There is no lack of efforts to crack this very thorny and controversial nut. For this Commentary nineteen contemporary, peer reviewed studies from major institutions have been reviewed … resulting in a rather murky conclusion with conflicting strong claims for superior effectiveness for both modalities. Is an answer to this question necessary?  Clearly, yes!, since men facing treatment decisions want, need, and deserve credible information on this basic issue. Is such a comparison possible? “Aye, there’s the rub.” (Hamlet)

Comparisons between treatment regimens are fraught with many problems, some of which are listed here:

  • Controversy exists as to the choice of an appropriate, agreed-upon metric for comparison. This disagreement concerns using PSA failure at >0.2 ng.mL for surgery vs. the Phoenix definition for radiotherapy failure at a post treatment PSA nadir + 2 ng/ml. Surgeons contend this method gives an unwarranted advantage to radiation outcome.
  • The dosing and quality of radiation therapy in different studies can be different.
  • Unavoidable, unrecognized unevenness in the quality and experience of the treating physicians can influence outcome.
  • There can be significant disparity in the demographic composition of the groups being compared, notably patient age and comorbidity.
  • Unequal durations of ADT among studies contribute to incomparability.
  • And, yes, bias toward a preferred modality can unintentionally color the outcome.

But probably the most important and worrisome aspect of this effort to compare is that the results reported in these studies are the product of treatment techniques that are incrementally (fortunately) being improved over the course of the longer period of observation required in these trial to achieve a meaningful endpoint.

What really counts for men are:  the adverse effects of treatment (to be discussed in the December issue); the duration of freedom from metastases; and the rate of prostate cancer-specific survival, — best if collected over 5 and optimally 10 years of observation.

What Has Changed?

During the last 10 years or so radiation doses have been gradually increased, achieving superior cell kill; surgeons are becoming more expert with the robot; and new and more accurate imagining techniques are influencing management decisions. Information from multi-parametric MRIs combined with “targeted” biopsies is challenging the solidity of standard risk categories that have been based on conventional clinical parameters.

Genomic classifiers are parsing a cancer’s hidden biologic behavior and altering treatment decisions. Recent advances in systemic therapy, and those yet to emerge, are extending, bit by bit, survival for prostate cancer patients.

The older 3-D conformal radiotherapy (used at the beginning of some study periods) has been replaced and improved by intensity modulated (IMRT), or even image guided radiotherapy. External beam radiation combined with a brachytherapy boost seems to be on its way to supplant IMRT monotherapy in intermediate- and high-risk prostate cancer. And high-dose hypo-fractionated radiotherapy, even extreme hypo-fractionation, (i.e,“CyberKnife”) delivered over 5 days with their significantly greater cell killing power may supplant IMRT because of greater effectiveness and convenience.

The train of progress is leaving the station while we are looking back to see where we have been.

But an effort at comparisons, even considering all these important caveats, is still worth a shot. So here it goes.

Examples from the Reviewed Studies

The focus of most of these was upon intermediate- and high-risk prostate cancer, since it is generally conceded that in low-risk disease (which currently might be managed with active surveillance) there is little difference in the outcome between surgery and radiation.

It is important to note that all of these are retrospective studies from single-institutions or meta-analyses analyzing either organ confined or locally advanced disease. Their comparisons are based on prostate cancer-specific survival and overall survival.

This article does not analyze each of these studies individually but will summarize that there is no clear consensus as to which treatment is superior. It is becoming almost inappropriate to compare surgery with IMRT, as was done with most articles, since hypofractionated radiation and brachytherapy are emerging as more effective treatments, as was demonstrated in the ASCENDE trial outcome (control+click link to open or visit

The following are samples of the conclusions of a few of the important studies just to give a flavor of the findings:

1)    Boorjian et al. from the Mayo Clinic, Urologic Onology, Oct. 2014: “Although comparisons between surgery and radiation in the setting of high-risk disease are comprised of retrospective analysis subject to the potential for significant bias, nevertheless these observational studies consistently have reported favorable oncologic outcome with surgery as the primary treatment modality.”

2)    Zelefsky et al. from Sloan-Kettering, Journal of Clinical Oncology, March 2010: “Metastatic progression is infrequent in men with low-risk prostate cancer treated with either radical prostatectomy (RP) or external beam radiotherapy (EBRT). RP patients with higher-risk disease treated had a lower risk of metastatic progression and prostate cancer specific death than EBRT patients. These results may be confounded by difference in the use and timing of salvage therapy.”

3)    Crook, British Columbia Cancer Agency, Brachytherapy. 2015: “For intermediate- and high-risk prostate cancer, brachytherapy provides superior long-term oncologic and functional outcomes. … High-risk patient do very well with multimodality treatment combining external bean radiotherapy, a brachytherapy boost, and androgen deprivation for 9 – 12 months.”

4)    Kibel et al. from Harvard and Cleveland Clinic, Journal of Urology. 2012: (comparing RP, EBRT and BT): “After adjusting for major confounders, radical prostatectomy was associated with a small but statistically significant improvement in overall and cancer-specific survival [cancer death at 10 years: 1.8%, 2.9%, and 2.3%, respectively]. These survival differences may arise from an imbalance of confounders, difference in treatment related mortality and/or improved cancer control when radical prostatectomy is performed initial therapy.”

5)    Wallis et al. from Sunnybrook (Toronto) and Mayo Clinic, European Oncology. 2015: “Radiotherapy for prostate cancer is associated with an increased risk of overall and cancer-specific mortality compared to surgery, based on observations data with low to moderate risk of bias. This data, combined with the forthcoming randomized data, may aid clinical decision making.”

Results of the Much Awaited Randomized “ProtecT Trial”, … which has now “forth come.”

Published in the New England Journal of Medicine in September 2016, the study was prospectively randomized and designed to arbitrate the comparative effectiveness of prostatectomy, external beam radiotherapy, and active surveillance in PSA detected localized prostate cancer. The study ran from 1999 to 2009 and reported 10-year outcomes for freedom from metastases, prostate cancer-specific and overall survival. Participants were skewed to lower-risk disease: the median PSA was 4.6ng/mL; 77% had Gleason Score 6; 29% had disease spreading beyond the prostate (pT3), and 76% had non-palpable cancer. In all arms androgen suppression was initiated at PSA >20 and bone scans were performed at PSA >10 ng/mL.

The authors themselves pointed up the major limitations of the study: it was conceived 20 years ago; surgical techniques have changed; there was no use of multiparametric MRI; the less effective conformal 3D technique was used in the earlier period; brachytherapy was not included; and there was no sub-stratification between favorable and unfavorable Gleason 7 disease.


1)    At a median follow-up of 10 years there was no difference in overall survival among the groups.

2)    Prostate cancer-specific mortality in the surgery arm ( 391 men) occurred at a rate of .9 men per 1000 person-years of observation compared to .7 for men in the radiotherapy arm (401 men) and 1.2 (482 men) in the surveillance arm. The prostate cancer-specific survival in all groups was 98.8%.

3)    Metastases were diagnosed in the RP arm at a rate of 2.4 men per 1000 person/years; 3.0 in the radiotherapy arm, and at a rate of 6.3 per 1000 person years in the active surveillance arm.

Conclusion: “At a median of 10 years, prostate cancer-specific mortality was low [~1%] irrespective of treatment assigned, with no significant difference among treatments.”

[Because of the acknowledged limitations of the study, the complexity of its schema, and its focus on lower-risk disease, it is unlikely by itself to settle the thorny “comparison” issue, although this study does make a useful contribution.]

A Credible Effort at Analyzing the “Comparison” issue:  

“Radical Prostatectomy Versus Radiation and Androgen Deprivation Therapy for Clinically Localized Prostate Cancer: How Good is the Evidence?” by Mack Roach III, MD Professor, Departments of Radiation and Urology, UCSF, et al., Int J Radiation Oncology, Biology, Physics, Aug 2015.

The article begins with a now-familiar caveat: “The optimal treatment of clinically localized prostate cancer is controversial. Most studies focus on biochemical (PSA) failure when comparing radical prostatectomy (RP) and radiation therapy (RT), but this endpoint has not been validated as predictive of overall survival (OS) or cause-specific survival (CSS).” Roach’s study reviewed 14 observational reports and focused on 10-year OS and CSS (as did all the studies cited above).

Since no randomized trials were available, the authors constructed a “reliability” scale (RS; range 5 – 18, detailed in the article) that allowed them to assign a weight as to the quality of each of the 14 studies. They defended the scale by saying: “We would argue that our analysis systematically takes on a whole host of biases not accounted for by any off the studies cited.”  [However, this system can easily evoke criticism from the urologic community.]

What Were Their Findings?

1)    At the median reliability scale of 12, “the median difference in 10-year OS and CSS favored RP over RT: 10% and 4%, respectively.

2)    “For studies with a RS > 12 (average RS 15.5), the 10-year OS and CSS median differences were 5.5% and 1%, respectively.

Conclusion:  “Reliable evidence that RP provides a superior CSS to RT with ADT is lacking. The most reliable studies suggest that the difference in 10-year CSS between RP and RT are small, possibly <1%.”

[Although PSA recurrence after initial treatment is not a validated metric for CSS or OS, none the less, it serves as a useful marker for some important management decisions: i.e., for considering the need of “salvage” radiotherapy following surgery; prompting a search for the source of the rising PSA with a view toward additional focal treatment; or as an alert to the possible need for hormonal intervention.]

BOTTOM LINE: Is This Controversial Issue Settled?  

Certainly not!  Improvements in techniques are emerging in all modalities of prostate cancer treatment that will influence outcome. Past studies have offered no clear consensus that radiation or surgery is the most effective management. However, in a very pertinent and important way these past studies provide only suggestive background data for men facing treatment decisions. Men are not being treated by a ‘collection’ of surgeons or radiation therapists, and clearly not by a ‘meta-analysis’ of physicians. Men are choosing or consulting with one particular practitioner equipped with his/her own unique skills and technique. This is where the rubber really meets the road. It is incumbent upon the patient to inquire about the outcomes in the men already treated by the physician he is facing in the consultation room. What is his experience history and outcome record? It is equally essential that the treating MD has collected his data and can discuss it with the patient. These must be components of an informed discussion …

… and this discussion of necessity involves the issue of the adverse effects of treatment on a man’s quality of life exacted by the treatment under consideration. This will be the subject of the December 2016 PCa Commentary.