PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Pain Symptom Manage. Author manuscript; available in PMC Oct 1, 2013.
Published in final edited form as:
PMCID: PMC3463773
NIHMSID: NIHMS373852
Trajectories and Predictors of Symptom Occurrence, Severity, and Distress in Prostate Cancer Patients Undergoing Radiation Therapy
Katie Knapp, RN, MS, Bruce Cooper, PhD, Theresa Koetters, RN, MS, Janine Cataldo, RN, PhD, Anand Dhruva, MD, Steven M. Paul, PhD, Claudia West, RN, MS, Bradley E. Aouizerat, PhD, MAS, and Christine Miaskowski, RN, PhD
Schools of Nursing (K.K., B.C., T.K., J.C., S.M.P., C.W., B.E.A., C.M) and Medicine (A.D.) and Institute for Human Genetics (B.E.A.), University of California at San Francisco, San Francisco, California, USA
Address correspondence to: Christine Miaskowski, RN, PhD, FAAN, Department of Physiological Nursing, University of California, 2 Koret Way – N631Y, San Francisco, CA 94143-0610, USA, chris.miaskowski/at/nursing.ucsf.edu
Context
Radiation therapy (RT) is a common treatment for prostate cancer. Despite available research, prostate cancer patients report that information about side effects is their most important unmet need. Additional research is needed that focuses on specific dimensions of the patient’s symptom experience.
Objectives
The study’s purposes were to evaluate the trajectories of occurrence, severity, and distress of the six most prevalent symptoms reported by patients undergoing RT for prostate cancer and to evaluate the effects of selected demographic and clinical characteristics on these trajectories.
Methods
Patients completed the Memorial Symptom Assessment Scale eleven times before, during, and after RT. For problems with urination, pain, lack of energy, feeling drowsy, difficulty sleeping, and diarrhea, the trajectories of occurrence, severity, and distress were evaluated using multilevel generalized linear models.
Results
Across all three dimensions, pain, lack of energy, feeling drowsy, and difficulty sleeping followed a decreasing linear trend. Problems with urination and diarrhea demonstrated more complex patterns of change over time.
Conclusion
Although longitudinal data on pain, lack of energy, feeling drowsy, and difficulty sleeping are limited, they are highly prevalent symptoms in these patients. In addition, diarrhea becomes a significant problem for these patients over the course of RT. A number of demographic and clinical characteristics affects the trajectories of these common symptoms differentially.
Keywords: Prostate cancer, radiation therapy, piecewise modeling, symptom trajectories, symptom predictors, occurrence, severity, distress, pain
Prostate cancer accounts for one in four new cancers diagnosed in men annually in the U.S., with an estimated 240,890 new diagnoses anticipated for 2011 alone (1). Treatments for prostate cancer included surgery, hormonal therapy, and radiation therapy (RT). Most of the longitudinal studies of side effects have focused on an evaluation of differences in urinary, bowel and sexual functioning among the various treatments for prostate cancer (27). However, no longitudinal studies have compared multiple symptom trajectories in the same sample of patients who underwent RT for prostate cancer or attempted to determine which symptoms are the most common as well as the most severe and distressing.
Changes in Bowel, Bladder, and Sexual Symptom Occurrence Rates Across Prostate Cancer Treatments
In studies that compared RT with radical prostatectomy (RP), patients who underwent RT had more bowel and bladder problems at the initiation of RT than patients who underwent radical prostatectomy (RP) (2, 5, 6). However, men who underwent RP had a sharp increase in urinary problems that decreased over time (36). In contrast, the occurrence of urinary problems after RT increased (2, 4, 5) or decreased then increased (6) over time. Additional work found that a higher percentage of patients who had a RP reported more urinary incontinence issues than those who had RT. However, a higher percentage of RT patients reported more obstructive and irritative symptoms (5). In terms of bowel dysfunction, higher occurrence rates were noted following RT compared with RP (35). Both treatments resulted in decreases in sexual function. However, patients who underwent RP reported a more precipitous decline in sexual function (3, 4, 6, 7).
Changes in Pain and Fatigue Occurrence and Severity Rates During and Following RT for Prostate Cancer
Changes in the occurrence and/or severity of pain and fatigue in prostate cancer patients during and after RT were reported in only five studies (812). In one study (9), pain severity did not change over the course of RT. In contrast, Lips et al. (10) reported that changes in pain intensity after one month of RT was dependent on the type of RT received. In terms of fatigue, in two studies (8, 9), fatigue increased over the course of RT but returned to baseline shortly after completion of treatment. In contrast, Monga et al. (12) found that fatigue scores remained significantly elevated for 12 months or more after RT. Of note, Miaskowski et al. (11), using hierarchical linear modeling, reported marked individual variability in fatigue severity during and after RT for prostate cancer.
Changes in Symptom Distress During and Following RT for Prostate Cancer
Most of the studies of symptom trajectories in patients with prostate cancer have evaluated the dimensions of occurrence or severity. However, an important dimension of the symptom experience is distress (13). To date, only six studies have reported on the trajectories of distress over the course of RT in patients with prostate cancer.(3, 4, 7, 1416) All of these studies evaluated distress associated with urinary, bowel, and/or sexual symptoms. The findings across these studies are inconsistent and warrant additional investigation.
In a longitudinal study of patients with localized prostate cancer (15), patients’ ratings of distress associated with urinary, bowel, and bladder dysfunction decreased over 48 months. In another study that compared patients who underwent RT with health controls (14), no between group differences in urinary bother were found at 15 years post-treatment. In a third study (7), no significant changes in urinary or sexual bother were found from baseline to 12 months after treatment. However, Krahn et al. (3) found that urinary and bowel bother increased significantly two months after treatment and remained elevated at one year post-treatment. In contrast, Namiki et al. (16) reported an increase in urinary bother at one month post RT that had returned to baseline levels at two months. Although not a direct measure of symptom distress, it is interesting that Yoshimura et al. (6) found no significant change in “overall satisfaction” with sexual function from baseline to two years after RT despite a significant decrease in sexual function. However, after two years, a significant decrease in overall satisfaction, as well as function, were reported. Of note, no studies evaluated distress associated with other symptoms in patients who underwent RT for prostate cancer.
Predictors of the Trajectories of Symptoms During and Following RT for Prostate Cancer
An evaluation of potential predictors of the trajectories of symptoms in prostate cancer patients undergoing RT is a way to identify patients who are at risk for poorer outcomes. For example, in one study (17), functional status prior to treatment was associated with better quality of life (QOL) after surgery or RT. In another study (8), a higher number of concomitant diseases and having no children independently predicted lower QOL two years after treatment.
Only five additional studies evaluated predictors of specific symptoms in patients with prostate cancer who underwent RT (11, 15, 16, 18, 19). In one study (11), younger age and higher levels of fatigue and depression prior to RT predicted higher levels of morning and/or evening fatigue during and after RT. In another study (18), poorer coping mechanisms were associated with higher levels of sleep disturbance over the course of RT.
In a study that compared U.S. and Japanese men who underwent RT for prostate cancer (16), time since treatment and baseline urinary function and bother scores, but not nationality, were independently associated with improved urinary bother scores at 24 months post-treatment. However, nationality was predictive of the slope of recovery of urinary function and bother in these patients. The trajectory was flat in the Japanese cohort and worsened somewhat in the U.S. cohort. Similarly, Gore et al. (15) found that worse urinary, bowel, and sexual function scores through 48 months post-treatment for prostate cancer were predictive of severe bother in these domains. In addition, time since treatment was inversely associated with urinary and bowel bother early in treatment but this relationship was no longer noted 48 months post-treatment. In a study of how neoadjuvant hormonal therapy affected bowel symptoms over the course of RT (19), men who received hormonal therapy had less rectal pain and tenesmus but more rectal mucus compared with those who did not. To our knowledge, no study has evaluated the effects of multiple predictors on changes in the multiple dimensions of the symptom experience (i.e., occurrence, severity, distress) during and after RT for prostate cancer.
Although numerous studies have focused on changes in the occurrence and severity of urinary, bowel and sexual symptoms during and following RT for prostate cancer, less is known about changes over time in other common symptoms experienced by these patients (e.g., pain, fatigue, sleep disturbance). In addition, no studies were found that evaluated for changes over time in multiple dimensions of the symptom experience (i.e., occurrence, severity, and distress) during and following RT in the same sample of patients with prostate cancer. Finally, limited information is available on the influence of a number of demographic and clinical characteristics on initial levels as well as the trajectories of common symptoms in these patients. Therefore, the purposes of this study were to evaluate the trajectories of occurrence, severity, and distress for the six most prevalent symptoms reported by prostate cancer patients undergoing RT and to evaluate the effects of selected demographic (i.e., ethnicity) and clinical (i.e., Karnofsky Performance Status (KPS) score, use of hormonal therapy prior to RT, presence of pain at the initiation of RT) characteristics on these trajectories.
Patients and Settings
This descriptive, longitudinal study is part of a larger study that evaluated the trajectories of multiple symptoms over the course of RT in outpatients with prostate cancer (n=82). Patients were recruited from two RT departments located in a comprehensive cancer center and a community-based oncology program at the time of the patient’s simulation visit.
Patients were eligible to participate if they were 18 years of age or older; were scheduled to receive primary or adjuvant RT for prostate cancer; were able to read, write, and understand English; gave written informed consent; and had a KPS score of ≥ 60. Patients were excluded if they had metastatic disease, more than one cancer diagnosis, or a diagnosed sleep disorder.
Instruments
Patients completed a demographic questionnaire, the KPS scale and the Memorial Symptom Assessment Scale (MSAS). The demographic questionnaire provided information about age, marital status, education, ethnicity, employment status, and the presence of a number of co-morbid conditions. The KPS is commonly used in the assessment of the functional status of cancer patients and has well-established validity and reliability (20).
The MSAS is a self-report questionnaire designed to measure the multidimensional experience of symptoms and has well-established reliability and validity (21, 22). It consists of a list of 32 physical and psychological symptoms that occur as a result of cancer or cancer treatment. Patients were asked to use the MSAS to systematically report whether or not they had experienced each of the symptoms in the past week (i.e., symptom occurrence). If the symptom was present, they were asked to rate its frequency, severity, and associated distress. Symptom frequency was measured using a four-point Likert scale (1=rarely, 2=occasionally, 3=frequently, 4=almost constantly). Symptom severity was measured using a four-point Likert scale (1=slight, 2=moderate, 3=severe, 4=very severe). Symptom distress was measured using a five-point Likert scale (0=not at all, 1=a little bit, 2=somewhat, 3=quite a bit, 4=very much).
Study Procedures
The study was approved by the Committee on Human Research at the University of California, San Francisco and at the second site. At the time of the simulation visit (approximately one week prior to the initiation of RT), patients were approached by a research nurse to discuss participation in the study. After obtaining written informed consent (at the baseline assessment), they were asked to complete the demographic questionnaire, the KPS, and the MSAS. Additional MSAS assessments were done every other week during the course of RT, every two weeks for two months, and once a month for two months following RT. Most patients completed 11 assessments. At the beginning and end of the study, medical records were reviewed for disease and treatment information.
Data Analysis
Data were analyzed using SPSS version 18.0 (SPSS Inc., Chicago, IL and Stata version 11.1 (StataCorp LP, College Station, TX). Descriptive statistics were used to characterize the sample and to determine the occurrence rates for the six most prevalent symptoms. As shown in Table 1, the occurrence rates for the 32 MSAS symptoms were evaluated at the middle, end, and one month after the completion of RT. The six symptoms that occurred in 40% or more of the patients at one or more time points (i.e., problems with urination, pain, lack of energy, feeling drowsy, difficulty sleeping, diarrhea) were evaluated in the subsequent longitudinal analyses.
Table 1
Table 1
Occurrence Rates for Symptoms at the Middle, End, and One Month After the Completion of Radiation Therapy
Using multilevel generalized linear models, these six symptoms were evaluated for changes over time in symptom occurrence, severity, and distress (a total of 11 assessments over six months). Symptom occurrence was coded as a binary variable (yes=1, no=0) and examined with multilevel logistic regression. Symptom severity items were coded as ordinal with 0=not present and with increasing severity reported as one to four. Symptom distress was coded as ordinal with 0=not present, 1=present but causing no distress, and with increasing distress reported as two to five. Therefore, changes in symptom severity and distress were examined with multilevel proportional odds ordinal logistic regression (also called cumulative odds logistic regression) (2327).
For both types of models, random intercepts were estimated, with the first assessment being treated as the baseline (or intercept) for the growth trajectory. Unconditional models were examined first to estimate the linear change in the symptom reports. Given the possibility that the growth trajectory might not be only linear, quadratic effects were examined. Further, the treatment period and a lengthy follow-up invited the examination of shifts (also called “discontinuities” (28)) in the growth trajectories. Therefore, piecewise models were examined. Three growth periods were examined with piecewise models: baseline to six weeks (P1), six weeks to 17 weeks (two months after RT; P2); and 17 weeks to 25 weeks (four months after RT; P3).
After identifying the best fitting growth trajectory for each symptom, conditional models were fit to examine the associations for each of four covariates (i.e., ethnicity [white versus non-white], KPS score [higher versus lower functional status scores], use of hormonal therapy prior to RT, and presence of pain [no pain versus pain] at the initiation of RT) on the reported symptoms at baseline and on the change in symptoms over time (cross-level interaction). These covariates were examined separately for each symptom.
The multilevel logistic regression models were estimated with Stata release 11.1 (program xtlogit) using mean and variance adaptive Gauss-Hermite quadrature with 12 integration points. The multilevel ordinal regression models were fit in Stata with gllamm (27), a program written for Stata. For these models, estimation was carried out with an ologit link and using adaptive Gauss-Hermite quadrature with 15 integration points. For both types of multilevel generalized linear models, estimation with adaptive quadrature was shown to be better than pseudo-likelihood estimation (27), as used in SPSS version 19, HLM version 6, and other programs. For all analyses, P-values of < 0.05 were considered statistically significant.
Demographic and Clinical Characteristics of the Sample
The demographic and clinical characteristics of the 82 patients are presented in Table 2. These men with prostate cancer were approximately 67 years of age, well educated, and had a KPS score of 95.7. Most of the patients were married or partnered (71.9%), white (76.8%), and not employed (54.4%). The distribution of clinical stage was 48.8% with T1, 42.5% with T2, and 8.8% with T3. Over 50% of the patients had received hormonal therapy prior to the initiation of RT and 47.6% reported pain at the initiation of RT.
Table 2
Table 2
Demographics, Disease, and Treatment Characteristics of Patients with Prostate Cancer (n=82)
Problems with Urination
The likelihood of reporting problems with urination increased from baseline to six weeks (P1), decreased from six to 17 weeks (P2), and then remained stable from 17 to 25 weeks (P3; Table 3, Fig. 1A). KPS score and the presence of pain at the initiation of RT influenced the occurrence of this symptom at baseline. For each 10-point decrease in KPS score, the odds of reporting problems with urination at the initiation of RT were 2.9 times greater. Patients who reported pain were 14.8 times more likely to report problems with urination at the initiation of RT. In addition, cross-level interactions were found between both ethnicity and pain and changes over time in the odds of reporting problems with urination (Table 3). As shown in Figs. 2A and B, the probability of reporting problems with urination increased more for non-whites than whites from baseline to six weeks (P1). As shown in Figs. 2C and D, given that patients with pain started with a higher likelihood of reporting problems with urination at baseline compared with patients without pain, the probability of reporting this symptom increased at a lesser rate in the patients with pain from baseline to six weeks (P1), then decreased at a lesser rate from six to 17 weeks (P2).
Table 3
Table 3
Results of the Multilevel Regression Analyses of Occurrence Ratings for Six Symptoms Reported by Patients With Prostate Cancer Over 25 Weeks
Fig. 1
Fig. 1
Observed (open circles) and predicted (filled circles) trajectories for the probability of occurrence of problems with urination (A), pain (B), lack of energy (C), feeling drowsy (D), difficulty sleeping (E), and diarrhea (F) across the 25 weeks of the (more ...)
Fig. 2
Fig. 2
Observed (open circles) and predicted (filled circles) trajectories for the probability of occurrence of problems with urination, across the 25 weeks of the study, in whites (A) compared to non-whites (B) and in patients who did (D) and did not (C) report (more ...)
Changes in overall severity ratings for problems with urination were nonlinear, first increasing sharply from baseline to six weeks, then decreasing sharply from six to 17 weeks, then increasing at a lower rate from 17 to 25 weeks (Table 4, Fig. 3A). KPS score influenced the severity rating of this symptom at baseline. For each 10-point decrease in KPS score, the odds of reporting a higher severity rating for problems with urination at the initiation of RT were 2.7 times greater. Presence of pain at the initiation of RT was the only predictor that influenced both the severity of this symptom at baseline as well as changes over time in the severity of this symptom. As shown in Table 4 and Fig. 4A and B, compared to patients without pain, patients with pain reported more severe problems with urination at baseline. However, the severity of their problems with urination increased at a lesser rate to six weeks, and then decreased less than those with no pain from six to 17, and from 17 to 25 weeks. Differences between the two pain groups were not significant for the second and third “pieces” of the model. As can be seen in Table 4, the odds of reporting more severe problems with urination at baseline were 6.2 times greater for the patients with pain. Given that they started with more severe problems with urination at baseline, the odds of reporting more severe problems for each additional week were 27% less over the first “piece” of their trajectory for the patients with pain ((odds ratio [OR]–1)*100 = (0.73 -1)*100 = -27%).
Table 4
Table 4
Results of the Multilevel Regression Analyses of Severity Ratings for Six Symptoms Reported by Patients With Prostate Cancer Over 25 Weeks
Fig. 3
Fig. 3
Observed (open symbols) and predicted (filled symbols) trajectories for the probability of severity ratings for problems with urination (A), pain (B), lack of energy (C), feeling drowsy (D), difficulty sleeping (E), and diarrhea (F) across the 25 weeks (more ...)
Fig. 4
Fig. 4
Observed (open symbols) and predicted (filled symbols) trajectories for the probability of severity ratings for problems with urination, across the 25 weeks of the study, in patients who did (B) and did not (A) report pain at the initiation of radiation (more ...)
Changes in overall distress ratings for problems with urination were nonlinear, first increasing sharply from baseline to six weeks, then decreasing sharply from six to 17 weeks, then remaining stable from 17 to 25 weeks (Table 5, Fig. 5A). KPS score was the only predictor that influenced the distress rating for this symptom at baseline. For each 10-point decrease in KPS score, the odds of reporting a higher distress rating for problems with urination at the initiation of RT were 2.7 times greater. Presence of pain influenced changes over time in the distress rating for this symptom. As shown in Table 5 and Figs. 6A and B, patients with pain reported that their distress associated with problems with urination increased at a lesser rate from baseline to six weeks, and then decreased less than those with no pain from six to 17, and from 17 to 25 weeks. However, the differences between the two pain groups were not significant for the second and third “pieces” of the model. As can be seen in Table 5, given that patients with pain started with more distress associated with problems with urination at baseline, the odds of reporting higher levels of distress for each additional week were 21% less over the first “piece” for the patients with pain.
Table 5
Table 5
Results of the Multilevel Regression Analyses of Distress Ratings for Six Symptoms Reported by Patients With Prostate Cancer Over 25 Weeks
Fig. 5
Fig. 5
Observed (open symbols) and predicted (filled symbols) trajectories for the probability of distress ratings for problems with urination (A), pain (B), lack of energy (C), feeling drowsy (D), difficulty sleeping (E), and diarrhea (F) across the 25 weeks (more ...)
Fig. 6
Fig. 6
Observed (open symbols) and predicted (filled symbols) trajectories for the probability of distress ratings for problems with urination, across the 25 weeks of the study, in patients who did (B) and did not (A) report pain at the initiation of radiation (more ...)
Pain
Pain occurrence rates decreased over time (Table 3, Fig. 1B). For each additional week, the odds of a patient reporting pain decreased by 4%. Ethnicity, KPS score, and hormonal treatment prior to RT had no effect on the occurrence of pain at baseline or changes in its occurrence over time (Table 3). Severity (Fig. 3B) and distress (Fig. 5B) ratings for pain decreased over time. For each additional week, the odds of a patient reporting a higher pain severity (Table 4) or distress (Table 5) score decreased by 5% for both dimensions. None of the predictors had an effect on severity or distress scores.
Lack of Energy
Occurrence rates for lack of energy decreased over time (Table 3, Fig. 1C). For each additional week, the odds of a patient reporting lack of energy decreased by 6%. Ethnicity, KPS score, and hormonal treatment prior to RT had no effect on the occurrence of lack of energy at baseline (Table 3). However, patients who reported pain were 6.9 times more likely to report lack of energy at the initiation of RT. In addition, cross-level interactions were found between both hormonal treatment and pain prior to RT and changes over time in the odds of reporting lack of energy. As shown in Table 3 and Figs. 7A and B, for each additional week, the odds of reporting lack of energy increased by 6% in patients who had hormonal treatment prior to RT. As shown in Table 3 and Figs. 7C and D, for each additional week, the odds of reporting lack of energy decreased by 9% in patients who reported pain.
Fig. 7
Fig. 7
Observed (open circles) and predicted (filled circles) trajectories for the probability of occurrence of lack of energy, across the 25 weeks of the study, in patients who were (B) and were not (A) on hormonal therapy and in patients who did (D) and did (more ...)
Changes in the overall severity ratings for lack of energy followed a decreasing linear trajectory over the 25 weeks, with an approximate 6% decrease in the odds of reporting higher severity ratings for lack of energy with each additional week (Table 4, Fig. 3C). KPS score was the only predictor that influenced the severity rating for this symptom at baseline. For each 10- point decrease in KPS score, the odds of reporting a higher severity rating for lack of energy at the initiation of RT were 2.7 times greater. Presence of pain influenced changes over time in the severity rating for this symptom. As shown in Table 4 and Figs. 8A and B, patients with pain started higher and reported a sharper decrease (7%) in the odds of reporting higher severity ratings for lack of energy, compared with patients without pain.
Fig. 8
Fig. 8
Observed (open symbols) and predicted (filled symbols) trajectories for the probability of severity ratings for lack of energy, across the 25 weeks of the study, in patients who did (B) and did not (A) report pain at the initiation of radiation therapy. (more ...)
Changes in the overall distress ratings for lack of energy followed a decreasing linear trajectory over the 25 weeks, with an approximate 4% decrease in the odds of reporting higher distress ratings for lack of energy with each additional week (Table 5, Fig. 5C). KPS score was the only predictor that influenced the distress rating for this symptom at baseline. For each 10- point decrease in KPS score, the odds of reporting a higher distress rating for lack of energy at the initiation of RT were 2.7 times greater. Hormonal treatment and presence of pain influenced changes over time in the distress ratings for this symptom. As shown in Table 5 and Figs. 9A and B, for each additional week, the probability of reporting a higher distress score for lack of energy increased by 5% in patients who had hormonal treatment prior to RT. As shown in Table 5 and Figs. 9C and D, patients with pain started higher and reported a sharper decrease (7%) in the odds of reporting higher distress ratings for lack of energy, compared with patients without pain.
Fig. 9
Fig. 9
Observed (open symbols) and predicted (filled symbols) trajectories for the probability of distress ratings for lack of energy, across the 25 weeks of the study, in patients who were (B) and were not (A) on hormonal therapy and in patients who did (D) (more ...)
Feeling Drowsy
Occurrence rates for feeling drowsy decreased over time (Table 3, Fig. 1D). For each additional week, the odds of a patient reporting feeling drowsy decreased by 5%. Only KPS score influenced the occurrence of this symptom at baseline. For each 10-point decrease in KPS score, the odds of reporting feeling drowsy at the initiation of RT were 3.1 times greater. Severity (Fig. 3D) and distress (Fig. 5D) ratings for feeling drowsy decreased over time. For each additional week, the odds of a patient reporting a higher severity (Table 4) or distress (Table 5) score for feeling drowsy decreased by 3% for both dimensions. KPS score was the only predictor associated with baseline severity and distress ratings for this symptom. For each 10-point decrease in KPS score, the odds of reporting higher severity and distress ratings for feeling drowsy at the initiation of RT were 3.4 and 3.5 times greater, respectively.
Difficulty Sleeping
Occurrence rates for difficulty sleeping decreased over time (Table 3, Fig. 1D). For each additional week, the odds of a patient reporting difficulty sleeping decreased by 5%. Only KPS score influenced the occurrence of this symptom at baseline. For each 10-point decrease in KPS score, the odds of reporting difficulty sleeping at the initiation of RT were 7.5 times greater.
Severity (Fig. 3E) and distress (Fig. 5E) ratings for difficulty sleeping decreased over time. For each additional week, the odds of a patient reporting a higher severity (Table 4) or distress (Table 5) score for difficulty sleeping decreased by 5% and 4%, respectively. KPS score was the only predictor associated with baseline severity and distress ratings for this symptom. For each 10-point decrease in KPS score, the odds of reporting higher severity and distress ratings for difficulty sleeping at the initiation of RT were 7.5 and 7.0 times greater, respectively.
Diarrhea
The likelihood of reporting diarrhea increased from baseline to six weeks (P1), decreased from six to 17 weeks (P2), and then increased from 17 to 25 weeks (P3; Table 3, Fig. 1F). KPS score influenced the occurrence of this symptom at baseline. For each 10-point decrease in KPS score, the odds of reporting diarrhea at the initiation of RT were 3.1 times greater.
Changes in severity (Table 4, Fig. 3F) and distress ratings (Table 5, Fig. 5F) for diarrhea were nonlinear, first increasing sharply from baseline to six weeks, then decreasing sharply from six to 17 weeks, then increasing at a lower rate from 17 to 25 weeks. Although significant for distress, the slight increase in severity from 17 to 25 weeks was not statistically significant. KPS score was the only predictor associated with baseline severity and distress ratings for this symptom. For each 10-point decrease in KPS score, the odds of reporting higher severity and distress ratings for diarrhea at the initiation of RT were 3.0 and 2.9 times greater, respectively.
This study is the first to examine changes in multiple dimensions of the symptom experience (i.e., occurrence, severity, and distress) for six of the most common symptoms reported by patients prior to, during, and following RT for prostate cancer. Whereas genitourinary and gastrointestinal symptoms were reported by 40% or more of these patients, four additional symptoms (pain, lack of energy, feeling drowsy, difficulty sleeping) emerged as significant clinical problems for these men with prostate cancer that are highly prevalent and are not well studied. Although fatigue is known to be a significant symptom during RT (2931), the potentially associated symptoms of feeling drowsy and difficulty sleeping are equally common, severe, and distressing to patients with prostate cancer during and after RT. These two symptoms may be more amenable to targeted interventions than fatigue and may reduce some of the fatigue reported by these patients. The finding of a relatively high prevalence of pain in these patients early in the course RT is important. Although the cause of the pain was not identified, the fact that nearly half of these patients reported its presence highlights the need for a thorough assessment and ongoing treatment of pain throughout the course of RT.
An evaluation of the trajectories for occurrence, severity, and distress for all six of the most common symptoms demonstrates that patient’s ratings across all three dimensions followed similar trends. Problems with urination and diarrhea were the only exceptions. Between weeks 17 and 25, occurrence and distress ratings related to problems with urination remained constant whereas severity ratings increased. The general trend for problems with urination (that is, an increase after treatment initiation with a subsequent decline) mirrors findings from previous studies (2, 4, 5). The subtle finding, that distress related to problems with urination is correlated with occurrence rather than severity suggests that continued emotional support, even after treatment ends, is warranted, at least as long as symptoms persist.
Consistent with previous reports (3, 10), diarrhea increased in frequency during treatment and subsequently decreased, with similar trajectories for ratings of severity and distress. However, the use of piecewise modeling revealed that this particular cohort reported a delayed lesser yet significant increase in the symptom, between 17 and 25 weeks after treatment. The reason for this increase in diarrhea is unclear. One explanation may be that patients received more frequent follow-up and better symptom management for up to 17 weeks after treatment. An equally plausible explanation is that patients may have adhered to special diets until roughly that time. Generally speaking, it is proposed that as time since treatment lengthens and usual lifestyles resume, chronic side effects become more apparent. This new finding, that all three dimensions of this symptom had a delayed significant increase, is important because it changes the expected trajectory of treatment-related diarrhea and provides evidence that patients need more long-term symptom management support.
For all three dimensions, the trajectories of the other four symptoms studied (pain, lack of energy, feeling drowsy, difficulty sleeping) followed a similar pattern. Occurrence, severity, and distress were highest at baseline and decreased by roughly 5% with each additional week. No other study has reported a steady downward pain trajectory. Results of previous studies suggest that pain remains stable, increases, or varies depending on the type of RT received (3, 9, 10). The reason that pain decreased over time in this study is unclear and warrants additional investigation.
Despite the fact that fatigue is known to be one of the most common and significant symptoms associated with RT, in this study, for all of the symptom dimensions the trajectories of lack of energy, feeling drowsy, and difficulty sleeping steadily improved over the course of treatment. Thomas et al. (18) found the equally puzzling result that despite reporting fewer hours of sleep through six months after treatment, all other measures of sleep (i.e., trouble falling asleep, sleep adequacy, sleep latency) improved or remained constant over time. Taken together, these findings suggest the need for additional research on these symptoms.
Consistent with the findings for problems with urination and diarrhea in this study, in the three studies that evaluated the impact of functional status on urinary, bowel, and sexual function (5, 15, 17), lower functional status was associated with worse symptom trajectories. No studies were found that evaluated the impact of functional status on other RT-related symptoms. In this study, the odds of reporting problems with urination, diarrhea, feeling drowsy, and difficulty sleeping at the beginning of treatment steadily increased as KPS score decreased. Interestingly, no association was found between the occurrence of pretreatment lack of energy and KPS score. Of note, baseline levels of symptom severity and distress had an inverse relationship with KPS score for all symptoms except pain. It is not entirely clear why KPS score affected only the baseline levels of these symptoms and not the trajectories of the various symptoms.
Ethnicity was found to have a significant interaction with only the occurrence of problems with urination between baseline and six weeks. Nonwhite patients were more likely than white patients to report an increase in problems with urination during this time. The reason for this finding is not entirely clear and warrants investigation in future studies.
It is interesting that none of the predictors evaluated affected any dimension of pain at baseline or the trajectories of pain. In contrast, pretreatment pain was predictive of occurrence of lack of energy, and occurrence and severity of problems with urination at baseline. Those patients with pretreatment pain experienced more rapid improvement in energy level over time. In addition, those patients with pretreatment pain had a less marked peak in problems with urination during treatment. These findings suggest that pain is not necessarily related to performance status and that better pain management, especially at the onset of treatment, may have a positive effect on symptom trajectories. Whereas Blesch et al. (32) found that pain was highly correlated with fatigue in cancer patients, the relationship between pain and other symptoms in prostate cancer patients undergoing RT is virtually unexplored and warrants further investigation.
Consistent with previous reports (8, 19), no interactions were found between hormonal treatment prior to RT and problems with urination or diarrhea. However, patients who received hormonal treatment prior to RT were more likely to report the occurrence of and distress related to lack of energy over time than those who had not received hormonal treatment. This information can be used to educate patients about risk factors for fatigue. In addition, it is interesting that distress level followed the same trajectory as occurrence rather than severity. This finding suggests that any complaint of lack of energy should be taken seriously, regardless of severity.
It is notable that patients in this study did not report their symptoms to be more than moderately severe. Furthermore, for all time points, with the exception of problems with urination, the greatest likelihood that patients would report even slight symptom severity or mild symptom distress was only approximately 50% or 40%, respectively. These findings suggest, that during and after RT, clinicians must listen carefully for reports of difficulty with symptom management because patients with prostate cancer are not likely to rate their symptoms as severe or very distressing.
Many studies have explored the effects of RT on sexual function in patients with prostate cancer. In this study, problems with sexual interest or activity were not found to be one of the most common symptoms. One explanation for this finding is that perhaps not all patients were sexually active. This finding is important because it suggests that a greater number of patients would benefit from future research on symptoms that are more common and not as well characterized (e.g., pain, feeling irritable).
Limitations of this study include the relatively small sample size and high proportion of white patients. Future studies that include more ethnically diverse samples are needed to further explore the relationship between ethnicity and symptom trajectories. Because patients in this study had an initial KPS score of ≥ 60 and no metastatic disease, its results cannot be generalized to prostate cancer patients with lower functional status or more extensive disease. Although the trajectories of multiple dimensions of pain were evaluated, as well as how its presence at the initiation of RT affected the trajectories of other common symptoms, the causes of pain were not assessed. More detailed information on the causes and nature of pain in this population would be useful in the application of these findings to the clinical setting.
In summary, this study is the first to report on which six symptoms are the most common in men with prostate cancer who underwent RT. In addition, it is the first to simultaneously examine the trajectories of occurrence, severity, and distress for these symptoms in the same sample of patients. Through an evaluation of how a number of predictors influence these trajectories, we determined that lower KPS score at the initiation of RT leaves one more likely to report, with increased severity and distress, problems with urination, diarrhea, feeling drowsy, and difficulty sleeping at that time. It is remarkable that no predictors for pain were identified, including KPS score. The presence of pretreatment pain was the only predictor of the presence of lack of energy before treatment. This finding suggests that better pain control, regardless of treatment phase, may lead to improvements in fatigue. We were only able to examine the trajectories of the six most frequent symptoms in this population but there are a number of other common symptoms that warrant further attention. For example, it is notable that roughly one- third of the men in this study reported feeling irritable, having difficulty concentrating, and sweats. These symptoms are prime areas for future patient-centered research, and results of this study might be used by others to further prioritize efforts.
Acknowledgments
This research was supported by a grant from the National Institute of Nursing Research (NR04835). Dr. Aouizerat was funded through the National Institutes of Health (NIH) Roadmap for Medical Research Grant (KL2 RR624130). Dr. Miaskowski is funded by the American Cancer Society as a Clinical Research Professor. Dr. Dhruva is funded through NIH Mentored Patient-Oriented Research Career Development Award (K23 AT005340).
Footnotes
Disclosures
The authors declare no conflicts of interest.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1. Siegel R, Ward E, Brawley O, Jemal A. Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. CA Cancer J Clin. 2011;61:212–236. [PubMed]
2. Eller LS, Lev EL, Gejerman G, et al. Prospective study of quality of life of patients receiving treatment for prostate cancer. Nurs Res. 2006;55:S28–S36. [PubMed]
3. Krahn MD, Bremner KE, Tomlinson G, Naglie G. Utility and health-related quality of life in prostate cancer patients 12 months after radical prostatectomy or radiation therapy. Prostate Cancer Prostatic Dis. 2009;12:361–368. [PubMed]
4. Litwin MS, Gore JL, Kwan L, et al. Quality of life after surgery, external beam irradiation, or brachytherapy for early-stage prostate cancer. Cancer. 2007;109:2239–2247. [PubMed]
5. Talcott JA, Manola J, Clark JA, et al. Time course and predictors of symptoms after primary prostate cancer therapy. J Clin Oncol. 2003;21:3979–3986. [PubMed]
6. Yoshimura K, Arai Y, Ichioka K, et al. A 3-y prospective study of health-related and disease-specific quality of life in patients with nonmetastatic prostate cancer treated with radical prostatectomy or external beam radiotherapy. Prostate Cancer Prostatic Dis. 2004;7:144–151. [PubMed]
7. Schapira MM, Lawrence WF, Katz DA, McAuliffe TL, Nattinger AB. Effect of treatment on quality of life among men with clinically localized prostate cancer. Med Care. 2001;39:243–253. [PubMed]
8. Geinitz H, Thamm R, Scholz C, et al. Longitudinal analysis of quality of life in patients receiving conformal radiation therapy for prostate cancer. Strahlenther Onkol. 2010;186:46–52. [PubMed]
9. Janda M, Gerstner N, Obermair A, et al. Quality of life changes during conformal radiation therapy for prostate carcinoma. Cancer. 2000;89:1322–1328. [PubMed]
10. Lips I, Dehnad H, Kruger AB, et al. Health-related quality of life in patients with locally advanced prostate cancer after 76 Gy intensity-modulated radiotherapy vs. 70 Gy conformal radiotherapy in a prospective and longitudinal study. Int J Radiat Oncol Biol Phys. 2007;69:656–661. [PubMed]
11. Miaskowski C, Paul SM, Cooper BA, et al. Trajectories of fatigue in men with prostate cancer before, during, and after radiation therapy. J Pain Symptom Manage. 2008;35:632–643. [PMC free article] [PubMed]
12. Monga U, Kerrigan AJ, Thornby J, et al. Longitudinal study of quality of life in patients with localized prostate cancer undergoing radiotherapy. J Rehabil Res Dev. 2005;42:391–399. [PubMed]
13. Jacobsen PB, Donovan KA, Trask PC, et al. Screening for psychologic distress in ambulatory cancer patients. Cancer. 2005;103:1494–1502. [PubMed]
14. Fransson P. Patient-reported lower urinary tract symptoms, urinary incontinence, and quality of life after external beam radiotherapy for localized prostate cancer—15 years’ follow-up. A comparison with age-matched controls. Acta Oncol. 2008;47:852–861. [PubMed]
15. Gore JL, Gollapudi K, Bergman J, et al. Correlates of bother following treatment for clinically localized prostate cancer. J Urol. 2010;184:1309–1315. [PubMed]
16. Namiki S, Kwan L, Kagawa-Singer M, et al. Urinary quality of life after prostatectomy or radiation for localized prostate cancer: a prospective longitudinal cross-cultural study between Japanese and U.S. men. Urology. 2008;71:1103–1108. [PubMed]
17. Chen RC, Clark JA, Talcott JA. Individualizing quality-of-life outcomes reporting: how localized prostate cancer treatments affect patients with different levels of baseline urinary, bowel, and sexual function. J Clin Oncol. 2009;27:3916–3922. [PubMed]
18. Thomas KS, Bower J, Hoyt MA, Sepah S. Disrupted sleep in breast and prostate cancer patients undergoing radiation therapy: the role of coping processes. Psychooncology. 2010;19:767–776. [PMC free article] [PubMed]
19. Tsai HK, Manola J, Abner A, et al. Patient-reported acute gastrointestinal toxicity in men receiving 3-dimensional conformal radiation therapy for prostate cancer with or without neoadjuvant androgen suppression therapy. Urol Oncol. 2005;23:230–237. [PubMed]
20. Karnofsky D, Abelmann WH, Craver LV, Burchenal JH. The use of nitrogen mustards in the palliative treatment of carcinoma. Cancer. 1948;1:634–656.
21. Portenoy RK, Thaler HT, Kornblith AB, et al. Symptom prevalence, characteristics and distress in a cancer population. Qual Life Res. 1994;3:183–189. [PubMed]
22. Portenoy RK, Thaler HT, Kornblith AB, et al. The Memorial Symptom Assessment Scale: an instrument for the evaluation of symptom prevalence, characteristics and distress. Eur J Cancer. 1994;30A:1326–1336. [PubMed]
23. Ananth CV, Kleinbaum DG. Regression models for ordinal responses: a review of methods and applications. Int J Epidemiol. 1997;26:1323–1333. [PubMed]
24. Lall R, Campbell MJ, Walters SJ, Morgan K. A review of ordinal regression models applied on health-related quality of life assessments. Stat Methods Med Res. 2002;11:49–67. [PubMed]
25. Long JS, Freese J. Regression models for categorical dependent variables using Stata. 2. College Station, TX: Stata Press; 2006.
26. O’Connell AA. Logistic regression models for ordinal response variables. Thousand Oaks, CA: Sage Publications; 2005.
27. Rabe-Hesketh S, Skrondal A. Multilevel and longitudinal modeling using Stata. 2. College Station, TX: Stata Press; 2008.
28. Singer JD, Willett JB. Applied longitudinal data analysis: Modeling change and event occurrence. New York: Oxford University Press; 2003.
29. Jereczek-Fossa BA, Marsiglia HR, Orecchia R. Radiotherapy-related fatigue. Crit Rev Oncol Hematol. 2002;41:317–325. [PubMed]
30. Hickok JT, Roscoe JA, Morrow GR, et al. Frequency, severity, clinical course, and correlates of fatigue in 372 patients during 5 weeks of radiotherapy for cancer. Cancer. 2005;104:1772–1778. [PubMed]
31. Lawrence DP, Kupelnick B, Miller K, Devine D, Lau J. Evidence report on the occurrence, assessment, and treatment of fatigue in cancer patients. J Natl Cancer Inst Monogr. 2004;32:40–50. [PubMed]
32. Blesch KS, Paice JA, Wickham R, et al. Correlates of fatigue in people with breast or lung cancer. Oncol Nurs Forum. 1991;18:81–87. [PubMed]