- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
External beam radiotherapy is one of the treatment options for localized and locally advanced prostate cancer. Several randomized trials showed an improvement in biochemical relapse-free survival rate with an increase in irradiation dose [1–3]. Further increases in dose are expected to lead to further improvement . Especially patients with an unfavourable prognosis might benefit from an increased radiation dose . By increasing the radiation dose, the risk of genitourinary and gastrointestinal toxicity also increases . To evaluate treatment, not only toxicity but also the impact on quality of life (QoL) is important .
Intensity-modulated radiotherapy (IMRT) can be used to increase the dose to the prostate while the dose to the organs at risk remains acceptable, because this technique provides better coverage of the target and sparing of the organs at risk. Furthermore, daily verification of the position of the prostate, using fiducial gold markers, improves the precision of the treatment, and therefore the organs at risk might receive less radiation dose . The theoretical advantages of IMRT dose distributions and the reduction of the positioning errors with fiducial markers has been established. However, not many clinical results of IMRT in combination with accurate position verification have been published .
Previously, we reported that improved technical possibilities prevented a deterioration in QoL when the radiation dose was increased. Patients treated with 76 Gy IMRT with accurate position verification even had clinically relevant improvements in QoL over a period of 6 months, compared with a group treated with 70 Gy conformal RT . Yoshimura et al. described comparable QoL outcomes between IMRT and conformal RT. However, concerns have been raised about the short-term follow-up of both studies, because late bladder and rectal toxicity can develop several years after completing RT . Late rectal toxicity usually occurs within 3 years of completing RT and bladder toxicity even continues to develop thereafter. Only one study reported the QoL scorest at 3 years after IMRT, but this involved only six patients . A few studies described the QoL scores 2 years after IMRT [13–15].
Good-quality QoL research has to meet certain conditions; not only are large groups of patients important, but it is also important to measure the patients’ perception of their health and ability to function in life, because physician assessment of treatment-induced complaints differs from morbidity reported by patients . Other conditions for good-quality QoL research include validated QoL questionnaires including an organ-specific module, and a longitudinal study design including baseline scores, which is essential in this older patient group .
In this prospective and longitudinal study we measured long-term QoL, using a prostate-specific module, of patients treated with IMRT of 76 Gy, using fiducial marker-based position verification. For this patient group we previously reported good QoL results at 1 and 6 months after RT, and in the present study we investigated the change in QoL between the measurement before RT (baseline) and 3 years afterward.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Between October 2003 and November 2004, 116 patients with mainly locally advanced prostate cancer were treated with IMRT. The prostate was delineated on CT and a margin of 8 mm was applied to the prostate and seminal vesicles to create a planning target volume (PTV). Patients received IMRT with a five-beam ‘step-and-shoot’ technique and 10-MV photons. A mean dose of 76 Gy in 35 fractions of 2.17 Gy was prescribed to the PTV and 95% of the prescribed dose (= 72 Gy) was prescribed to 99% of the PTV. The dose in the part of the PTV overlapping the rectum and bladder was limited so that ≤5% of the rectum and 10% of the bladder would receive a dose of ≥72 Gy [18,19]. For position verification, three fiducial gold markers were transrectally implanted inside the prostate, using antibiotic prophylaxis. During treatment the fiducial gold markers were visualized using portal images of the first fields of all five beam directions, using the iView-GT amorphous silicon flat-panel detector (Elekta Ltd, Crawley, UK). The position of the fiducial gold markers can be easily and reliable detected, allowing for fast and accurate determination of the position of the prostate. Daily imaging of the fiducial markers was used for off-line position verification, using an adapted ‘shrinking action level’ protocol .
General health-related QoL was measured using the RAND-36 generic health survey , cancer-specific QoL using the European Organization for Research and Treatment of Cancer (EORTC) core questionnaire (QLQ-C30(+3)) , and the prostate tumour-specific QoL using the EORTC prostate cancer module (QLQ-PR25) . All questionnaires are well validated and widely used in oncology trials .
Scales and items of these questionnaires range in score from 0 to 100. For RAND-36 and for the ‘global’ QoL and functional scales of the EORTC questionnaires, a high score represents a high level of QoL and better functioning. For the symptom/problem items of the EORTC questionnaires, higher scores represent a higher level of symptoms and consequently a worse QoL. A change of 10% (or in general, 0.5 sd) of the scale breadth is perceptible to patients as a meaningful change, and a change in QoL of ≥10 points is therefore considered clinically relevant .
The first questionnaire was presented to the patient before treatment at the department of radiation oncology; at 1, 6 and 36 months after completing the treatment the measurements were repeated. For all patients treated between 2001 and 2004 at our department with 76 Gy IMRT and fiducial marker-based position verification, the pretreatment, acute and late toxicity was scored . Ninety-five patients were included in the study, because 15 were lost to follow-up and six died within the 3 years of follow-up. All questionnaires had a high response rate, except for questions about sexual activity.
Because the data were normally distributed a paired sample t-test was used to examine differences in QoL scores between the time points (baseline vs 3 years). The QoL changes between baseline, 1 and 6 months were analysed previously . Test results with a P ≤ 0.01 were considered statistically significant, to account for multiple comparisons. Cronbach’s coefficient α was calculated to determine internal consistency and reliability of the questionnaires.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
The characteristics of the 95 patients included in the study are shown in Table 1; most had locally advanced prostate cancer (tumour stage T3). At the time of the study entry, no national guidelines for hormonal treatment were available, and therefore, only 36 patients received adjuvant androgen-deprivation therapy (ADT), which was divided into short-term (≈6 months) and long-term hormonal treatment (≤36 months). All patients had a WHO performance score of ≤2.
Table 1. The characteristics of the 95 patients
|Characteristic||Mean (range) or n (%)|
|Age at baseline, years||68 (46–79)|
|PSA level, ng/mL|| |
| ≤10||29 (31)|
| 10–20||36 (38)|
| >20||30 (32)|
|Gleason score|| |
| 2–6||48 (51)|
| 7||28 (30)|
| 8–10||19 (20)|
|Tumour stage|| |
| T1|| 9 (10)|
| T2|| 4 (4)|
| T3||82 (86)|
| T4|| 0|
|Hormonal treatment|| |
| None||59 (62)|
| Short-term (≈6 months)||26 (27)|
| Long-term (≤36 months)||10 (11)|
|TURP before radiotherapy|| |
| Yes|| 8 (8)|
| No||87 (92)|
During RT four patients developed grade 3 genitourinary toxicity, consisting of urinary frequency/urgency, urinary retention, haematuria and dysuria. One of these patients already had grade 3 complaints of urinary frequency before RT. There was no acute grade ≥3 gastrointestinal toxicity. Furthermore, one patient developed a UTI during RT that required i.v. antibiotic (grade 3).
Within 3 years after RT two patients developed grade 3 genitourinary toxicity and one developed severe radiation proctitis requiring several argon-plasma coagulations (grade 3).
Table 2 shows the mean (sd) of all QoL items before treatment (baseline), and 1, 6 and 36 months after treatment. Two-related samples tests between the QoL scores at baseline and after 3 years resulted in statistically significant differences between these time points for several QoL items. Figure 1 shows the course in QoL of the QoL items with a significant difference after 3 years, vs baseline scores. The RAND-36 questionnaire at 3 years showed a better QoL for ‘emotional role restriction’, ‘mental health’ and ‘change in health’, and worse ‘physical functioning’ than the baseline measurement. Only ‘emotional role restriction’ showed a change in QoL over time of >10 points and was therefore considered as the only clinically meaningful change in QoL for this questionnaire (Fig. 1A). The three statistically significant changes in QoL over time for the EORTC QLQ-C30 items were <10 points and therefore not clinically relevant (Fig. 1B). Figure 1C shows worse QoL for ‘bowel symptoms/function’ and ‘sexual activity’. ‘Sexual activity’ showed a clinically relevant decrease of 12 points after 3 years; this QoL item was worse at 1 month after treatment and that persisted after 6 months and 3 years.
Table 2. The mean (sd) scores for scales and single items of each questionnaire
|Item||Baseline||1 month||6 months||3 years||P for difference, baseline vs 3 years|
|No. of patients||95||85||85||95|| |
|Physical functioning||87 (14)||85 (17)||86 (16)||83 (19)|| 0.007|
|Social functioning||83 (18)||85 (18)||90 (16)||88 (18)||ns|
|Physical role restriction||80 (35)||71 (40)||84 (30)||83 (32)||ns|
|Emotional role restriction||79 (35)||87 (30)||92 (25)||90 (24)|| 0.01*|
|Mental health||77 (15)||78 (13)||81 (14)||82 (15)|| 0.001|
|Vitality||70 (20)||68 (19)||71 (19)||71 (22)||ns|
|Pain||89 (17)||86 (18)||92 (15)||90 (19)||ns|
|General health||68 (16)||67 (16)||70 (18)||66 (19)||ns|
|Change in health||44 (13)||53 (24)||63 (24)||52 (16)|| 0.002|
|Physical functioning||89 (13)||88 (13)||89 (14)||87 (16)||ns|
|Role functioning||88 (18)||86 (20)||90 (18)||88 (22)||ns|
|Emotional functioning||80 (15)||88 (16)||89 (16)||88 (16)||<0.001|
|Cognitive functioning||91 (14)||87 (18)||87 (18)||85 (18)||<0.001|
|Social functioning||90 (14)||92 (14)||95 (12)||93 (17)||ns|
|Global health/QoL||79 (14)||79 (12)||82 (13)||80 (15)||ns|
|Fatigue||19 (19)||23 (21)||18 (18)||21 (20)||ns|
|Nausea and vomiting|| 2 (8)|| 2 (7)|| 1 (5)|| 1 (6)||ns|
|Pain||11 (18)||12 (20)|| 8 (16)||13 (21)||ns|
|Dyspnoea|| 8 (18)||11 (21)||13 (21)||13 (21)||ns|
|Insomnia||21 (25)||24 (25)||15 (24)||14 (22)|| 0.01|
|Appetite loss|| 5 (13)|| 2 (8)|| 2 (7)|| 3 (9)||ns|
|Constipation|| 3 (11)|| 5 (14)|| 6 (17)|| 6 (14)||ns|
|Diarrhoea|| 5 (16)||13 (21)||11 (22)|| 8 (18)||ns|
|Financial difficulties|| 3 (11)|| 4 (10)|| 2 (7)|| 2 (9)||ns|
|Urinary symptoms/problems||17 (13)||21 (18)||15 (14)||16 (13)||ns|
|Bowel symptoms/function|| 5 (7)|| 8 (11)|| 8 (11)|| 8 (12)|| 0.004|
|Treatment-related symptoms|| 8 (12)||13 (13)||11 (12)||10 (10)||ns|
|Sexual functioning||23 (21)||24 (20)||24 (21)||26 (21)||ns|
|Sexual activity†||68 (24)||57 (28)||54 (22)||56 (26)|| 0.01*|
Figure 1. Changes in QoL for the RAND-36 (A), the EORTC core QLQ-C30 (B), and the EORTC-QLQ-PR25 (C) QoL items, with a significant difference between baseline and 3 years. Error bars represent the 95% CI. In the RAND-36, a higher score reflects better health. In EORTC QLQ-C30(+3) and QLQ-PR25, a higher score reflects a high level of symptoms or functioning or QoL.
Download figure to PowerPoint
A subanalysis showed that there were no significant differences in QoL for ‘sexual activity’ between patients with and without ADT at baseline, with a mean (sd) QoL score of 54 (25) vs 72 (22) at baseline and at 3 years after RT of 54 (32) vs 56 (25). However, patients with no ADT had decreased ‘sexual activity’ scores between baseline and 3 years (P = 0.004), while patients with ADT had no deterioration in ‘sexual activity’ after 3 years compared with baseline. Unfortunately ‘sexual activity’ scores were only available for seven patients with ADT and 25 without.
We also compared the change in QoL for patients with and with no increase in toxicity between baseline and 3 years after RT. There were no clinically relevant and significant differences in the change in QoL for patients with an increase in gastrointestinal and genitourinary complaints compared with those with a decrease or no change in toxicity.
For sufficient internal consistency, a Cronbach’s α coefficient of ≥0.70 is required. Reliability analysis gave a Cronbach’s α of ≥0.70 for all scales, except ‘nausea’ (0.31–0.73) and ‘cognitive’ and ‘social functioning’ (0.52–0.84) in the EORTC QLQ-C30(+3), and ‘bowel function’ (0.33–0.64), ‘treatment-related symptoms/problems’ (0.50–0.64) and ‘sexual activity’ (0.58–0.77) in the QLQ-PR25.
The present study, including 95 patients, was well powered to detect QoL differences of 10 points. In 95% of the scales the sd was 14–19 (Table 2). The power to detect a difference of 10 points between baseline and 3 years as statistically significant with 95 patients and a two-sided α of 0.01 is 100% (when using a sd of 14) and 99.7% (when using a sd of 19).
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
To our knowledge, this is the first study to evaluate the change in QoL for a large group of patients 3 years after 76 Gy IMRT using gold-fiducial marker-based position verification. Previously we reported that, compared with 70-Gy conformal RT, there was no deterioration in QoL after 6 months. In the present study there was no meaningful deterioration of QoL 3 years after completing RT from the QoL before RT, except for the QoL item ‘sexual activity’. Better planning of the target volumes and sparing of the organs at risk using IMRT and adequate position verification might have resulted in little radiation damage and consequently no clinically relevant effect of the side-effects on QoL.
The previous evaluation of the short-term QoL changes after 76 Gy IMRT  showed a significant and meaningful increase in ‘emotional role restriction’ after 6 months. This increase in QoL score remained after 3 years, with a difference of 11 points from baseline. A decreasing fear of recurrence and death, and having a long time to adapt to the situation, might be involved in the persistent increase of this QoL item.
The short-term analysis also showed that after 6 months there was an increase in ‘change in health’ score . The increase in QoL for ‘change in health’ between baseline and 3 years was significant (P = 0.002), but not clinically relevant (<10 points). However, the change in QoL between 6 months and 3 years was significant, with a clinically relevant decrease of 11 points. The improvement in ‘change in health’ after 6 months was attributed to the ‘response shift’ mechanism (referring to a changed internal standard on which patients base their perception, due to a life-threatening disease ). The worsening of ‘change in health’ thereafter might be due to ageing, as Aaronson et al. reported lower QoL scores for older respondents (>70 years) than younger respondents in The Netherlands.
QoL for ‘sexual activity’ had already decreased by 1 month after RT. The persistent deterioration in QoL might be due to late radiation effects and has been reported previously [13,27,28], but also the patients’ age and comorbities might have influenced sexual activity . Previous reports suggest an arteriogenic pathology as the main cause for erectile dysfunction after RT . The Dutch translation of the EORTC QLQ-PR25 requests patients to complete the questions about sexual activity only when they were sexually active in the last 4 weeks. For that reason the sexual activity scores were available for only 32 patients, which could have led to under- or overestimated QoL scores.
Sanda et al. reported long-lasting symptoms involving sexuality due to androgen suppression of limited duration. In the present study patients treated with ADT had no deterioration in ‘sexual activity’ between baseline and 3 years, but they seemed to have worse ‘sexual activity’ scores before RT. At 3 years after RT there was no difference in ‘sexual activity’ between patients with and without ADT. This might be due to the few patients treated with ADT in our study. However, Yoshimura et al. even reported an improvement in sexual function during the follow-up, probably resulting from reduced sexual function before the start of RT, as a result of neoadjuvant hormone therapy.
In the present study the increase in gastrointestinal and genitourinary toxicity did not significantly affect the change in QoL. This could be because the incidence of severe toxicity was very low, but the absence of a correlation between QoL and toxicity was reported previously [16,31]. The increase in toxicity might be too small to be detected by the QoL questionnaires, but another explanation might be that patients had time to adapt to the RT-induced complaints and developed coping skills .
After a follow-up of 6 months there were no differences in QoL scores between IMRT and conformal RT, or improved QoL scores in patients treated with IMRT [9,10]. At ≈2 years after RT, Kupelian et al. reported no differences in QoL between a dose of 70 Gy in 2.5 Gy fractions using IMRT and conformal RT. However, a less solid cross-sectional design was used and only 24 patients were included. In accordance with our findings, Namiki et al. reported, for 12 patients at 2 years after monotherapy with 76 Gy IMRT, no significant difference in QoL from the baseline level. Recently Sanda et al. reported worse bowel and sexual QoL scores 2 years after external beam RT (either IMRT or conformal RT) than baseline scores. Unfortunately no technical specifications and treatment dose were described. Only Junius et al. measured the long-term QoL after IMRT; at 3 years after treatment with 66 Gy in 2.64 fractions there were no clinically relevant QoL changes for six patients, except an increase in emotional and decrease in cognitive functioning, which was also seen in the present patients.
We used a high RT dose of 76 Gy because it is expected to lead to high biochemical control rates [1–3]. In the present study we conclude that our treatment is well tolerated and causes no significant clinical deterioration in QoL, but a longer follow-up is required to determine the biochemical or clinical failure in our patient group.
In conclusion, 3 years after IMRT with 76 Gy using fiducial gold marker-based position verification, all QoL items, except ‘sexual activity’, had no significant and clinically relevant deterioration compared with the QoL scores before the start of RT. The QoL for ‘emotional role restriction’ improved after 6 months and remained high 3 years after RT. The satisfactory QoL results 3 years after treatment might be a consequence of the few side-effects after treatment with IMRT and accurate position verification. IMRT is especially suitable for avoiding organs at risk, e.g. rectum and bladder, while a high dose of radiation is delivered to the prostate. In combination with daily position verification using fiducial markers, the organs at risk might have received little irradiation, leading to fewer complaints during and after treatment.