Disease-specific symptoms and general quality of life of patients with prostate carcinoma before and after primary three-dimensional conformal radiotherapy

Abstract

BACKGROUND

Approximately 189,000 men are diagnosed with prostate carcinoma each year and more than 1 million are living with the disease. Good prognoses and undesirable sequelae accompany each of several available primary and adjuvant treatment options. The current study explored the effects of primary three-dimensional conformal radiotherapy with or without neoadjuvant hormonal therapy on urinary, bowel, and sexual symptoms and health-related quality of life (HRQOL).

METHODS

A prospective, repeated-measures design study included 100 patients. Data from the Medical Outcomes Study Short Form Health Survey (a measure of general HRQOL) and a 12-item symptom questionnaire were collected before the start of radiotherapy, approximately 1–3 months after completion of treatment, and again approximately 5–10 months after completion of treatment for follow-up.

RESULTS

Patients reported few urinary symptoms after treatment. Bowel frequency and urgency were reported more frequently posttreatment and at follow-up. Erectile difficulties, which were common pretreatment, were reported with increased frequency posttreatment and at follow-up. General HRQOL scores were higher than age-related general population norms for men at all three data collection times, but there were significant losses posttreatment for patients' physical functioning and vitality. At the 5–10-month follow-up, physical functioning remained lower but vitality scores regained some of the losses. A more extended follow-up is needed. Neoadjuvant therapy, which was received before the pretreatment data collection, had a deleterious effect on erectile functioning but no interactive effects with the radiotherapy on symptoms or HRQOL.

CONCLUSIONS

Although patients with a diagnosis of prostate carcinoma experienced increased bowel and sexual dysfunction and decreased vitality after radiotherapy, their HRQOL scores remained at or above age-related general population norms. Cancer 2003. © 2003 American Cancer Society.

Prostate carcinoma is often identified as the second leading cause of cancer death among men. According to estimates from the American Cancer Society, nearly 30,000 deaths will occur during 2003.1 A more dramatic statistic, however, is the number of men living with prostate carcinoma. The National Cancer Institute estimates that more than 1.69 million men diagnosed with prostate carcinoma within the last 25 years are still alive.2 With approximately 221,000 new cases estimated for 2003 and a 5-year survival rate of 97% (compared with a 67% survival rate 25 years ago), the number of men living with the disease and its treatments certainly will increase during the next 5–10 years.1

The 221,000 men diagnosed this year will face a challenging decision. Patients with prostate carcinoma have several treatment options (radical prostatectomy, external beam radiation, brachytherapy, and watchful waiting are the most common), but there are no clear guidelines to help them make a treatment decision. Several explicit attempts to reach a consensus, by the Prostate Cancer Clinical Guideline panel of the American Urological Association and by others, have yielded no definitive evidence that any one treatment is superior.3–5

Without clear evidence of the superiority of any one treatment and with the potential for many years of survival, patients with prostate carcinoma have reason to be concerned. They need information pertaining to the side effects and limitations to quality of life they will experience with different treatments.6, 7

During the last 7 years, investigators have explored the effects of prostate carcinoma and its treatments on patients' health-related quality of life (HRQOL). The studies often employed a retrospective, cross-sectional design.8–10 In these studies, patients treated for prostate carcinoma were surveyed at one point in time at various intervals after receiving treatment and were compared with those treated only by observation or watchful waiting and/or with age-matched control individuals without prostate carcinoma. The studies focused on urinary, bowel, and sexual functioning, areas particularly relevant to prostate carcinoma and treatment, and on general HRQOL.8–10 Time frames and specific symptom definitions and their measurement differed but the results of those studies were similar. Compared with men without prostate carcinoma, patients reported increased bowel, urinary, and sexual symptoms but little difference in general HRQOL measures.8 The retrospective cross-sectional design has been criticized for lacking control of previous conditions and for being too distant in time from the treatment.11, 12 More recent studies have made use of prospective longitudinal designs.13–20

Of particular relevance to the current investigation are two studies that focused on external beam radiation. A large, population-based study presented data on urinary, bowel, and sexual functioning of approximately 500 men at 3 points in time, 6 months, 12 months, and 24 months after primary treatment of prostate carcinoma with external beam radiation alone.21 A comparative baseline standard was collected at 6 months by recall shortly before malignant disease was diagnosed. Compared with the recalled baseline, the proportion of patients experiencing moderate-to-severe bowel problems (such as pain or urgency) increased at 6 months (from 7% to 14%) and then decreased (to 9%) by 24 months. With regard to sexual function, the proportion of men experiencing moderate-to-severe problems progressively increased, from 26% at baseline to 40% at 24 months. There was no increase in urinary problems. The exploration of symptoms across an extended period of time posttreatment in that study21 was an especially valuable contribution. The use of a recall baseline rather than a prospective pretreatment assessment, however, may pose some limitations on the interpretation of the results. The authors had previously presented evidence for its use, but the validity is weaker when the symptom is more prevalent, particularly for the sexual function symptoms.22

An older study by Beard et al.23 was one of the few to include a crucial factor in the exploration of symptoms and HRQOL after radiotherapy for prostate carcinoma, whether the planning and delivery technique for the radiotherapy was whole-pelvis, small-field, or (what was then a new technique) computer-assisted three-dimensional (3-D) conformal radiation. The investigators evaluated 121 men (n = 25, n = 60, and n = 36 in the whole-pelvis, small-field, and computer-assisted 3-D conformal radiation groups, respectively) pretreatment and at 3 and 12 months posttreatment. They found an increase of bowel symptoms at 3 months that declined again at 12 months, but a more steady increase in sexual symptoms. The investigators also reported findings that disfavored whole-pelvis therapy or favored conformal therapy with respect to a number of symptom measures, but the differences were not statistically significant.

A more recent study by Wei et al.24 added another important but often ignored dimension to the study of quality of life after radiotherapy for prostate carcinoma—neoadjuvant hormonal therapy. This additional therapy has been recommended to improve the clinical effects of radiation. Therefore, it is an important part of patients' decision processes.25, 26 A large cross-sectional study involving more than 1000 patients who had received a prostatectomy, brachytherapy, or 3-D conformal radiation within the previous 4 years was conducted. Men in the community without prostate carcinoma served as age-matched control individuals. As expected, the patients who received radiotherapy had worse bowel and sexual functioning compared with the control individuals. The subset of patients who had received either adjuvant or neoadjuvant hormonal therapy reported significantly worse functioning on a set of symptoms related to androgen deprivation (e.g., hot flashes, lack of energy, breast tenderness, body weight changes)27 but otherwise did not differ from patients receiving primary therapy only.

The current study was undertaken to explore the short-term (up to 1 year) effects of 3-D conformal radiotherapy with and without neoadjuvant hormonal ablation on disease-specific and general HRQOL.

MATERIALS AND METHODS

Study Population

Patients were recruited from a total of 225 men treated from April 1998 to April 2000 with external beam radiation for prostate carcinoma at a radiation oncology unit of an urban regional hospital's cancer center. Only patients diagnosed with localized disease without any evidence of metastasis were eligible. One hundred eighty-one patients were known to have met this eligibility criterion. Of these, 157 patients initially consented to be in the study but only 125 patients returned baseline surveys. The remaining 32 men were lost to follow-up. Chart reviews and information from the patients revealed that of the 125 patients, 16 patients received either adjuvant or salvage radiotherapy after a radical prostatectomy and 9 patients received adjuvant brachytherapy within the period of their participation in the study. Due to the small numbers in the two groups and the desire to focus on radiation as a primary therapy, these patients were not included in the final analyses. The remaining 100 patients received primary 3-D conformal radiation at 7000–7400 centigrays for a period of 7–8 weeks. Of the 100 patients, 68 received neoadjuvant hormonal ablation, whereas 32 did not. At the time of the study, the standard of care for neoadjuvant hormonal ablation to external beam radiation at the Helan & Harry Gray Cancer Center was 6 months (i.e., 2 injections at 3-month intervals) commencing 6–8 weeks before initiation of radiotherapy.

Data Collection and Measures

Study Instruments

The Medical Outcomes Study Short Form Health Survey (SF-36) was used as the measure of general HRQOL.28, 29 The specific measures included the physical (PCS) and mental (MCS) component summary scores and normed scores for the eight subscales. A self-report symptom checklist was developed to assess disease-specific symptoms (i.e., urinary, bowel, and sexual symptoms). The self-report measure comprised 12 questions covering bowel frequency, urgency, bleeding, cramps and impairment of routine functioning; urinary frequency (day and night), urgency, flow (stream), and impairment of routine functioning; and the existence of erectile difficulties and how disturbing they were. The SF-36 is similar in content to previously used instruments11, 30 but was developed independently for ease of administration and minimal response burden. Each item on the scale is treated individually because interitem consistency was not sufficiently high for the creation of bowel (Cronbach alpha = 0.55), urinary (Cronbach alpha = 0.75), and sexual scales (Cronbach alpha = 0.62). In addition, self-report responses were used to establish the existence of comorbidities such as diabetes, previous pelvic surgery, bladder disease, and bowel disease. A chart review was conducted to determine the diagnostic prostate-specific antigen (PSA) level and Gleason grade. A combination of self-report responses and chart review information was used to determine the receipt of neoadjuvant hormonal therapy.

Data Collection

After being fully informed about the study and obtaining their consent, patients completed the symptom checklist and the SF-36 just before beginning their radiotherapy. The pretreatment baseline level was not established before the neoadjuvant therapy was administered. Posttreatment information was collected 10–20 weeks after the initiation of treatment, approximately 1–3 months after completion of the 7–8-week radiotherapy schedule. Follow-up information was collected 6 months to 1 year after the initiation of treatment or approximately 5–10 months after its completion. Only patients with at least 90 days between the posttreatment and follow-up data collection points were included in the study population. For the 100 patients, the average posttreatment collection time was 99 ± 16.5 days after treatment began and the average follow-up data collection time was 266 ± 23.4 days after initiation of treatment. Questionnaires were mailed to participating patients at their homes. Of the 100 patients, 69 provided posttreatment data and 74 provided follow-up data. Sixty-one patients provided data at all three time points. Most of the analyses were focused on the patients with all three data points.

Statistical Analysis

The main outcomes measures were the change over time for reported symptoms and HRQOL. Friedman's rank tests for correlated samples were used to evaluate differences in pretreatment, posttreatment, and follow-up symptoms. In addition, a descriptive ordinal variable was created for each symptom; this variable used the reported symptoms at the three time points to categorize each patient as having experienced 1) improvement; 2) no change; 3) an increase in symptoms posttreatment with a subsequent decrease toward pretreatment levels at follow-up; and 4) an increase in reported symptom levels posttreatment and at follow-up for each symptom. General linear model (GLM) repeated- measures analyses were used for the general HRQOL measures—i.e., two summary measures (PCS, MCS) and the normed scores on the eight subscales of the SF-36.

The effect of neoadjuvant hormonal therapy before treatment was analyzed by Wilcoxon rank-sum tests for each preradiation symptom measure. Interactive effects of the neoadjuvant hormonal treatment with the primary radiation treatment were explored by Wilcoxon rank-sum tests on the descriptive ordinal classification variable described above created for each symptom.

Due to the variability in the timing of the posttreatment and follow-up data collection, additional correlational analyses were performed to determine if the timing was a confounding variable in the interpretation of the analyses of the main outcomes. Possible confounding effects of comorbidities were explored using chi-square tests of proportion for symptoms and GLM with comorbidities as a between-subjects factor for the HRQOL analyses.

RESULTS

Pretreatment Patient Characteristics

Table 1 outlines the demographic and health characteristics of the 100 patients at baseline. Most of the men were Caucasian (78%) and more than one-half were older than 70 years. Diagnostic PSA scores ranged from less than 1 to 36.7, with a mean of 10.8. Fifty percent of the patients had PSA scores that fell between 4 and 10. Gleason scores were moderate, with scores of 6 and 7 being most common (37% and 39% of the sample, respectively). There were few reported comorbidities: 10 patients had diabetes, 10 patients had received pelvic surgery, 6 patients reported bowel disease, and 5 patients reported bladder disease. In total, 26 patients reported the presence of 1 or more conditions before treatment. The demographics and health characteristics of the patients who received/did not receive hormonal ablation therapy were compared. The only significant difference between these two groups of patients was that a larger proportion of the patients who received hormonal ablation were Caucasian compared with patients whi did not receive hormonal ablation (89% vs. 69%).

Table 1. Distribution of Demographic and Health Characteristics

Characteristic	All study patients (n = 100) (%)	With hormonal ablation (n = 68) (%)	Without hormonal ablation (n = 32) (%)
PSA: prostate-specific antigen.
Age (yrs)
< 60	12 (12.0)	8 (11.8)	4 (12.5)
60–69	31 (31.0)	19 (27.9)	12 (37.5)
≥ 70	57 (57.0)	41 (60.3)	16 (50.0)
Ethnicity/race
Caucasian	78 (83.0)	58 (89.2)	20 (69.0)
African American	7 (7.4)	2 (3.1)	5 (17.2)
Latino	6 (6.4)	3 (4.6)	3 (10.3)
Other	3 (3.2)	2 (3.1)	1 (3.4)
Unknown	6	3	3
Diagnostic PSA
≤ 4.0	8 (8.1)	6 (8.8)	2 (6.5)
4+ through 10	50 (50.5)	34 (50.0)	16 (51.6)
10+ through 20	33 (33.3)	22 (32.4)	11 (35.5)
> 20	8 (8.1)	6 (8.8)	2 (6.5)
Unknown	1		1
Gleason score
5 or 6	39 (39.4)	24 (35.3)	15 (48.4)
7	39 (39.4)	27 (39.7)	12 (38.7)
8+	21 (21.2)	17 (25.0)	4 (12.9)
Unknown	1		1
Comorbidity
Diabetes	10 (10.6)	6 (9.5)	4 (12.9)
Previous pelvic surgery	10 (10.8)	8 (12.9)	2 (6.5)
Bowel disease	6 (6.5)	6 (9.4)	0 (0)
Bladder disease	5 (5.5)	3 (4.9)	2 (6.7)

Reported Urinary, Bowel, and Sexual Symptoms

Patients reported significantly different degrees of severity at pretreatment, posttreatment, and follow-up assessment for 8 of the 12 bowel, urinary, and sexual symptoms: bowel frequency, urgency, and interference with normal routine; urinary frequency (day and night), urgency, and stream difficulty; and erectile difficulties. There are different patterns of individual change among the symptoms as indicated by the ordinal classification variable for symptom change (Table 2). More than 30% of the patients reported a decrease in 3 of the urinary symptoms: frequency during the day, urgency, and difficulty starting the urinary stream. The effect of treatment on urinary stream was related to the timing of the posttreatment data collection (r = −0.318). Patients who reported their symptoms soon after treatment were more likely to report urinary stream problems than patients who completed the questionnaire at a later time point. In contrast to the general decrease in urinary symptoms, greater than 30% of the patients reported a sustained increase in bowel frequency and bowel urgency at posttreatment and follow-up. Difficulties with bowel bleeding were related to later timing of the follow-up data collection (r = 0.300; P < 0.02). For example, 7 of the 17 (41%) patients who reported bowel bleeding more than 278 days after the initiation of treatment (highest quartile) had difficulty, compared with 1 of 18 (6%) patients who reported bowel bleeding less than 253 days after the initiation of treatment (lowest quartile). Almost 25% of the patients reported a sustained increase in erectile difficulties through follow-up. The increase was more dramatic for patients who reported sufficient erectile functioning for intercourse before radiation compared with patients who did not. Of the 42 men in that category, 20 provided follow-up data. Fifty percent of these men reported a sustained increase in symptoms posttreatment and at follow-up. Due to the low incidence of comorbidities reported among the study population, patients reporting any comorbidity were compared with those with no reported comorbidities. There were no differences in the proportion of patients reporting an increase posttreatment in any of the symptoms.

Table 2. Change in Disease-Specific Symptoms Reported before Treatment, after Treatment, and on Follow-Up

Symptoms	n	Patients reporting symptoms (%)
		Decreased	Did not change	Increased and then decreased	Increased
a Pretreatment, posttreatment, and follow-up distribution determined to be significantly different by Friedman rank-sum test for matched samples (P < 0.05). b Pretreatment, posttreatment, and follow-up distribution determined to be significantly different by Friedman rank-sum test for matched samples (P < 0.01).
Bowel frequencya	58	10.3	48.3	10.3	31.0
Bowel urgencyb	56	12.5	35.7	16.1	35.7
Bowel bleeding	56	5.4	66.1	10.7	17.9
Bowel cramping	56	14.3	60.7	8.9	16.1
Impaired bowel functioningb	55	3.6	61.8	14.5	20.0
Urinary frequency (day)b	57	35.1	35.1	19.3	10.5
Urinary frequency (night)a	58	20.7	48.3	19.0	12.1
Urinary urgencyb	60	40.0	26.7	15.0	18.3
Difficulty starting urine flowa	59	32.2	44.1	10.2	13.6
Impaired urinary functioning	57	10.5	68.4	8.8	12.3
Erectile dysfunctiona	53	3.8	64.2	7.5	24.5
Sufficient function at baselineb	20	0.0	40.0	10.0	50.0
Disturbed by erectile dysfunction	56	12.5	58.9	3.6	25.0

Changes in Health-Related Quality of Life

Table 3 shows that from the pretreatment period through follow-up, patient scores for PCS (46.8–49.2) and MCS (54.0–54.9) were higher by several points compared with the norms among the general population for men in either the 55–64-year-old or ≥ 65-year-old age groups (norms for PCS, 46.90 and 41.95 for the two age groups, respectively; MCS, 51.60 and 52.51 for the two age groups, respectively).29 There were significant pretreatment to posttreatment changes with respect to several of the HRQOL measures. The PCS and two physical subscales, Physical Functioning and Role Physical, showed a significant decline from pretreatment to posttreatment that continued through the period of follow-up. A fourth measure, the patient's self-reported level of vitality, also significantly declined from the pretreatment period to the posttreatment period but increased again to a level indistinguishable from the pretreatment level on follow-up. The MCS and the other subscales showed no significant change after treatment. Patients reporting 1 or more comorbidities (n = 14) had significantly lower scores (4–6 points) on the Physical Functioning subscale at all 3 time points (P < 0.05) and a greater decline in scores at follow-up for the General Health and Role Emotional subscales compared with patients who did not report comorbidities. The timing of the posttreatment and follow-up data collection was not related to any differences in the reported HRQOL.

Table 3. Change in Health-Related Quality of Life Normed Scores: Pretreatment to Posttreatment and Follow-Up

Measure	n	Preradiation	Posttreatment	Follow-up	Norm
					Males ages 55–64 yrs	Males age 65+ yrs
PCS: physical component summary score; MCS: mental component summary score. a Preradiation score differs from postradiation and follow-up scores (P < 0.05). b Preradiation score differs from postradiation and follow-up scores (P < 0.01). c Preradiation score differs from postradiation score only (P < 0.01).
PCSa	54	49.5	46.8	47.2	46.90	41.95
MCS	54	54.9	54.8	54.0	51.60	52.51
Physical Functioningb	60	49.4	47.0	46.7
Role Physicala	61	49.1	45.1	45.6
Bodily Pain	59	53.4	52.5	51.3
General health	57	50.0	50.6	49.1
Vitalityc	58	52.6	49.2	51.1
Social Functioning	58	53.2	51.6	51.7
Role Emotional	58	52.0	51.0	49.7
Mental Health	58	53.8	55.0	53.6

Effects of Neoadjuvant Hormonal Ablation

Patients responded to the symptom checklist before radiotherapy but not before any hormonal ablation shots were received. Therefore, the effects of the neoadjuvant treatment were explored in a comparison of pretreatment symptoms for the two groups of patients (Table 4). Analyses of these distributions as ordinal scales using the Wilcoxon rank-sum test found that sexual functioning was most affected by the hormonal ablation shots. The two sexual functioning symptoms showed significant differences. In longitudinal analysis, there were no interactive effects of the presence of hormonal ablation with any longitudinal effects of the primary radiation treatment.

Table 4. Effects of Hormonal Ablation Therapy Before Radiation on Disease-Related Symptoms

Symptoms	Without hormonal ablation (n = 31)				With hormonal ablation (n = 67)
	No problem or least frequent (%)	Some, slight or occasional frequency (%)	Moderate level or frequency (%)	Worst level or most frequent (%)	No problem or least frequent (%)	Some, slight or occasional frequency (%)	Moderate level or frequency (%)	Worst level or most frequent (%)
a Distribution across all categories significantly different chi-square (P < 0.05), and Wilcoxon rank-sum test indicated differences in ordinal distribution (P < 0.01). b Distribution across all categories significantly different chi-square (P < 0.01), and Wilcoxon rank-sum test indicated differences in ordinal distribution (P < 0.005).
Bowel frequency	9.4	40.6	46.9	3.1	7.4	58.8	29.4	4.4
Bowel urgency	74.2	25.8	0.0	0.0	53.7	44.8	0.0	1.5
Bowel bleeding	83.9	12.9	0.0	3.2	86.4	13.6	0.0	0.0
Bowel cramping	87.1	12.9	0.0	0.0	77.3	22.7	0.0	0.0
Impaired bowel functioning	93.8	3.1	3.1	0.0	91.2	7.4	1.5	0.0
Urinary frequency (day)	28.1	40.6	28.1	3.1	34.3	41.8	22.4	1.5
Urinary frequency (night)	16.1	48.4	19.4	16.1	9.1	54.5	33.3	3.0
Urinary urgency	25.0	46.9	15.6	12.5	21.5	46.2	30.8	1.5
Difficulty starting urine flow	67.7	16.1	12.9	3.2	54.5	30.3	12.1	3.0
Impaired urinary functioning	80.6	12.9	3.2	3.2	83.3	13.6	1.5	1.5
Erectile dysfunctiona	48.4	12.9	16.1	22.6	20.6	15.9	20.6	42.9
Disturbed by erectile dysfunctionb	76.6	13.3	n.a.	10.0	47.7	36.9	n.a.	15.4

DISCUSSION

The pattern of symptoms reported by the patients in the current study is consistent with previous studies of patients undergoing primary radiotherapy. There were few urinary problems after treatment. In fact, a substantial number of patients reported improvement in several urinary symptoms, including urinary urgency, daytime frequency, and flow. Several bowel symptoms increased in severity. For example, more than 1 in 6 men reported significant bowel urgency problems 5–10 months after treatment. This is consistent with the 15% incidence rate reported by Madalinska et al.15 and the 16% rate reported by Beard et al.23 for the conformal radiation subsample at 6 and 12 months posttreatment. The finding that patients who reported their follow-up symptoms at a later point in time (closer to a year) were more likely to report difficulties with bowel bleeding and the finding by Hamilton et al.21 that the incidence of these symptoms declined at 24 months suggest that radiation may have a delayed effect, peaking at approximately 1–2 years after treatment, with respect to this symptom.

Forty-seven percent of patients reported erectile difficulties at follow-up. This rate probably can be attributed, at least in part, to the pretreatment rate of erectile difficulties (40%) in the sample. However, before treatment decisions are made, patients should be made aware that approximately 50% of patients with self-reported erectile functioning sufficient for intercourse before radiation report a worsening of function after radiotherapy.

As found in the earlier studies, the men in the current study report HRQOL scores higher than general population norms for men in their age categories despite the presence of disease and treatment protocols. The significant pretreatment-to-posttreatment differences found for Physical Functioning, Role Physical, Vitality, and overall PCS are consistent with the commonly held belief that radiation affects stamina and feelings of fatigue. All four significant differences indicate a diminution of energy after the course of radiation. The increase toward pretreatment levels on the Vitality subscale at follow-up suggests that these effects might be temporary. Patients need to be educated before treatment that they may experience this decline in energy soon after treatment but that vitality is likely to increase again.

Comparisons of the two patient subgroups suggest that hormonal ablation therapy primarily affected sexual functioning. There were few effects on urinary or bowel symptoms or more general HRQOL. The sexual effects were in response to the hormonal ablation received before radiotherapy, with no interaction with the effects of the primary radiotherapy.

What does this tell the patient with prostate carcinoma who needs to make a treatment decision and his physicians, who are trying to inform his choice? Radiotherapy is unlikely to result in sustained urinary frequency, urgency, or initiation symptoms. If the patient is having trouble with frequency, urgency, or urinary flow due to prostate carcinoma, treatment might offer some relief. However, there is an increased risk of bowel frequency and urgency problems throughout the first year after radiation. Erectile functioning, if present when radiation begins, is also likely to be diminished for at least 6 months to 1 year after radiation. However, despite these specific symptoms and a more general loss in physical functioning and vitality, the ratings of general HRQOL remain relatively high. For patients making a secondary treatment decision regarding whether to receive adjuvant hormonal therapy, the results suggest that they will need to balance an immediate deleterious effect on sexual functioning against any assumed clinical benefit of the treatment. Otherwise, hormonal treatment is likely to have little effect on other symptoms of radiation or on HRQOL.

The current study has two limitations. The choice of a newly developed instrument for symptom measurement was made for the instrument's brevity and lack of patient burden but at the loss of proven reliability and validity. The instrument's similarity in content to other instruments was noted earlier. The similarity of the pattern of symptoms reported by patients using the current measure strengthens the suggestion that the instrument was consistent with the other measures.

There also are issues regarding the timing of the baseline and follow-up data collection. The enrollment and first data collection were timed to the beginning of the radiotherapy and therefore were performed after hormonal ablation therapy had begun. The data collection served as an appropriate baseline for the effects of radiotherapy but was limited in that capacity for the neoadjuvant therapy. The follow-up data collection occurred 5–10 months after completion of radiotherapy. The current data set has shown that several symptoms attenuate after a few months but the recovery time for other symptoms may be longer. For example, bowel bleeding appears to be a delayed symptom, and the effects of hormonal ablation therapy may linger after the 6-month dose, making them incompatible with the timing of the current study. In both cases, the information collected is valuable but must be considered to be preliminary in terms of timing until additional studies with expanded time frames can be conducted. The first data collection should be positioned earlier in time so that an absolute baseline (i.e., one established before all treatments) can be established. Data on symptoms should be collected at later time points to determine whether symptoms persist for an extended time before improvement or whether they become permanent. Given the number of years that most patients with prostate carcinoma can expect to live after primary treatment, the measurement of symptoms and HRQOL at 1, 2, and 5 years posttreatment clearly is an important issue to researchers, clinicians, and, most of all, patients making crucial treatment decisions.

With the emerging consensus of findings across studies, additional issues should be explored in future HRQOL research for patients with prostate carcinoma. First, more attention should be given to the effects of neoadjuvant and adjuvant hormonal therapy for patients with localized prostate carcinoma, as it is becoming a common practice. The current study was one of the few that had a high proportion of patients receiving this treatment and explicitly evaluated its effects. In addition, instruments such as the one introduced by Wei et al.,27 which included additional symptoms that might result from hormonal changes, such as hot flashes and weight changes,. should be considered.

Despite the existence of many varieties of erectile aids (e.g., injections, vacuum pumps, sildenafil), most studies that explored the effects of prostate carcinoma treatment on erectile difficulties ignored their use and their impact on sexual functioning and HRQOL. In an early study, Lim et al.31 found that few patients reported the use of any erectile functioning aids, but that study predated the introduction of sildenafil by several years. In a more recent study Schover et al.32 found that a substantial proportion of patients attempted one or more treatments for erectile dysfunction. Questionnaires on sexual functioning after prostate carcinoma treatment should include questions about patients' use of erectile aids and should allow responses to questions about sexual functioning to be answered for both situations (with and without aids). The responses regarding functioning without aids will more truly reflect the effect of treatment, whereas the responses regarding functioning with erectile aids when needed may more truly reflect the life situation of the patient and may be more closely related to the patient's quality of life.

In summary, the current study offers evidence that for the time frame studied, patients undergoing radiotherapy for prostate carcinoma may face bowel and sexual functioning symptoms and some temporary loss of vitality but their quality of life remains high. These findings should be verified via a longer period of data collection and should be expanded to account for the effects of neoadjuvant hormonal therapy and improved measures of erectile functioning, including the important role of erectile aids.

Acknowledgements

The authors thank the staff of the Helen & Harry Gray Cancer Center, Hartford Hospital (Hartford, Connecticut), especially Janet Lynch, and the support staff of the Institute for Outcomes Research and Evaluation, Hartford Hospital.