Mikio Sugimoto md, Department of Urology, Kagawa University Faculty of Medicine, Kita-gun, Kagawa 761-0793, Japan. Email: email@example.com
Objectives: To evaluate health related quality of life (HRQOL) using the Medical Outcomes Study 8-items Short Form Health Survey (SF-8) questionnaire in Japanese patients with early prostate cancer.
Methods: A cross-sectional analysis was done in 457 patients with prostate cancer treated with radical prostatectomy, external beam radiotherapy, brachytherapy, androgen deprivation therapy, and watchful waiting or a combination these therapies. General HRQOL was measured using the Japanese version of the SF-8 questionnaire and disease-specific HRQOL was assessed using the Japanese version of the Extended Prostate Cancer Index Composite.
Results: The external beam radiotherapy group reported significantly lower values for the physical health component summary score (PCS) in comparison to the radical prostatectomy and brachytherapy groups (P < 0.05). In the analysis of both the PCS and the mental health component summary score (MCS) over time after treatment, higher scores with time were found in the radical prostatectomy group. No significant change over time after androgen deprivation therapy in the PCS was found. In contrast, the MCS was found to deteriorate in the early period, showing a significant increase over time.
Conclusions: SF-8 in combination with the Extended Prostate Cancer Index Composite has shown to be a helpful tool in the HRQOL assessment of Japanese patients treated for localized prostate cancer.
With the increase in early stage prostate cancer, various treatment modalities such as radical prostatectomy (RPx), external beam radiotherapy (EBRT), brachytherapy (BR), androgen deprivation therapy (ADT), and watchful waiting (WW), are utilized and generally provide equivalent treatment outcomes. In patients with low risk prostate cancer, it is assumed that there are few differences in recurrence rate and survival between the therapeutic modalities. Based on comparable treatment outcomes, factors influencing the patient's choice of primary therapy include consideration of general health-related quality of life (HRQOL) which is becoming an important determinant in the choice of treatment not only for patients but also for clinicians.
General HRQOL has been assessed using several instruments including the Medical Outcomes Study 36 Items Short Form Health Survey (SF-36), the Medical Outcomes Study 12 Items Short Form Health Survey (SF-12), the Study protocol for the World Health Organization project to develop a quality of life assessment instrument (WHOQOL), the King's Health Questionnaire (KHQ), the Functional Assessment of Cancer Therapy General (FACT-G) and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) to date. FACT-G and EORTC QLQ-C30 were specifically developed for the assessment of general HRQOL with cancer patients. On the other, the University of California, Los Angels Prostate Cancer Index (UCLA-PCI) and the Functional Assessment of Cancer Therapy-Prostate (FACT-P) were used as prostate cancer specific QOL questionnaires. In particular, UCLA-PCI was the first to be developed for the assessment of symptoms related to the treatment for localized prostate cancer. However, it cannot fully evaluate irritation and obstructive urinary symptoms (an important adverse event for BR), hematuria, or irritable bowel symptoms after radiotherapy. The assessment of the psychosomatic influence of androgen deprivation is also needed in localized prostate cancer patients.1 The Expanded Prostate Cancer Index Composite (EPIC) was developed in 2000 as an expanded version of UCLA-PCI to adopt the current treatment modalities.2
In 2004, we initiated the development of the Japanese version of EPIC and completed a validation study.3 The Japanese version of EPIC has already been used for prospective, large-scale cohort studies including Japanese Prostate Cancer Outcome Study by Permanent I-125 Seed Implantation (J-POPS) and in the Japan Study Group of Prostate Cancer (J-CAP).
In many health-related quality of life studies, general HRQOL and disease specific HRQOL are simultaneously analyzed. From a practical point of view, it is not feasible to use too many questions. EPIC consists of 50 items, so SF-12 not SF-36 is accompanied by the original EPIC. Unfortunately, a well-validated Japanese version of SF-12 has not been available, so the Japanese version of Medical Outcomes Study 8 Items Short Form Health Survey (SF-8) is linked to the Japanese version of EPIC.
There have been many studies that have evaluated general HRQOL in prostate cancer patients assessed using SF-36. The aim of this study is to evaluate the general HRQOL in Japanese patients with localized prostate cancer treated with current multidisciplinary modalities using SF-8. Physical and mental summary scores were compared between different treatment modalities. Furthermore, changes in both of the summary scores over time after treatment were evaluated. This is the first study that evaluates the general HRQOL with SF-8 in Japanese patients with localized prostate cancer undergoing a variety of treatment modalities.
During the period between 4 January and 31 July in 2005, 457 patients (average age: 69.9) with localized prostate cancer treated at 10 institutions in Japan were enrolled in the present study. After obtaining written informed consent from all of the participants using a common instruction approved by institutional review boards of the 10 institutions, HRQOL was measured using the instruments described below. Serum testosterone levels were also measured at the time of enrollment. Patients' characteristics are shown in Table 1. Forty-nine of 170 patients who had undergone radical surgery were treated with ADT due to biochemical failure at the time of this survey. All of the patients who underwent brachytherapy (n = 92) had been treated with ADT for several months prior to the seed implantation. All EBRT patients were treated by 3D conformal methods without ADT.
Table 1. Characteristics of participants (n = 457)
A cross-sectional analysis of general HRQOL and disease-specific HRQOL was conducted with the Japanese version of Medical Outcomes Study 8 Items Short Form Health Survey (SF-8) and the Japanese version of EPIC, respectively. Validity and reliability of the Japanese version of both instruments have already been confirmed, which has been described elsewhere.4,5 SF-8 comprises 8 subscales (PF, Physical functioning; RP, Role limitation due to physical problems; BP, Body pain; GH, General health; VT, Vitality; SF, Social functioning; RE, Role limitation due to emotional problems; MH, Mental health) which are compatible with SF-36. In this study, the score of each 8 subscale, physical health component summary score (PCS), and mental health component summary score (MCS) were measured using the Norm-Based Scoring (NBS) method, which was based on a large-scale population study conducted in Japan.5 QOL scores are shown as mean scores with 95% confidence interval (95% CI) with higher scores representing better QOL. In order to enhance the potential for unbiased answers, participants responded anonymously to all questions and returned them using pre-stamped envelopes.
Analyses of covariance (ancova) were used to analyze the mean differences in each SF-8 subscale among treatment groups. We also used ancova to estimate means in SF-8 summary scores (PCS score and MCS score) among treatment groups. These models were performed after adjustment for age, comorbidities, and periods from termination of treatment. We calculated the least square means and the corresponding 95% confidence intervals (CI).
We performed ancova to assess changes in summary scores over time after treatment. These models were performed after adjustments for age and comorbidities. All statistical analyses were done with SPSS version 12 for Windows (SPSS Inc, Chicago, IL). P-values less than 0.05 were considered to be statistically significant.
Difference in SF-8 subscales between treatment groups
The comparisons of subscales in the SF-8 between the EBRT groups and other treatments groups are listed in Table 2. This date was adjusted for age, comorbidity and duration from each treatment. In the EBRT group, PF scores were significantly lower than scores in the BR group (P = 0.004). RP scores in the EBRT group were also low compared to scores in the BR (P = 0.001) and ADT (P = 0.038) groups. GH scores in the EBRT group were also low when compared to scores in the RPx (P = 0.027) groups. Other domains, including BP, VT, SF and RE showed no significant differences between treatment groups.
Table 2. Comparison of SF-8 subscale between treatments by ancova
ADT, androgen deprivation therapy; BP, Body pain; BR, permanent125 I seed implantation, brachytherapy; RE, Role limitation due to emotional problems; EBRT, external beam radiation therapy; GH, General health; MH, Mental health; PF, Physical functioning; RP, Role limitation due to physical problems; RPx, radical prostatectomy; SF, Social functioning; VT, Vitality; WW, watchful waiting.
Difference in PCS and MCS between treatment groups
Figure 1 shows the comparison of post treatment physical and mental health summary scores between treatment groups adjusted for age, comorbidity and duration from treatment. PSC and MCS were compared between each treatment. There were no significant differences between treatment groups in MCS. In contrast, the external beam radiotherapy group reported significantly lower scores in PCS than the radical prostatectomy group (P = 0.041) and the brachytherapy group (P = 0.011).
PCS and MCS scores with time after treatment
In the RPx group, PCS and MCS showed high scores over time and demonstrated statistical significance with a score exceeding 50 (Norm based score) at 12 or 24 months after surgery (Fig. 2a). In the salvage hormonal therapy (HT) after radical prostatectomy group, there was no significant difference between the early treatment group (less than 12 months) and the late treatment group (more than 12 months) (Fig. 2b). In the EBRT group, there were no significant differences over time in PCS and MCS (Fig. 2c).
There was no significant difference in PCS between the early treatment group (less than 3 months) and the late treatment group (more than 12 months) in BR. MCS showed a higher score in the late treatment group compared with the early treatment group (Fig. 2d).
In the ADT group, no significant difference was observed over time in PCS. Notably, MCS was substantially low in early periods, but showed a significantly higher score over time. At 6 months after treatment, MCS exceeded 50 (Fig. 2e). It was interesting to note the difference in changes between PCS and MCS in the ADT group.
Because of the insufficient number of cases, we can not discuss changes over time in the watchful waiting group.
Assessment of general health status after treatment for cancers including prostate neoplasm is as important as that of disease-specific HRQOL in the medical outcome study. Until now, SF-36 has been the most frequently used instrument that evaluated the general HRQOL after treatment for localized prostate cancer, and has provided us with useful information as to change in general HRQOL in prostate cancer patients.6–15 The number of questions in SF-36, however, is too large (n = 36) when we attempt to use it simultaneously with EPIC, which is a standard instrument for the assessment of disease-specific HRQOL in patients with localized prostate cancer treated with the most current modalities including brachytherapy.2 The original version of EPIC therefore has been linked to SF-12 (questionnaire number: 12), although the Japanese version of SF-12 has not been available. Instead, the Japanese version of SF-8 has been linked to the newly developed Japanese version of EPIC because it has already been fully validated and the norm-based scoring is available based on a large Japanese population study.
The aim of this study is to evaluate general HRQOL using the Japanese version of SF-8 in a relatively large Japanese prostate cancer cohort to clarify whether it can reasonably assess difference in general HRQOL between different treatment modalities.
The first interesting result was the difference in PCS and MCS between treatment groups (Fig. 1). There were no significant differences in MCS. However, in reviewing comparisons of PCS, PCS in the EBRT group was significantly inferior to the RPx and BR groups. This data indicates that the influence is not so much in the mental health component of the general HRQOL in EBRT patients, but as it compares to other treatments. We also assume that the importance given to physical health by patients is more important. The EBRT group deteriorates in the PF, RP, GH and PCS domains and may indicate a need for additional analysis of each subscale. As for EBRT, a burden that is more physical than expected may be indicated.
The second interesting result was the scores in PCS and MCS over time after treatment, especially in the EBRT group. Although, in current study, there were no significant statistical differences in comparison with duration after treatment, it seemed that in the early period after treatment, PCS score was low, but over 12 months, it showed higher scores by 24 months. It is assumed that patients treated with EBRT experienced deterioration in HRQOL to varying degrees. Clark et al. demonstrated changes of SF-36 values in EBRT patients at the time of their clinical consultation prior to deciding on primary therapy, and at 3 and 12 month follow-up periods. The result was that the SF-36 demonstrated a pattern of decline at 3 months and recovery to baseline at 12 months.9 Monika et al. reported a change of SF-36 in 43 EBRT patients in the RE domain that improved from baseline up to 30 weeks. They assumed that there was no change in the other domains.10
Several previous studies have identified the influence on HRQOL after RPx when assessed with SF-36. These showed that the deterioration of HRQOL after surgery was temporary and that returned to baseline or exceeded baseline at 1 year or more after the surgery.11–13 Findings from other investigators were similar to the results of this study using SF-8, such as the mental health score, which decreased, then gradually improved in the months after radical prostatectomy to a value of 50 (Fig. 2a).
In this study, we sampled two groups having brachytherapy. The surveys were based on postoperative periods, with the groups identified as the early group (less than 12 months) and the late group (more than 12 months), respectively. There was no significant difference in PCS between the early treatment group and the late treatment group. In contrast, MCS improved significantly over time. There are still few reports where patients were assessed with SF-36 after BR, but results seem to be similar in available studies.
In the ADT group, there was no significant difference in PCS with respect to follow-up duration after treatment. Of note, MCS in the early treatment group was significantly low, but like other findings, it markedly improved with time. This trend is clinically significant. We must consider the deterioration of mental condition in the early ADT treatment period. Lubeck et al. also reported that there were no significant changes in all domains of SF-36 in comparison to before treatment and after 6 months in 67 ADT patients enrolled in the study.15 Kakehi et al.4 demonstrated that sexual function scores, for patients with salvage ADT who showed biochemical or clinical progression after radical prostatectomy, were negative (P = 0.001), as were Hormonal subscales (P < 0.001), hormonal function scores (P < 0.001) and hormonal bother scores (P < 0.001) using EPIC. In our study, PCS and MCS were stable in the salvage hormonal therapy after radical prostatectomy group (Fig. 2b). We have described the conditions that are impacted by hormonal therapies, but to validate the true impact on HRQOL, several more years of study are necessary. In additional, large numbers of patients are necessary to fully understand the impact of each treatment on HRQOL.
From the results of this study, we demonstrate that SF-8 accompanied with EPIC accurately evaluates the HRQOL in Japanese patients with localized prostate cancer. Use of the SF-8 questionnaire and its findings may help clinicians prepare their patients for expected changes in HRQOL after treatment.
We investigated the general health related quality of life in Japanese men after treatment using the SF-8 survey instrument and a cross-sectional analysis. We were able to evaluate HRQOL assessed with SF-8. The EBRT group reported significantly lower scores in PCS than the RPx and BR groups. Similar findings from other investigators using other instruments were observed after treatment for comparable time periods. The deterioration of MCS was marked in the early period after ADT, but there was a definite recovery trend over time.
SF-8 combined with EPIC is helpful in understanding HRQOL in patients treated for localized prostate cancer. Our findings demonstrate that to evaluate HRQOL in Japanese patients with prostate cancer, we must employ both the EPIC and the SF-8 survey instruments. A prospective, longitudinal, longer-term study with larger numbers of patients is warranted to further elucidate quality of life outcomes in Japanese men undergoing treatment for prostate cancer.
This study was supported by the Committee for Developing a Japanese Version of EPIC as described below, and the Committee has contributed much to the present study design and enrollment of participants. The Committee members (excluding authors) are: Ogawa, O. Kamoto, T. Yoshimura, K. (Department of Urology, Kyoto University Graduate School of Medicine), Arai, Y. Namiki, S. (Department of Urology, Tohoku University Graduate School of Medicine), Saito, S. (Department of Urology, Tokyo Medical Center, National Hospital Organization), Okuyama, A., Nonomura, N., Nakayama, M. (Department of Urology, Osaka University Graduate School of Medicine), Hara, I. (Division of Urology, Department of Organ Therapeutics, Faculty of Medicine, Kobe University Graduate School of Medicine), Miki, T., Okihara, K. (Department of Urology, Kyoto Prefectural University of Medicine), Horie, S. (Department of Urology, Teikyo University School of Medicine), Suzuki, K. (Department of Urology Gunma University Graduate School of Medicine), Nonomura, K., Shinohara, N., Harabayashi, T. (Department of Urology, Hokkaido University Graduate School of Medicine), Kuwata Y (Department of Urology, Kagawa University Faculty of Medicine).