Fax: (617) 566-3475
The effect of postprostatectomy external beam radiotherapy on quality of life
Results from the Cancer of the Prostate Strategic Urologic Research Endeavor
Article first published online: 15 JUN 2006
Copyright © 2006 American Cancer Society
Volume 107, Issue 2, pages 281–288, 15 July 2006
How to Cite
Hu, J. C., Elkin, E. P., Krupski, T. L., Gore, J. and Litwin, M. S. (2006), The effect of postprostatectomy external beam radiotherapy on quality of life. Cancer, 107: 281–288. doi: 10.1002/cncr.21980
- Issue published online: 5 JUL 2006
- Article first published online: 15 JUN 2006
- Manuscript Accepted: 21 FEB 2006
- Manuscript Revised: 29 JAN 2006
- Manuscript Received: 7 NOV 2005
- TAP Pharmaceutical Products, Inc.
- external beam radiotherapy;
- health-related quality of life;
- radical prostatectomy;
- salvage therapy
Postprostatectomy salvage radiotherapy may improve prostate-specific antigen (PSA) progression-free survival, but little is known about its effect on quality of life.
From the Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) data base, 1289 patients who had undergone radical prostatectomy (RP) without neoadjuvant or adjuvant hormone therapy completed validated health-related quality of life (HRQOL) questionnaires. Of these, 69 patients also received salvage radiotherapy at a median of 14 months after RP. The University of California–Los Angeles Prostate Cancer Index and the 36-item short form SF-36 questionnaire were used to compare HRQOL 12 to 18 months after external beam radiotherapy or 26 to 32 months after RP alone. Those responses also were compared with HRQOL responses from 55 men with data prior to and 12 to 18 months after primary radiotherapy. Multivariate regression identified differences between treatment groups.
Men who underwent salvage radiotherapy were younger (P = .03) and had lower incomes (P = .01) than men who underwent RP alone; they also were younger than men who underwent primary radiotherapy (P < .01). In addition, men who received salvage radiotherapy were more likely than men who underwent RP alone to have clinically high-risk prostate cancer (P < .01). Multivariate analyses revealed that men who received salvage radiotherapy experienced more marked decrements in sexual function (P = .01) and bowel function (P = .03) than men who underwent RP alone. Salvage radiotherapy led to less impairment of sexual function (P < .01) and less sexual bother (P = .04) than primary radiotherapy.
Although salvage radiotherapy is associated with unclear survival benefits, it adversely affects sexual and bowel function. Until randomized clinical trials demonstrate disease-specific survival benefits for salvage radiotherapy, the HRQOL detriments of additional therapy must be weighed against improved PSA progression-free survival. Cancer 2006;. © 2006 American Cancer Society.
Although there is no gold standard for the treatment of patients with clinically localized prostate cancer, conservative management of clinically insignificant tumors (those that are not expected to affect survival) is advocated to avoid impairment of health-related quality of life (HRQOL). Given the typically indolent nature of prostate cancer, most series report short-term biochemical progression-free survival rather than long-term disease-specific survival. Within 10 years after radical prostatectomy (RP), 1 in 3 men will develop biochemical recurrence1–4; and, within 5 years of RP, 1 in 4 men will receive salvage radiotherapy.5 Furthermore, treatment for biochemical recurrence after primary therapy is controversial. Although recent studies have identified clinical characteristics that portend a durable biochemical response to salvage radiotherapy,6, 7 it remains unclear whether additional treatment lengthens disease-specific survival.
Numerous longitudinal studies have demonstrated the impact of RP and primary radiotherapy on HRQOL; however, few studies have addressed the quality-of-life sequelae of salvage radiotherapy. We defined salvage radiotherapy as pelvic external beam radiotherapy after RP. We sought to chronicle the impact on quality of life of undergoing RP and receiving subsequent external beam radiotherapy and to compare it with the quality of life after either treatment alone.
MATERIALS AND METHODS
Data from the Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) registry were used for this analysis. CaPSURE is a longitudinal, observational registry of men with biopsy-proven prostate cancer in which data are collected from participant questionnaires and medical records at 34 community-based practices and 6 academic or Veterans Administration clinical sites across the United States. Clinical data (medical history, tumor stage, and prostate-specific antigen [PSA] levels4) are provided by the urologist at clinical encounters. Participants themselves report baseline HRQOL data when they enter the registry and every 6 months thereafter, and the a response rate ranges from 73% to 79%.
We identified men with localized prostate cancer who underwent RP without receiving neoadjuvant or adjuvant hormone therapy between 1990 and 2002 (Fig. 1) and who subsequently received salvage radiotherapy a median of 16 months after surgery. We then identified control groups who had undergone retropubic RP or external beam radiotherapy alone. We restricted our analysis to men who had HRQOL data collected at baseline and at 12 to 18 months of follow-up for each of the 3 treatment groups (Fig. 2). To make these assessments as similar as possible, baseline HRQOL was assessed before radiotherapy in the primary and salvage radiotherapy groups; whereas, in the RP alone group, baseline HRQOL was assessed 14 months postprostatectomy.
We measured general HRQOL with the RAND 36-Item Health Survey, version 1.0,8 a 36-item questionnaire that quantifies physical and mental HRQOL as component summaries. The scores are standardized to population norms, with a mean of 50 and a standard deviation of 10 for each summary scale. We assessed disease-specific HRQOL with the University of California–Los Angeles (UCLA) Prostate Cancer Index,9 a self-administered, 20-item questionnaire that quantifies prostate cancer-specific HRQOL in 6 domains, including urinary, sexual, and bowel function and bother. Domains are scored from 0 to 100. For all of these measures, higher scores represent better outcomes. We also identified men from each treatment group who scored from 80 to 100 points on the domains of the UCLA Prostate Cancer Index. It has been demonstrated that scores in this range have strong concordance with definitions of excellent continence, potency, and bowel function.10
Differences between the treatment groups in clinical and sociodemographic characteristics and the proportion of men who achieved scores from 80 to 100 in disease-specific HRQOL domains were evaluated with the chi-square test. The Student t test was used to evaluate differences in HRQOL between groups, and Student t tests for paired data were used to evaluate statistical changes between baseline and follow-up for change scores. Multivariate linear regression models were developed with each of the HRQOL change scores as the outcome and included clinical and sociodemographic variables that differed between the treatment groups at P < .20. P values were considered statistically significant at P < .05 without adjustment for multiple comparisons. All analyses were performed using version 9 of SAS.
Table 1 presents the sociodemographic characteristics of the study group. Men who received salvage radiotherapy were younger than men who received primary radiotherapy (P < .01.) and had lower incomes than men who underwent RP (P = .01). Men who received salvage radiotherapy were more likely than men who received primary radiotherapy to be unpartnered (P = .01). Compared with men who underwent RP alone, men who received salvage radiotherapy were more than twice as likely to have high-risk tumors (P < .01). Finally, there was a trend for men who received salvage radiotherapy to undergo nonnerve-sparing RP more often than men who underwent RP alone (P = .06).
|Characteristic||No. of patients (%)||P|
|Surgery||Salvage RT||Primary RT||Surgery vs. Salvage RT||Salvage RT vs. Primary RT|
|Age at diagnosis, y||.24||<.01|
|<60||450 (37)||19 (28)||1 (2)|
|60–64||312 (26)||22 (32)||7 (13)|
|65–69||311 (25)||16 (23)||13 (24)|
|≥70||147 (12)||12 (17)||34 (62)|
|Age at posttreatment questionnaire, y||.03||<.01|
|<60||296 (24)||11 (16)||1 (2)|
|60–64||302 (25)||14 (20)||2 (4)|
|65–69||322 (26)||16 (23)||14 (25)|
|≥70||300 (25)||28 (41)||38 (69)|
|Year of diagnosis||<.01||<.01|
|1990–1997||312 (25)||37 (54)||14 (25)|
|1998–2002||909 (73)||32 (46)||41 (50)|
|Not white||101 (8)||2 (3)||2 (4)|
|White||1119 (92)||67 (97)||53 (96)|
|Unknown||94 (8)||4 (6)||0|
|<HS||97 (8)||12 (17)||10 (18)|
|HS graduate||281 (23)||18 (26)||14 (25)|
|Some college||224 (18)||10 (14)||8 (15)|
|College graduate||524 (43)||25 (36)||23 (42)|
|Unknown||188 (15)||11 (16)||4 (7)|
|<$50,000||488 (40)||39 (57)||17 (31)|
|$50–75,000||232 (19)||11 (16)||7 (13)|
|>$75,000||335 (27)||8 (12)||6 (11)|
|Unknown||114 (9)||7 (10)||0|
|Married/partner||1043 (85)||60 (87)||49 (89)|
|Single||63 (5)||2 (3)||6 (11)|
|No. of comorbidities||.63||.15|
|Unknown||96 (8)||3 (4)||1 (2)|
|None||241 (20)||13 (19)||4 (7)|
|1||375 (31)||19 (28)||11 (20)|
|2||295 (24)||18 (26)||20 (36)|
|≥3||213 (17)||16 (23)||19 (35)|
|Prostate cancer risk||<.01||.34|
|Unknown||43 (4)||3 (4)||0|
|Low||561 (46)||24 (35)||22 (40)|
|Intermediate||475 (39)||24 (35)||22 (40)|
|High||141 (12)||18 (26)||11 (20)|
|Yes||825 (68)||39 (57)||—|
|No||395 (32)||30 (43)||—|
Table 2 presents HRQOL outcomes in men who underwent surgery alone, men who also received salvage radiotherapy, and men who received only radiotherapy, both before and 12 to 18 months after radiotherapy or at comparable time points after surgery. Physical component summary scores before and after salvage radiotherapy were worse than scores among men who underwent RP at similar time points after surgery (47.7 vs. 52.0 [P < .01; 46.0] vs. 51.2 [P < .01], respectively). Men who received salvage radiotherapy had worse urinary function scores before treatment (68 vs. 75; P = .02) and after treatment (69 vs. 78; P < .01) compared with men who underwent RP alone, respectively. Similarly, urinary bother scores were worse for men after salvage radiotherapy than after surgery alone (74 vs. 82, respectively; P < .01). Overall sexual function (23 vs. 32, P < .01; and 33 vs. 44, P = .03) and sexual function change scores (−2.5 vs. +5.8; P < .01) were worse in men after salvage radiotherapy than after RP alone, respectively, at comparable times. Bowel function and bother scores also were worse in men after salvage radiotherapy than after RP alone at comparable times (86 vs. 89, respectively; P < .01; 85 vs. 90, respectively; P = .04). Finally, bowel function (−3.3 vs. +0.2; P = .03) and bother (−4.7 vs. +0.2; P = .04) worsened after salvage radiotherapy, whereas men who underwent RP had stable bowel function and bother during the same interval.
|HRQOL domain||Mean ± SD||P|
|Surgery||Salvage RT||Primary RT||Surgery vs. Salvage RT||Salvage RT vs. Primary RT|
|Pretreatment||51.96 ± 8.22||47.71 ± 9.54||48.57 ± 8.99||<.01||.62|
|Posttreatment||51.18 ± 8.20||46.04 ± 10.84||45.33 ± 10.12||<.01||.72|
|Change||−0.85 ± 6.31||−1.16 ± 6.12||−3.72 ± 7.87||.70||.06|
|Pretreatment||53.50 ± 8.29||53.92 ± 8.06||53.14 ± 9.44||.69||.63|
|Posttreatment||54.33 ± 7.98||53.98 ± 10.30||51.26 ± 10.63||.73||.16|
|Change||0.90 ± 6.84||−0.01 ± 9.02||−1.38 ± 8.74||.32||.42|
|Pretreatment||26.73 ± 23.62||24.89 ± 25.24||44.48 ± 27.45||.53||<.01|
|Posttreatment||32.48 ± 27.22||22.75 ± 23.22||27.99 ± 26.02||<.01||.24|
|Change||5.84 ± 17.22||−2.54 ± 14.66||−16.50 ± 18.76||<.01||<.01|
|Pretreatment||35.09 ± 35.20||34.85 ± 35.59||58.49 ± 37.96||.96||<.01|
|Posttreatment||44.96 ± 37.29||38.08 ± 35.93||50.00 ± 39.77||.15||.10|
|Change||9.92 ± 30.26||5.16 ± 36.2||−9.69 ± 30.96||.23||.02|
|Pretreatment||74.97 ± 23.23||68.19 ± 24.40||90.19 ± 13.49||.02||<.01|
|Posttreatment||77.64 ± 21.56||69.16 ± 23.49||84.53 ± 17.44||<.01||<.01|
|Change||2.62 ± 14.98||0.70 ± 19.30||−5.90 ± 14.31||.32||.04|
|Pretreatment||79.24 ± 25.55||75.00 ± 27.18||79.17 ± 26.93||.19||.40|
|Posttreatment||81.79 ± 24.58||73.91 ± 27.93||74.52 ± 32.27||.01||.91|
|Change||2.60 ± 21.43||−1.12 ± 21.95||−3.92 ± 28.88||.17||.55|
|Pretreatment||89.15 ± 13.50||89.02 ± 14.30||86.51 ± 12.64||.94||.31|
|Posttreatment||89.30 ± 12.95||85.58 ± 15.74||79.80 ± 18.10||.02||.06|
|Change||0.18 ± 12.43||−3.28 ± 15.45||−6.53 ± 16.18||.03||.26|
|Pretreatment||89.84 ± 19.63||89.86 ± 22.00||81.94 ± 28.07||.99||.08|
|Posttreatment||90.03 ± 19.12||85.14 ± 25.12||71.08 ± 30.97||.04||.01|
|Change||0.23 ± 19.80||−4.71 ± 21.12||−10.00 ± 24.74||.04||.21|
Table 2 also presents HRQOL comparisons between men who received salvage radiotherapy and primary radiotherapy. At baseline, men who received salvage radiotherapy had worse urinary function (68 vs. 90; P < .01), sexual function (25 vs. 44; P < .01), and sexual bother (35 vs. 58; P < .01) scores. Urinary function scores remained worse after salvage radiotherapy than after primary radiotherapy (69 vs. 84, respectively; P < .01). Furthermore, whereas men who received primary radiotherapy saw a decrease in their urinary function scores, men who received salvage radiotherapy experienced no change (−6 vs. +0.7, respectively; P = .04). Similarly, men who received radiotherapy faired worse in sexual function than men who received salvage radiotherapy (−16 vs. −2.5, respectively; P < .01). During this period, sexual bother scores improved in men who received salvage radiotherapy but worsened in men who received primary radiotherapy (+5 vs. −10, respectively; P = .02). Finally, whereas bowel bother scores were similar in men prior to salvage or primary radiotherapy, they were substantially better after salvage than primary radiotherapy (85 vs. 71, respectively; P = .01).
Table 3 shows men who achieved HRQOL domain scores of 80 to 100, signifying excellent function.10 Slightly greater than 50% of men achieved excellent urinary function at 14 months and at least 26 months after RP alone, whereas just over 33% of men achieved excellent urinary function both before (P = .02) and after (P = .01) salvage radiotherapy at similar time points. In contrast to salvage radiotherapy, a greater proportion of men before (P < .01) and after (P < .01) primary radiotherapy reported excellent urinary function. In addition, a greater proportion of men achieved excellent bowel function after RP alone than after salvage radiotherapy (P = .04). Finally, a lower proportion of men after primary than after salvage radiotherapy reported excellent bowel function (P = .05).
|HRQOL scale and time point||Surgery (%)||Salvage RT (%)||Primary RT (%)||P value|
|Surgery vs. Salvage RT||Salvage RT vs. Primary RT|
|No. of patients||1220||69||55|
After controlling for age, race/ethnicity, education, income, relationship status, comorbidities, diagnosis year, and nerve-sparing technique in the multivariate analysis, differences in sexual and bowel function change scores persisted between the salvage radiotherapy and RP alone groups. The average adjusted change scores for the salvage radiotherapy group were 6.1 points less for sexual function (P = .01) and 3.7 points less for bowel function (P = .03) compared with the RP alone group (Table 4). In addition, multivariate analyses demonstrated that the average adjusted change scores for sexual function and bother for the salvage radiotherapy group were 15 points (P < .01) and 16 points (P = .04) greater than the primary radiotherapy group, respectively (Table 4).
|HRQOL scale||Standard error of β coefficient*||P|
|Salvage RT vs. RP alone (reference group)||Salvage RT vs. Primary RT (reference group)||Salvage RT vs. RP alone||Salvage RT vs. Primary RT|
|PCS change||−0.18 (0.87)||0.91 (1.68)||.83||.59|
|MCS change||−0.76 (0.96)||−0.04 (2.23)||.43||.98|
|SF change||−6.10 (2.22)||14.91 (3.62)||.01||<.01|
|SB change||−0.88 (4.13)||16.49 (7.94)||.83||.04|
|UF change||−2.09 (2.02)||3.37 (3.90)||.30||.39|
|UB change||−3.08 (2.82)||0.73 (5.78)||.28||.90|
|BF change||−3.70 (1.65)||−1.96 (3.69)||.03||.60|
|BB change||−4.60 (2.59)||0.39 (5.58)||.08||.95|
The current study produced several important findings. First, men who underwent salvage radiotherapy after RP had worse physical function before and after radiotherapy than men before or after surgery alone. This is consistent with other studies, which have shown worse SF-36 physical function scores in cancer survivors who experience disease recurrence. Northouse et al. demonstrated physical impairment in patients with recurrent breast cancer as early as 1 month after cancer detection.11 Camilleri-Brennan examined the impact of colon cancer recurrence on quality of life and observed lower SF-36 domain scores in patients who had disease recurrence, with the greatest decrement noted for physical function. Finally, patients who had prostate cancer with PSA recurrence had significantly worse physical function than men after RP without biochemical recurrence. Three years after surgery, men who had biochemical recurrence had significantly worse physical functioning and pain (P < .02) than similar men without PSA recurrence.13
Second, sexual and bowel functions were markedly worse in men who received salvage radiotherapy than in men who underwent RP alone. Although, to our knowledge, ours is the first study to use validated quality-of-life instruments to assess men longitudinally after prostatectomy, salvage radiotherapy, and primary radiotherapy, in 1 retrospective chart review, it was reported that 10% of salvage radiotherapy recipients had spontaneous erections that were adequate for sexual intercourse, and 67% were completely continent.14 Furthermore, other retrospective studies ascribed a 1% to 26% rate of urinary incontinence and a 9% rate of rectal symptoms to salvage radiotherapy.15–17 In contrast, in the current study, only 3% and 39% of men after salvage radiotherapy scored between 80 and 100 points on their sexual and urinary function domains, respectively, correlating with excellent function. By comparison, 55% versus 39% of men after RP versus salvage radiotherapy, respectively, had urinary function scores of 80 to 100, a 16% difference in continence that may have been caused by salvage radiotherapy. The marked variation in continence and potency rates between our study and retrospective chart reviews may have been because of differences in patient self-reporting with validated instruments versus provider-reported outcomes.18, 19
Third, men who received primary radiotherapy experienced a greater decline in sexual function than men who received salvage radiotherapy. This observation probably is because of a floor effect of RP on sexual function scores prior to salvage radiotherapy: Sexual function scores in men prior to salvage radiotherapy, on average, were 20 points worse than the scores in men prior to primary radiotherapy. Finally, sexual bother worsened after primary radiotherapy but improved after salvage radiotherapy. This may be explained by the longer adjustment time for men who underwent RP followed by salvage radiotherapy compared with the adjustment time for men who received primary radiotherapy.
Our study has several important limitations. First, an ideal longitudinal assessment of HRQOL would include baseline questionnaire data prior to surgery for the RP alone and salvage radiotherapy groups; however, this was unavailable for a significant number of our men who received salvage radiotherapy. Therefore, we defined baseline in these 2 groups at a median of 14 months postprostatectomy, allowing for presumed recovery of urinary and sexual function.20, 21 In multivariate analyses to determine the timing impact of salvage radiotherapy on HRQOL change scores, 23 men who received salvage radiotherapy <1 year after surgery demonstrated improvements in sexual function and bother and in urinary function and bother, whereas 46 men who received salvage radiotherapy ≥1 year after surgery experienced worsening change scores in these domains. However, it noteworthy that we had only 7% power to detect a 20% or greater difference in change scores because of the large standard deviations relative to the means and the small numbers in each group. In addition, the relatively lower numbers of patients in the salvage and primary radiotherapy groups may have under-powered our analysis, resulting in under-detection of small differences that may exist between treatment groups. Also, the dose and type of radiotherapy administered (e.g., whole pelvis, conventional, and conformal) varied in treatment groups as radiotherapy techniques were modified over the course of our study period; this may have had an impact on HRQOL outcomes. Finally, our study population was predominantly white and well educated, limiting the generalizability of our findings to other ethnic groups.
In conclusion, given the unclear disease-specific survival benefits, men must consider the quality-of-life impact of salvage radiotherapy on sexual, urinary, and bowel function. Furthermore, although it is perceived that delaying treatment until recovery of continence may attenuate the adverse effects of salvage radiotherapy, empirical evidence from studies that use validated HRQOL instruments are lacking.