Five-year follow-up of health-related quality of life after primary treatment of localized prostate cancer
Version of Record online: 30 MAR 2005
Copyright © 2005 Wiley-Liss, Inc.
International Journal of Cancer
Volume 116, Issue 2, pages 291–296, 20 August 2005
How to Cite
Korfage, I. J., Essink-Bot, M.-L., Borsboom, G. J.J.M., Madalinska, J. B., Kirkels, W. J., Habbema, J. D. F., Schröder, F. H. and de Koning, H. J. (2005), Five-year follow-up of health-related quality of life after primary treatment of localized prostate cancer. Int. J. Cancer, 116: 291–296. doi: 10.1002/ijc.21043
- Issue online: 24 MAY 2005
- Version of Record online: 30 MAR 2005
- Manuscript Accepted: 18 JAN 2005
- Manuscript Received: 30 JUN 2004
- Dutch Cancer Society, Amsterdam. Grant Numbers: EUR 2000-2329, EUR 95-11.43
- quality of life;
- prostatic neoplasms;
- prostate-specific antigen;
Although with earlier detection of prostate cancer more men face the long-term consequences of primary treatment, studies on the impact of treatment on long-term health-related quality of life (HRQoL) are scarce. We followed 314 men with newly diagnosed localized prostate cancer from 1 month before until 5 years after radical prostatectomy (n = 127) or external beam radiotherapy (n = 187; median follow-up = 52 months). Questionnaires addressing disease-specific (UCLA PCI) and generic (SF-36, EQ-5D) HRQoL were sent 1 month before and 6, 12 and 52 months after treatment. Repeated-measures modeling was used to study HRQoL over time. Regular urinary leakage was reported by 12% of prostatectomy patients before treatment and by 31% at the 52-month assessment. Erectile dysfunction before treatment was reported by 31% of prostatectomy patients and by 40% of radiotherapy patients; at the 52-month assessment, these percentages were 88% and 64%, respectively. Erectile dysfunction present at 1 year posttreatment can be considered permanent. Prostatectomy patients reported better generic functioning both before and after treatment than radiotherapy patients, who were on average 5.9 years older and had more comorbid conditions. General physical functioning of prostatectomy patients slightly improved over time, but declined in radiotherapy patients. The relation between age and physical scores was found to be nonlinear. The long-term physical decline in radiotherapy patients partly resulted from aging and its nonlinear impact on health, although treatment effects cannot be excluded. Scores of both patient groups remained above those of norm populations. Innovative graphs describing disease-specific and generic functions after treatment can help patients and physicians in their treatment choices. © 2005 Wiley-Liss, Inc.
In Western countries, prostate cancer is the most prevalent malignancy in males.1 Due to early detection and treatment, the number of men with long posttreatment survival of localized prostate cancer is increasing. However, the growing tendency for prostate screening and the concomitant overdetection and overtreatment2 warrant careful evaluation of treatment effects on health-related quality of life (HRQoL). Prostatectomy is reported to lead to better disease-specific survival than watchful waiting,3 but data are lacking concerning radical prostatectomy versus external beam radiotherapy, which are the most commonly used primary therapies. A randomized controlled treatment trial in the United Kingdom is currently comparing survival and HRQoL after radical prostatectomy, external beam radiotherapy and monitoring.4
While awaiting results from randomized trials, longitudinal studies with long-term follow-up are important to assess the effects of treatment on HRQoL, defined as a patient's physical, psychologic and social functioning and well-being. A longitudinal prospective cohort study on the effects of primary therapy in localized prostate cancer was started in the Netherlands in 1996: the Rotterdam study,5, 6 within the context of the European Randomized Study of Screening for Prostate Cancer (ERSPC).7 The principal aims of the Rotterdam study were to assess the frequency of side effects of primary therapy and to determine HRQoL in men with localized prostate cancer up to 5 years after primary treatment with radical prostatectomy or external beam radiotherapy.
Material and methods
Prostate cancer patients from 4 hospitals (University Medical Center Rotterdam; St. Franciscus Gasthuis; Medical Center Rotterdam Zuid, locations “Clara” and “Zuider”) participated in the Rotterdam study. Patients were enrolled between June 1996 and January 1999, on average 1 month before the start of nonrandomly allocated primary treatment, consisting of radical prostatectomy (intentionally nerve-sparing) or external beam radiotherapy (comprising an average of 33 radiation sessions over 7 weeks). For details of patient recruitment and first-year results, see Madalinska et al.5, 6 Written informed consent was obtained from all respondents. The medical ethics committees of all 4 hospitals approved the study design.
Respondents completed postal self-assessment questionnaires on 4 occasions: 1 month before and 6, 12 and 52 months after treatment. The following disease-specific and generic (i.e., comprehensive and nondisease-specific) HRQoL measures were used.
One, the University of California, Los Angeles (UCLA), Prostate Cancer Index (PCI) measures disease-targeted HRQoL in men treated for early-stage prostate cancer.8 Following formalized procedures, 4 of its 6 scales were adapted for use in Dutch: urinary function (5 items, including frequency of urinary leakage, number of pads used to control leakage), bowel function (4 items, including frequency of rectal urgency and of crampy pain in abdomen or pelvis), urinary bother and bowel bother (1 item each). Psychometric properties (feasibility, score distribution, reliability, construct validity and responsiveness to change) of the Dutch version of the UCLA PCI have been described earlier.9 SF-36 missing item procedures were used for imputation of missing responses in PCI items.10 Each scale is scored from 0 to 100, with higher scores representing better outcomes. Differences of more than 5–10 points are considered clinically meaningful.11
Two, sexual function was defined as regularly or often having problems in achieving or maintaining an erection if wished to, or not being sexually active because of erectile problems. This was measured with 12 Dutch single items derived from Slob et al.9, 12
Three, generic HRQoL was measured with the RAND 36-item Short-Form Health Survey (SF-36) and the EuroQol classification (EQ-5D). Both measures are available in over 50 languages and are considered applicable in a wide range of populations, health conditions and treatments. The SF-36 consists of 8 scales: physical functioning (10 items), role limitations due to physical problems (role-physical, 4 items), bodily pain (2 items), general health perception (5 items), vitality (4 items), social functioning (2 items), role limitations due to emotional problems (role-emotional, 3 items) and mental health (5 items). Scales are transformed to ranges of 0–100 with higher scores indicating better functioning.13 Differences of at least 6.5 points in the physical dimensions and 7.9 points in the mental dimensions are considered clinically meaningful.14 Procedures concerning imputation of missing responses in SF-36 items were conducted according to the guidelines of the SF-36 Health Survey Manual.10
The EQ-5D classification consists of 5 items (mobility, self-care, usual activities, pain/discomfort and anxiety/depression) complemented by a visual analogue scale (VAS) for valuation of own health. Classification scores can be linked to a utility score.15
Additionally, information on age, marital status, educational level and profession was obtained. Educational level was classified as low (primary school or lower technical education), high (college/university degree), or intermediate. To assess comorbidity, respondents were also asked which of 28 chronic conditions on a standardized list they were experiencing or had experienced over the past year (Dutch Health Interview Survey, Statistics Netherlands).
Baseline clinical information on tumor stage (Tumor-Node-Metastasis clinical classification),16 histopathologic tumor (biopsy) grade and urologic treatment history was obtained from the Regional Cancer Registry. Data on the clinical or biochemical progression at the time of the respondent's long-term HRQoL assessment were obtained from the treating physicians. Biochemical recurrence was defined as a prostate-specific antigen (PSA) level of at least 0.2 ng/mL after prostatectomy, confirmed on repeat testing, or a rise in PSA level of at least 0.5 ng/mL after radiotherapy, confirmed on a repeat testing.
Between-group differences in background characteristics and descriptive statistics for the HRQoL scales were calculated using SPSS for Windows, release 10.0.7. A p-value less than 0.01 (referring to 2-sided statistical tests) was considered significant. The chi-square test was used for categorical variables and the Mann-Whitney U-test for continuous variables.
The development over time of disease-specific and generic function scores in the 2 groups was analyzed with repeated-measures analysis of variance (ANOVA) using proc mixed from the SAS system for Windows release 8.2. Random intercept models were applied that allowed for the use of all available data, including incomplete records. The models comprised the main effects of treatment and time and the interaction between treatment and time. Time was included as a factor with 4 levels (1 for each assessment) to account for possible nonlinearities in the development of HRQoL scores. A nonresponse analysis was performed; details on this are available from the authors.
Between June 1996 and January 1999, 415 eligible men were identified, of whom 387 consented to participate in this study (Fig. 1). Primary treatment consisted of prostatectomy (n = 127) or radiotherapy (n = 187). Men referred to watchful waiting (n = 25) or advanced disease therapy (n = 48) were excluded from analysis.
There was a significant difference in age, comorbidity and average PSA levels between the 2 treatment groups. Compared to radiotherapy patients, prostatectomy patients were on average 5.9 years younger (p < 0.01; Table I), had on average 0.4 fewer comorbid conditions (p < 0.01) and had a lower PSA level (p < 0.01). TNM stages were more favorable in prostatectomy patients (p = 0.06).
|Prostatectomy (n = 127)||Radiotherapy (n = 187)||p-value|
|Age at baseline (years)||0.00|
|Average ± SD (range)||62.3 ± 5.2 (49–74)||68.2 ± 5.8 (49–82)|
|Educational level (%)||0.08|
|Marital status (%)||0.92|
|Married or cohabiting||87%||87%|
|Average number of conditions||0.6||1.0|
|PSA level before treatment, ng/mL||0.00|
|Average ± SD||9.6 ± 15.6||15.4 ± 24.3|
|Tumor stage before treatment||0.06|
|Tumor grade before treatment||0.84|
The overall response rate to all 4 questionnaires was 76%, representing 224 of 294 men still alive at the 52-month assessment. Median and mean time to long-term follow-up was 52 months (range, 45–58 months).
Information on recurrence was available for 94% of the 52-month respondents. There was clinical evidence of recurrence in at least 7% (7/97) of the prostatectomy patients and in 22% (28/127) of the radiotherapy patients.
Average age differed considerably between treatment groups (± 6 years). Because the relation between age and physical functions in particular was found to be nonlinear (as the decline with aging was generally steeper for older subjects than for younger ones), age adjustment would not have been appropriate. Therefore, we decided to present the HRQoL scores over time in graphs with the differences in age between the treatment groups presented on the x-axes.
The impact of prostatectomy had a marked effect on urinary function, urinary bother and erectile dysfunction (Table II). The proportion of prostatectomy patients experiencing regular urinary leakage (i.e., at least several days a week) was 12% (15/127) before treatment, 37% (46/123) at the 12-month assessment and 31% (30/97) at the 52-month assessment. Of all prostatectomy patients, 7% reported wearing pads for incontinence before treatment; this increased to 34% at the 12-month assessment and to 25% at the 52-month assessment. Radiotherapy had a marked effect on bowel function and bowel bother scores (Table II).
|Radical prostatectomy||External beam radiotherapy|
|Pretreatment (n = 127)||6 months (n = 123)||12 months (n = 123)||52 months (n = 97)||Pretreatment (n = 187)||6 months (n = 182)||12 months (n = 179)||52 months (n = 127)|
|Urinary function||93 ± 16||70 ± 27||77 ± 24||79 ± 24||95 ± 12||92 ± 15||93 ± 15||91 ± 16|
|Every day||10 (8%)||46 (37%)||28 (23%)||23 (24%)||7 (4%)||12 (7%)||12 (7%)||5 (4%)|
|Several days a week||5 (4%)||14 (11%)||18 (15%)||7 (7%)||8 (4%)||11 (6%)||9 (5%)||12 (9%)|
|No control||3 (2%)||7 (6%)||5 (4%)||2 (2%)||3 (2%)||3 (2%)||1 (1%)||3 (2%)|
|Frequent dribbling||4 (3)||19 (15%)||14 (11%)||10 (10%)||11 (6%)||11 (6%)||10 (6%)||11 (9%)|
|Use of pads|
|3 or more pads a day||3 (2%)||21 (17%)||12 (10%)||7 (7%)||1 (1%)||3 (2%)||2 (1%)||5 (4%)|
|1–2 pads a day||6 (5%)||42 (34%)||30 (24%)||17 (18%)||1 (1%)||11 (6%)||11 (6%)||3 (2%)|
|Urinary bother||90 ± 23||71 ± 36||79 ± 31||82 ± 28||89 ± 24||85 ± 27||89 ± 24||86 ± 25|
|Big problem||3 (2%)||17 (14%)||10 (8%)||4 (4%)||4 (2%)||7 (4%)||4 (2%)||3 (2%)|
|Moderate problem||4 (3%)||6 (5%)||5 (4%)||4 (4%)||8 (4%)||9 (5%)||7 (4%)||6 (5%)|
|Bowel function||90 ± 16||91 ± 15||90 ± 16||89 ± 13||91 ± 12||81 ± 22||81 ± 21||84 ± 19|
|Loose or liquid stools|
|Always||1 (1%)||1 (1%)||1 (1%)||1 (1%)||9 (5%)||4 (2%)|
|Usually||4 (3%)||3 (2%)||6 (5%)||2 (2%)||8 (4%)||19 (10%)||13 (7%)||10 (8%)|
|Several times a day||1 (1%)||1 (1%)||2 (2%)||2 (2%)||3 (2%)||11 (6%)||6 (3%)||3 (2%)|
|About once a day||6 (5%)||1 (1%)||3 (2%)||2 (2%)||3 (2%)||4 (2%)||5 (3%)||3 (2%)|
|Bowel bother||94 ± 18||94 ± 19||93 ± 20||95 ± 14||95 ± 17||81 ± 27||78 ± 29||81 ± 29|
|Big problem||1 (1%)||2 (2%)||2 (2%)||2 (1%)||6 (3%)||9 (5%)||4 (3%)|
|Moderate problem||4 (3%)||4 (3%)||3 (2%)||1 (1%)||3 (2%)||9 (5%)||10 (6%)||10 (8%)|
|Erectile dysfunction (ED)|
|Sexually active and ED||32 (25%)||39 (32%)||47 (38%)||41 (42%)||40 (21%)||52 (29%)||46 (26%)||36 (28%)|
|Sexually inactive because of ED||7 (6%)||71 (58%)||61 (50%)||44 (45%)||34 (18%)||42 (23%)||50 (28%)||45 (35%)|
|Total ED1||39 (31%)||110 (89%)||108 (88%)||85 (88%)||74 (40%)||94 (52%)||96 (54%)||81 (64%)|
Prostatectomy patients had higher scores on all 8 SF-36 scales, higher EQ-5D utility scores and higher valuation of own health than radiotherapy patients (Table III). Differences in mean scale scores between groups were smallest for mental health and largest for role-physical. Comparison with Dutch reference scores17 showed that the pretreatment SF-36 scale scores of prostatectomy patients were higher except for the mental health scores, which were significantly lower (data not shown). Radiotherapy patients had lower scores for mental health and in general health perceptions, similar scores for vitality and social functioning and higher scores for the remaining SF-36 scales. At the 52-month follow-up assessment, the prostatectomy patients scored significantly better than the general population on all SF-36 scales, whereas radiotherapy patients scored equal to or significantly better than (role-physical, bodily pain and role-emotional) the general population. Both treatment groups scored equal to or higher than the U.S. national age-adjusted norms for males10 except for pretreatment mental health (prostatectomy and radiotherapy group) and general health (radiotherapy group).
|Radical prostatectomy||External beam radiotherapy|
|Pretreatment (n = 127)||6 months (n = 123)||12 months (n = 123)||52 months (n = 97)||Pretreatment (n = 187)||6 months (n = 182)||12 months (n = 179)||52 months (n = 127)|
|Physical functioning||89 ± 16||87 ± 15||86 ± 17||87 ± 15||77 ± 23||76 ± 23||74 ± 24||68 ± 27|
|Role-physical||83 ± 33||82 ± 32||87 ± 30||90 ± 26||70 ± 40||72 ± 39||71 ± 41||67 ± 44|
|Bodily pain||89 ± 18||92 ± 17||90 ± 17||88 ± 18||80 ± 25||82 ± 24||80 ± 24||79 ± 23|
|General health||69 ± 17||74 ± 18||73 ± 17||69 ± 20||58 ± 18||63 ± 20||61 ± 19||61 ± 22|
|Vitality||78 ± 18||75 ± 19||76 ± 19||76 ± 16||69 ± 19||69 ± 21||68 ± 21||67 ± 21|
|Social functioning||87 ± 19||89 ± 18||90 ± 19||91 ± 16||81 ± 23||85 ± 21||84 ± 23||84 ± 24|
|Role-emotional||84 ± 32||88 ± 27||90 ± 25||89 ± 27||77 ± 36||84 ± 32||83 ± 32||80 ± 35|
|Mental health||72 ± 16||82 ± 18||81 ± 17||82 ± 16||68 ± 18||80 ± 19||79 ± 19||80 ± 19|
|Summary||53 ± 7||52 ± 7||52 ± 7||52 ± 6||48 ± 9||47 ± 10||46 ± 10||44 ± 11|
|Mental component||51 ± 9||54 ± 9||55 ± 9||54 ± 8||50 ± 10||54 ± 10||54 ± 10||54 ± 10|
|Utility||89 ± 15||91 ± 16||90 ± 17||88 ± 18||81 ± 20||83 ± 21||82 ± 20||76 ± 23|
|Valuation of own||79 ± 17||84 ± 12||81 ± 13||81 ± 13||72 ± 17||76 ± 17||73 ± 16||74 ± 16|
HRQoL over time
Results of the repeated-measures model are shown in Figure 2. After prostatectomy, there was a steep decline in scale scores for urinary function and urinary bother (Fig. 2a and b), and after radiotherapy, in scores for bowel function and bowel bother (Fig. 2c and d). Generic physical function scores (Fig. 2e and f) of radiotherapy patients declined between the 12- and 52-month assessment, whereas those of prostatectomy patients increased slightly. The scores for bodily pain and valuation of own health (Fig. 2g and j) showed similar patterns in both treatment groups; the average scores, however, were lower in the radiotherapy group. Mental health score patterns were identical for both groups, showing no influence of either age or treatment choice (Fig. 2h).
Cross-sectional studies have shown that localized prostate cancer treatments can have serious consequences for urinary, bowel and sexual function.18 Longitudinal follow-up studies, however, provide better insight in the effect of treatment on disease-specific and generic function than cross-sectional studies, especially when a pretreatment assessment is included. However, the number of published cohort studies is limited.11, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28 In the present longitudinal prospective cohort study, we followed 314 men with localized prostate cancer from pretreatment until 5 years after treatment; the overall response rate was 76%. Average follow-up was longer and the number of patients larger than in most other prospective cohort studies. In one study on health outcomes after prostatectomy or radiotherapy, 1,591 prostate cancer patients participated, but this study did not include a pretreatment assessment.22
Previous longitudinal cohort studies found that prostatectomy mainly affected urinary and sexual functioning, whereas radiotherapy had consequences for bowel and sexual functioning. These results were confirmed by the present study.
In our study, there were some important baseline differences between the prostatectomy and radiotherapy patients. Prostatectomy patients had fewer comorbid conditions, a lower PSA level, a more favorable TNM stage and were on average 5.9 years younger than radiotherapy patients. These differences were due to the nonrandom treatment allocation. Randomized controlled trials may ensure comparable groups at baseline; however, only a selected group of prostate cancer patients will be eligible for random allocation to either treatment option.
We found a nonlinear relation between age and physical function: cross-sectional HRQoL scores at baseline showed a steep decrease with increasing age of the patient. Assuming that the effect of illness on HRQoL at baseline was small and similar in both treatment groups, this decrease with age was probably the result of aging. It is likely that a similar effect of age occurred in the within-person or longitudinal changes in HRQoL. Such an effect, however, could have been confounded by the effects of having prostate cancer and of treatment, both of which may also vary with age. Because of this nonlinear relation, adjustment for age would not have been appropriate. Therefore, we have presented the results as a descriptive analysis of the average course of HRQoL over age in 2 treatment groups. The innovative graphs in our study illustrate disease-specific and generic function in the years following treatment and provide a useful tool for patients and physicians in their treatment choices.
The unclear overall benefit of PSA screening and the high prevalence of side effects of primary treatment complicate decision-making about PSA testing and prostate cancer treatment. Increasing attention is currently paid to the process of shared or informed decision-making. Information on the development of disease-specific and generic functions over the years as well as the nonlinear relationship with age is useful for both physicians and patients in this process of decision-making.
At the 52-month assessment, we observed a decrease in general physical function of radiotherapy patients, whereas physical function and role-physical scores of prostatectomy patients showed no decline over time in spite of aging. The decrease in radiotherapy patients may be explained by advancing age and its nonlinear impact on health, but treatment effects or the higher rate of recurrence in the radiotherapy arm may also have played a role.29
In both treatment groups, mental health scale scores showed a striking increase at the 6-month assessment. At the pretreatment assessment, the men had just been informed that they had prostate cancer, whereas 6 months later, they had undergone treatment for this disease. We hypothesize that the diagnosis of prostate cancer is the main stressor (not the treatment) and that treatment reduces distress.
Another interesting finding is that in both treatment groups, the urinary and bowel scale scores before treatment in both treatment groups were better than those in an age-related sample of U.S. men without prostate cancer (n = 134; mean age = 66.4 years).30 SF-36 scores, both before and after treatment, were equal to or higher than Dutch and U.S. age- and sex-adjusted norms. The better generic function of prostate cancer patients before treatment may be the result of selection bias, since socioeconomic status is positively associated with prostate cancer incidence.31 Possible explanations for the higher generic scores after treatment include insensitivity of generic instruments for changes induced by prostate cancer diagnosis and treatment and response shift.32 Patients may reevaluate the value of life itself when (after intentionally curative treatment for cancer) they are still alive and in a reasonably good functional state, i.e., they may perceive urinary, bowel, or sexual dysfunction as an inevitable consequence of treatment.
Confronted with similar results, Krahn et al.33 concluded that although sexual, urinary and bowel dysfunctions are common and important, their impact on the health status of prostate cancer patients may be overstated. We believe, however, that even if patients accept the side effects of treatment and are prepared to put up with them, these side effects should never be taken lightly in the context of screening evaluation. Further study is required to establish what men perceive as the benefits or disadvantages of screening.
The authors are grateful to all patients for their participation in the study. They also thank the departments of urology of the 4 participating hospitals for their cooperation and J.H. Groenewoud for critical comments on the manuscript.
- 7Large-scale randomized prostate cancer screening trials: program performances in the European randomized screening for prostate cancer trial and the prostate, lung, colorectal and ovary cancer trial. Int J Cancer 2002; 97: 237–44., , , , , , , , , , , , et al.
- 10SF-36 health survey: manual and interpretation guide. Boston: Health Institute, New England Medical Center, 1993., , , .
- 16TNM classification of malignant tumours. Berlin: Springer-Verlag, 1992., .