RRP vs. RT:comparative analyses
Zelefsky and colleagues from Memorial Sloan-Kettering Cancer Center recently performed a retrospective comparison of patients with cT1-T3b PCa treated at their center by highly-experienced clinicians with intensity-modulated external beam RT (n=1,062) or RRP (n=1,318).13 All RT patients received a dose level of ≥81Gy, and only a minority of RRP patients received postoperative RT (6%) or adjuvant ADT (1%).13 These investigators found that the eight-year probability of freedom from metastatic progression was 97% for RRP patients and 93% for RT patients.13
Moreover, after controlling for clinicopathologic variables, treatment with RRP was associated with a reduced risk of metastases (HR 0.35; p<0.001) and cancer-specific mortality (HR 0.32; p=0.015).13 The improvement in metastatic progression was more substantial for patients with unfavorable-risk disease (7.8% difference at eight years) than for patients with favorable-risk disease (1.9% difference at eight years).13
Importantly, however, high-risk tumors accounted for only 17% of the overall cohort.13 Furthermore, adjuvant androgen deprivation therapy (ADT) was not routinely used in the high-risk patients treated with RT, despite randomized clinical trial data which has demonstrated a benefit to long-term (i.e., 24-36 months) adjuvant ADT + RT for patients with high-risk PCa.14-19
To assess the comparative outcomes of surgery and radiation specifically for men with high-risk PCa, the Mayo Clinic Prostatectomy Registry and the Fox Chase Cancer Center (FCCC) Radiation Oncology Database were reviewed to identify 1,847 patients with high-risk PCa treated with definitive local therapy in the form of RRP (n=1,238), EBRT alone (n=265), or external-beam RT with long-term (median 22.8 months) adjuvant ADT (n=344) between 1988-2004.20
For this analysis, high-risk disease was defined according to NCCN guidelines21 as: PSA ≥20 ng/mL or clinical stage ≥ T3N0M0 or biopsy Gleason score 8-10. Median follow-up was 10.2 years after RRP, 6.0 years after RT + ADT, and 7.2 years after RT alone. These investigators found that, when controlling for case mix and patient variables, patients treated with RT + ADT had a significantly increased risk of all-cause mortality compared to patients who underwent RRP (HR 1.60, p=0.0002).20 Indeed, the 10-year overall survival after RRP was 77%, versus 67% following RT + ADT and 52% after RT alone.20
One potential explanation for this finding is that ADT, which was given more frequently to patients receiving RT than patients treated with RRP, adversely impacted men treated with RT. ADT has been associated with an increased risk of cardiac death, particularly in men with coronary artery disease.22 While the exact mechanism of this interaction is unknown, the metabolic effects of ADT may contribute to the comorbidities and risk factors for cardiac death such as diabetes, hypercholesterolemia, and hypertension.23
One potential benefit of surgery versus RT for patients with high-risk PCa is therefore the ability to obtain pathological staging, which, as has been suggested previously,24 may guide the selective application of secondary therapies, and may, for example, delay or avoid the need for ADT. Indeed, Meng et al found that patients with high-risk PCa treated with radiation therapy were 3.5 times more likely to receive ADT than patients treated with RRP.25
As increasing data have emerged on the adverse consequences of ADT on the quality of life3 and non-cancer morbidity26 of men with PCa, the ability to delay if not avoid ADT may represent a potential advantage to surgery for these patients. RRP affords the ability to facilitate the selective application of secondary therapies in patients confirmed to have pathologically aggressive disease features.27-29 As several independent series24,30 have demonstrated, up to 55% of patients classified as having high-risk disease are in fact found to have organ-confined tumors at surgery.
These men may not therefore require additional therapy and may be spared the cost and potential side effects of secondary treatment. The median time from RRP to salvage ADT in the comparative Mayo Clinic/FCCC series was 10.3 years.20 Although the timing of salvage ADT was at the discretion of the surgeon and is not an indicator of treatment success, this suggests that high-risk patients treated with surgery may have long intervals of ADT-free survival.
Assessing the comparative outcomes after surgery and RT for patients with high-risk disease remains particularly clinically relevant, for although over the course of the PSA era the proportion of newly-diagnosed PCa patients who would be characterized as having high-risk disease has declined, nevertheless up to 15% of patients continue to present with high-risk tumor features.30,31
In fact, the management of patients with high-risk cancers represents one of the biggest challenges in PCa today, with little consensus on the optimal treatment. A trend toward increasing use of RT and ADT, with a corresponding lower application of RRP, has been noted with increasing PCa disease risk.31
Interestingly, however, the previous retrospective series comparing the outcomes after surgery and radiation for high-risk tumors demonstrated widely disparate results, with several reporting improved outcomes following RRP,32-36 others finding better results following radiation,37,38 and a few,39-41 including a small prospective trial,42 noting equivalent efficacy. However, these studies involved different definitions of high-risk cancer, evaluated disparate outcome measures, and included a relatively limited number of patients, often with short-term follow-up. Therefore, it remains important, as in the recent studies reviewed here,13,20 to evaluate robust datasets with long-term follow-up to assess the impact of treatment on the endpoint of mortality.
Concerns exist regarding the ability to extrapolate results from tertiary referral centers and other treatment settings. Thus, it is important to review outcomes from population-based datasets, which have evaluated comparative survival following surgery and radiation. For example, Cooperberg et al recently analyzed treatment outcomes from the CaPSURE dataset, a national disease registry that is primarily community-based.35
After adjusting for patient age and disease risk, these investigators noted an increased cancer-specific mortality after radiation versus surgery, such that the hazard ratio for death from PCa after RT relative to RRP was 2.21.35 Again, the magnitude of improvement in survival for surgery versus RT increased with increasing disease risk.35
Two other recent series8,43 evaluated comparative outcomes following RRP and RT using the population-based Surveillance, Epidemiology, and End Results (SEER) database, which covers approximately 26% of the U.S. population. Specifically, Liu and colleagues, using SEER-Medicare linked data, identified 5,845 men diagnosed with local/regional PCa at age 65-74 in 1992, with a comorbidity score < 2, who they defined as potential candidates for RRP.8 Of these patients, 2,567 underwent RRP, 2,006 received RT, 302 underwent RRP plus RT, and 970 were managed with watchful waiting.8 The authors found that the 10-year cancer-specific survival was significantly better following RRP (98.1%) than RT (93.8%, p<0.0001).8
Likewise, the 10-year all-cause survival was also higher after RRP (81%) than RT (60.5%, p<0.0001).8 Not surprisingly, men who underwent RRP had significant differences with regard to tumor stage, Gleason score, and comorbidity status compared with men treated with RT.8 Nevertheless, on multivariate analyses controlling for these factors, the hazard ratio for the risk of death from PCa among patients who underwent surgery compared with watchful waiting was 0.17 (95% CI 0.10, 0.28), versus 0.56 (95% CI 0.37, 0.85) for patients who received RT.8
A separate study by Abdollah et al likewise evaluated PCa mortality and other-cause mortality following RRP, RT, and observation using the SEER dataset, from 1988-2006.43 Using a competing-risks survival analysis in 404,604 patients with clinically-localized disease, these investigators found that, for men with low- to intermediate-risk PCa, the lowest cancer-specific and overall mortality rates for men < 80 years old were recorded in RRP patients.43
Moreover, for men with high-risk disease who were < 70 years old, RRP was again associated with the lowest cancer-specific and overall mortality rates.43 For example, the 10-year rate of death from prostate cancer for a man 60-69 years old with a high risk tumor was 7.2% following surgery, compared to 11.3% after RT (p<0.001).43 Interestingly, however, the authors also noted that, between 1988 and 2006, the number of patients treated with RRP decreased from 57.4% to 38.3%, while more patients were treated with RT (28.8-38.1%).43 No meaningful differences were identified for treatment choice according to PCa risk category, suggesting that tumor variables were not the main determinant of treatment modality in this dataset.43 Further investigation into the factors which are responsible for patients’ decision-making with regard to prostate cancer treatment is therefore needed.