Radical Prostatectomy Versus Radiotherapy for Prostate Cancer

Stage-, age-, and frailty-specific cancer-control outcomes of 2600 patients

Background: Both radical prostatectomy (RP) and radiotherapy (RT) are recommended as standard treatments for prostate cancer. The prospective comparisons available to date provide only limited information.

Methods: We used data from the database of our university cancer center to compare the metastasis-free (MFS), cancer-specific (CSS) and overall survival (OS) of all patients with prostate cancer who underwent either RP or RT in the period 2014–2024. Stage-, age-, and frailty-specific sensitivity analyses were carried out.

Results: Of 2685 patients with prostate cancer, 1999 (74%) underwent RP and 686 (26%) underwent RT. The RP patients were younger (66 vs. 74 years); a higher percentage of the RP patients than of the RT patients had high-risk prostate cancer (60% vs. 43%), stage cT3 (47% vs. 9.6%), and stage cN1 (11 vs. 5.2%), while the RT patients more commonly had ECOG status 1–2 (16% vs. 6.7%; p<0.001 for all comparisons). Univariate analyses of MFS mostly favored RT, while univariate analyses of OS mostly favored RP. These differences, however, were no longer seen after adjustment for patient and tumor characteristics in multivariable Cox regression models, nor were they seen in sensitivity analyses of D’Amico risk groups, age categories, or ECOG status. Lastly, in 2:1 propensity-score-matched analyses, no differences between RP and RT were found in any of the oncological outcome measures (p≥0.15 for MFS, CSS, and OS).

Conclusion: The findings of this real-world study of prostate cancer patients who underwent either RP or RT suggest equally effective cancer control by the two methods when the statistical analysis is adjusted for patient and tumor characteristics. At present, patients with high-risk prostate cancer und unfavorable disease stages more commonly undergo RP.

Cite this as: Wenzel M, Burdenski K, Tselis N, Rödel C, Brandts C, Ahrens M, Koellermann J, Graefen M, Humke C, Siech C, Hoeh B, Banek S, Chun FKH, Mandel P: Radical prostatectomy versus radiotherapy for prostate cancer: Stage-, age-, and frailty-specific cancer-control outcomes of 2600 patients. Dtsch Arztebl Int 2025; 122: 495–500. DOI: 10.3238/arztebl.m2025.0089

Prostate cancer (PCa) is the most commonly occurring form of cancer in Europe and the second-ranked cancer worldwide. The decision as to the most appropriate treatment for PCa is often challenging (1, 2). Radical prostatectomy (RP) and radiation therapy (RT) are well established and guideline-recommended treatments for all localized and locally advanced stages of PCa (2). However, the two treatments differ regarding potential adverse effects and may not always be offered as equivalent treatment options to all patients.

The currently available prospective data comparing cancer-control outcomes of RP versus RT for PCa stem predominantly from patients with low/intermediate risk or from studies comparing functional outcomes (3, 4). Therefore, most of the pertinent literature relies on large retrospective databases such as the American National Cancer Database or SEER, in which studies show varying results for comparisons of RP versus RT, but usually similar or better survival rates with RP (5, 6, 7, 8, 9). However, the principal limitation of such epidemiological databases is the overall quality of registry data in the USA, with differences in hospitals’ surgical volume, surgeons’ experience, or the surgical procedure used for RP, together with differences in the availability and technical refinement of RT. Therefore, these epidemiological data reflect merely the average oncological outcome quality of RP versus RT in North American PCa patients across rural areas and urban regions, not the surgical RP or the advanced RT techniques currently used for treatment of European PCa patients. Setting out to address this gap in knowledge, we used the database of the university cancer center (UCT) of a German tertiary-care hospital to compare cancer-control outcomes such as metastasis-free survival (MFS), cancer-specific survival (CSS), and overall survival (OS) between RP and RT in patients with PCa. We hypothesized that differences in oncological outcomes may exist between the two treatment types regarding stage-, age-, and frailty-specific analyses and that some subgroups of patients may benefit more from one treatment than the other.

Material and methods

Study population

The patient data for this study were provided by UCT Frankfurt. The study was approved by the institutional review boards of the UCT and University Hospital Frankfurt (project number: SUG-4–2024). Retrospective analyses were performed of all PCa patients treated curatively at the departments of Urology or Radiation Oncology of University Hospital Frankfurt, Germany between 2014 and 2024. In all cases, the PCa treatment was recommended by a multidisciplinary tumor board. Patients with metastatic disease and those with unknown primary treatment were excluded. The treatment decision was based on patient preference, comorbidities, age, and physicians’ recommendation. Application of these criteria resulted in a total cohort of 2685 PCa patients.

Radical prostatectomy

RP was performed using either an open or a robot-assisted approach, depending on the patients’ comorbidities, tumor characteristics, stage, and preferences. Lymph node dissection was performed in accordance with the recommendations of the EAU guidelines. Other factors with a potential influence on oncological outcomes, such as nerve-sparing surgical techniques were performed as previously described (10, 11, 12).

Radiation therapy

RT was delivered as either external beam radiation therapy (EBRT), brachytherapy, or a combination of both, depending on patients’ comorbidities, tumor characteristics, stage, and preferences, in accordance with the version of the EAU-ESTRO Guidelines prevailing in the year of treatment. RT to the lymph drainage pathways and concomitant androgen deprivation therapy for primary or adjuvant/salvage purposes also adhered to these guidelines (2).

Statistical analyses

Kaplan–Meier curves were constructed for MFS, OS, and CSS outcomes in all RP versus RT patients. Sensitivity analyses stratified outcomes by stage (D’Amico low, intermediate, high risk), age (< 65, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, > 75 years), and frailty (Eastern Cooperative Oncology Group Performance Status [ECOG] 0 versus 1–2). Due to low event numbers, CSS was analyzed only in the overall cohort. All figures include censoring information. Univariable and multivariable Cox regression models were applied, adjusting for baseline patient (age, ECOG) and tumor characteristics (PSA, cT stage, cN stage, biopsy Gleason score). The OS/CSS follow-up was obtained through queries to the cancer registry, ensuring complete recording of vital status. Metastatic progression is largely captured due to the minimum documentation requirements for certified urological centers. The duration of follow-up is determined not only by overall survival but also by the time elapsed since primary treatment.

Finally, to validate all findings, 2:1 propensity score matching (PSM) was performed, matching for age, PSA, D’Amico risk category, Gleason score, cT stage, cN stage, and ECOG status. Analyses were conducted using R (version 3.4.3).

Results

Baseline characteristics

Of the total 2685 patients with PCa included, 1999 (74%) underwent RP and 686 (26%), RT. The RP patients were younger (66 versus 74 years) and had lower PSA (7.8 versus 8.7 ng/mL) than the RT patients (Table 1). Moreover, RT patients were more often classified as ECOG 1–2 (16% versus 6.7%). Conversely, patients undergoing RP were more frequently classified as high-risk PCa (60% versus 43%), cT3 stage (47% versus 9.6%) or cN1 stage (11% versus 5.2%). The Gleason score distribution did not differ. The median duration of follow-up was 21 months.

Table 1

Patient characteristics of 2685 patients with prostate cancer stratified by radical prostatectomy (RP) or radiation therapy (RT)

Cancer-control outcomes: RP versus RT

In MFS analyses, differences between RP versus RT were observed (Figure 1a), with a hazard ratio (HR) of 0.28, 95% confidence interval [0.16; 0.50] and 60-month MFS rates of 88.2% versus 96.2%. However, after multivariable adjustment for tumor and patient characteristics, no differences were observed (HR 0.71 [0.33; 1.50], p = 0.4; Table 2).

Figure 1

Kaplan–Meier curves

Table 2

Univariable and multivariable Cox regression models predicting metastasis-free survival, cancer-specific survival, and overall survival for radical prostatectomy (reference) versus radiation therapy

Also in CSS analyses, no differences were observed between RP and RT (Figure 1b), with 60-month CSS rates of 98.8% versus 98.8% (HR: 1.06 [0.40; 2.79]). After multivariable adjustment, still no differences were found (HR 1.91 [0.48; 7.61], p = 0.4; Table 2).

OS analyses showed differences between RP and RT (Figure 1c), with a HR of 1.47 [1.08; 1.99] and 60-month OS rates of 88.2% versus 83.8% for RP and RT respectively. After additional multivariable adjustment, however, no differences were observed (HR 1.19 [0.76; 1.87], p = 0.4; Table 2).

Cancer-control outcomes by risk category

In analyses of RP-treated versus RT-treated patients stratified according to risk groups, no differences in MFS and OS rates were observed for low risk (eFigure 1a, 1d) and intermediate risk (eFigure 1b, 1e). There were also no differences after multivariable adjustment.

For high-risk PCa (eFigure 1c, 1f), MFS differences were observed between RP and RT with a HR of 0.37 [0.18; 0.65] and 60-month MFS of 83.3% versus 93.4% for RP and RT respectively. Conversely, the 60-month OS rates were 86.0% versus 77.2% with a corresponding HR of 1.90 [1.31; 2.75]. However, the multivariable Cox regression models showed no differences in OS and MFS (both p ≥ 0.3).

Cancer-control outcomes by age category

In age-stratified analyses of RP-treated versus RT-treated PCa patients, no differences in MFS or OS were observed for patients aged < 60 years (eFigure 2a, 2d). In patients aged > 60–75 years, however, there were differences regarding MFS, with a HR of 0.18 [0.06; 0.48] and 60-month MFS rates of 88.5% versus 97.6% for RP and RT respectively, despite the absence of differences in OS (eFigure 2b, 2e).

No differences were observed regarding MFS and OS in patients aged > 75 years. As in the analyses described above, multivariable adjustment also revealed no differences in any age-stratified analyses.

Cancer-control outcomes by ECOG performance status

In ECOG 0 patients, MFS differences were observed between RP and RT, with a HR of 0.40 [0.21; 0.78] and 60-month MFS of 91.2% versus 95.4% (eFigure 3a). Likewise, a HR of 0.06 [0.01; 0.47] was recorded in ECOG 1–2 patients, with 60-month MFS of 72.7% versus 100% (eFigure 3b). However, no MFS differences were observed after multivariable adjustment. In the OS analyses, no differences between RP and RT patients were observed for ECOG 0 or ECOG 1–2 (eFigure 3c/3d).

PSM analyses: RP versus RT

Finally, 2:1 PSM analyses (eFigure 4, eTable), matched for patient and tumor characteristics in order to achieve maximal adjustment for differences in baseline characteristics, compared the cancer-control outcomes between RP and RT. These analyses revealed no differences in MFS, CSS, and OS between RP and RT patients with 60-month rates of 95.4% versus 96.1%, 99.1% versus 98.6%, and 91.8% versus 91.5% for MFS, CSS, and OS respectively (Figure 2).

Figure 2

Kaplan–Meier curves

eTable

Patient characteristics of 1374 patients with prostate cancer stratified by radical prostatectomy (RP) or radiation therapy (RT) after propensityscore matching

Discussion

The study presented here used routine clinical data to compare cancer-control outcomes in patients with PCa treated with of RP versus RT and to identify any relevant differences regarding stage-, age- and frailty-specific subgroups. For this, we relied on the UCT database of a German tertiary-care hospital. Several important observations were made.

First, we observed substantial differences in patient and tumor characteristics between the RP and RT patients. For example, the median age of RT patients was 8 years higher (74 versus 66 years) than RP patients, consistent with previous studies. A recently published study based on the American SEER database reported a 7-year median age difference between RP and RT patients (64 versus 71 years) (6). The findings of an Italian report of registry data were similar (65 versus 71 years) (13). One reason why younger patients tend to be treated with RP may be that younger patients have a longer residual life span with the risk of biochemical/local recurrence after primary treatment, and second-line salvage RT is associated with fewer complications than salvage RP—especially given the poor functional outcome of salvage RP (14). Moreover, RT patients exhibited more comorbidities, based on the proportion of patients with ECOG performance status 1–2 relative to those treated with RP.

Conversely, RP patients had worse primary tumor properties, such as higher rates of cT3–4 or cN1 stage or high-risk PCa. This observation differs from an Italian report of higher rates of high-risk PCa in the RT group than in the RP group (13). Such differences may reflect regional variations in treatment algorithms. This hypothesis is supported by a previous report showing differences in treatment distribution between RP and RT even within Germany (15).

Second, in the overall cohort, RT patients showed better MFS outcomes than RP patients. This trend was also seen in subgroups of high-risk PCa, patients aged > 60–75 years, and patients with ECOG 1–2 performance status. Nevertheless, both treatments yielded high 60-month MFS rates (88.2% for RP versus 96.2% for RT). The lower MFS rate for RP may be due to the initially more unfavorable tumor characteristics, such as higher rates of cT3 or cN1 tumors. Moreover, due to their lower age at diagnosis, RP patients may have longer life expectancy and therefore more time to develop metastasis, while the RT patients tended to have more comorbidities and may have died due to causes other than PCa. This hypothesis is consistent with findings from two merged prospective trials presented at ASCO GU 2025, in which more favorable MFS outcomes for RT were computed in high risk PCa (16).

The results of the multivariable adjusted analyses support the hypothesis that differences in MFS can be explained by disparities in baseline characteristics, as no differences in MFS remained after multivariable adjustment—neither in the entire cohort nor in subgroup analyses. This was supported by the PSM analyses, which allowed an even more nuanced adjustment to baseline patient and tumor characteristics. Here, we also observed no difference in MFS, with 60-month MFS rates of 95.4% for RP versus 96.1% for RT. Comparing these results with previous reports, similarities can be found: for example, a recently published systematic review also shows no MFS difference between RP- and RT-treated PCa patients (17). The low long-term 10-year MFS rates in that review (2.3% versus 2.9%) may be explained largely by the inclusion of mostly low- and intermediate-risk patients.

Finally, important findings were also made with regard to survival outcomes, with no differences in the overall cohort concerning CSS , but better OS rates for RP than for RT. These OS results also appeared in subgroup analyses of high-risk patients. Nevertheless, 60-month CSS and OS rates (98.8% versus 98.8% and 88.2% versus 83.8%) were very high for both groups. Moreover, OS disadvantages in RT patients probably reflect the more unfavorable baseline characteristics (higher age, higher rates of ECOG performance status 1–2) or comorbidities such as cardiovascular disease, which may translate into higher rates of non-cancer mortality. This assumption is supported by the observation of identical CSS rates in the two groups and the absence of differences after multivariable adjustment. To further validate these findings we used a PSM procedure, which also revealed no difference in CSS and OS rates between RP versus RT. These findings are in agreement with several previously published reports. Two systematic reviews and meta-analyses also found no differences in OS and CSS rates between RP and RT (17,18). Furthermore, clinical and registry-based studies reported a CSS and OS advantage of RP over RT in univariable models, which similarly disappeared after adjustment for baseline characteristics and tumor properties or in PSM analyses. This trend was also seen in subgroup analyses of high-risk PCa or locally advanced tumor stages (5, 8, 19, 20, 21, 22, 23, 24, 25). Moreover, several studies support these findings by providing higher non-cancer mortality rates based on concomitant comorbidities in RT patients (26, 27, 28). It thus appears that the observed differences in OS are more likely to be explained by a selection effect for treatment than by a superiority of one form of treatment. Comparable findings were observed in the ProtecT trial, where RP and RT showed similar cancer-control outcomes, although the study was not designed for direct comparison (3).

The retrospective, single-center design of our study imposes limitations. Unmeasured confounders such as RT modality, adjuvant/salvage therapies, and ADT may have influenced the outcomes, as may recent technical advances, e.g., changes in staging modalities. Moreover, introduction of a certain selection bias by physicians cannot be ruled out. Follow-up is censored for MFS and OS analyses if the primary treatment took place not long ago and the observation time is therefore short. Finally, unfortunately no data were available on complications or quality of life.

In conclusion, this study of over 2600 PCa patients at a German tertiary-care hospital has provided important information about the basic characteristics of RP and RT patients and on their cancer-control outcomes. Specifically, RP patients are younger and have a better ECOG performance status than RT patients. Conversely, RP patients have less favorable tumor properties, resulting in higher rates of high-risk PCa. After application of statistical methods such as adjustments or PSM, however, all analyses showed no differences between the two treatment modalities in the cancer-control outcomes MFS, CSS and OS. These findings indicate that RP and RT can be equally effective at a specialized center. Therefore, the decision between RP and RT should be made predominantly on the basis of patient characteristics, comorbidities, age, and patients’ preferences, as well as the expected functional results. Nonetheless, randomized trials are needed to provide further evidence in the oncological comparison between RP and RT.

Acknowledgments
This study was part of the EPIC-REAP project (Enhancing Prostate Cancer Care In Germany Combining Real-World Data and AI for Enhanced Analysis and Precision) supported by the Mildred Scheel Center for Young Scientists (Mildred-Scheel Nachwuchszentrum) Frankfurt.

Conflict of interest statement

The authors declare that no conflict of interest exists.

Manuscript received on 19 October 2024, revised version accepted on 13 May 2025

Corresponding author
PD Dr. med. Mike Wenzel

References

1. European Comission: https://ecis.jrc.ec.europa.eu/sites/default/files/2023-12/prostate_cancer_En-Nov_2021.pdf (last accessed on 25 July 2025).

2. Cornford P, Tilki D, van den Bergh RCN et al.: EAU guidelines on prostate cancer. Edn. presented at the EAU annual congress Paris 2024.

3. Hamdy FC, Donovan JL, Lane JA, et al.: ProtecT Study Group. Fifteen-year outcomes after monitoring, surgery, or radiotherapy for prostate cancer. N Engl J Med 2023; 388: 1547–58

4. van As N, Yasar B, Griffin C, et al.: Radical prostatectomy versus stereotactic radiotherapy for clinically localised prostate cancer: Results of the PACE-A randomised trial. Eur Urol 2024; 86: 566–76

5. Chierigo F, Wenzel M, Würnschimmel C, et al.: Survival after radical prostatectomy versus radiation therapy in high-risk and very high-risk prostate cancer. J Urol 2022; 207: 375–84

6. Chierigo F, Borghesi M, Würnschimmel C, et al.: Survival after radical prostatectomy versus radiation therapy in clinical node-positive prostate cancer. Prostate 2022; 82: 740–50

7. Reitblat C, Fleishman A, Kaplan IA, et al.: Radical prostatectomy versus external beam radiation therapy for high-grade, clinically localized prostate cancer: Emulation of a target clinical trial. Urol Oncol 2021; 39: 785.e1–e10

8. Knipper S, Palumbo C, Pecoraro A, et al.: Survival outcomes of radical prostatectomy vs. external beam radiation therapy in prostate cancer patients with Gleason Score 9–10 at biopsy: A population-based analysis. Urol Oncol 2020; 38: 79.e9–79.e14. doi:10.1016/j.urolonc.2019.09.01

9. Goy BW, Burchette R, Soper MS, Chang T, Cosmatos HA: Ten-year treatment outcomes of radical prostatectomy vs external beam radiation therapy vs brachytherapy for 1503 patients with intermediate-risk prostate cancer. Urology 2020; 136: 180–9

10. Preisser F, Theissen L, Wild P, et al.: Implementation of Intraoperative frozen section during radical prostatectomy: Short-term results from a German tertiary-care center. Eur Urol Focus 2021: 95–101

11. Wenzel M, Preisser F, Theissen LH, et al.: The effect of adverse patient characteristics on perioperative outcomes in open and robot-assisted radical prostatectomy. Front Surg 2020; 7: 584897

12. Hoeh B, Wenzel M, Hohenhorst L, et al.: Anatomical fundamentals and current surgical knowledge of prostate anatomy related to functional and oncological outcomes for robotic-assisted radical prostatectomy. Front Surg 2021; 8: 825183

13. Antonelli A, Palumbo C, Noale M, et al.: Overview of potential determinants of radical prostatectomy versus radiation therapy in management of clinically localized prostate cancer: Results from an Italian, prospective, observational study (the Pros-IT CNR study). Minerva Urol Nefrol 2020; 72: 595–604

14. Moschovas MC, Bravi CA, Dell’Oglio P, et al.: Outcomes of salvage robotic-assisted radical prostatectomy in the last decade: Systematic review and perspectives of referral centers. Int Braz J Urol 2023; 49: 677–87

15. Medenwald D, Ferencz J, Vordermark D: Predictors of the regional variation of prostatectomy or radiotherapy: Evidence from German cancer registries. J Cancer Res Clin Oncol 2020; 146: 1197–1204

16. Roy S, Sun Y, Estham JA, et al.: Radical prostatectomy (RP) versus radiotherapy (RT) in high-risk prostate cancer (HR-PCa): Emulated randomized comparison with individual patient data (IPD) from two phase III randomized trials (RCTs). https://ascopubs.org/doi/10.1200/JCO.2025.43.5_suppl.309 (last accessed on 27 July 2025)

17. Hekman L, Barrett A, Ross D, et al.: A systematic review of clinical trials comparing radiation therapy versus radical prostatectomy in prostate cancer. Clin Genitourin Cancer 2024; 22: 102157

18. Aydh A, Motlagh RS, Abufaraj M, et al.: Radiation therapy compared to radical prostatectomy as first-line definitive therapy for patients with high-risk localised prostate cancer: An updated systematic review and meta-analysis. Arab J Urol 2022; 20: 71–80

19. Bandini M, Marchioni M, Preisser F, et al.: Survival after radical prostatectomy or radiotherapy for locally advanced (cT3) prostate cancer. World J Urol 2018; 36: 1399–407

20. Berg S, Cole AP, Krimphove MJ, et al.: Comparative effectiveness of radical prostatectomy versus external beam radiation therapy plus brachytherapy in patients with high-risk localized prostate cancer. Eur Urol 2019; 75: 552–5

21. Hayashi N, Osaka K, Muraoka K, et al.: Outcomes of treatment for localized prostate cancer in a single institution: comparison of radical prostatectomy and radiation therapy by propensity score matching analysis. World J Urol 2020; 38: 2477–84

22. Bandini M, Pompe RS, Marchioni M, et al.: Radical prostatectomy or radiotherapy reduce prostate cancer mortality in elderly patients: A population-based propensity score adjusted analysis. World J Urol 2018; 36: 7–13

23. Hoeh B, Hohenhorst JL, Flammia R, et al.: Cancer-specific mortality after radical prostatectomy vs external beam radiotherapy in high-risk Hispanic/Latino prostate cancer patients. Int Urol Nephrol 2022; 54: 81–7

24. Wenzel M, Würnschimmel C, Chierigo F, et al.: Pattern of biopsy Gleason Grade Group 5 (4 + 5 vs 5 + 4 vs 5 + 5) predicts survival after radical prostatectomy or external beam radiation therapy. Eur Urol Focus 2022; 8: 710–7

25. Würnschimmel C, Wenzel M, Chierigo F, et al.: Presence of biopsy Gleason pattern 5 + 3 is associated with higher mortality after radical prostatectomy but not after external beam radiotherapy compared to other Gleason Grade Group IV patterns. Prostate 2021; 81: 778–84

26. Knipper S, Pecoraro A, Palumbo C, et al.: A 25-year period analysis of other-cause mortality in localized prostate cancer. Clin Genitourin Cancer 2019; 17: 395–401

27. Wenzel M, Würnschimmel C, Chierigo F, et al.: Non-cancer mortality in elderly prostate cancer patients treated with combination of radical prostatectomy and external beam radiation therapy. Prostate 2021; 81: 728–35

28. Würnschimmel C, Wenzel M, Chierigo F, et al.: Radiation therapy after radical prostatectomy is associated with higher other-cause mortality. Cancer Causes Control 2022; 33: 769–77