Men with newly diagnosed localized prostate cancer face a decision between prostatectomy and radiotherapy, treatments deemed similarly effective but with well-established trade-offs in terms of treatment-related morbidity. Numerous clinical trials and other prospective studies, from both academic and community settings, have consistently demonstrated that prostatectomy causes more urinary incontinence and erectile dysfunction,1,2 with a 0.5% risk for perioperative mortality;3 whereas radiation causes more urinary irritation and bowel/rectal symptoms,2,4 with a small risk of second malignancy.5
Both treatments carry a risk for cancer recurrence. It is also true that older patients, with more underlying comorbid conditions and with more aggressive cancers, tend to receive radiation therapy, whereas younger, healthier patients with less aggressive cancers tend to receive prostatectomy.6 The study by Nam et al7—reviewed in this issue of The ASCO Post—must be interpreted with caution against the background of this well-established knowledge.
Key Bias
First, the reader should notice that this study selectively excluded the most common consequences of radical prostatectomy—namely, urinary incontinence, erectile dysfunction, and cancer recurrence (by eliminating all patients who received radiotherapy after surgery). The latter is especially important.
When a patient is deciding between prostatectomy and radiotherapy, he and his physician are unable to predict whether postprostatectomy radiotherapy will ultimately be needed. One-third or more of those who undergo a radical prostatectomy will have an indication for adjuvant or salvage radiotherapy.8 Eliminating patients who require postprostatectomy treatment, and the associated effects from these treatments, thus creates an artificial study cohort that does not reflect clinical reality or provide meaningful guidance for treatment decision-making.
In contrast, the study specifically includes postradiotherapy prostate biopsies as “events”; these are actually diagnostic procedures to determine whether a local recurrence has occurred in radiotherapy patients. When a comparative morbidity study selectively disregards the most common consequences of one treatment option, as has occurred here, the other option will inevitably appear to have higher rates of the measured outcomes.
Canadian Context
The outcome of “hospital admissions” in this article needs to be viewed in context of the health-care environment in Ontario. The reader may not be aware, but it is relatively common practice in Ontario to admit patients one night for procedures usually performed on an outpatient basis in the United States. Importantly for this article, this includes routine colonoscopies.
At first glance, the finding that one-third of radiotherapy patients experienced “hospitalization” as a result of treatment-related morbidity (Figure A in the article) contradicts the clinical experience of most clinicians who treat prostate cancer. Not surprisingly, if one counts only admissions that lasted more than 1 day, the number of events dropped almost 10-fold to something more realistic and recognizable.
As each admission must be associated with a “diagnosis,” it would be helpful to know how diagnosis is coded for the large number of 1-day, procedure-only admissions. A validation study using chart review of a select number of these patients would have provided significant insight to address this fundamental methodologic problem.
Other Confounding Factors
Another important methodologic issue that must be addressed in all observational studies, and that is especially pertinent for studies comparing surgery and radiation, is that of patient selection and confounding. Patients who are young and healthy enough to receive radical prostatectomy can, in fact, have better “outcomes” than the general noncancer population!
As reported by Nam et al,7 the standardized incidence ratio for development of a second malignancy comparing radical prostatectomy patients to noncancer controls was 0.8. On the other hand, the vast majority of second cancers after prostate radiotherapy were in body areas not directly in the treatment field, including lung, head/neck, central nervous system, and breast. It is difficult to conclude that a radical prostatectomy reduced a patient’s risk for second cancers by 20%, and similarly difficult to believe that prostate radiotherapy caused numerous lung cancers as early as 5 years after treatment. Rather, these results strongly suggest unbalanced patient characteristics not fully accounted for by the analysis of this study.
While sophisticated analytic approaches such as propensity score and instrumental variable methods can be used to minimize bias in observational studies,9 these approaches were not described in the article. Indeed, this study lacked detailed information on key confounding factors, most notably smoking history (which presumably was higher in radiotherapy patients and directly causes lung, head/neck, and other cancers) and prostate cancer diagnostic information (more aggressive cancers are usually treated by radiotherapy instead of surgery; some metastatic patients were likely included in the analysis due to lack of staging information).
The inability to account for these well-established confounding factors can easily explain the relatively small absolute differences between surgical and radiation patients in open surgeries (0.3% difference), hospital admissions more than 1 day (1.8%), and second cancers. Regarding the outcomes of open surgeries and hospital admissions, the authors again selectively excluded the known outcomes of radical prostatectomy (100% risk of open surgery and hospitalizations more than 1 day, both directly attributable to the treatment), compared to the 1.1% open surgery and 3.3% admission rates of radiotherapy patients, which have an unclear causal link to treatment.
We must add that the patient decision nowadays often revolves around the choice of robotic surgery or intensity-modulated radiation therapy, two new approaches that certainly alter, and arguably improve, outcomes. This study may have lost some contemporary relevance by excluding robotic surgery and by being unable to specifically study intensity-modulated radiation treatment due to lack of necessary data.
Conclusions
The authors spoke the truth when they stated that “patients want to know the frequencies and severities of various complications associated with different treatments.” Unfortunately, this study was unable to address this central goal, which is to inform treatment decision-making.
In light of several important methodologic limitations, the authors reached too far to suggest causality between prostatectomy/radiotherapy and the “treatment-related complications.” It is not at all clear that the event rates from this study are due to treatment (although an unknown proportion certainly is). In light of these considerations, patients with localized prostate cancer and their physicians considering modern radical prostatectomy or radiotherapy options must be cautious when reviewing findings of this study. ■
Disclosure: Drs. Chen and Zietman reported no potential conflicts of interest.
References
1. Alemozaffar M, Regan MM, Cooperberg MR, et al: Prediction of erectile function following treatment for prostate cancer. JAMA 306:1205-1214, 2011.
2. Resnick MJ, Koyama T, Fan KH, et al: Long-term functional outcomes after treatment for localized prostate cancer. N Engl J Med 368:436-445, 2013.
3. Alibhai SM, Leach M, Tomlinson G, et al: 30-day mortality and major complications after radical prostatectomy: Influence of age and comorbidity. J Natl Cancer Inst 97:1525-1532, 2005.
4. Chen RC, Clark JA, Talcott JA: Individualizing quality-of-life outcomes reporting: How localized prostate cancer treatments affect patients with different levels of baseline urinary, bowel, and sexual function. J Clin Oncol 27:3916-3922, 2009.
5. Murray L, Henry A, Hoskin P, et al: Second primary cancers after radiation for prostate cancer: A systematic review of the clinical data and impact of treatment technique. Radiother Oncol. January 30, 2014 (early release online).
6. Chen RC, Carpenter WR, Hendrix LH, et al: Receipt of guideline-concordant treatment in elderly prostate cancer patients. Int J Radiat Oncol Biol Phys 88:332-338, 2014.
7. Nam RK, Cheung P, Herschorn S, et al: Incidence of complications other than urinary incontinence or erectile dysfunction after radical prostatectomy or radiotherapy for prostate cancer: A population-based cohort study. Lancet Oncol 15:223-231, 2014.
8. Thompson IM, Valicenti RK, Albertsen P, et al: Adjuvant and salvage radiotherapy after prostatectomy: AUA/ASTRO Guideline. J Urol 190:441-449, 2013.
9. Concato J: Is it time for medicine-based evidence? JAMA 307:1641-1643, 2012.
Dr. Chen is Assistant Professor in the Department of Radiation Oncology, University of North Carolina at Chapel Hill. Dr. Zietman is Associate Director of the Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston.
