Abstract
Purpose:
We aim to compare patient-reported mental health outcomes for men undergoing treatment for localized prostate cancer, longitudinally over 5 years.
Materials & Methods:
We conducted a prospective population-based analysis using the Comparative Effectiveness Analysis of Surgery and Radiation study (CEASAR) study. Patient-reported depressive symptoms (Centers for Epidemiologic Studies Depression, CES-D) and domains of the Medical Outcomes Study 36-item short form survey evaluating emotional well-being and energy/fatigue, were assessed through 5 years after treatment with surgery, radiotherapy (with or without androgen deprivation therapy) and active surveillance. Regression models were adjusted for outcome-specific baseline function, demographic and clinicopathologic characteristics, and treatment approach.
Results:
2742 men (median [quartiles] age 64 [59–70]) met inclusion criteria. Baseline depressive symptoms, as measured by the CES-D, were low (median 4, quartiles 1–8) without differences between groups. We found no effect of treatment modality on depressive symptoms (p= 0.78), though older age, poorer health, being unmarried, and baseline CES-D score were associated with declines in mental health. There was no clinically meaningful association between treatment modality and scores for either emotional well-being (p=0.81) or energy/fatigue (p=0.054).
Conclusions:
This prospective, population-based cohort study of men with localized prostate cancer showed no clinically important differences in mental health outcomes including depressive symptoms, emotional well-being, and energy/fatigue according to the treatment received (surgery, radiotherapy, or surveillance). However, we identified a number of characteristics associated with worse mental health outcomes including: older age, poorer health, being unmarried, and baseline CES-D score which may allow for early identification of patients most at risk of these outcomes following treatment.
Keywords: prostate cancer, mental health, quality of life
Introduction
For patients diagnosed with localized prostate cancer, guideline-recommended treatment options include active surveillance, radical prostatectomy, and radiotherapy, based on risk-stratification. Absolute differences in cancer-specific and overall mortality between these approaches are small1,2, thus, treatment-related morbidity is paramount in treatment decision making. While health-related quality of life outcomes following prostate cancer treatments such as urinary symptoms, erectile dysfunction, and bowel symptoms are well reported3–6 and the burden of other interventions to manage treatment-related complications is increasingly recognized7–10, less attention has been paid to the association between treatment for localized prostate cancer and mental health outcomes.
While mental health outcomes of treatment have been poorly explored among patients with prostate cancer, one small (n=368) cross-sectional study of Black men with prostate cancer found a relatively high prevalence of major depressive symptoms (33%), with an increased likelihood among those who underwent radiotherapy (odds ratio 2.38, 95% confidence interval 1.02–5.51). However, these data are limited by the lack of generalizability, difficulties with causation with a cross-sectional study design, and the need for reproducibility11. In contrast, the association between treatment and mental health outcomes is better established in both breast cancer and colorectal cancer12,13, in which sexual dysfunction and body image concerns contribute to emotional distress and worsening psychosocial function over time.
In this context, we evaluated the association between prostate cancer treatment and patient-reported depression and emotional well-being over time applying previously validated Centers for Epidemiologic Studies Depression (CES-D) and Medical Outcomes Study 36–item Short Form survey (SF-36) scores to data from the prospective population-based Comparative Effectiveness Analysis of Surgery and Radiation study (CEASAR) study.
Materials & Methods
Study Population
The CEASAR study enrolled men with clinically localized prostate cancer (cT1-cT2, PSA<50ng/dL) from 5 population-based Surveillance, Epidemiology and End Results (SEER) Program registries and the observational Cancer of the Prostate Strategic Urologic Research Endeavor prostate cancer registry from 2011–2012, as previously described14,15. Institutional review board approval was obtained from Vanderbilt University Medical Center (coordinating center) and from each participating site.
Participants completed surveys at baseline, 6 months and 1-, 3-, and 5-years following enrollment (with the final survey being completed in September 2017). Data regarding tumor characteristics, treatment choice, and treatment dates were obtained from medical chart abstraction 1 year after enrollment14,15. Any treatment received after 1 year was based on patient report. Survival was determined from vital status follow-up data obtained from the SEER and Prostate Strategic Urologic Research Endeavor registries.
Exposure
The exposure of interest was the primary treatment modality, categorized as active surveillance, surgery, radiation with androgen deprivation therapy (ADT), and radiation without ADT. Treatment modality was determined primarily based on 1-year chart abstraction, supplemented by patient report.
Outcomes
Depressive symptoms were assessed by using a 10-question version of the previously validated Centers for Epidemiologic Studies Depression scale (CES-D).16,17 The CES-D10 has been validated across diverse populations18,19. To reduce respondent burden, the CES-D was modified to a nine question version. We adjusted the overall CES-D score to reflect this difference (from a standard score of 30 points to 27 points.) Domain scores ranged from 0–27, where 27 indicated more severe depressive symptoms. Notably, to our knowledge, a clinically meaningful difference on this scale has not been described though scores above 19 have strong specificity and positive predictive value for major depression.
The validated Medical Outcome Study 36-item Short Form Survey (SF-36) was used to evaluate emotional well-being and energy and fatigue20,21. For the SF-36, domain scores range from 0–100 with 100 indicating the best function. The results of these domain scores were analyzed over time for each treatment modality. We interpreted results based on previously determined minimally clinical important differences of 6 and 9 for emotional well-being and energy and fatigue, respectively22,23.
Statistical Analysis
Clinical and sociodemographic characteristics were evaluated by treatment modality, categorized as active surveillance, surgery, radiation with ADT, and radiation without ADT. Treatment group differences were assessed using Kruskal-Wallis tests and χ2 tests for continuous and categorical variables, respectively.
The study endpoints of CES-D score and SF-36 domain scores for emotional well-being and energy and vitality were reported as adjusted mean score differences (with 95% confidence intervals). To further evaluate the associations between treatments and measures of mental health over time, using the longitudinal survey data, we fit multivariable longitudinal linear regression models for CES-D, emotional well-being and energy and vitality adjusting for time since treatment (continuous, restricted cubic splines using 4 knots), age at diagnosis (continuous, restricted cubic splines using 3 knots), race (White, Black, Hispanic, Asian, other), education (less than high school, high school graduate, some college, college graduate, graduate/professional school), marital status (not married, married), comorbidity as measured with total burden index for prostate cancer - TIBI-CaP24 (categorical: 0–2, 3–4, 5 or more), income (less than $30,000, $30,001-$50,000, $50,001-$100,000, more than $100,000), insurance status (Medicare; private or HMO; and VA, military, Medicaid, other, or uninsured), D’Amico risk category (low, intermediate, high), site (Utah, Atlanta, Los Angeles, Louisiana, NJ, CaPSURE), baseline physical functioning (continuous, linear), baseline general health (continuous, linear), baseline social support (continuous, linear), baseline participatory decision-making scale (continuous, linear), baseline sexual function score (continuous, linear), time-varying sexual function scores (at 6 months, 1, 3, and 5 years, continuous, linear) and corresponding baseline value of the outcome (continuous, restricted cubic splines using 3 knots). Covariates were obtained from patient-reported surveys and chart abstraction, as appropriate.
Comparing between treatment modalities, we utilized active surveillance as the referent. To allow for variable estimation of treatment effect at different time points, we included the interaction terms between treatment and time since treatment in the models. In all models, to account for the correlation due to repeated measures obtained on the same subjects from multiple time points, the Huber-White method25,26 was implemented by robcov function in rms R package to estimate the variance-covariance matrices. The results of models were reported as mean differences between treatment groups and the associated 95% confidence intervals (CI). All missing covariate values were imputed 10 times using the MICE (multiple imputation using chained equations) implemented by aregImpute function in rms R package. The EPIC-26 sexual function score was missing at 4, 9, 10, 20, 28% at each time point, respectively. Income was missing in 11%; however, all other variables had less than 5% missing. To graphically represent the trends in CES-D and SF-36 scores, we fit simpler models that included time since treatment start and treatment modality, along with their interaction terms. Statistical significance was considered for all two-sided p values < 5%. All analyses were conducted using R version 4.0.2.
Results
Among 2,742 patients included in the analysis, 372 (13.7%) underwent active surveillance, 1,419 (51.8%) underwent surgery, 630 (23.0%) underwent radiation therapy without ADT and 321 (11.7%) underwent radiation with ADT (Figure 1). Differences in baseline characteristics across the treatment groups are in keeping with prior reports: men who underwent surgery as their primary treatment type were younger, with fewer comorbidities while men undergoing active surveillance were more likely to have low risk disease characteristics, including lower PSA, clinical stage T1, and a low D’Amico risk category. Men with features of high-risk disease (Gleason 8, 9, 10, T2, high D’Amico risk category) were more likely to undergo radiation therapy with ADT (Table 1). Baseline urinary function, urinary incontinence, and bowel function domain scores were similar across all treatment groups. In contrast, baseline sexual function domain scores were higher for men treated with radical prostatectomy compared to those treated with radiation or active surveillance (Table 1).
Figure 1:
Diagram of the Assembly of the Comparative Effectiveness Analyses of Surgery and Radiation (CEASAR) Study Cohort and Final Analytic Cohort (CES-D10, Emotional Well-Being, Energy/Vitality cohort)
Table 1:
Baseline demographic, socioeconomic, and disease characteristics by treatment type
| Active Surveillance | Surgery | Radiation with ADT | Radiation without ADT | p-value | |
|---|---|---|---|---|---|
| N=372 | N=1419 | N=321 | N=630 | ||
| Age at diagnosis, median (IQR) | 67 (62–72) | 62 (57–66) | 70 (65–74) | 67(62–72) | <0.001 |
| Race, no (%) | |||||
| White | 296 (80) | 1069 (76) | 220 (69) | 481 (76) | 0.001 |
| Black | 39 (11) | 165 (12) | 56 (18) | 103 (16) | |
| Hispanic | 22 (6) | 113 (8) | 25 (8) | 27 (4) | |
| Asian | 9 (2) | 40 (3) | 13 (4) | 10 (2) | |
| Other | 5 (1) | 21 (1) | 4 (1) | 8 (1) | |
| TIBI-CaP categories, no (%) | |||||
| 0--2 | 95 (26) | 452 (34) | 49 (16) | 133 (22) | <0.001 |
| 3--4 | 142 (39) | 586 (44) | 114 (37) | 253 (43) | |
| 5 or more | 126 (35) | 307 (23) | 147 (47) | 206 (35) | |
| Income, no (%) | |||||
| <$30,000 | 63 (18) | 211 (17) | 101 (36) | 124 (22) | <0.001 |
| $30,001–50,000 | 79 (23) | 213 (17) | 67 (24) | 129 (23) | |
| $50,001–100,000 | 105 (31) | 422 (33) | 67 (24) | 174 (31) | |
| >$100,000 | 97 (28) | 422 (33) | 47 (17) | 126 (23) | |
| Education, no (%) | |||||
| Less than high school | 25 (7) | 113 (8) | 52 (17) | 59 (10) | <0.001 |
| High school graduate | 66 (18) | 271 (20) | 67 (22) | 121 (21) | |
| Some college | 77 (21) | 298 (22) | 68 (22) | 145 (25) | |
| College graduate | 91 (25) | 324 (24) | 57 (19) | 135 (23) | |
| Graduate/professional school | 104 (29) | 332 (25) | 62 (20) | 129 (22) | |
| Marital status, no (%) | |||||
| Not married | 67 (19) | 216 (16) | 70 (23) | 153 (26) | <0.001 |
| Married | 294 (81) | 1120 (84) | 235 (77) | 437 (74) | |
| Employment Status, no (%) | |||||
| Full time | 160 (43) | 876 (62) | 80 (25) | 250 (40) | <0.001 |
| Retired/part time/unemployed | 210 (57) | 530 (38) | 235 (75) | 371 (60) | |
| Insurance Status, no (%) | |||||
| Medicare | 214 (58) | 464 (33) | 221 (69) | 388 (62) | <0.001 |
| Private/HMO | 142 (38) | 884 (62) | 78 (24) | 211 (34) | |
| VA/military/medicaid/none | 16 (4) | 68 (5) | 22 (7) | 29 (5) | |
| PSA at diagnosis, corrected, median (IQR) | 5.2 (3.9–7.0) | 5.1 (4.2–6.8) | 7.0 (4.9–11.3) | 5.5 (4.3–7.3) | <0.001 |
| Clinical tumor stage | |||||
| T1 | 309 (84) | 1069 (75) | 218 (68) | 491 (78) | <0.001 |
| T2 | 57 (16) | 347 (25) | 103 (32) | 138 (22) | |
| Biopsy gleason score | |||||
| 6 or less | 330 (89) | 717 (51) | 44 (14) | 334 (53) | <0.001 |
| 3 + 4 | 33 (9) | 437 (31) | 113 (35) | 197 (32) | |
| 4 + 3 | 6 (2) | 147 (10) | 57 (18) | 58 (9) | |
| 8,9,10 | 1 (0) | 114 (8) | 107 (33) | 36 (6) | |
| D’Amico risk category | |||||
| Low risk | 293 (79) | 614 (43) | 28 (9) | 305 (49) | <0.001 |
| Intermediate risk | 67 (18) | 598 (42) | 144 (45) | 261 (42) | |
| High risk | 10 (3) | 205 (14) | 149 (46) | 60 (10) | |
| Site | |||||
| Utah | 56 (15) | 119 (8) | 24 (7) | 52 (8) | <0.001 |
| Atlanta | 47 (13) | 189 (13) | 25 (8) | 188 (30) | |
| Los Angeles | 116 (31) | 409 (29) | 76 (24) | 100 (16) | |
| Lousiana | 93 (25) | 356 (25) | 141 (44) | 143 (23) | |
| New Jersey | 28 (8) | 241 (17) | 36 (11) | 128 (20) | |
| CaPSURE | 32 (9) | 105 (7) | 19 (6) | 19 (3) | |
| SF-36 Physical function, median (IQR) | 95 (80–100) | 100 (85–100) | 90 (65–100) | 90 (75–100) | <0.001 |
| SF-36 Emotional Well-Being, median (IQR) | 88 (72–92) | 84 (68–92) | 88 (72–92) | 84 (72–92) | 0.008 |
| SF-36 Energy & Fatigue, median (IQR) | 75 (60–85) | 75 (60–85) | 70 (55–85) | 75 (58–85) | <0.001 |
| Depression (CES-D10) Score, median (IQR) | 3 (1–6) | 4 (1–8) | 4 (2–8) | 4 (1–8) | 0.05 |
| Social support, median (IQR) | 95 (75–100) | 95 (75–100) | 95 (75–100) | 95 (70–100) | 0.046 |
| Participatory decision making, median (IQR) | 86 (68–96) | 86 (71–93) | 79 (64–89) | 79 (68–92) | <0.001 |
| Prostate cancer specific burden, baseline, median (IQR) | 14.3 (0.0–37.1) | 22.9 (8.6–45.7) | 22.9 (5.7–42.9) | 20 (5.7–42.9) | <0.001 |
| EPIC-26 sexual function, baseline, median (IQR) | 75 (42–89) | 80 (38–95) | 50 (12–80) | 65 (32–85) | <0.001 |
| EPIC-26 urinary incontinence, baseline, median (IQR) | 100 (85–100) | 100 (79–100) | 100 (79–100) | 100 (85–100) | 0.2 |
| EPIC-26 urinary irritative, baseline, median (IQR) | 88 (75–94) | 88 (75–100) | 88 (75–94) | 88 (75–100) | 0.011 |
| EPIC-26 bowel function, baseline, median (IQR) | 100 (92–100) | 100 (96–100) | 100 (92–100) | 100 (92–100) | 0.012 |
| EPIC-26 hormonal domain score at baseline, median (IQR) | 95 (85–100) | 95 (85–100) | 90 (75–95) | 95 (85–100) | <0.001 |
At baseline, the median CES-D score in this cohort was 4 (interquartile range 1–8), indicating a low prevalence of depressive symptoms (Table 1). We found no evidence of a clinically meaningful treatment-related effect on longitudinal assessments of depressive symptoms measured with the CES-D score, whether assessed continuously or dichotomized at 9 (Table 2). In addition to higher baseline CES-D score (p<0.001), on multivariable analysis, significant predictors of decline in CES-D score were older age (p=0.001), higher comorbidity (p<0.001), poorer overall health (p=0.001) and physical function (p=0.008), being unmarried (p=0.02), lower income (p=0.002), and lower baseline participatory decision-making score (p=0.003). Interestingly, social support (p=0.39) and education (p=0.12) were not associated with worsening CES-D scores, nor were race (p=0.38), insurance status (p=0.95), D’Amico risk group (p=0.99), or registry site (p=0.11).
Table 2:
The effect of treatment modality on longitudinal assessment of CES-D and SF-36 emotional well-being and energy/fatigue scores at 6 months and 1-, 3-, and 5-years following index, adjusted for the effect of patient demographic, tumor, and baseline functional characteristics.
| Time | Surgery vs. AS* | Radiation (+ ADT) vs. AS | Radiation (no ADT) vs. AS | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Effect | 95% CI | P-value | Effect | 95% CI | P-value | Effect | 95% CI | P-value | ||
| CESD10 | ||||||||||
| 0.5 | −0.1 | [−0.8, 0.5] | 0.669 | 0.2 | [−0.5, 1.0] | 0.5 | 0.4 | [−0.3, 1.0] | 0.264 | |
| 1 | −0.5 | [−0.9, −0.1] | 0.024 | −0.3 | [−0.9, 0.3] | 0.351 | 0 | [−0.5, 0.4] | 0.881 | |
| 3 | −0.3 | [−0.7, 0.2] | 0.316 | 0.1 | [−0.6, 0.9] | 0.713 | 0 | [−0.6, 0.5] | 0.873 | |
| 5 | −0.5 | [−1.0, 0.0] | 0.067 | −0.1 | [−0.9, 0.7] | 0.78 | 0 | [−0.6, 0.6] | 0.959 | |
| Emotional Well Being | ||||||||||
| 0.5 | 1.9 | [−0.1, 4.0] | 0.065 | 0.6 | [−17, 3.0] | 0.598 | 0.4 | [−17, 2.5] | 0.721 | |
| 1 | 2.4 | [1.0, 3.8] | <0.001 | 1.8 | [−0.3, 3.9] | 0.094 | 0.6 | [−0.9, 2.2] | 0.432 | |
| 3 | 1.7 | [0.0, 3.3] | 0.044 | 0 | [−2.4, 2.4] | 0.988 | 0.8 | [−10, 2.6] | 0.376 | |
| 5 | 1.2 | [−0.6, 3.0] | 0.187 | 0.4 | [−2.4, 3.2] | 0.769 | 0.6 | [−1.4, 2.6] | 0.562 | |
| Energy/Fatigue | ||||||||||
| 0.5 | 4.7 | [2.2, 7.2] | <0.001 | −1 | [−4.0, 2.0] | 0.528 | 0.3 | [−2.3, 2.9] | 0.84 | |
| 1 | 3 | [1.4, 4.6] | <0.001 | −0.9 | [−3.3, 1.5] | 0.455 | −1.5 | [−3.3, 0.2] | 0.085 | |
| 3 | 2.1 | [0.3, 3.9] | 0.023 | −2.2 | [−5.0, 0.5] | 0.113 | −0.2 | [−2.2, 1.8] | 0.845 | |
| 5 | 1.5 | [−0.4, 3.4] | 0.134 | −1.1 | [−4.0, 1.8] | 0.455 | −1.8 | [−4.0, 0.5] | 0.128 | |
| CESD>9 (binary) | Odds Ratio (95% CI) | Odds Ratio (95% CI) | Odds Ratio (95% CI) | |||||||
| 0.5 | 0.9 | [0.6, 1.6] | 0.829 | 1.3 | [0.7, 2.2] | 0.448 | 1.4 | [0.8, 2.5] | 0.181 | |
| 1 | 0.9 | [0.6, 1.3] | 0.556 | 1 | [0.6, 1.5] | 0.858 | 1.1 | [0.8, 1.6] | 0.528 | |
| 3 | 0.8 | [0.6, 1.3] | 0.404 | 1.4 | [0.8, 2.3] | 0.236 | 0.9 | [0.5, 1.3] | 0.484 | |
| 5 | 0.8 | [0.6, 1.2] | 0.364 | 1 | [0.6, 1.8] | 0.983 | 1.3 | [0.8, 2.0] | 0.323 | |
AS=Active Surveillance
Assessing emotional well-being and energy/fatigue domains of the SF-36, we found that, while baseline scores were overall quite high, lower scores were reported among those undergoing radiotherapy (with or without ADT), a difference which persisted over time (Figure 2). However, on adjusted and unadjusted analysis, treatment type was not associated with a clinically significant difference in longitudinal assessments of SF-36 emotional well-being or energy/fatigue scores at 6 months or 1-, 3-, or 5-years following treatment initiation, despite statistically significant differences (Table 2, Supplementary Table 1). Minor declines in SF-36 energy/fatigue scores at 6 and 12 months among those treated with radiation and ADT (Figure 2) are unlikely to be clinically meaningful and were not statistically significant (Table 2). Notably, there was a statistically significant difference in the SF-36 energy/fatigue domain between patients who underwent surgery compared to radiotherapy though this failed to meet the threshold for a clinically important difference and, also, diminished with time. Further pairwise testing between surgery and radiotherapy, and between radiotherapy with and without ADT demonstrated no clinically meaningful differences in the captured mental health related outcomes (Supplementary Table 2). In addition to baseline SF-36 emotional well-being (p<0.001), significant predictors of decline in SF-36 emotion well-being scores were older age (p<0.001), higher comorbidity (p<0.001), lower income (p=0.004), general health (p=0.02) and physical function (p<0.001), social support (p=0.001), and baseline participatory decision making scores (p=0.002) Notably, unlike for changes in CES-D score, marital status was not associated with changes in emotional well-being measured by SF-36 (p=0.17), nor was race (p=0.26), insurance status (p=0.88), D’Amico risk group (p=0.62), registry site (p=0.24)
Figure 2:
Trend in unadjusted CES-D and SF-36 scores by treatment modality over time
Discussion
In this population-based, prospective cohort study of men with localized prostate cancer, we found no clinically meaningful association between treatment approach (including active surveillance, radical prostatectomy and radiotherapy) and measures of mental health including depressive symptoms (captured using the validated CES-D) and emotional well-being and energy/fatigue (captured as domains of the validated SF-36). These findings were consistent with our hypothesis, that treatment type would not impact overall mental health outcomes in men with localized prostate cancer. However, on multivariable analysis, we did identify baseline characteristics associated with declining emotional well-being following prostate cancer treatment including older age, poor overall health, unmarried status, and worse baseline depression and emotional well-being symptoms. These characteristics may allow clinicians to identify patients most at risk of declines in mental health following prostate cancer diagnosis and treatment to target interventions to address these issues.
To our knowledge, this is the first prospective evaluation of mental health outcomes in men with localized prostate cancer. While previous work has demonstrated an increased utilization of anti-depressants following diagnosis for patients who received surgery or radiotherapy, but not active surveillance, this relied on administrative records and prescriptions as a proxy for symptoms27. In contrast, and in keeping with our findings, there does not appear to be an increased risk of suicide amongst patients diagnosed with prostate cancer, whereas there are increased risks of suicide amongst patients with other malignancies28.
Although we did not demonstrate a relationship between treatment type and mental health outcomes in men with localized prostate cancer, we did find several factors predictive of declining mental health outcomes. Consistent with research in other malignancies, older, unmarried patients had declining emotional well-being in our analysis. In the bladder cancer population, prior research has demonstrated those at risk for suicidal death were typically elderly, unmarried men29. An additional vulnerable population we identified was those with poor depression and poor emotional well-being symptoms prior to treatment. It is imperative that urologists seize available opportunities to identify and intervene in patients with mental health concerns both at the time of diagnosis and during follow-up. In addition to the importance of addressing patient distress and morbidity associated with these symptoms, prior work has demonstrated that significant mental health care utilization is independently associated with worse cancer specific and all-cause mortality30. While psychologic interventions are beyond the scope of most clinicians treating prostate cancer, we ought to appropriately screen and subsequently refer those at risk.
Our findings should be considered in the context of several limitations. First, as this is an observational study, treatment choice is non-random and thus there is the potential for confounding by indication. However, given the baseline similarities between groups and the longitudinal nature of assessment, and control for clinical factors, it is unlikely that this explains the findings. Second, due to the nature of the survey employed, we utilized a modified CES-D 10 with patients completing only nine questions rather than ten questions on the recommendation of our psychometrician in order to reduce respondent burden as other included instruments captured overlapping concepts. We adjusted the overall CES-D score to reflect this difference (from a standard score of 30 points to 27 points.) Similarly, the SF-36 and CES-D have been validated in a general population, but may not detect minute differences in our population of overall healthy men, and should not be used in isolation for diagnosing depression or mental health changes. Third, in the context of a finding of no significant differences, we must consider the potential for type II error. However, given the small differences noted which did not meet established threshold for clinically meaningful differences, increases in sample size are unlikely to change study conclusions. Additionally, many patients with low-risk disease in the CEASAR study received active intervention which, while common at the time, is not reflective of current practice patterns which now favor active surveillance in this cohort. Finally, we must not underestimate the mental health burden of being diagnosed with prostate cancer including for those who choose active surveillance as their primary treatment strategy. Thus using the active surveillance group as the referent group may contribute to the limited impact of treatment modality on mental health outcomes. Perhaps a more appropriate referent group, and consideration for future work, would be a group of healthy age-matched men without a cancer diagnosis.
These limitations notwithstanding, in this population-based, prospective cohort study of men with localized prostate cancer, we found no clinically meaningful association between treatment approach (including active surveillance, radical prostatectomy and radiotherapy) and measures of mental health including depressive symptoms (captured using the validated CES-D) and emotional well-being and energy/fatigue (captured as domains of the validated SF-36). We further identified characteristics of patients with a higher likelihood of declining mental health following prostate cancer diagnosis, independent of treatment approach, including older age, being unmarried, worse overall health and worse baseline mental health.
Conclusions
Careful evaluation of patients at risk for adverse mental health outcomes is warranted among all treatment groups, and appropriate psychiatric assistance should be provided to these patients to optimize the comprehensive care we provide to prostate cancer patients.
Supplementary Material
Acknowledgements:
This is written on behalf of all CEASAR investigators.
Funding:
This work was supported by the Agency for Healthcare Research and Quality (1R01HS019356, 1R01HS022640, R01CA230352); the Patient-Centered Outcomes Research Institute (CE-12-11-4667); and the National Cancer Institute (NIH/NCI grant 5T32CA106183). Data management was facilitated by Vanderbilt University’s Research Electronic Data Capture (REDCap) system, which is supported by the Vanderbilt Institute for Clinical and Translational Research grant (UL1TR000011 from NCATS/NIH).
Role of Funder:
The funders had no role in the study design, data collection, and analysis, decision to publish, or preparation of the manuscript.
Notes:
Author Disclosures: The authors have no disclosures relevant to this work.
Disclaimers: There are no relevant disclaimers.
References:
- 1.Hamdy FC, Donovan JL, Lane JA, et al. 10-Year Outcomes after Monitoring, Surgery, or Radiotherapy for Localized Prostate Cancer. The New England journal of medicine. 2016. [DOI] [PubMed]
- 2.Wallis CJ, Saskin R, Choo R, et al. Surgery Versus Radiotherapy for Clinically-localized Prostate Cancer: A Systematic Review and Meta-analysis. European urology. 2015. [DOI] [PubMed]
- 3.Wallis CJD, Glaser A, Hu JC, et al. Survival and Complications Following Surgery and Radiation for Localized Prostate Cancer: An International Collaborative Review. European urology. 2018;73(1):11–20. [DOI] [PubMed] [Google Scholar]
- 4.Resnick MJ, Koyama T, Fan KH, et al. Long-term functional outcomes after treatment for localized prostate cancer. The New England journal of medicine. 2013;368(5):436–445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Barocas DA, Alvarez J, Resnick MJ, et al. Association Between Radiation Therapy, Surgery, or Observation for Localized Prostate Cancer and Patient-Reported Outcomes After 3 Years. JAMA : the journal of the American Medical Association. 2017;317(11):1126–1140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hoffman KE, Penson DF, Zhao Z, et al. Patient-Reported Outcomes Through 5 Years for Active Surveillance, Surgery, Brachytherapy, or External Beam Radiation With or Without Androgen Deprivation Therapy for Localized Prostate Cancer. JAMA : the journal of the American Medical Association. 2020;323(2):149–163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wallis CJ, Herschorn S, Saskin R, et al. Complications After Radical Prostatectomy or Radiotherapy for Prostate Cancer: Results of a Population-based, Propensity Score-matched Analysis. Urology. 2015;85(3):621–628. [DOI] [PubMed] [Google Scholar]
- 8.Wallis CJ, Cheung P, Herschorn S, et al. Complications following surgery with or without radiotherapy or radiotherapy alone for prostate cancer. British journal of cancer. 2015;112(6):977–982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Wallis CJ, Mahar A, Cheung P, et al. New Rates of Interventions to Manage Complications of Modern Prostate Cancer Treatment in Older Men. European urology. 2015; 10.1016/j.eururo.2015.10.043. [DOI] [PubMed]
- 10.Wallis CJ, Mahar AL, Cheung P, et al. Hospitalizations to Manage Complications of Modern Prostate Cancer Treatment in Older Men. Urology. 2016. [DOI] [PubMed]
- 11.Kinlock BL, Parker LJ, Howard DL, et al. Prevalence and Correlates of Major Depressive Symptoms among Black Men with Prostate Cancer. Ethn Dis. 2017;27(4):429–436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Congard A, Christophe V, Duprez C, et al. The self-reported perceptions of the repercussions of the disease and its treatments on daily life for young women with breast cancer and their partners. J Psychosoc Oncol. 2019;37(1):50–68. [DOI] [PubMed] [Google Scholar]
- 13.Reese JB, Handorf E, Haythornthwaite JA. Sexual quality of life, body image distress, and psychosocial outcomes in colorectal cancer: a longitudinal study. Support Care Cancer. 2018;26(10):3431–3440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Barocas DA, Chen V, Cooperberg M, et al. Using a population-based observational cohort study to address difficult comparative effectiveness research questions: the CEASAR study. J Comp Eff Res. 2013;2(4):445–460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cooperberg MR, Broering JM, Litwin MS, et al. The contemporary management of prostate cancer in the United States: lessons from the cancer of the prostate strategic urologic research endeavor (CapSURE), a national disease registry. The Journal of urology. 2004;171(4):1393–1401. [DOI] [PubMed] [Google Scholar]
- 16.Irwin M, Artin KH, Oxman MN. Screening for depression in the older adult: criterion validity of the 10-item Center for Epidemiological Studies Depression Scale (CES-D). Arch Intern Med. 1999;159(15):1701–1704. [DOI] [PubMed] [Google Scholar]
- 17.Smarr KL, Keefer AL. Measures of depression and depressive symptoms: Beck Depression Inventory-II (BDI-II), Center for Epidemiologic Studies Depression Scale (CES-D), Geriatric Depression Scale (GDS), Hospital Anxiety and Depression Scale (HADS), and Patient Health Questionnaire-9 (PHQ-9). Arthritis Care Res (Hoboken). 2011;63 Suppl 11:S454–466. [DOI] [PubMed] [Google Scholar]
- 18.Zhang W, O’Brien N, Forrest JI, et al. Validating a shortened depression scale (10 item CES-D) among HIV-positive people in British Columbia, Canada. PloS one. 2012;7(7):e40793. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Baron EC, Davies T, Lund C. Validation of the 10-item Centre for Epidemiological Studies Depression Scale (CES-D-10) in Zulu, Xhosa and Afrikaans populations in South Africa. BMC Psychiatry. 2017;17(1):6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.McHorney CA, Ware JE Jr., Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Medical care. 1993;31(3):247–263. [DOI] [PubMed] [Google Scholar]
- 21.Ware JE Jr., Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Medical care. 1992;30(6):473–483. [PubMed] [Google Scholar]
- 22.Skolarus TA, Dunn RL, Sanda MG, et al. Minimally important difference for the Expanded Prostate Cancer Index Composite Short Form. Urology. 2015;85(1):101–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Jayadevappa R, Malkowicz SB, Wittink M, et al. Comparison of distribution- and anchor-based approaches to infer changes in health-related quality of life of prostate cancer survivors. Health services research. 2012;47(5):1902–1925. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Litwin MS, Greenfield S, Elkin EP, et al. Assessment of prognosis with the total illness burden index for prostate cancer: aiding clinicians in treatment choice. Cancer. 2007;109(9):1777–1783. [DOI] [PubMed] [Google Scholar]
- 25.White H. A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity. Econometrica. 1980;48:817–838. [Google Scholar]
- 26.Huber PJ. The behavior of maximum likelihood estimates under nonstandard conditions. Proceedings of the fifth Berkeley symposium on mathematical statistics and probability. University of California. 1967:221–233.
- 27.Matta R, Wallis CJD, Goldenberg MG, et al. Variation and Trends in Antidepressant Prescribing for Men Undergoing Treatment for Nonmetastatic Prostate Cancer: A Population-based Cohort Study. European urology. 2019;75(1):3–7. [DOI] [PubMed] [Google Scholar]
- 28.Klaassen Z, Wallis CJD, Chandrasekar T, et al. Cancer diagnosis and risk of suicide after accounting for prediagnosis psychiatric care: A matched-cohort study of patients with incident solid-organ malignancies. Cancer. 2019;125(16):2886–2895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Klaassen Z, Goldberg H, Chandrasekar T, et al. Changing Trends for Suicidal Death in Patients With Bladder Cancer: A 40+ Year Population-level Analysis. Clin Genitourin Cancer. 2018;16(3):206–212 e201. [DOI] [PubMed] [Google Scholar]
- 30.Klaassen Z, Wallis CJD, Goldberg H, et al. The impact of psychiatric utilisation prior to cancer diagnosis on survival of solid organ malignancies. British journal of cancer. 2019;120(8):840–847. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.