional strength of this analysis is that the Kattan and CAPRA scoring systems assign different relative weights to the various prognostic factors included, reducing the likelihood that the outcome of the model depends on the specific risk stratification system. In the study by Zelefsky et al, likewise, different considerations of risk did not substantially modify the outcomes.39
In a multi-institutional, prospective cohort of men with prostate cancer, we observed a low overall risk of CSM. After rigorous case-mix adjustment and multiple sensitivity analyses, however, we identified roughly 2-fold and 3-fold increases in the risk of cancer mortality among those who received external-beam radiation or primary androgen deprivation, respectively, compared with those who underwent radical prostatectomy, and the greatest differences were observed for higher risk patients. These findings should be verified with randomized trial data when available and with longer follow-up in CaPSURE and other large registries as more men ultimately reach mortality endpoints.
CONFLICT OF INTEREST DISCLOSURES
CaPSURE is supported in part by Abbott Laboratories (Abbott Park, Ill) and also is funded internally by the University of California, San Francisco Department of Urology. This work also was supported by National Institutes of Health/National Cancer Institute University of California-San Francisco Specialized Program of Research Excellence P50CA89520.