SEARCH

SEARCH BY CITATION

Keywords:

  • external beam radiotherapy;
  • Japanese men;
  • prostate cancer;
  • quality of life;
  • radical prostatectomy

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Abstract  Background:  We performed a retrospective survey of general and disease specific health-related quality of life (HRQOL) after radical prostatectomy (RP) and external beam radiotherapy (XRT) in Japanese men.

Methods:  A total of 186 patients underwent RP and 78 underwent XRT for clinically localized prostate cancer between 2000 and 2002. We measured the general and disease specific HRQOL with the MOS 36-Item Health Survey and the University of California, Los Angeles Prostate Cancer Index, respectively. Each treatment group was further divided into four subgroups according to the time scale.

Results:  Patients from the RP group were significantly younger than those from the XRT group. The tumor characteristics differed significantly in their distributions among the treatment groups. Patients undergoing XRT had low scores in most of the general measures of HRQOL just after treatment, but after 6 months there were no differences between the treatment groups, except for the physical domains. The RP group was associated with worse urinary function, whereas the XRT group had worse bowel function and bother during the first 6 months after treatment. Thereafter, however, urinary and bowel domain did not differ between the groups. Both groups reported poor sexual function, although the RP group scored lower sexual bother.

Conclusion:  The patients who underwent RP had significantly worse urinary and better bowel function than those treated with XRT. Both treatment groups had decrements in sexual function throughout the post-treatment period; careful attention should be paid to this side-effect in preoperative counselling, especially in younger patients, regardless of the primary treatments.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

The incidence of early stage prostate cancer in Japanese men is clearly increasing, although the rate is still lower than that in Western countries.1 This fact is known to us because of the widespread use of the serum prostate-specific antigen (PSA) test. Because prostate cancer is being increasingly diagnosed at early stages, survival outcomes are more favorable and the basis on which patients select primary therapy has shifted toward considerations of the health-related quality of life (HRQOL).2,3 Other factors beyond ‘cancer cure’, such as possible side-effects, are also an important motivation for choosing a procedure.4 Accordingly, HRQOL assessment has become an important form of outcome-based research that might weigh heavily in the justification for health care expenditures and in treatment selection by patients.

Some aspects of prostate cancer and its treatment might also have a profound impact on both general and disease-specific HRQOL domains.5,6 In a cross-sectional analysis, Litwin and colleagues found no differences in the general measures of HRQOL among patients with localized disease treated with radical prostatectomy, irradiation, or observation alone. However, disease-targeted domains were found to differ among the treatment groups, with radiotherapy having a greater impact on bowel function compared with other treatments and radical prostatectomy being associated with more substantial changes in urinary and sexual function.7 In a study of radical prostatectomy and radiotherapy, Lim et al. suggested that the radical prostatectomy group had worse sexual function and urinary incontinence, and the radiotherapy group had worse bowel function.8 Similar results were reported by Shrader-Bogen and colleagues in a cross-sectional analysis of patients who underwent radical prostatectomy and radiotherapy.9 Because the patients of these HRQOL studies were predominantly white men, the results might not be readily extrapolated to those of other ethnic backgrounds, such as Japanese men.

To our knowledge, there have been few well-constructed studies that measured Japanese men's HRQOL after surgery or irradiation therapy by internationally validated methods, such as the MOS 36-Item Health Survey (SF-36) and the University of California, Los Angeles Prostate Cancer Index (PCI). This is the first report that compares the HRQOL between radical surgery and irradiation therapy in Japanese. Our goal is to provide data on such differences in HRQOL to assist physicians in counselling patients on the selection of a preferred course of treatment for prostate cancer. Armed with this information, patients are likely feel more confident about their treatment choices.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Patient population

Between January 2000 and December 2002, 206 patients were treated with retropubic radical prostatectomy (RP) and 94 were treated with external beam radiation (XRT). All patients were informed of their cancer diagnosis before being asked to fill out the HRQOL questionnaires. Every patient who agreed to participate in this study received from their urologists a questionnaire, an informed consent form, and a prepaid envelope for returning the questionnaire. The patients who suffered from localized prostate cancer (T1-T3N0M0) were treated at Tohoku University Hospital and its three affiliated hospitals and Kurashiki Central Hospital. Tumors were staged clinically according to the 1992 TNM classification.10 Patients who received both radiation therapy and surgery were excluded from the study. For our purposes, the patients were divided into four groups according to study time scales, as follows: time 1 (group t1), less than 3 months after treatment; time 2 (group t2), 3–6 months after treatment; time 3 (group t3), 6–12 months after treatment; time 4 (group t4), more than 12 months after treatment. Moreover, to assess the influence of age on disease specific QOL, each treatment group was categorized into two subgroups (younger or older than median age).

In general, RP was recommended if the patient was aged 75 years or younger and had a tumor with a clinical stage of cT2 or lower, and XRT with neoadjuvant therapy was recommended if the patient was older than 75 years and/or had a tumor with a clinical stage of cT3a or higher. However, the final determination of the treatment method was made by the patient after thorough discussion of the options.

Radical prostatectomy was performed at Tohoku University hospital, Miyagi Cancer Center, Sendai Shakaihoken Hospital, and Kurashiki Central Hospital. Radiation therapy was performed at Tohoku University Hospital, Miyagi Cancer Center, Furukawa City Hospital and Kurashiki Central Hospital.

Quality-of-life method

We measured the general and prostate specific HRQOL using two types of instruments. The general HRQOL was assessed with SF-36.11,12 The general scales cover eight domains, four physical and four emotional. The eight scales are scored separately from zero to 100, a higher score representing a better quality of life. The prostate specific HRQOL was assessed with the PCI, developed by Litwin and colleages.13 The disease-specific items encompass urinary, bowel, and sexual problems and the extent of bother from problems in each area. Both questionnaires have already been translated into Japanese and their validity and reliability were previously tested.14,15

Statistics

Quality of life scores for the various domains are shown as mean ± SD in scales of 0–100, with a higher score always representing a better HRQOL. Differences in the distributions of the background variables were evaluated by non-parametric procedures (χ2 or Mann–Whitney tests). P < 0.05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Demographic characteristics

The surveys were returned by 186 (90%) and 78 (83%) of men who underwent RP and XRT, respectively. Table 1 presents information on the background characteristics of the patients with localized prostate cancer who subsequently underwent RP or XRT. We analyzed the survey responses of 264 Japanese men (Tohoku University Hospital, 58 patients; Miyagi Cancer Center, 78 patients; Sendai Shakaihoken Hospital, 40 patients; Kurashiki Central Hospital, 61 patients; Furukawa City Hospital, 27 patients) with localized prostate cancer. The XRT was delivered to the prostate using the conventional four-field box (anterior, posterior and right and left laterals) technique in 51 patients. The remaining 27 patients were treated with 3-D conformal external-beam radiation therapy. The median dose was 70 Gy (range, 66–72 Gy).

Table 1.   Demographic and clinical characteristics of study population
Treatment groupRPXRTP-value RP vs XRT
  1. PSA, prostate specific antigen; RP, radical prostatectomy; XRT, external beam radiotherapy. *, Mann-Whitney U-test; **, χ2 test.

Number of patients18678 
Age at survey (years)
 Mean ± SD66.7 ± 5.672.9 ± 5.1P < 0.001*
 Median 6773
 Range48–7757–86
Follow-up time (months)
 Mean ± SD16.7 ± 8.612.8 ± 9.2P = 0.001*
 Median 1510
 Range2–341–35
Distribution of follow-up time
 <3 months (t1) 4319P = 0.085**
 3–6 months (t2) 5917
 6–12 months (t3) 4515
 >12 months (t4) 3927
Working status
 Full-time worker 7548P = 0.006**
 Part-time woker 8220
 Retired/no job 20 9
 Unknown  9 1
Marital status
 Married18074P = 0.373**
 Unmarried 5 4
 Unknown 1 0
Selected conditions
 Diabetes 19 5P = 0.425**
 Cardiovascular disease 9 9
 Other cancer 13 8
 Hypertension 4118
 Cerebral disease 9 6
 Ulcer 1611
Pretreatment tumor stage
 T1 9120P < 0.001**
 T2 7930
 T3 1629
Histologic tumor grade
 Well differentiated 27 4P = 0.085**
 Moderately differentiated13158
 Poorly differentiated 2514
 Unknown 3 2
Neoadjuvant therapy ablation
 Number of patients 8468P < 0.001**
 % 4487
PSA at diagnosis (ng/ml)
 Mean ± SD14.3 ± 18.234 ± 49.7P < 0.001*
 Median   8.516.9
 Range3.3–1902.7–260
Nerve sparing
 Bilateral 21  
 Unilateral 69  
 None 99  
Adjuvant therapy ablation
 Number of patients 1848P < 0.001**
 %   9.561

Patients from the RP group were significantly younger than those from the XRT group (P < 0.001). The median follow-up periods after treatment were 16.7 months and 12.7 months, respectively (P = 0.001). A comparison of the RP group and XRT group based on the time distribution of the study did not show any statistically significant differences. The two groups showed similar levels of comorbidity. Fewer RP treated patients had PSA values greater than 10 ng/mL compared with those who underwent XRT (42%vs 63%). The tumor characteristics differed significantly in their distributions between the RP group and the XRT group, with the latter group having a significantly higher clinical stage than T2a (39%vs 64%). However, differences in tumour grade between the two groups were not statistically significant.

General HRQOL

The mean scores for each general HRQOL (SF-36 scales) domain at t1–t4 are listed in Table 2. In the early post-treatment term, at t1 and t2, the XRT group tended to score lower, relative to the RP group, for physical function (P = 0.0077, t1), general health perception (P = 0.029, t1), mental health (P = 0.012, t1 and P = 0.0088, t2), role limitations due to emotional problems (P = 0.047, t2), social function (P = 0.046, t2) and vitality (P = 0.0064, t2). In the late post-treatment term, at t3 and t4, significant differences were found on two of eight SF-36 scales. Relative to the RP group, the XRT group reported more limitations because of physical function (P = 0.041, t4) and role limitations due to physical problems (P = 0.036, t3).

Table 2.   Sf-36 scores of patients in the posttreatment groups
Time groupProstatectomyRadiation therapyP-value*
  1. Data are presented as mean ± SD. Each domain is scored from 0 to 100 with higher scores representing better quality of life. SF-36, MOS 36-Item Health Survey. *, Mann-Whitney U-test.

Physical function
 t185.0 ± 12.274.7 ± 15.60.007
 t284.7 ± 15.377.1 ± 23.40.256
 t386.5 ± 17.282.0 ± 20.20.497
 t487.3 ± 16.078.9 ± 20.60.041
Role limitations due to physical problems
 t173.1 ± 25.465.5 ± 27.40.265
 t280.3 ± 24.270.2 ± 28.70.122
 t387.1 ± 22.977.5 ± 22.40.036
 t485.9 ± 23.179.2 ± 24.40.092
Bodily pain
 t174.9 ± 20.869.0 ± 19.80.233
 t279.5 ± 20.971.4 ± 24.00.182
 t382.1 ± 21.687.0 ± 15.00.609
 t481.3 ± 16.981.6 ± 24.30.485
General health perception
 t167.5 ± 13.857.8 ± 18.30.029
 t260.7 ± 14.657.6 ± 19.50.443
 t363.1 ± 13.458.7 ± 15.50.361
 t460.0 ± 14.756.8 ± 11.90.162
Mental health
 t176.5 ± 17.366.1 ± 12.80.012
 t279.5 ± 16.466.6 ± 15.10.008
 t379.2 ± 16.577.7 ± 14.40.471
 t476.6 ± 19.274.8 ± 18.60.592
Role limitations due to emotional problems
 t173.1 ± 25.071.0 ± 27.20.474
 t284.3 ± 21.869.8 ± 25.40.047
 t387.3 ± 21.278.9 ± 17.80.052
 t488.7 ± 21.679.9 ± 26.30.203
Social function
 t184.3 ± 19.179.6 ± 14.60.213
 t284.7 ± 21.072.1 ± 24.40.046
 t389.9 ± 15.590.8 ± 12.90.939
 t487.5 ± 19.088.0 ± 17.80.994
Vitality
 t167.3 ± 19.357.6 ± 17.30.056
 t272.7 ± 18.656.6 ± 21.10.006
 t371.3 ± 19.070.8 ± 19.60.782
 t471.0 ± 20.266.6 ± 19.90.493

Disease specific HRQOL

Just after treatment, at t1, the RP group showed a significantly lower urinary function score than the XRT group (P = 0.0045), but thereafter there was no significant difference. No difference was observed in urinary bother throughout the follow-up time. The XRT group had lower bowel function than the RP group at t1 and t2 (P = 0.015 and P = 0.040, respectively), but the difference was not significant at t3 and t4. Similarly, a difference in bowel bother was observed only in the early post-treatment period, at t1 (P = 0.017). In the sexual function domain, both treatment groups showed low scores throughout the follow-up time. At t4, however, the RP group had better sexual function than the XRT group (P = 0.017). In contrast, the sexual bother scores of the XRT group were better than those of the RP group at t1, t3 and t4 (P = 0.0054, P = 0.028, and P < 0.001, respectively) (Table 3).

Table 3.   UCLA PCI scores of patients in the post-treatment groups
Time groupProstatectomyRadiation therapyP-value*
  • *

    , Mann–Whitney U-test. Data are presented as mean ± SD. Each domain is scored from 0 to 100 with higher scores representing better quality of life (QOL). UCLA PCI, the University of California, Los Angeles Prostate Cancer Index.

Urinary function
 t165.7 ± 22.683.6 ± 20.60.004
 t269.8 ± 27.181.8 ± 22.20.102
 t384.4 ± 23.390.1 ± 17.60.529
 t481.7 ± 19.988.5 ± 17.50.066
Urinary bother
 t180.8 ± 25.576.3 ± 28.20.585
 t280.1 ± 28.582.4 ± 30.30.671
 t388.6 ± 23.485.0 ± 24.60.597
 t482.7 ± 25.188.0 ± 21.20.382
Bowel function
 t186.0 ± 16.677.3 ± 18.80.015
 t286.8 ± 15.280.3 ± 11.60.040
 t391.6 ± 12.088.3 ± 11.70.295
 t487.0 ± 17.085.3 ± 16.20.615
Bowel bother
 t195.3 ± 12.584.2 ± 19.00.017
 t291.5 ± 17.794.1 ± 14.10.755
 t391.3 ± 16.891.7 ± 20.40.963
 t487.8 ± 23.690.7 ± 15.70.756
Sexual function
 t1 4.5 ± 5.3 8.1 ± 14.20.674
 t2 7.6 ± 10.9 6.4 ± 9.50.489
 t3 7.2 ± 8.7 8.1 ± 12.10.681
 t410.5 ± 13.7 5.8 ± 12.10.017
Sexual bother
 t149.4 ± 36.077.6 ± 29.90.005
 t255.9 ± 33.969.1 ± 42.00.118
 t360.3 ± 35.683.3 ± 24.90.028
 t441.7 ± 36.078.7 ± 36.5<0.001

Impact of age on disease specific HRQOL

Figure 1 shows the disease specific HRQOL scores of the younger and older men of the RP group at t1–t4. The younger men had higher urinary function and bother scores at t4. There was no difference in sexual function between the two groups. In sexual bother, however, the scores of the younger men were worse than those of the older men at t1–t3. In the XRT group, as in the RP group, the younger patients had a lower score of sexual bother than the older ones, despite their similarly poor sexual function score. In bowel function, the older men scored lower than the younger men at t2, as shown in Figure 2.

image

Figure 1.  Comparison of the mean University of California, Los Angeles prostate cancer index (UCLA PCI) scores of the six scales (urinary function [URF], urinary bother [URB], bowel function [BWF], bowel bother [BWB], sexual function [SXF], sexual bother [SFB]) between younger men (white bars) and older men (black bars) at t1–t4 in the radical prostatectomy (RP) group. The Y-axis shows the PCI scores, with higher scores representing better outcomes. *, Differences between the groups are statistically significant (P < 0.05).

Download figure to PowerPoint

image

Figure 2.  Comparison of the mean prostate cancer index (PCI) scores of the six scales (urinary function [URF], urinary bother [URB], bowel function [BWF], bowel bother [BWB], sexual function [SXF], sexual bother [SFB]) between younger men (white bars) and older men (black bars) at t1–t4 in the external beam radiotherapy (XRT) group. The Y-axis shows the PCI scores, with higher scores representing better outcomes. *Differences between the groups are statistically significant (P < 0.05).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

The present study showed that the patients who underwent XRT for prostate cancer had low HRQOL scores just after treatment in most general measures, but after 6 months the general HRQOL appeared to be unaffected by the type of treatment. Physical function and role limitations due to physical problems, classified as the physical domain, scored lower in the XRT group than in the RP group. However, the patients of the XRT group were older than those of the RP group, suggesting a possible influence of age on HRQOL. In contrast, substantial deteriorations of role limitations due to emotional problems, social function, vitality and mental health, classified as the mental domain, were observed right after XRT, but these diminished with time. In a longitudinal study, Leubeck et al. showed that the general HRQOL decreased 3 months after XRT, but returned to the baseline by 12 months.16 Physicians might find it helpful to share this risk information with patients before selection of a therapy.

In the present study, our PCI findings are consistent with those in many of the previous reports on disease specific HRQOL after prostate cancer treatments, showing lower sexual and urinary function after RP, and lower bowel and sexual function after XRT.

Urinary incontinence, one of the most disabling complications of RP, has become an increasingly rare complication with technical advances such as the anatomical approach to RP and nerve sparing procedures. According to our survey, a certain degree of recovery from the postoperative incontinence can be anticipated, and urinary function appears to approach the same level as that of the post-irradiation group within a year. The two types of prostate cancer treatment examined in the present study had different patterns of side-effects on the urinary tract. Patients who underwent RP reported lower urinary function scores. However, there was no significant difference in urinary bother between the groups suggesting that, if urinary incontinence is present after RP, its effects are only minimal. It is important to remember that the urinary function scale of the PCI is designed to measure the urinary leakage symptoms often seen after surgery and not the irritating symptoms often seen after radiation.17 Wei and colleague recently developed a new HRQOL instrument (Expanded Prostate Index Composite) that enables a more comprehensive assessment of the prostate cancer-related HRQOL.18 As an extension of the present study, we are now collecting data on disease-specific QOL using PCI and the International Prostate Symptom Score.

Findings from several studies appear to indicate that bowel problems such as increased frequency, increased urgency, diarrhea, and bleeding with movements are reported more often by men treated with XRT than by men treated with RP.5,19 Recent longitudinal studies have shed light on the trajectory of bowel dysfunction over time. Potosky et al. found that men treated with XRT and those treated with RP experienced declines in bowel function at 4 months after the initiation of treatment, and showed improvements by 10 months post-treatment.20 Takahashi and colleagues reported that XRT is a safe method for managing clinically localized prostate cancer in Japan; adverse events were generally mild and acceptable.21 In the present study, the decrease in bowel function and bowel bother scores was observed only in the early post-treatment period. This finding suggests that XRT can induce complications immediately after treatment, but these complications reduce with time.

Although most of the patients reported good general HRQOL, a significant deterioration of sexual function was observed in both treatment groups. In the second year after treatment, however, the RP group had better sexual function than the XRT group. Our findings confirm the longitudinal trends in sexual function observed by Litwin et al., in which the patients undergoing XRT for early stage prostate cancer began to show declining sexual function during the second year after treatment, whereas the RP group continued to improve.22

In line with the results of several previous studies, results from the present study show that both treatments for localized prostate cancer do not affect general HRQOL outcomes in the long run. In previous studies we reported that older patients (older than 65 years of age) who underwent RP felt less sexual bother despite their lower sexual function, when compared with younger patients (65 years old or younger).23,24 With the XRT group, we found the same trend: younger patients experienced a similar amount of bother from sexual dysfunction as the RP group, although the XRT group was older than the RP group. Although the small number of patients might limit the extent to which we can generalize, it would appear that erectile dysfunction was a burden to the younger patients regardless of the primary treatments. The reason why the patients of the XRT group had better sexual bother score despite their low sexual function score might be related to the fact that most of them were not only more than 70 years old, but they also received neoadjuvant and adjuvant therapy ablation. Accordingly, greater emphasis needs to be placed on the risk of postoperative erectile dysfunction in preoperative counselling.

The present study has several limitations. First, pretreatment differences between surgery and radiotherapy patients are unavoidable because urologists in Japan tend to prefer RP for the majority of patients, relying on XRT for older patients who have existing complications that might make general anesthesia hazardous.25 These clinical differences were apparent in the present study and might partially account for the fact that patients of the XRT group were significantly older and had lower scores in physical domains of general HRQOL, but had better sexual bother scores. Second, because the present study was retrospective, there was a lack of information about the pretreatment functioning of the participants. Third, we did not distinguish between those patients who had undergone nerve sparing procedures, nor those who had received hormone therapy before or after treatment. Finally, patients who chose not to participate in the study might have HRQOL outcomes that were either better or worse than those in this study.

Despite these limitations, we found that early HRQOL outcomes differed according to treatment method for patients with localized prostate cancer. Overall, men who underwent RP experienced more urinary and sexual problems compared with those who were treated with XRT. Alternatively, patients of the XRT group maintained good urinary control but had more bowel dysfunction. Our study using the SF-36 and the PCI, which are both widely used instruments, could provide useful information for patient-centered outcome evaluations. Ethnic or cultural differences in the human costs related to localized prostate cancer treatment also warrant examination. A prospective longitudinal study with a larger patient population of Japanese men is warranted to further elucidate differences in outcome between surgery and radiation for early stage prostate cancer. As a follow up to the present study, we are now collecting prospective multi-institutional outcome data on RP, comparing nerve or non-nerve sparing procedures, perineal prostatectomy, laparoscopic prostatectomy, and radiotherapy.

Life will be forever changed once prostate cancer is effectively diagnosed and treated. However, by characterizing the impact of this disease on HRQOL, we can provide our future patients with a better picture of how their lives will be altered by treatment.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

We performed a retrospective study of general and disease specific HRQOL after RP and XRT in Japanese men. The patients who underwent RP had significantly worse urinary and better bowel function than those treated with XRT. However, these problems diminished within 12 months after treatment. Both treatment groups had decrements in sexual function throughout the post-treatment period; careful attention should be paid in preoperative counselling, especially in younger patients, regardless of the primary treatments.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

This work was supported in part by a grant from the Ministry of Health and Welfare of Japan (11–10). We thank R. Toda for editorial assistance.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References