As the long-term efficacy of SBRT becomes established and other prostate cancer treatment approaches are refined and improved, continued research into QOL after prostate cancer treatment is critical in driving both patient and clinical treatment decisions. Indeed, 5-year follow-up for SBRT has yielded disease control that approximates that produced by surgery and other conventionally fractionated radiation therapy treatments
. Thus, preservation of QOL may become the primary basis upon which patients choose a prostate cancer treatment. It is in this context that we present the first study to compare QOL after SBRT and radical prostatectomy, with QOL assessed at approximately the same times pre- and post-treatment and using the same validated QOL instrument (EPIC). In this study, the largest differences in QOL occurred in the first 1–6 months after treatment, demonstrating larger declines following surgery in urinary and sexual QOL as compared to SBRT, and a larger decline in bowel QOL following SBRT as compared to surgery. Overall, these results may have implications for patient and physician clinical decision making which are often influenced by QOL.
Urinary and sexual function are the primary concerns for men undergoing prostate cancer treatment
[9, 14, 16]. Urinary QOL is typically altered by several types of urinary complications. Incontinence, leakage of urine, is the predominant urinary issue after surgery
[14, 16] with 3 month continence rates of 51-74% for radical prostatectomy
[23–25] that improve to 80-94% by 12 months
[23, 26]. Incontinence is not typically a concern following SBRT
[4, 5, 27, 28], more typical are urinary toxicities such as urinary retention, urgency or hesitancy at rates of 2-10%
[3–5, 27–29]. A major urinary factor in surgery is that the urethra is cut in half during surgical removal of the prostate and then reconnected during the surgery, whereas the urethra remains completely intact and is identified and avoided during SBRT. As a result surgery patients require a catheter that stays in place following discharge from the hospital for 7 to 10 days after treatment and often patients must use pads to address urine leakage in the months following treatment. In this analysis, the overall EPIC urinary domain was analyzed, but the EPIC urinary sub-domain scores for incontinence and obstruction were not analyzed. For the SBRT patients, only the overall urinary summary score for each patient was recorded which prevented analysis of the sub-domain scores in this analysis. A prior examination of this EPIC surgery data using the obstruction and incontinence sub-domains showed urinary obstructive QOL had a transient decline at one month followed by return to baseline with even improvement for patients with preexisting urinary irritative-obstructive symptoms, whereas urinary incontinence QOL significantly impacted surgery patients at all time points
In the current study, SBRT patients had significantly higher sexual QOL after treatment than surgery patients with no clinically significant decline in sexual QOL at any of the study’s follow-up time points. In contrast, the surgery patients, whether or not the nerve-sparing technique was used, had clinically significant declines in sexual QOL at all time points. This is not surprising given the difficulty in identifying the neurovascular bundle in the operating room and the mechanical and thermal injury that may occur to the nerves during surgery. While robotic assisted laparoscopic prostatectomy is more commonly used in the United States, the surgical patients in this study received open radical retropubic prostatectomy, as laparoscopic surgery was not routinely offered in the Spanish hospitals at the time of their study. It remains unknown how significant of an impact laparoscopic (either robotic or not) has on sexual QOL as published functional outcomes vary greatly. For example, potency rates at 12 months for robotic-assisted laparoscopic prostatectomy range from 61–97%
[30–34] with patient selection, potency definition, surgeon experience and use of penile rehabilitation primarily accounting for the large variation in rates
. Indeed, a recent review by Finkelstein et al. at NYU in which they examine multiple comparative publications states “it is difficult to determine if one approach is superior to the other for the preservation of neurovascular bundles and sexual function”
Baseline mean EPIC sexual QOL scores for the patients receiving SBRT and nerve-sparing surgery in the current study were similar whereas the patients receiving non-nerve-sparing surgery had significantly lower baseline mean EPIC sexual QOL scores. This low baseline sexual QOL likely reflects why these patients were not chosen for the nerve-sparing procedure in that their sexual function was already grossly impaired. Mean EPIC sexual scores were observed to drop dramatically in the short term after surgery independent of surgical approach. While patients receiving nerve-sparing surgery had a greater degree of recovery neither group recovered back to baseline levels. In contrast, mean EPIC scores after SBRT declined slowly over time but were not clinically significantly different than the baseline levels. Failure to return to baseline sexual function is not uncommon following prostatectomy. Levinson et al., found 27% of 568 patients receiving laparoscopic radical prostatectomy returned to baseline EPIC sexual QOL at 24 months
. Similarly, while in a study of potent patients Willis et al. observe a greater return to baseline function for patients receiving robotic-assisted laparoscopic prostatectomy the mean EPIC sexual QOL score at 12 months remained 66.2 and 73.7% of the baseline score for robotic-assisted laparoscopic prostatectomy and laparoscopic prostatectomy, respectively
Wiegner and King at Stanford University assessed sexual QOL using EPIC for 32 low-risk prostate cancer patients who received SBRT (36.25 Gy in 5 fractions)
. The mean baseline sexual QOL score of 67.5 declined to 56.4, 50.7 and 37.4 at 12, 20 and 50 months, respectively. McBride et al. report sexual QOL using EPIC for 26 patients treated with SBRT (primarily 35 or 36.26 Gy in 5 fractions) as part of a multi-institutional study
. The median baseline sexual QOL score of 43 declined to 17 at 36 months. The small number of patients in both of these studies may reflect upon the larger observed declines in sexual QOL. These studies also consisted of patients treated very early in the groups’ SBRT experience and prior to adoption of more advanced and currently used SBRT techniques, such as advanced collimation that have been shown to spare more normal tissue during treatment
[39, 40]. In addition, the Stanford report did not incorporate the use of MRI which is routinely used to definitively identify the neurovascular bundles in treatment planning.
Mild to moderate bowel toxicity consisting of proctitis and rectal bleeding has been reported following SBRT
[4, 29, 41]. In contrast, very low rates of bowel injury have been reported during prostatectomy
[42, 43], as bowel complications are not typically an issue for patients undergoing surgery. In the current study, bowel QOL was minimally affected by surgery whereas bowel QOL fell after SBRT but returned to near baseline 6–12 months post-treatment. This is consistent with studies showing the incidence of Grades 2 and 3 bowel toxicity after SBRT (as measured using the RTOG scale) range from 4–24% in the acute phase, but falls to 1-12% in the late phase
[4, 5, 28, 29, 41, 44–46].
The present study is limited by the retrospective nature of the analysis despite the prospective data collection for the two treatment groups. First, treatment groups differed at baseline in ways that might be expected of such a retrospective analysis, and which might have contributed to post-treatment differences between groups. SBRT patients were older, on average, than surgery patients. Although the influence of age was considered by including it in the GEE models, adjusting for age is not sufficient since we do not have data for surgery patients older than 74 years. Differences in the use of treatment for erectile dysfunction may affect sexual outcomes; for example, baseline sexual scores of the older patients in the SBRT group were as good or better than the surgery patients, suggesting a possible relationship with higher sildenafil intake among SBRT patients. However, this could not be included as a baseline adjustment because patients started sildenafil use at various timepoints throughout the study. Other lifestyle factors such as smoking, marital status, and education, and comorbidities were not explicitly balanced between groups so their contribution cannot be assessed. General QOL at baseline was also not assessed. Second, nerve-sparing techniques were not widely applied in the surgery group (28% of patients treated with prostatectomy), and therefore, our findings from the GEE model for the prostatectomy group as a whole could overestimate their adverse sexual effects. Third, both the SBRT and surgery groups had small numbers of patient responses early in their studies. For the surgery patients, the 1-month follow-up had the fewest responses whereas for the SBRT group the 6 month follow-up had the fewest. For the sexual domain, GEE analysis excluding the 1- and 6-month time points showed similar results, suggesting these fluctuations in questionnaire completion did not greatly impact the results. Fourth, there are QOL domains that are not assessed by EPIC which may differ between groups, e.g., fatigue and general quality of life. A more extensive analysis of a broader range of QOL variables may permit more complete comparisons. Some aspects of treatment could not be rigorously controlled, including surgeon expertise which is known to affect disease-control and toxicity outcomes. These issues of heterogeneity across groups could only be addressed effectively in the context of a randomized design. Finally, prostate treatment is an evolving practice; results such as these are based on a limited snapshot of current practices. The emphasis on QOL in prostate cancer patients demands continuing re-analyses of QOL outcomes as treatments evolve.