In 2008, the estimated number of new prostate cancer cases worldwide was almost 900,000; this burden is expected to increase to 1.7 million by 2030 due to the growth and aging of the global population [
The burden associated with prostate cancer diagnosis is high and stems from the diagnosis, the disease itself, and the varying impact of the available treatment options. The majority (90%) of men with low-risk prostate cancers receive radical intervention, with 50–60% undergoing radical prostatectomy as their primary treatment [
This was a prospective, 1-year, observational, pan-European (Germany, France, Spain, Italy, and Sweden) study of men aged 50–75 years with prostate cancer of low-to-moderate risk of progression (Gleason score ≤ 7, PSA ≤ 20 ng/mL and clinical staging T1c–T2b according to D’Amico criteria of low-intermediate risk [
Patients were asked to complete three validated QoL questionnaires: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Prostate Cancer 25 (EORTC QLQ-PR25); EORTC Quality of Life Questionnaire—Cancer 30 (EORTC QLQ-C30); and EuroQoL-5D (EQ-5D) [
The primary study endpoint was the change in QLQ-PR25 urinary symptoms subscale score from baseline to the assessment at 3 months after the start of prostate cancer treatment. Secondary endpoints were changes in other QLQ-PR25 subscale scores from baseline to the 3-month assessment; changes in all QLQ-PR25 subscale scores from baseline to the assessment after 12 months’ treatment; changes in QLQ-C30 scores from baseline to the 3- and 12-month assessments; difference from normative data (based on the UK general population) in baseline EQ-5D and HADS scores; changes in EQ-5D scores from baseline to the 3- and 12-month assessments; changes in HADS scores from baseline to the 3- and 12-month assessments; and cost assessment for prostate cancer subjects based on type of treatment administered, resource utilisation (visits/treatment), and indirect costs captured using the WPAI.
A total of 134 patients per country were needed in order to have 90% power to detect a difference of 4.9 from baseline to 3 months in QLQ-PR25 urinary symptoms subscale scores. The Full Analyses Set (FAS) consisted of all patients who completed the baseline and the 3-month QLQ-PR25 questionnaire and was used for the primary outcome analysis. The Baseline Analyses Set (BAS) consisted of all patients who completed the baseline EQ-5D and HADS questionnaire. The primary endpoint was assessed for the FAS population (observed cases) using a repeated measures analysis of variance with the following covariates: age, centre, initial treatment received, Gleason score, T-stage, PSA test result, education status, and whether the subject had a progressive benign prostatic hyperplasia (BPH) diagnosis (defined by acute urinary retention [AUR]/BPH surgery). For the study endpoints, two-sided 95% confidence intervals for the adjusted mean changes from baseline in the questionnaire scores were calculated along with
A total of 672 patients were enrolled, of whom 603 completed the baseline EQ-5D and HADS questionnaires (BAS population; 86 patients (18 centres) in France, 132 (20 centres) in Germany, 131 each in Italy (9 centres) and in Spain (11 centres), and 123 (9 centres) in Sweden) and 404 completed the baseline and 3-month QLQ-PR25 questionnaires (FAS population); 326 patients completed the 12-month QLQ-PR25 questionnaire. Information on why patients did not complete the questionnaire was not collected. For the BAS population, the median study duration was 398 days (range: 1 to 787 days). Demographic and baseline characteristics are summarised in Table
Baseline demographics and clinical characteristics (BAS population).
Overall |
|
---|---|
Demographic characteristics | |
Age (yrs), |
603 |
Mean ± SD |
|
Median (range) | 66.0 (50–75) |
Education, |
|
Less than high school | 242 (40%) |
High school | 179 (30%) |
Some college/university | 69 (11%) |
College/university graduate | 46 (8%) |
Post graduate (M.S., Ph.D.) | 15 (2%) |
Do not care to answer | 52 (9%) |
Family history of prostate cancer, |
104 (17%) |
Father | 60 (10%) |
Brother | 38 (6%) |
Grandfather | 13 (2%) |
Uncle | 13 (2%) |
Son | 0 |
|
|
Disease characteristics | |
Total Gleason score, |
603 |
≤6 | 400 (66%) |
7 | 203 (34%) |
PSAa (ng/mL), |
603 |
Mean ± SD |
|
Median (range) | 6.35 (0.01–19.30) |
≤10 ng/mL, |
505 (84%) |
11–20 ng/mL, |
98 (16%) |
Clinical staging, |
603 |
T1a | 8 (1%) |
T1b | 5 (<1%) |
T1c | 355 (59%) |
T2a | 123 (20%) |
T2b | 112 (19%) |
Clinical BPHb diagnosis, |
603 |
Yes | 225 (37%) |
Progressive BPH diagnosisc, |
225 |
Yes | 34 (15%) |
The most frequently used initial treatment for prostate cancer was radical prostatectomy (71% of patients). Other treatments were external beam radiotherapy (9%), brachytherapy (3%), combined hormonal therapy/radiotherapy (2%), hormonal therapy alone (1%), and radical prostatectomy followed by salvage radiotherapy (<1%). Ten percent of patients were subjected to active surveillance/watchful waiting, and 2% received other treatment. A total of 176 patients were receiving concomitant medications for genitourinary conditions, the most common of which were alprostadil (7%), tadalafil (7%), bicalutamide (6%), and tamsulosin (6%).
QLQ-PR25 urinary symptoms subscale score was significantly increased at 3 months (
Change from baseline in QLQ-PR25 urinary symptoms subscale score (FAS population, observed cases).
QLQ-PR25 urinary symptomsa,b |
|
Adjusted mean ± SE |
Adjusted mean change |
|
---|---|---|---|---|
Baseline | 401 |
|
||
Month 3 | 403 |
|
9.36 (7.47, 11.25) | <0.001 |
Month 12 | 326 |
|
4.43 (2.70, 6.16) | <0.001 |
Other QLQ-PR25 subscale score endpoints are reported in Table
Change from Baseline in other QLQ-PR25 Subscale Scores (FAS Population, Observed Cases).
QLQ-PR25 subseta |
|
Adjusted mean ± SE |
Adjusted mean change from baseline (95% CI) |
|
---|---|---|---|---|
Incontinence aid poblemsc | ||||
Baseline | 49 |
|
||
Month 3 | 202 |
|
15.81 (5.76, 25.85) | 0.003 |
Month 12 | 124 |
|
9.63 (−1.01, 20.27) | 0.075 |
Bowel symptomsc | ||||
Baseline | 399 |
|
||
Month 3 | 399 |
|
0.60 (−0.24, 1.45) | 0.159 |
Month 12 | 318 |
|
0.84 (−0.14, 1.81) | 0.093 |
Treatment-related symptomsc | ||||
Baseline | 371 |
|
||
Month 3 | 376 |
|
5.42 (4.40, 6.43) | <0.001 |
Month 12 | 306 |
|
5.11 (4.00, 6.21) | <0.001 |
Sexual Functioningb | ||||
Baseline | 293 |
|
||
Month 3 | 221 |
|
−26.54 (−30.57, −22.50) | <0.001 |
Month 12 | 194 |
|
−27.67 (−31.43, −23.91) | <0.001 |
Sexual activityb | ||||
Baseline | 397 |
|
||
Month 3 | 401 |
|
−10.29 (−12.99, −7.59) | <0.001 |
Month 12 | 323 |
|
−7.39 (−10.34, −4.44) | <0.001 |
For the QLQ-C30 questionnaire, global health status/QoL score significantly decreased at month 3 but was not different from baseline by month 12. Scales for physical, role and social functioning, and fatigue, showed significant deterioration at 3 and 12 months. No significant change was observed in cognitive functioning, while emotional functioning significantly improved. Pain score was significantly worse compared with baseline at month 3 but not at month 12. Nausea and vomiting score was largely unaffected (Table
Change from baseline in QLQ-C30 scales (FAS population, observed cases).
QLQ-C30 subseta |
|
Adjusted mean ± SE |
Adjusted mean change |
|
---|---|---|---|---|
Global health status/QoL scaleb | ||||
Baseline | 400 |
|
||
Month 3 | 401 |
|
−3.19 (−5.26, −1.12) | 0.003 |
Month 12 | 323 |
|
−0.02 (−2.25, 2.22) | 0.987 |
Physical functioning scaleb | ||||
Baseline | 397 |
|
||
Month 3 | 394 |
|
−3.81 (−5.03, −2.59) | <0.001 |
Month 12 | 322 |
|
−1.95 (−3.18, −0.72) | 0.002 |
Role functioning scale b | ||||
Baseline | 400 |
|
||
Month 3 | 400 |
|
−8.81 (−11.05, −6.58) | <0.001 |
Month 12 | 324 |
|
−3.77 (−5.65, −1.88) | <0.001 |
Emotional functioning scaleb | ||||
Baseline | 396 |
|
||
Month 3 | 401 |
|
3.20 (1.26, 5.13) | 0.001 |
Month 12 | 323 |
|
6.26 (4.37, 8.14) | <0.001 |
Cognitive functioning scaleb | ||||
Baseline | 302 |
|
||
Month 3 | 305 |
|
−1.21 (−2.92, 0.49) | 0.162 |
Month 12 | 245 |
|
−1.51 (−3.15, 0.13) | 0.071 |
Social functioning scaleb | ||||
Baseline | 399 |
|
||
Month 3 | 400 |
|
−6.89 (−8.92, −4.85) | <0.001 |
Month 12 | 323 |
|
−3.83 (−5.74, −1.92) | <0.001 |
Fatigue scalec | ||||
Baseline | 399 |
|
||
Month 3 | 399 |
|
5.22 (3.53, 6.90) | <0.001 |
Month 12 | 322 |
|
2.78 (1.26, 4.30) | <0.001 |
Nausea and vomiting scalec | ||||
Baseline | 402 |
|
||
Month 3 | 401 |
|
0.06 (−0.79, 0.92) | 0.882 |
Month 12 | 325 |
|
0.20 (−0.67, 1.07) | 0.652 |
Pain scalec | ||||
Baseline | 399 |
|
||
Month 3 | 400 |
|
3.36 (1.26, 5.47) | 0.002 |
Month 12 | 326 |
|
−0.38 (−2.06, 1.29) | 0.652 |
There was no significant change from baseline in EQ-5D scores at 3 and 12 months following treatment for prostate cancer (data not shown). Compared with age-matched normative data (UK general population), the health status (EQ-5D) of the study population was similar to the general population at baseline and month 12, but significantly worse 3 months after starting treatment.
Anxiety score (HADS anxiety subscale) was significantly reduced (
Approximately 25% of patients were employed at the time of entering the study. Fewer patients remained in work after they received a prostate cancer treatment (19% at month 3 and 16% at month 12). On average, 32 working hours in the previous week were reported at baseline; these hours slightly decreased following initial treatment. The average missed working hours due to prostate cancer in the previous week was 6 hours at baseline and month 3, and 2 hours at month 12. Based on the overall WPAI scores, diagnosis and treatment of prostate cancer had no impact on working productivity and on regular daily activities over the course of the study.
Medical costs including resource utilisation associated with prostate cancer diagnosis and/or treatment were analysed. Of 603 subjects in the BAS population, 96% (
In this observational study, the majority of patients recruited were due to receive treatment of curative intent such as radical prostatectomy, external radiotherapy, brachytherapy, and hormone therapy, despite having tumours of low-to-moderate risk. For example, almost three-quarters of patients (71%) underwent radical prostatectomy as primary treatment. Overall, treatment of prostate cancer had negative effects on all six domains of the prostate cancer-specific QLQ-PR25 questionnaire. Urinary symptoms were generally worse after 3 months and although they improved after 12 months, they remained significantly worse than at baseline. Incontinence aid problems were also worse after 3 months than after 12 months (only significant versus baseline at month 3). Unlike urinary symptoms, sexual functioning did not tend to improve after one year compared with 3 months, suggesting that the impact of treatment on sexual function may be longer-lasting and more profound compared with the effect on urinary symptoms. On the other hand, sexual activity score did show a slight improvement after 12 months compared with 3 months, although it remained significantly lower compared with baseline. Bowel symptoms scores worsened, particularly among patients whose initial treatment included radiotherapy (data not shown), although overall the change from baseline was not statistically significant. Worsening of urinary symptoms among patients managed with active surveillance may suggest an age-related natural decline, although age was taken into account in our analyses as a covariate.
The primary endpoint for the study was the change in QLQ-PR25 urinary symptoms subscale score from baseline to the assessment at 3 months after the start of prostate cancer treatment. For some prostate cancer interventions (e.g., radiation therapy), the adverse effects are not immediately evident; the 3-month time point was therefore selected on the basis that this would enable the adverse impact of most interventions to become apparent. The urinary symptoms subscale score was chosen in order to have a primary endpoint that was common across all prostate cancer interventions. The number of evaluable patients was less than planned, which had the potential to adversely affect power. However, despite this, statistically significant differences were observed for the primary endpoint, both overall and in each individual country.
Using the more general cancer QoL QLQ-C30 questionnaire, physical, role, and social functioning significantly worsened following treatment, as did fatigue and pain symptoms. In contrast, emotional functioning improved after treatment. Changes in QLQ-PR25 and C30 questionnaires were not strongly correlated with health status (as assessed by EQ-5D scores), which did not significantly change from baseline. One possible explanation for this lack of correlation is that the EQ-5D assessment may not have been sensitive enough to reflect the changes in health-related QoL that occur following interventions for prostate cancer.
Patients’ anxiety and depression levels were relatively unaffected at the time of diagnosis, as mean scores were within the normal range of the HADS questionnaire. These scores might have been expected to deteriorate over time, as other studies have shown a high incidence of anxiety, depression, and distress [
The majority of patients (75%) were not working at study entry, thereby limiting the data on the impact of prostate cancer on work productivity. However, for those subjects who did work our data suggest that any impact on work productivity may be transitory, and improves over time, following prostate cancer treatment.
Our study primarily examined the burden of low-to-moderate risk prostate cancer from the perspective of the disease state. Nevertheless, our findings are generally in line with those from other studies that have assessed the reported impact of prostate cancer treatment on various aspects of physical functioning. Radical prostatectomy, radiotherapy, hormonal therapy, and watchful waiting have all been shown to have a negative impact on sexual function, urinary function, and bowel function of patients treated for prostate cancer [
A potential limitation of our study is that the results rely on patient recall after 3 months and after 12 months. Our study may also have benefited from a longer follow-up period than 12 months. In addition, because of the observational nature of the study, there may be differences in factors such as age, Gleason score, and tumour stage according to initial treatment received; this may restrict comparison of the data according to initial treatment type, and also prevented any further subgroup analyses according to initial treatment received. Another possible limitation is the high proportion of patients treated with radical prostatectomy, which may limit the external validity of our findings. However, it is important to note that interventions were not restricted in the study protocol, with treatment decisions left to the physician/patient based on individual patient circumstances. In this respect, our study population reflects real-life clinical practice. Further, the proportion of patients treated by surgery in this study is consistent with other published data that show radical prostatectomy is the most common treatment in men with low-to-moderate risk prostate cancer [
In the present study, statistically significant differences compared with baseline were demonstrated for several of the QoL subscales assessed; however, this does not necessarily translate to clinically important effects. There is currently no definition available as to what would represent a clinically meaningful change in QLQ-PR25 scores, although some information is available on the minimal clinically important difference in QLQ-C30 scores [
QoL considerations are important in helping guide treatment decision-making in patients with prostate cancer, especially those at low-to-moderate risk of progression. The majority of these patients have a favourable prognosis and are not destined to die of their disease even in the absence of treatment. However, overtreatment of indolent disease may be a problem given that the various treatment options can have a significant negative impact on a patient’s health status and QoL. This study showed that, across France, Germany, Italy, Spain, and Sweden, low-to-moderate risk prostate cancer is usually treated with radical therapies. This treatment strategy had a negative impact on various dimensions of QoL in these patients during the one-year observation period. Prostate cancer treatment was associated with a decrease in urinary and sexual functioning and increase in hormonal treatment-related symptoms. With some exceptions, the impact on treatment-related functioning scales and symptoms tended to be higher at month 3 and to have some degree of improvement at month 12. Treatment of prostate cancer had minimal effects on depression and anxiety, and a limited impact on productivity among active workers.
In conclusion, low-to-moderate risk prostate cancer may have a substantial effect on the QoL of affected patients within one year following treatment. Our study provides further supportive information on the QoL impact of treatments for low-to-moderate risk prostate cancer and will help physicians to tailor discussions with patients and guide decision making for disease management, particularly with regard to the primary treatment chosen.
Laure Benjamin, Matthew Somerville, Juan-Manuel Palacios, Libby Black, and Ramiro Castro are employees of GlaxoSmithKline. Javier Burgos, Anders Bjartell, and Cesare Selli have no potential conflicts to declare in relation to this paper.
This study was funded by GlaxoSmithKline. Medical writing support was provided by Tony Reardon of Spirit Medical Communications Limited and funded by GlaxoSmithKline. The authors thank the following investigators and their patients for participating in the study: (France) Ahmadraseen Atassi, Didier Ayuso, Jacques Benchetrit, Frédéric Boutemy, Franck Bruyere, Eric Chartier, Antony Cicco, Laurent Dahmani, Alexandre De La Talle, Marc Fourmarier, Olivier Haillot, Didier Hollard, Mahmoud Kahil, Olivier Lan, Richard Mallet, Vincent Ravery, Jean Paul Regin, Jacques Schlosser, and Xavier Stefaniak; (Germany) Lothat Bauer, Thomas Benusch, Ralk Eckert, Christian Girke, Petra Groeschel, Tom Henschel, Karin Herrman, Toralf Kellner, Rainer Klammert, Tilo Koettig, Ullrich Matz, Stefan Mohr, Detlef Nietzsch, Detlef Quast, Peter Rothe, Mattias Solga, Thomas Walter, Wolfgang Warnack, Joerg Willgerodt, and Alexander Von Keitz; (Italy) Giampaolo Bianchi, Giorgio Carmignani, Bruno Frea, Vincenzo Gentile, Paolo Gontero, Vincenzo Mirone, Francesco Montorsi, and Arcangelo Pagliarulo; (Spain) Javier Angulo Cuesta, Pedro Arrosagaray, Jaime Bachiller Burgos, José Manuel Cozar Olmo, Javier Extramiana, Eladio Franco, Jordi Huguet, Jose Maria Martinez Sagarra, Bernardino Miñana, and Manuel Sanchez Chapado; (Sweden) Jan-Erik Damber, Per Folmerz, Eirikur Gudmundson, Ali Khatani, Börje Ljungberg, Elisabeth Nelson, Åke Paradis, and Yu-Hui Wang.