|Home | About | Journals | Submit | Contact Us | Français|
The health-related quality of life (HRQOL) after treatment of prostate cancer is examined using a new HRQOL tool. HRQOL, based on the expanded prostate cancer index composite (EPIC) and SF-8 questionnaires, was prospectively compared after either a radical retropubic prostatectomy (RRP) or a permanent prostate brachytherapy (PPB) at a single institute.
Between October 2005 and June 2007, 96 patients were treated by an RRP and 88 patients were treated by a PPB. A HRQOL survey was completed at baseline, and at 1, 3, 6 and 12 months after treatment, prospectively.
The general HRQOL in the RRP and PPB groups was not different after 3 months. However, at baseline and 1 month after treatment, the mental component summary was significantly better in the PPB group than in the RRP group. Moreover, the disease-specific HRQOL was worse regarding urinary and sexual functions in the RRP group. Urinary irritative/obstructive was worse in the PPB group, but urinary incontinence was worse in the RRP group and had not recovered to baseline after 12 months. The bowel function and bother were worse in the PPB group than in the RRP group after 3 months. In the RRP group, the patients with nerve sparing demonstrated the same scores in sexual function as the PPB group.
This prospective study revealed the differences in the HRQOL after an RRP and PPB. Disease-specific HRQOL is clarified by using EPIC survey. These results will be helpful for making treatment decisions.
There are many treatment choices for localized prostate cancer, especially low-risk patients. The treatment outcomes between radical retropubic prostatectomy (RRP) and radiotherapy for low-risk localized prostate cancer are the same by recently published retrospective studies (1,2). Therefore, the decision of treatment depends on the patient's or the oncologist's preference. However, it is difficult for patients to decide which treatment should be selected. In Japan, permanent prostate brachytherapy (PPB) was started at 2003 and this treatment is now rapidly expanding (3,4). So decision of treatment is made more difficult. One consideration in the treatment selection is the quality of life (QOL) after treatment, and the QOL is a very important factor for patients. Previously, we reported the QOL comparing RRP and PPB (5). In our study, the disease-specific QOL in prostatectomy was worse than in brachytherapy, especially the urinary and sexual functions, although the general health-related QOL (HRQOL) recovered by 3 months following a prostatectomy. Our reports were used for UCLA-PCI (6), so disease-specific QOL after PPB was not explained exactly. Recently, new disease-specific QOL, expanded prostate cancer index composite (EPIC) (7,8), was developed to make up for the disadvantage of UCLA-PCI. However, there have so far been very few reports describing the QOL using EPIC in Japanese subjects, especially in regard to a prospective study. In this study, we evaluated the QOL after RRP and PPB using EPIC.
Between October 2005 and June 2007, RRP was performed in 114 patients and PPB in 88 patients at our hospital. The indications for PPB were limited in the patients with low and intermediate risk. If the prostate-specific antigen (PSA) level was between 10 and 20 ng/ml or the primary Gleason pattern was 4 among intermediate risk, additional external beam radiotherapy (EBRT) was recommended and considered. Androgen deprivation therapy (ADT) was not used except for volume reduction. The patients treated with PPB received 145 Gy to the prostate with an I-125 seed using a modified peripheral loading technique via a transrectal ultrasound guided transperineal approach (3). Until December 2005, we performed PPB by the pre-planning method and after January 2006, by the intra-operative planning method.
On the other hand, the indications for RRP were patients with any risk but aged 75 years or younger. Clinical stage T3 was also indicated for surgery but the Gleason score and PSA level were carefully considered and informed to the patients. The nerve-sparing technique was performed if the patient wanted to preserve sexual function. The indications for a nerve-sparing procedure depended on the pre-operative (number and Gleason score of the positive biopsy cores, PSA level or patient preference) and intra-operative factors, prioritizing cancer control. The RRP was performed using Walsh's technique by two staff urologists or under their supervision. ADT was not used for neoadjuvant and adjuvant settings. The patients who indicated RRP or PPB selected their own therapy themselves after we informed each therapy.
We measured the general and disease-specific HRQOL using two types of instruments. The general HRQOL was assessed with the Japanese version of Medical Outcome Study 8-Items Short Form Health Survey (SF-8) (9). The SF-8 is an HRQOL assessment consisting of eight scales and generates two summary measures, a physical component summary (PCS) and a mental component summary (MCS). The disease-specific HRQOL was assessed with the EPIC, a 50-item questionnaire that quantifies the prostate cancer-specific HRQOL in eight separate domains (10). In this study, domains of hormone function and bother were omitted because there were few patients with ADT and ADT was used for neoadjuvant setting only. All patients were informed of their cancer diagnosis before being asked to complete the HRQOL questionnaires. No interviews were conducted. The questionnaires were administered at five time points. The baseline survey was conducted within 1 week before surgery and PPB. The follow-up survey was conducted in person at the scheduled study visits at 1, 3, 6 and 12 months after treatment. This study was approved by the Institutional Review Board in our hospital. Written informed consent was obtained from all patients before the initiation of treatment.
QOL scores were shown as mean scores with standard deviation. PCS and MCS are calculated by weighting each SF-8 item using the norm-based scoring method. All scales of the SF-8 above and below 50 were above and below the average in the general Japanese population. All scores of EPIC were linearly transformed to a scale of 0 (lowest) to 100 (highest). Each group comparison was made using the Mann–Whitney U-test and χ2 test. Two-tailed P values of <0.05 were considered statistically significant. SPSS software ver.16.0J (Tokyo, Japan) was used for all statistical analyses.
Eight patients in RRP were disagreed for this study. The surveys were performed for 96 and 88 patients who underwent RRP and PPB, respectively. The average answer rate of each survey was 82.1% in RRP and 89.1% in PPB. Background of each group is shown in Table 1. The median age of the RRP and PPB groups was 66 and 69 years, respectively. As for the age, the RRP group was younger than the PPB group. As for the clinical stage and PSA level, the PPB group had a lower stage and a lower average PSA level. In addition, the Gleason score was lower in the PPB group. ADT was performed in 13 patients from the PPB group and in 3 patients from the RRP group. All of these patients discontinued hormone therapy after either RRP or PPB. Nerve-sparing surgery was performed in 12 patients. Six patients added EBRT after PPB. After 12 months, five patients demonstrated recurrence in RRP and no patient in PPB. The former was PSA failure only (PSA > 0.2 ng/ml) (Table 1). However, the general and disease-specific QOL among the patients with recurrence or EBRT were the same as in patients without the recurrence or EBRT. We therefore included all cases in this analysis.
The SF-8 scores are listed in Table 2. There is a significant difference for the baseline QOL scores in physical function, role physical, social functioning, role emotional and mental health among each group. These baseline scores were worse in the RRP group than in the PPB group. In the RRP group, the QOL scores at 1 month were worse than the baseline score in physical functioning, role physical, body pain, vitality and social functioning. These worse scores recovered until 3 months except body pain. The score of body pain recovered at 6 months. The scores of role physical, general health, role emotional and mental health improved over the baseline after 6 and 12 months. In the PPB group, the QOL scores in role physical and social functioning at 1 month were worse than the baseline score. These scores recovered until 3 months. The mental health improved above the baseline at 12 months. Comparing the RRP and PPB groups, there were significant differences in all scores except for general health at 1 month after treatment. These QOL scores were better in the PPB group than in the RRP group. The general health was the only same score at 1 month in both groups. After 3 months, some QOL scores were remained worse in RRP but there were no significant differences at 6 months between the RRP and PPB groups.
The SF-8 scores for PCS and MCS are summarized in Table 3. The score of PCS in RRP was worse than in PPB at 1 month, but recovered until 3 months. The baseline difference in MCS scores was significantly worse and continued to 1 month after surgery. It was recovered at 3 months. The MCS score in the high-risk RRP group was poor in comparison to the low- to intermediate-risk group; however, the difference was not significant.
The results of the EPIC are listed in Table 4. For the baseline score, there was significant difference in urinary incontinence, bowel function and sexual bother between the RRP and PPB groups. These scores were worse in the RRP group at baseline. At 1 month, the urinary incontinence and sexual function were more worsen in RRP. However other scores were worse from baseline, there was no significant difference between RRP and PPB. The urinary incontinence in RRP was recovered after 3 months, but it was worse than in the PPB group until 12 months. The urinary irritative/obstructive score at 1 month was worse from baseline in both groups, but in the RRP group, the score was recovered after 3 months and better than baseline. On the other hand, in the PPB group, the score was remained worse until 6 months and it was significant comparing the RRP group. The urinary bother at 1 month was also worse in both groups from baseline. In the RRP group, it was recovered until 3 months, but in the PPB group, it remained worse until 6 months significantly. Bowel function and bother were worse in the PPB group at 3 and 6 months than in the RRP group significantly. Bowel function in PPB at 12 months was worse than baseline and RRP group. The sexual function and sexual bother in the RRP group were worse than in the PPB group until 12 months. The sexual function did not recover in the RRP group. In the PPB group, the sexual function and bother did not change in the PPB group. Moreover, no significant factor correlated with the EPIC score and the EPIC did not correlate with SF-8.
Concerning sexual function and bother, the effect of nerve sparing was examined. In the RRP group, if nerve sparing was performed, sexual function was worse than baseline but recovered gradually better than in the non-nerve-sparing group. However, the nerve-sparing group did not recover to the baseline at 12 months. On the other hand, sexual bother was worse in the nerve-sparing group than in the non-nerve-sparing group, although it was not significant. Comparing the PPB group, the sexual function after treatment was the same in the nerve sparing, although baseline score was better in the RRP group. Sexual bother was also worse in each group but there was no significant difference (Table 5).
The assessment of the treatment for prostate cancer was made, not only regarding the duration of survival, but also the HRQOL. Therefore, the HRQOL after treatment becomes important and there are many QOL reports after treatment of prostate cancer, i.e. RRP, EBRT or PPB (11–15). Litwin et al. (11) reported the QOL after both RRP and PPB. He reported the general QOL to be the same in both groups but that urinary function was better in the PPB group. Moreover, sexual function was better in the PPB group but bowel function was worse. Other investigators reported the same results (12–15). Previously, we reported the results of QOL after RRP and PPB (5). In that study, our results in the RRP patients were the same as those of the other investigators (11–15). Most of investigators, including us, had used UCLA-PCI for disease-specific QOL. For urinary function, the UCLA-PCI focused mainly on urinary incontinence. Therefore, urinary irritability might have been underestimated. Moreover, bowel function in UCLA-PCI does not include rectum bleeding and irritability, and there are few item numbers in it. Because of these contents, disease-specific QOL in RRP was inferior to PPB. Because of these disadvantages, EPIC is developed for QOL after treatment. Recently, the QOL survey varied from the UCLA-PCI to the EPIC (16–18). The EPIC contained more questions for urinary and bowel functions, including, for example, urinary and bowel irritation, than UCLA-PCI. The results from the EPIC were emphasized for urinary and bowel functions. However, there are a few reports of QOL using EPIC. In this study, we used EPIC for disease-specific QOL. To our knowledge, this report is the first report of longitudinal results using EPIC in Japanese.
The RRP group was worse about urinary function even if EPIC was used. However, urinary irritative had worse in PPB and urinary incontinence had worse in RRP among urinary function. A difference in urinary irritative is a major difference of EPIC from UCLA-PCI. As for the difference in incontinence and irritative, incontinence was bigger. So, urinary function in RRP was worse than PPB. These results indicated that different problem existed in RRP and PPB and emphasized these differences. On the other hand, bowel function was worse in PPB than RRP. Among bowel function, pain, frequency and diarrhea were factors of worse function but bleeding was not a factor in PPB. This point became clear for the first time by using EPIC. Our previous study using UCLA-PCI did not become clear in these points. On the other hand, sexual function and bother were the same between EPIC and UCLA-PCI scores.
Frank et al. (16) reported the disease-specific QOL after RRP, PPB and EBRT by using EPIC. In their study, PPB had a significantly worse bowel function and bother than RRP (19). Although the RRP had significantly worse urinary incontinence than the PPB, the PPB had more urinary irritation than the RRP. The PPB had a significantly better sexual function than the EBRT or RRP. However, their report was a cross-sectional study, and QOL survey was performed at only one point after treatment. So, their report did not clear about the change of QOL. On the other hand, Ferrer et al. (18) reported the longitudinal QOL after treatment by using EPIC. They examined QOL before and after treatment (1, 3, 6, 12 and 24 months). They reported that the RRP group had worse EPIC sexual summary and urinary incontinence scores compared with the PPB group, and the RRP group had significant better EPIC urinary irritation scores than PPB. Moreover, they reported the change of QOL after treatment. The change of their QOL scores was the same as our results. Other investigators reported the same results.
In this study, we used SF-8 for general HRQOL. SF-8 consisted of only eight questions and obtained the same results as SF-36. General HRQOL and disease-specific QOL are simultaneously analyzed. Because EPIC has too much questions, 50 items, SF-8 is often linked to EPIC. Our report is the first report of longitudinal results using SF-8 in Japanese, to our knowledge. According to SF-8, PCS was worse after 1 month in the RRP group, but the PPB group did not change. PCS in the RRP group was recovered until 3 months after treatment. Therefore, it was thought that the PCS score improves as urinary incontinence is restored. On the other hand, MCS at baseline in RRP was significantly worse than PPB, and this worse score continued until 1 month. After all, MCS scores were the same in both groups at 3 months. In RRP, there were more advanced stage and higher PSA and Gleason score compared with PPB. It was thought that these worse clinical factors and the attitude for operation influenced worse MCS score at baseline in RRP and patients with RRP felt stress. The MCS score tended to be poor in the high-risk group but no significant difference was observed. After treatment, MCS score recovered in the RRP group because patients with RRP felt free of stress. Sahai et al. (19) reported that there was no significant difference in PCS, but the MCS score showed a significant increase after surgery. In their study, surgery was laparoscopic upper urinary tract surgery. Surgery was influenced to MCS, even though it was different from RRP.
In this study, the PPB group had better QOL compared with the RRP group, except for urinary irritative and bowel function. However, there are some limitations associated with our study. There were some differences in the backgrounds between RRP and PPB. This study is a prospective study but not a randomized study. Indeed, the PPB group tended to have a high age and the RRP group included more high-risk patients. If there are many high-risk patients, then recurrence and the need to perform additional treatment would also increase, thus influencing the QOL. However, because this study focused on the outcomes only up until 12 months from the start of treatment, only a few patients with recurrence were thus observed and they did not affect the results. Second, this survey was only up until 12 months and the observation period was thus short. The results of the QOL may therefore change during the long-term follow-up. Recently, Namiki et al. (20) reported the QOL after 5 years. In their report, the urinary function in RRP continued to recover gradually but never returned to baseline after 5 years. Long-term follow-up of QOL is needed in both groups.
Despite these limitations, the change and difference in the QOL up until 12 months became clear in this study. This difference in disease-specific QOL has become more clearer by using EPIC. These results and other published results will therefore be useful and provide important information when selecting the optimal treatments for localized prostate cancer, although long-term observation is necessary.
This study was supported in part by a Grant-in Aid Cancer Research (17-10) from the Ministry of Health, Labour and Welfare, Japan.