Hormonal Therapy Use for Prostate Cancer and Mortality in Men With Coronary Artery Disease–Induced Congestive Heart Failure or Myocardial Infarction FREE

Patients with localized prostate cancer have many options available for curative treatment.¹ The use of brachytherapy both as monotherapy and in conjunction with external beam radiation therapy has been demonstrated to exhibit both long-term efficacy and safety.^2- 5 Neoadjuvant hormonal therapy (HT) is used as a means for prostate gland cytoreduction in order to eliminate pubic arch interference, thereby facilitating the ability to perform brachytherapy.⁶ Although HT has been associated with adverse effects including the development of insulin resistance⁷ and unfavorable lipid profiles,⁸ a decrease in muscle mass and bone mineral density,⁹ and an increased risk of cardiovascular death,^10- 13 its combined use with external beam radiation therapy has resulted in improvements in both cancer-specific and overall survival compared with external beam radiation therapy alone in men with locally advanced or localized, unfavorable-risk prostate cancer.^14- 18

Recently, a postrandomization hypothesis-generating analysis of a trial comparing external beam radiation therapy alone vs external beam radiation therapy plus 6 months of HT involving men with localized, unfavorable-risk prostate cancer addressed whether the extent of preexisting comorbidity could affect the overall survival benefit.¹⁹ The results suggested that men with no or minimal comorbidity had a survival benefit with combination treatment, whereas there was no significant difference in survival for men with moderate to severe comorbidity. However, the extent to which any particular comorbid condition contributed toward this loss in survival benefit remains unknown. Therefore, the purpose of this study was to assess whether neoadjuvant HT use in men with prostate cancer treated with brachytherapy affects the risk of all-cause mortality in men with known coronary artery disease–induced conditions including congestive heart failure and myocardial infarction; coronary artery disease risk factors including diabetes mellitus, hypercholesterolemia, and hypertension; or no comorbidity.

This single-institution, retrospective practice experience comprised 5077 men with clinical stage T1 to T3 N0 M0 adenocarcinoma of the prostate consecutively treated between 1997 and 2006 at the Chicago Prostate Cancer Center, Westmont, Illinois. All men were observed for a minimum of 1 year during which time no deaths occurred. Men underwent either a magnetic resonance imaging (MRI) or computed tomographic (CT) scan of the pelvis to rule out pathological lymph nodes. Prostate needle biopsies were reviewed by pathologists with expertise in genitourinary oncology. The study was performed with the approval of an independent institutional review board, IntegReview, which is fully accredited by the Association for the Accreditation of Human Research Protection Programs. Each patient signed an informed consent at the time of initial consultation allowing their deidentified clinical and prostate cancer–related information to be collected and entered into a secure, password-protected database for subsequent outcomes analysis. Men were referred to this center based on their interest and possible candidacy for brachytherapy. Those electing for radical prostatectomy, external beam radiation therapy alone, or active surveillance were screened out before being referred to this center.

The collection of data for risk-factor assessment was made by the treating physician at the time of initial consultation and recorded. For each patient, as part of the past medical history, the treating physician thoroughly and systematically assessed for the presence or absence of coronary artery disease risk factors including diabetes mellitus, hypercholesterolemia, or hypertension and coronary artery disease–induced conditions including congestive heart failure or myocardial infarction. This information was initially ascertained from medical records provided by referring physicians and then confirmed through verbal communication with each patient at the time of initial consultation. Although no specific diagnostic criteria or threshold laboratory levels were enforced by the study, it is assumed that each patient's primary care physician or cardiologist used up-to-date, evidence-based diagnostic criteria set forth by national, consensus panel statements for each comorbid condition.^20- 29

When making decisions about optimal treatment, the treating physicians adhered to the 1999 American Brachytherapy Society guidelines for permanent seed implantation.³⁰ These guidelines stipulate the use of brachytherapy alone in the setting of low-risk disease and the use of supplemental external beam radiation therapy when there is a significant risk of extraprostatic extension. Neoadjuvant HT was used in the setting of unfavorable-risk disease³¹ and only used in the setting of favorable-risk disease³¹ for downsizing of the prostate to eliminate pubic arch interference⁶ in order to allow patients to be eligible for brachytherapy.

Computed tomography–based 3-dimensional conformal or intensity modulated radiation therapy treatment planning was used for supplemental external beam radiation therapy. The pelvic lymph nodes were not included in the treatment volume. For those men receiving combination treatment, a total dose of 45 Gy of supplemental external beam radiation therapy was delivered to the prostate and seminal vesicles in 25 fractions 5 days a week for 5 weeks. Two to 4 weeks later, an interstitial brachytherapy boost consisting of ultrasound-guided implantation of radioactive seeds into the prostate was performed. A peripheral-loading technique using preloaded isotopes was administered in conjunction with preplanned dosimetry. Adhering to uniformly consistent and accepted standards within the United States, the prescribed peripheral doses when used alone were 144 Gy iodine, 108 Gy palladium, and 132 Gy cesium and when used with supplemental external beam radiation therapy, 108 Gy iodine, 90 Gy palladium, and 100 Gy cesium.

Neoadjuvant HT consisted of both a luteinizing hormone-releasing hormone (LHRH) agonist (leuprolide or goserelin) and a nonsteroidal antiandrogen (bicalutamide or flutamide). Injections of leuprolide (7.5 mg/mo) or goserelin (3.6 mg/mo) were given in 1-month formulations. Bicalutamide (50 mg/d) or flutamide (250 mg every 8 hours) was taken orally starting at 1 to 3 days before LHRH agonist initiation to block transient testosterone surge.

Follow-up began on the day of interstitial brachytherapy and concluded on July 1, 2008, or the date of death, whichever was earlier. No patient was lost to follow-up. Routine follow-up included a serum prostate-specific antigen (PSA) measurement followed by a digital rectal examination generally every 3 months for 1 year, then every 6 months for an additional 4 years, and annually thereafter. At the time of PSA failure, patients underwent routine follow-up assessment, a pelvic CT or MRI scan, and a bone scan. Salvage therapy consisted of an LHRH agonist initiated after PSA recurrence and always before symptomatic or radiographic progression.

Descriptive statistics were used to define the patient population and tumor and treatment characteristics at baseline. The Mantel-Haenszel χ² test³² was used to make comparisons of categorical baseline factors for men in each comorbidity group stratified by neoadjuvant HT use. The Wilcoxon rank sum test³³ was used to make comparisons of the median values and distributions of continuous factors including age and pretreatment PSA level for men in each comorbidity group stratified by neoadjuvant HT use.

A propensity analysis was performed to account for potential biases in treatment effect arising from nonrandom allocation of patients to different treatment groups.³⁴ Propensity analysis required calculation of a conditional probability for the 2 treatment groups (brachytherapy alone or brachytherapy with supplemental external beam radiation therapy) using a multivariable logistic regression model. The propensity score for the selection of supplemental external beam radiation therapy with brachytherapy was then used in the Cox regression models for the 3 comorbidity groups.

The primary end point of the study was the risk of all-cause mortality. The date of death was ascertained and documented by each patient's attending oncologist or urologist. The National Death Index was initially used to determine whether a patient was alive or deceased. If deceased, this was further verified by the patient's family, physician, and death certificate. Survival time was defined as the time from the date of brachytherapy to the date of death or to the date of last follow-up of living patients.

For each comorbidity group, Cox regression univariable and multivariable analyses³⁵ were performed to assess whether neoadjuvant HT use affected the risk of all-cause mortality, adjusting for age, year of treatment, supplemental external beam radiation therapy use, treatment propensity score, and known prostate cancer prognostic factors, including pretreatment PSA level, biopsy Gleason score, and clinical tumor (T) category.³⁶ Neoadjuvant HT and supplemental external beam radiation therapy use, Gleason score, and clinical T category were analyzed as categorical variables, whereas age, PSA level, and year of brachytherapy were considered as continuous variables. Categorical variables were defined prior to analyzing the data and included neoadjuvant HT use, supplemental external beam radiation therapy use, Gleason score of 7 and from 8 to 10, and clinical category T2 and T3. Baseline groups were treatment without neoadjuvant HT or without supplemental external beam radiation therapy as well as Gleason score 6 or less and clinical category T1. For the Cox regression analyses, the assumptions of the proportional hazards model were tested using the time-dependent covariate method³⁵ and assessed via various residuals.³⁵ No evidence was found to suggest that these assumptions were violated. All statistical tests were 2 sided and P < .05 was considered statistically significant. Unadjusted and adjusted hazard ratios (HRs) for the risk of all-cause mortality and the associated 95% confidence intervals (CIs) were calculated for each covariate.³⁵

For the purpose of illustration, age-adjusted estimates³⁷ of the risk of all-cause mortality stratified by neoadjuvant HT use were determined for men in each comorbidity group and graphically displayed. The method of Kaplan and Meier³⁸ was used to calculate overall survival estimates and all-cause mortality estimates were defined as 1 minus Kaplan-Meier estimates of overall survival. Five-year point estimates were provided with 95% CIs. Pairwise comparisons of the distribution of all-cause mortality estimates were assessed using a log-rank test.³⁹ A Bonferroni correction⁴⁰ was used to adjust for multiple comparisons. Given 3 comorbidity groups and 2 variables (treatment with or without neoadjuvant HT), there were a total of 3 pairwise comparisons and so a Bonferroni-corrected significance level of 0.05/3 = 0.017 was used. All statistical analyses were performed using SAS version 9.2 (SAS Institute Inc, Cary, North Carolina).

Five thousand seventy-seven men in the study cohort—median age 69.5 years (interquartile range [IQR], 63.3-74.2 years)—received brachytherapy. Of those, 1521 (30.0%) received neoadjuvant HT for a median of 4 months (IQR, 3-4 months) and 3556 (70.0%) did not. Five hundred fifty-five men (10.9%) underwent supplemental external beam radiation therapy and 4522 (89.1%) did not. Two thousand six hundred fifty-three patients (52.3%) had no history of a comorbidity; 2168 (42.7%) had a coronary artery disease risk factor—179, diabetes mellitus; 326, hypercholesterolemia, and 1663, hypertension; and 256 (5.0%) had known coronary artery disease—136, congestive heart failure and 120, myocardial infarction. When men in each comorbidity group were further stratified by neoadjuvant HT use, those who received it were significantly more likely to have unfavorable prostate cancer prognostic factors than men who did not (Table 1).

In all, 419 men died. Of those, 200 men had no underlying comorbidity, 176 had one coronary artery disease risk factor, and 43 had a history of known coronary artery disease resulting in congestive heart failure or myocardial infarction. When further stratified, mortality was greater among those who received neoadjuvant HT than among those who did not: 75 (9.6%) vs 125 (6.7%) had no underlying comorbidity; 69 (10.7%) vs 107 (7.0%) had 1 coronary artery disease risk factor; and 25 (26.3%) vs 18 (11.2%) had a history of known coronary artery disease. For the entire cohort of 5077 men, neoadjuvant HT use was not associated with an increased risk of all-cause mortality (11.1% vs 7.0%; adjusted HR, 1.08, 95% CI; 0.88-1.33; P = .46) after a median follow-up of 4.8 years (IQR, 2.8-6.9 years). When considering comorbidity groups separately, neoadjuvant HT use was not associated with an increased risk of all-cause mortality in men with no comorbidity (9.6% vs 6.7%; adjusted HR, 0.97; 95% CI, 0.72-1.32; P = .86) or a single coronary artery disease risk factor (10.7% vs 7.0%; adjusted HR, 1.04; 95% CI, 0.75-1.43; P = .82) after median follow-ups of 5.0 years (IQR, 3.0-7.1 years) and 4.4 years (IQR, 2.7-6.4 years), respectively (Table 2).

However, for men with coronary artery disease–induced congestive heart failure or myocardial infarction, after a median follow-up of 5.1 years (IQR, 3.2-7.1 years), neoadjuvant HT use was significantly associated with an increased risk of all-cause mortality (26.3% vs 11.2%; adjusted HR, 1.96; 95% CI, 1.04-3.71; P = .04). Based on the sample number of 256, event rate of 0.17, and adjusted HR of 1.96, this study is estimated to have approximately 60% power to observe such a difference.⁴¹ Increasing age was also significantly associated with an increased risk of all-cause mortality in men with no comorbidity (adjusted HR, 1.09; 95% CI, 1.07-1.13; P < .001), a single coronary artery disease risk factor (adjusted HR, 1.09; 95% CI, 1.06-1.13; P < .001), and coronary artery disease-induced congestive heart failure or myocardial infarction (adjusted HR, 1.12; 95% CI, 1.03-1.21; P = .01). Furthermore, when tests for interaction were performed for the entire cohort of 5077 men, a significant interaction existed between neoadjuvant HT use and comorbid coronary artery disease–induced congestive heart failure or myocardial infarction (adjusted HR, 2.11; 95% CI, 1.08-4.13, P = .03) but not for diabetes mellitus, hypercholesterolemia, or hypertension (adjusted HR, 1.14; 95% CI, 0.75-1.74; P = .53).

As illustrated in the Figure, no significant differences existed in the age-adjusted estimates of all-cause mortality stratified by neoadjuvant HT use in men with no underlying comorbidity or a single coronary artery disease risk factor. However, for men with a history of known coronary artery disease resulting in congestive heart failure or myocardial infarction, treatment with neoadjuvant HT led to higher age-adjusted estimates of all-cause mortality than treatment without neoadjuvant HT. The 5-year age-adjusted point estimates for percent all-cause mortality stratified by treatment with or without neoadjuvant HT were 6.5 (95% CI, 4.6-8.4) vs 7.2 (95% CI, 5.8-8.7) for men with no comorbidity, 10.6 (95% CI, 7.4-13.8) vs 8.6 (95% CI, 6.7-10.5) for men with 1 coronary artery disease risk factor, and 24.4 (95% CI, 15.5-33.3) vs 10.7 (95% CI, 3.9-17.6) for men with known coronary artery disease resulting in congestive heart failure or myocardial infarction.

Our results demonstrate that after adjusting for age, treatment year, supplemental external beam radiation therapy use, treatment propensity score, and known prostate cancer prognostic factors, the use of neoadjuvant HT is associated with an increased risk of all-cause mortality in men with a history of coronary artery disease–induced congestive heart failure or myocardial infarction but not in those with no comorbidity or a single coronary artery disease risk factor. Although the effect of neoadjuvant HT on the risk of all-cause mortality in men treated with brachytherapy has been previously reported,^42- 43 no studies to date have accounted for specific cardiovascular risk factors and conditions. Given recent data from a postrandomization analysis¹⁹ associating moderate to severe comorbidity with a loss in survival benefit from the addition of HT to external beam radiation therapy in men with localized, unfavorable-risk prostate cancer, our current findings identify specific comorbid conditions that may directly contribute to this observed loss in survival benefit. It is also important to note that the population of men in whom the use of neoadjuvant HT may be detrimental was limited to 5% (256 of 5077) in this community-based study cohort. This latter point may explain why there has been a survival benefit observed in the major randomized trials comparing HT plus external beam radiation therapy to external beam radiation therapy alone.^14- 18

The clinical significance of this finding is that for men with favorable-risk prostate cancer and a history of congestive heart failure or myocardial infarction who require neoadjuvant HT solely to eliminate pubic arch interference,⁶ alternative strategies such as active surveillance or treatment with external beam radiation therapy or prostatectomy should be considered. However, for men with unfavorable-risk prostate cancer³¹ who require HT in addition to radiation therapy to take advantage of its survival benefit,^14- 18 appropriate medical evaluation prior to initiation should facilitate clinicians in balancing the relative risks against the benefits of HT use. Moreover, these data suggest that in the design of future clinical trials for men with prostate cancer with HT representing any component of the treatment, prerandomization stratification based on the presence or absence of coronary artery disease–induced congestive heart failure or myocardial infarction should be considered in order to ascertain if the treatment effect varies in men with and without a history of known coronary artery disease.

Several points require further consideration. First, although the adjusted HR for all-cause mortality with neoadjuvant HT use was less than 1.0 in men with no comorbidity, the absolute death rate was higher in men who received neoadjuvant HT (9.6%) than in men who did not (6.7%), which can be explained by the significantly higher median age and more unfavorable prostate cancer prognostic factors for the group receiving neoadjuvant HT as shown in Table 1. Second, although men who received neoadjuvant HT were more likely to have had unfavorable prognostic factors that can lead to higher rates of all-cause mortality because of increased prostate cancer–specific mortality, these prognostic factors were adjusted for in the multivariable analyses.

Third, in our study, coronary artery disease risk factors included diabetes mellitus, hypercholesterolemia, and hypertension. It is well-known that a positive family history for coronary artery disease and a personal history of tobacco use also represent important coronary artery disease risk factors.^44- 45 Future studies are needed to determine whether these additional factors alone or any combination of all of the factors affect the risk of all-cause mortality in men receiving HT compared with those who do not. Fourth, a potential limitation of this study is that we assumed that the primary care physicians or cardiologists had made diagnoses based on up-to-date, evidence-based diagnostic criteria set forth by national consensus panel statements.^20- 29 Future studies should ensure the use of these guidelines in defining individual comorbidities.

Fifth, although the purpose of this particular study was to assess the effect of cardiovascular risk factors and disease on all-cause mortality in men treated with or without neoadjuvant HT, future studies may rely on a standardized comorbidity index like the Adult Comorbidity Evaluation 27,⁴⁶ which can account for multiple clinical measures at once. Sixth, larger multi-institutional studies with longer follow-up will be needed to ensure generalizability and more precisely estimate the magnitude of the increased risk to men with coronary artery disease–induced congestive heart failure or myocardial infarction.

Finally, in our cohort the median duration of HT was 4 months and consisted of both an LHRH agonist and an antiandrogen. However, men with locally advanced prostate cancer are frequently treated with 2 to 3 years of HT in combination with external beam radiation therapy and part of that therapy consists of an LHRH agonist plus an antiandrogen, whereas the majority of treatment is typically treated with an LHRH agonist alone. Given our current findings, future studies assessing the effect of both the duration and extent of HT on the risk of all-cause mortality in men with known coronary artery disease are needed.

In conclusion, this study should heighten awareness about the potential for harm with neoadjuvant HT use in select men. Specifically, the use of neoadjuvant HT is significantly associated with an increased risk of all-cause mortality among men with a history of coronary artery disease−induced congestive heart failure or myocardial infarction but not among men with no comorbidity or a single coronary artery disease risk factor.