Health-Related Quality of Life up to Six Years After 125I Brachytherapy for Early-Stage Prostate Cancer

Introduction 

Iodine-125 (125I) brachytherapy for early-stage prostate cancer shows excellent tumor control and survival rates 1, 2, 3, 4. The results are equivalent to those achieved by radical prostatectomy or external beam radiotherapy (5). Hence, toxicity and health-related quality of life (HRQOL) are considered important endpoints that should be taken into account in the decision of a treatment modality.

Although many studies have reported HRQOL after 125I prostate brachytherapy 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, most were limited by being cross-sectional, by having a short duration of follow-up, or by lacking extended and validated HRQOL questionnaires. Because long-term side effects can occur >3 years after treatment (16), long-term follow-up of HRQOL is required. Furthermore, a baseline QOL measure is required to analyze QOL changes over time. To the best of our knowledge, no study has prospectively analyzed HRQOL ≥2 years after 125I prostate brachytherapy.

Some reports have described long-term prostate symptom scores such as the International Prostate Symptom Score (IPSS). Ash et al. (15) reported that IPSS scores 9 years after treatment did not differ compared with the baseline IPSSs. However, in addition to somatic functioning, HRQOL also includes patients' perceptions of their social and psychological functioning and well-being 17, 18. An appraisal of physical and psychosocial complications after prostate cancer treatment can differ when evaluated by patients or physicians (19). Therefore, HRQOL assessments should be performed using internationally validated questionnaires 20, 21.

Previously, we published our HRQOL data of 127 patients treated with 125I brachytherapy for localized prostate cancer up to 1 year after treatment (22). In the present study, we present the long-term HRQOL outcomes 6 years after implantation.

Results 

The mean scores and standard deviations of the HRQOL items at five different points are listed in Table 2. For most items, an increase in symptoms or a decreased level of functioning or QOL was seen 1 month after treatment. Subsequently, the symptoms gradually diminished and the functioning scores improved. Comparing the HRQOL scores at 6 years after treatment with those at baseline resulted in statistically significant differences for several HRQOL items. However, by comparing the HRQOL scores 6 years after treatment with those at 1 year after treatment found no statistically significant differences were found for any of the HRQOL items (data not shown).

Table 2. Mean scores ± standard deviation of HRQOL items at different time-points before and after 125I prostate brachytherapy

	Variable	Baseline (n = 127)	1 mo (n = 125)	6 mo (n = 118)	1 y (n = 91)	6 y (n = 102)	Difference between baseline and 6y (p-value)
	RAND-36
	Physical functioning	89 ± 14	84 ± 19	85 ± 19	87 ± 16	84 ± 20	.01∗
	Social functioning	86 ± 18	73 ± 23	83 ± 20	88 ± 17	88 ± 17	NS
	Physical role restriction	84 ± 31	64 ± 41	74 ± 37	77 ± 37	84 ± 34	NS
	Emotional role restriction	80 ± 32	78 ± 36	82 ± 33	91 ± 23	88 ± 28	NS
	Mental health	77 ± 17	79 ± 16	81 ± 14	81 ± 15	82 ± 14	.004∗
	Vitality	70 ± 19	67 ± 21	69 ± 18	70 ± 19	72 ± 16	NS
	Pain	94 ± 13	82 ± 19	87 ± 19	90 ± 17	91 ± 15	.003∗
	General health	69 ± 15	67 ± 17	68 ± 19	67 ± 18	70 ± 17	NS
	Change in health	47 ± 17	41 ± 17	47 ± 19	56 ± 19	50 ± 12	NS
	EORTC QLQ-C30
	Physical functioning	92 ± 12	90 ± 13	89 ± 14	90 ± 13	88 ± 14	.006∗
	Role functioning	92 ± 17	79 ± 25	85 ± 21	88 ± 20	89 ± 20	NS
	Emotional functioning	79 ± 18	84 ± 17	86 ± 16	88 ± 15	89 ± 14	<.001†
	Cognitive functioning	86 ± 16	88 ± 16	88 ± 16	88 ± 17	84 ± 21	NS
	Social functioning	92 ± 15	81 ± 20	90 ± 15	91 ± 15	94 ± 13	NS
	Global health/QOL	80 ± 14	73 ± 16	76 ± 16	78 ± 16	82 ± 12	NS
	Fatigue	19 ± 19	25 ± 22	23 ± 20	20 ± 19	22 ± 19	NS
	Nausea and vomiting	1 ± 5	2 ± 7	2 ± 5	1 ± 4	1 ± 3	NS
	Pain	6 ± 14	17 ± 19	14 ± 20	10 ± 18	9 ± 17	.002∗
	Dyspnea	10 ± 18	12 ± 22	15 ± 22	13 ± 21	10 ± 17	NS
	Insomnia	20 ± 27	29 ± 33	21 ± 29	16 ± 25	17 ± 24	.01∗
	Appetite loss	4 ± 13	3 ± 13	3 ± 12	3 ± 11	2 ± 7	NS
	Constipation	2 ± 10	13 ± 24	5 ± 14	4 ± 13	3 ± 9	NS
	Diarrhea	6 ± 15	12 ± 21	11 ± 19	6 ± 18	5 ± 15	NS
	Financial difficulties	1 ± 5	4 ± 12	3 ± 9	2 ± 8	1 ± 6	NS
	EORTC QLQ-PR25
	Urinary symptoms/problems	13 ± 12	40 ± 23	26 ± 18	17 ± 15	18 ± 17	.002∗
	Bowel symptoms/problems	3 ± 5	8 ± 11	7 ± 10	4 ± 7	6 ± 8	.002∗
	Treatment-related symptoms	7 ± 9	8 ± 8	9 ± 10	7 ± 8	6 ± 8	NS
	Sexual functioning	34 ± 26	25 ± 22	33 ± 24	37 ± 23	39 ± 24	NS
	Sexual activity	77 ± 24	61 ± 24	64 ± 27	63 ± 25	62 ± 25	<.001†

Abbreviations: HRQOL = health-related quality of life; RAND-36 = RAND Medical Outcomes Study Short Form 36-item Health Survey; NS = not statistically significant; EORTC QLQ-C30 = European Organization for Research and Treatment of Cancer core questionnaire; EORTC QLQ-PR25 = tumor-specific EORTC prostate cancer module. In RAND-36, higher score represents better health; in EORTC QLQ-C30 and QLQ-PR25, higher score represents more symptoms or better level of functioning or quality of life.

∗ Statistically significant. † Clinically relevant (≥10-point difference).

A statistically significant deterioration after 6 years compared with baseline was seen for physical functioning, pain, urinary symptoms, bowel symptoms, and sexual activity (p ≤.01). A statistically significant improvement after 6 years was seen for mental health, emotional functioning, and insomnia (p ≤.01). For all other HRQOL items, no significant difference was seen. The only clinically relevant changes after 6 years were seen for emotional functioning (10-point increase) and sexual activity (15-point decrease).

The most reported adverse events after prostate brachytherapy are urinary and bowel symptoms. Figure 1 shows the mean change over time in the urinary symptom scores (EORTC QLQ-PR25) up to 6 years after 125I prostate brachytherapy. After an initial increase at 1 month after treatment, the urinary symptoms gradually improved and had reached baseline levels at 1 year after treatment. The urinary symptoms then remained stable up to 6 years after treatment. The same trend was seen for the bowel symptom scores (EORTC QLQ-PR25) (Fig. 2). At 6 years after treatment, 36% of the patients had less urinary symptoms compared with baseline, 47% had more urinary symptoms, and 17% had no change in urinary symptoms. For bowel symptoms, 14% of the patients had less symptoms compared with baseline, 32% had more symptoms, and 54% had no change in symptoms.

View Large Image Download to PowerPoint

Fig. 1 Urinary symptom scores (mean and 95% confidence intervals (CIs)) of the European Organization for Research and Treatment of Cancer prostate cancer module (EORTC QLQ-PR25) up to 6 years after 125I prostate brachytherapy. Higher score represents more symptoms or more problems.

View Large Image Download to PowerPoint

Fig. 2 Bowel symptom scores (mean and 95% confidence intervals (CIs)) of the European Organization for Research and Treatment of Cancer prostate cancer module (EORTC QLQ-PR25) up to 6 years after 125I prostate brachytherapy. Greater score represents more symptoms or more problems.

Figure 3 shows the clinically relevant improvement over time in emotional functioning (EORTC QLQ-C30). The development of sexual activity scores (EORTC QLQ-PR25) over time is shown in Fig. 4. At 6 years after treatment, 70% of the patients had diminished sexual activity compared with baseline, 12% had improved sexual activity, and 18% had no change in sexual activity.

View Large Image Download to PowerPoint

Fig. 3 Emotional functioning scores (mean and 95% confidence intervals (CIs)) of the European Organization for Research and Treatment of Cancer core questionnaire (EORTC QLQ-C30) up to 6 years after 125I prostate brachytherapy. Higher score represents greater level of emotional functioning.

View Large Image Download to PowerPoint

Fig. 4 Sexual activity scores (mean and 95% confidence intervals (CIs)) of the European Organization for Research and Treatment of Cancer prostate cancer module (EORTC QLQ-PR25) up to 6 years after 125I prostate brachytherapy. A higher score represents a higher level of sexual activity.

Multivariate linear regression analysis was performed to identify predictors for diminished HRQOL at 6 years after treatment. Predictors for urinary symptoms were hormonal therapy (β, 8.1; 95% confidence interval [CI], −0.07 to 16.2) and initial prostate-specific antigen level (β, −0.19; 95% CI, −0.48 to 0.09). Prostate volume (β, 0.11; 95% CI, −0.03 to 0.25) was identified as a predictor for bowel symptoms. Older age (β, −1.37; 95% CI, −2.26 to −0.48) was associated with diminished sexual activity at 6 years. Age (β, −1.50; 95% CI, −2.13 to −0.86), hormonal therapy (β, 14.2; 95% CI, 3.8 to 24.6) and initial prostate-specific antigen level (β, −0.53; 95% CI, −0.89 to −0.16) were predictors for sexual functioning.

Discussion 

To the best of our knowledge, this is the first study prospectively assessing long-term HRQOL after 125I brachytherapy for early-stage prostate cancer. Long-term QOL is required, because toxicity after prostate brachytherapy can develop more than 3 years after treatment (16). In the present study, we compared HRQOL approximately 6 years after treatment with the baseline values. Previously, we published our HRQOL data at up to 1 year after treatment, showing significantly worse HRQOL at 1 month after treatment compared with other points (22). Our long-term data show that overall HRQOL at 6 years after treatment did not differ significantly from that at baseline. Thus, 125I brachytherapy does not adversely affect patients' long-term HRQOL.

Although a statistically significant change between 6 years after treatment and baseline was seen for some HRQOL items, most changes were not clinically relevant. A statistically significant deterioration was seen for urinary symptoms, bowel symptoms, pain, physical functioning, and sexual activity. A statistically significant improvement was seen for emotional functioning, mental health, and insomnia. Only for emotional functioning and sexual activity a clinically relevant change was seen.

In an additional analysis, we compared the HRQOL at 6 years after treatment with that at 1 year after treatment. For none of the investigated HRQOL items was a statistically significant difference found, indicating that the QOL remains stable after 1 year post-treatment. Therefore, our data would suggest that measuring patients' HRQOL after prostate brachytherapy could be limited to 1 year after treatment. However, these data need to be confirmed.

Urinary dysfunction is the most common adverse event associated with prostate brachytherapy. The increase in symptoms 1 month after treatment has been frequently described 6, 15, 22. Figure 1 shows that 1 year post-treatment the urinary symptoms had returned to approximately baseline level and remained stable for up to 6 years after treatment.

Although numerous studies have reported the short-term urinary symptoms after prostate brachytherapy, to date, only a few studies have evaluated the long-term urinary QOL 8, 12, 15, 29. Ash et al. (15) described the IPSS scores up to 9 years after 125I brachytherapy. However, HRQOL is a multidimensional concept that includes more than toxicity alone 17, 18. The more extended Expanded Prostate Cancer Index Composite urinary symptom scores were also evaluated but were limited by a short follow-up of 2 years. At 1 year after treatment, the mean Expanded Prostate Cancer Index Composite score had returned to the pretreatment level. This finding is similar to the results from Merrick et al. (8), who found no significant difference in long-term urinary QOL when brachytherapy patients were compared with a matched control group. Although that study was limited by its cross-sectional design, the results are similar to our data.

The results from studies that evaluated short-term urinary QOL were similar to ours as well. Lee et al. (6) described a return to baseline levels for IPSS and Functional Assessment of Cancer Therapy–Prostate scores 1 year after prostate brachytherapy. Downs et al. (9) found a return of University of California, Los Angeles, Prostate Cancer Index scores to baseline at 18–24 months after treatment. Feigenberg et al. (10) reported no return of IPSSs to baseline levels 1 year after treatment for 60% of patients. Nevertheless, using the Functional Assessment of Cancer Therapy–Prostate, two-thirds of the men reported urinary function compared with baseline. Only Caffo et al. (11) did not found a return to baseline levels for urinary symptoms until 3 years after treatment. Some other published studies were primarily designed to compare different treatment modalities 7, 13, 14, 16. A difficulty in comparing these studies is the use of different questionnaires, making interpretation of the results a challenge.

Fewer studies have been published concerning bowel symptoms after prostate brachytherapy. In our study, bowel symptoms returned to the baseline level at 1 year and remained stable for up to 6 years after treatment (Fig. 2). Although a statistically significant worsening of symptoms was found at 6 years compared with baseline, no clinically relevant change was seen. However, we must interpret these results with caution, because van Andel et al. (28) recently described unsatisfactory clinical validity for the EORTC QLQ-PR25 bowel function item. Also, in our analysis, the Cronbach α coefficient for this item was limited. Despite this possible limitation, our results are in accordance with other published data. Merrick et al. (30) demonstrated that prostate brachytherapy adversely affects bowel symptoms in approximately 10% of patients and, in most patients, the changes are minimal and slowly resolve with time. Others have confirmed the absence of relevant long-term bowel morbidity after prostate brachytherapy 15, 29.

Sexual activity after prostate brachytherapy was also evaluated. A clinically relevant 15-point decrease in sexual activity was seen after 6 years, without a significant worsening in sexual functioning. Feigenberg et al. (10) confirmed this discrepancy in sexual functioning and sexual activity. At 1 year, 78% of the patients stated that they could achieve an erection with or without assistance; however, almost 50% reported a decrease in sexual activity (10). This is important, because most studies reporting sexual toxicity after cancer therapies focus only on erectile function. Stock et al. (31) reported actuarial decreases in erectile function in 29% of patients 1 year after prostate brachytherapy, and Potters et al. (32) reported a potency rate of 67% at 5 years. However, the decrease in sexual activity might be explained, not only by the possibility of achieving an erection, but also by aging, the loss of a partner, or the detrimental effect of cancer treatment on a patient's frequency of sexual activity.

As mentioned, in the present study, we did not find long-term sexual dysfunction. This might have been because of our longer follow-up, indicating that improvement in sexual functioning could continue for several years after treatment. Ash et al. (15) also found a moderate improvement in sexual function 2 years after prostate brachytherapy. Feigenberg et al. (10) demonstrated that the negative effect of prostate brachytherapy on sexuality was significant only in the immediate posttreatment period. Finally, one could argue that the Dutch translation of the EORTC QLQ-PR25 requests an answer to the questions concerning sexual activity only when the patients had been sexually active in the previous 4 weeks, thus causing a low response rate for this scale and perhaps leading to an under- or overestimated QOL in terms of sexual activity. However, in our analysis, the response rate to sexual activity was 71% and was even 100% for sexual functioning.

We found a clinically relevant 10-point improvement in emotional functioning 6 years after prostate brachytherapy. This improvement after cancer therapy has been described for other cancer types as well and can be explained by patients having had time to adapt to the situation, a response shift mechanism, and a decreasing fear of recurrence and death with time (33). The same mechanisms could apply to the statistically significant improvement in mental health.

Although HRQOL at 6 years after treatment was not significantly different compared with baseline for the whole cohort, we found some predictors for diminished HRQOL. In particular, the effect of neoadjuvant hormonal therapy on HRQOL has been questioned in published studies. Our data showed that hormonal therapy seemed to predict for more urinary symptoms and worse sexual functioning at 6 years after treatment. Because our data at 6 years hardly showed any clinically relevant changes in QOL compared with baseline, we did not expect any correlation with dose. Therefore, we did not include dosimetric parameters in the analysis. Acute urinary retention, relapse, and salvage therapy are also factors that might be associated with HRQOL. In ongoing research, we will explore further the predictive factors for short- and long-term HRQOL and validate them.

The present study had several limitations. First, we did not have a control group, which precludes a correction for age-related morbidity. Second, patients received the first questionnaire after their diagnosis. The knowledge of having cancer might have influenced their HRQOL. Third, our study described the experience of a single center. Morbidity might vary from center to center, depending on the treatment techniques used. Fourth, social and demographic items were not evaluated and these might have influenced patients' HRQOL. The study population was a white Dutch population, and the QOL might differ from those of other countries and cultures.

Conclusion 

This is the first study presenting prospective HRQOL data up to 6 years after 125I prostate brachytherapy. The HRQOL scores had returned to approximately baseline values at 1 year and remained stable up to 6 years after treatment. 125I prostate brachytherapy did not adversely affect patients' long-term HRQOL.