EUROPEAN UROLOGY 65 (2014)

Transcription

1 available at journal homepage: Prostate Cancer Neoadjuvant Hormonal Therapy Use and the Risk of Death in Men with Prostate Cancer Treated with Brachytherapy Who Have No or at Least a Single Risk Factor for Coronary Artery Disease Akash Nanda a, *, Ming-Hui Chen b, Brian J. Moran c, Michelle H. Braccioforte c, Daniel Dosoretz d, Sharon Salenius d, Michael Katin d, Rudi Ross d, Anthony V. D Amico e a Department of Radiation Oncology, M.D. Anderson Cancer Center Orlando, Orlando, FL, USA; b Department of Statistics, University of Connecticut, Storrs, CT, USA; c Prostate Cancer Foundation of Chicago, Westmont, IL, USA; d 21st Century Oncology, Fort Myers, FL, USA; e Department of Radiation Oncology, Brigham and Women s Hospital/Dana-Farber Cancer Institute, Boston, MA, USA Article info Article history: Accepted August 30, 2012 Published online ahead of print on September 7, 2012 Keywords: Prostate cancer Neoadjuvant hormonal therapy Brachytherapy Risk factors Coronary artery disease Diabetes mellitus Hypercholesterolemia Hypertension Abstract Background: Neoadjuvant hormone therapy (NHT) use is associated with an increased risk of all-cause mortality (ACM) in men with a history of coronary artery disease (CAD) induced congestive heart failure (CHF) or myocardial infarction (MI). However, its effect in men with no or at least a single risk factor for CAD stratified by prostate cancer (PCa) aggressiveness is unknown. Objective: To assess whether NHT use affects the risk of ACM in men with low-, intermediate-, and high-risk PCa treated with brachytherapy who have no or at least a single risk factor for CAD. Design, setting, and participants: This retrospective study cohort consisted of 5411 men with low-risk PCa (prostate-specific antigen [PSA] <10 ng/ml, Gleason score 6, and clinical stage T1 T2a); 4365 men with intermediate-risk PCa (PSA ng/ml or Gleason score <8 or clinical stage <T3); and 1360 men with localized or locally advanced, high-risk PCa consecutively treated in a community-based, multi-institutional setting between 1991 and CAD risk factors included at least a history of diabetes mellitus, hypercholesterolemia, or hypertension. The median follow-up for men with low-, intermediate-, and high-risk PCa were 4.1, 4.4, and 4.6 yr, respectively. Interventions: Men were treated with or without a median duration of 4 mo of NHT followed by brachytherapy with or without supplemental external-beam radiation therapy (EBRT). Outcome measurements and statistical analysis: Cox regression multivariable analyses were performed to assess whether NHT use affected the risk of ACM in men with low-, intermediate-, and high-risk PCa, adjusting for age; year of brachytherapy; supplemental EBRT use; the presence of CAD risk factors; treatment propensity score; and known PCa prognostic factors, including pretreatment PSA level, biopsy Gleason score, and clinical stage. Results and limitations: NHT use was associated with a significantly increased risk of ACM in men with low-risk PCa (adjusted hazard ratio [HR]: 1.27; 95% confidence interval [CI], ; p < 0.01) but not in men with intermediate-risk (adjusted HR: 1.13; 95% CI, ; p = 0.15) or high-risk PCa (adjusted HR: 0.86; 95% CI, ; p = 0.28). Using an interaction model for the low-risk group, NHT use was associated with a significantly increased risk of ACM in the subgroup of men with at least a single CAD risk factor (adjusted HR: 1.36; 95% CI, ; p = 0.01) but not for men with no CAD risk factors (adjusted HR: 1.19; 95% CI, ; p = 0.13). * Corresponding author. Department of Radiation Oncology, M.D. Anderson Cancer Center Orlando, 1400 S. Orange Ave., MP 760, Orlando, FL 32806, USA. Tel ; Fax: address: akash.nanda@orlandohealth.com (A. Nanda) /$ see back matter # 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved.

2 178 Conclusions: For men with no or at least a single risk factor for CAD, NHT use is associated with an increased risk of ACM in the setting of low-risk but not intermediateor high-risk PCa. This effect is driven by the subgroup of men with at least a single risk factor for CAD. These results warrant prospective validation given the widespread use of NHT for prostate downsizing prior to brachytherapy. # 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved. 1. Introduction Comorbidity has emerged as an important clinical parameter that can be used to decide whether to use neoadjuvant hormone therapy (NHT) in the management of men with prostate cancer (PCa) [1 3]. Recent data demonstrate an interaction between NHT use and comorbidity status such that men with a history of congestive heart failure (CHF) or myocardial infarction (MI) who received NHT were observed to experience a higher risk of all-cause mortality (ACM) compared to those who did not [4]. No studies to date have assessed whether NHT use affects mortality in men with no or at least a single risk factor for coronary artery disease (CAD) stratified by PCa aggressiveness. Because it is well known that HT use can lead to the development of metabolic syndrome [5] and other comorbidities [6 8], we hypothesize that NHT use might negatively affect mortality in men with PCa who present with single or multiple CAD risk factors. However, this effect may be confounded because of the multitude of data demonstrating efficacy of HT use [9 12] within the setting of unfavorable-risk PCa and a lack of evidence supporting efficacy in the setting of low-risk PCa. Therefore, the purpose of the current study was to assess whether NHT use was associated with the risk of ACM in men with low-, intermediate-, and high-risk PCa treated with brachytherapy who have no or at least a single risk factor for CAD. 2. Patients and methods 2.1. Patient characteristics and treatment This retrospective study consisted of men with clinical stage T1 T3,N0,M0 adenocarcinoma of the prostate consecutively treated between September 1991 and September 2006 either at the Chicago Prostate Cancer Center or in one of 20 community-based medical centers within the 21st Century Oncology establishment located in Florida, New York, and North Carolina. Five thousand four hundred forty-one men were categorized as having low-risk disease based on their PCa prognostic factors (prostate-specific antigen [PSA] <10 ng/ml, Gleason score 6, and 2002 American Joint Commission on Cancer [AJCC; 13] clinical stage T1 T2a disease); 4365 men had intermediate-risk disease (PSA ng/ml or Gleason score <8 or clinical stage <T3). The remaining 1360 men had high-risk disease (PSA 20 ng/ml or Gleason score 8 or clinical stage T3). The study was performed with the approval of the institutional review board of the participating institutions. Comorbidity assessment, treatment, and follow-up of patients within this cohort have been previously described [4,14]. Men in the study had no or at least a single risk factor and no documented history of CAD at the time of enrollment. The extent of cardiovascular comorbidity for men in the study ranged from a single CAD risk factor, including diabetes mellitus, hypercholesterolemia, or hypertension alone, to any two or all three of these. All men were treated with or without NHT followed by brachytherapy with or without supplemental external-beam radiation therapy (EBRT) Statistical analysis Comparison of baseline characteristics for each prostate cancer risk group stratified by neoadjuvant hormone therapy use Descriptive statistics were used to define the baseline patient population and tumor and treatment characteristics. For categoric factors, the Mantel-Haenszel x 2 test was used to make comparisons, and the Wilcoxon rank sum test was used for continuous factors Propensity analysis A propensity analysis was performed to account for potential biases in treatment effect arising from nonrandom allocation of patients to different treatment groups [15]. The methodology for the propensity analysis has been previously described [4] Predictors of the risk of all-cause mortality The primary end point of the study was the risk of ACM. The date and cause of death were determined by the attending oncologist or urologist following each patient. Initially, the National Death Index was used to determine whether a patient were deceased. A deceased status was then further verified through a patient s family, physician, and death certificate. The time of survival was defined as the time from the date of brachytherapy to the date of death or, in living patients, last follow-up. Men who died within 1 yr of brachytherapy were excluded to minimize confounding from other, non treatment-related causes of death. For each PCa risk group, Cox regression univariable and multivariable analyses [16] were performed to assess whether NHT use affected the risk of ACM, adjusting for age; year of treatment; supplemental EBRT use; the presence of CAD risk factors; treatment propensity score; and known PCa prognostic factors, including pretreatment PSA level, biopsy Gleason score, and 2002 AJCC [13] clinical stage. NHT and supplemental EBRT use, Gleason score, clinical stage, and the presence of CAD risk factors were analyzed as categoric variables, whereas age, PSA level, and year of brachytherapy were considered continuous variables. groups consisted of treatment without NHT or without supplemental EBRT as well as Gleason score 6 disease, clinical category T1, and no CAD risk factors. An interaction model with NHT use and the presence or absence of CAD risk factors was also used for men with low-risk PCa. For the Cox regression analyses, the assumptions of the proportional hazards model [16] were tested using the time-dependent covariate methodology and assessed via various residuals [16]. No evidence was found to suggest that these assumptions were violated. All statistical tests were two-sided, and p < 0.05 was considered significant. Unadjusted and adjusted hazard ratios (HRs) for the risk of ACM and the associated 95% confidence intervals (CIs) were calculated for each covariate [16] Estimates of all-cause mortality for each prostate cancer risk group stratified by neoadjuvant hormone therapy use The method of Kaplan and Meier [17] was used to calculate overall survival (OS), and estimates of ACM were defined as 1 minus the Kaplan- Meier estimates of OS. For the purpose of illustration, estimates of ACM (adjusted for age and use of supplemental EBRT [18]) stratified by NHT use were graphically displayed. Pairwise comparisons of the distribution

3 Table 1 A comparison of patient, tumor, and treatment characteristics at baseline for the men in the study cohort by prostate cancer risk status and stratified by hormone therapy use Clinical characteristic PCa Risk Status * Low-risk PCa (n = 5441) Intermediate-risk PCa (n = 4365) High-risk PCa (n = 1360) Without HT (n = 3517) With HT (n = 1924) Without HT (n = 2225) With HT (n = 2140) Without HT (n = 353) With HT (n = 1007) PSA level in ng/ml, no. (%) (13.2) 204 (10.6) < (6.1) 149 (7.0) (4.3) 58 (5.8) 0.12 >4 to (86.8) 1720 (89.4) 1027 (46.2) 1002 (46.8) 107 (30.3) 311 (30.9) >10 to (47.7) 989 (46.2) 53 (15.0) 189 (18.8) > (50.4) 449 (44.6) 2002 AJCC [13] clinical stage, no. (%) T (76.0) 1482 (77.0) (57.6) 1091 (51.0) < (47.0) 401 (39.8) <0.01 T2 845 (24.0) 442 (23.0) 943 (42.4) 1049 (49.0) 162 (45.9) 465 (46.2) T3 25 (7.1) 141 (14.0) Gleason score, no. (%) (100.0) 1924 (100.0) 953 (42.8) 823 (38.5) < (31.7) 184 (18.3) < (57.2) 1317 (61.5) 60 (17.0) 226 (22.4) 8 to (51.3) 597 (59.3) Age, no. (%) 60 yr 613 (17.4) 131 (6.8) < (11.2) 105 (4.9) < (7.7) 85 (8.4) yr 1399 (39.8) 758 (39.4) 775 (34.8) 673 (31.5) 117 (33.1) 290 (28.8) yr 1418 (40.3) 954 (49.6) 1078 (48.5) 1207 (56.4) 175 (49.6) 547 (54.3) >80 yr 87 (2.5) 81 (4.2) 122 (5.5) 155 (7.2) 34 (9.6) 85 (8.4) Year of brachytherapy, no. (%) 1999 or sooner 720 (20.5) 535 (27.8) < (24.1) 615 (28.7) < (35.4) 264 (26.2) (60.0) 1241 (64.5) 1216 (54.7) 1347 (62.9) 172 (48.7) 638 (63.4) 2005 or later 687 (19.5) 148 (7.7) 473 (21.3) 178 (8.3) 56 (15.9) 105 (10.4) Treatment without supplemental EBRT, no. (%) 3158 (89.8) 1382 (71.8) < (77.4) 962 (45.0) < (63.7) 253 (25.1) <0.01 Treatment with supplemental EBRT, no. (%) 359 (10.2) 542 (28.2) 504 (22.7) 1178 (55.1) 128 (36.3) 754 (74.9) No CAD risk factors, no. (%) 1731 (49.2) 947 (49.2) (49.7) 1097 (51.3) (54.7) 511 (50.7) 0.97 Diabetes only, no. (%) 89 (2.5) 66 (3.4) 74 (3.3) 92 (4.3) 16 (4.5) 51 (5.1) Hypertension only, no. (%) 1033 (29.4) 626 (32.5) 667 (30.0) 637 (29.8) 79 (22.4) 299 (29.7) Hypercholesterolemia only, no. (%) 191 (5.4) 75 (3.9) 108 (4.9) 66 (3.1) 13 (3.7) 28 (2.8) Any two CAD risk factors, no. (%) 430 (12.2) 193 (10.0) 250 (11.2) 226 (10.6) 46 (13.0) 108 (10.7) All three CAD risk factors, no. (%) 43 (1.2) 17 (0.9) 21 (0.9) 22 (1.0) 6 (1.7) 10 (1.0) Median age in years, no. (IQR) 69.0 ( ) 71.0 ( ) < ( ) 73.0 ( ) < ( ) 72.6 ( ) 0.32 Median PSA in ng/ml, no. (IQR) 6.0 ( ) 6.1 ( ) < ( ) 9.1 ( ) ( ) 14.9 ( ) 0.53 PCa = prostate cancer; HT = hormone therapy; PSA = prostate-specific antigen; AJCC = American Joint Committee on Cancer; EBRT = external-beam radiation therapy; CAD = coronary artery disease; IQR = interquartile range. * Percentages may not sum to 100% because of rounding. 179

4 180 of ACM estimates were assessed using a log-rank test [19]. A Bonferroni correction [20] was used to adjust for multiple comparisons. Given three PCa risk groups and two variables (treatment with or without NHT), there were a total of three pairwise comparisons; therefore, a Bonferroni-corrected significance level of 0.05/3 = was used. All statistical analyses were performed using SAS v.9.2 software (SAS Institute Inc., Cary, NC, USA). diabetes mellitus, hypercholesterolemia, or hypertension alone (n = 4210), any two of these factors (n = 1253), or all three factors (n = 119). No patient had a prior history of CAD. A comparison of patient, tumor, and treatment characteristics at baseline for the men in the study cohort by PCa risk status and stratified by HT use is outlined in Table Results 3.2. Predictors of the risk of all-cause mortality 3.1. Baseline characteristics for men in each prostate cancer risk group stratified by neoadjuvant hormone therapy use The study cohort consisted of men with a median age of 71 yr of age (interquartile range [IQR]: 66 75) treated with brachytherapy; 45.4% (n = 5071) of men also received NHT, and 31.0% (n = 3465) of men received supplemental EBRT. The median duration of NHT use was 4 mo; 5584 men had no CAD risk factors. CAD risk factors consisted of A total of 1531 (13.7%) men died. For men with low-, intermediate-, and high-risk disease, 596, 642, and 293 deaths occurred, respectively, of which 21 (3.5%), 37 (5.8%), and 44 (15.0%), respectively, were the result of PCa. The median follow-up times for men with low-, intermediate-, and high-risk PCa were 4.1 (IQR: ), 4.4 (IQR: ), and 4.6 (IQR: ) yr, respectively. As shown in Tables 2 4, NHT use was associated with a significantly increased risk of ACM in men with low-risk PCa (adjusted Table 2 Unadjusted and adjusted hazard ratios and associated 95% confidence intervals describing the risk of all-cause mortality for patient, tumor, and treatment characteristics from the Cox regression univariable and multivariable analyses for the low-risk prostate cancer risk group Clinical characteristic Low-risk PCa * No. of men No. of events Univariable analysis Multivariable analysis HR Adjusted HR (95% CI) (95% CI) No HT HT < <0.01 ( ) ( ) Age (in years) on day of brachytherapy per-year increase < <0.01 ( ) ( ) No CAD risk factors One or more CAD risk factors ( ) ( ) Log (PSA in ng/ml) per unit increase ( ) ( ) Gleason score AJCC [13] clinical stage T T ( ) ( ) T3 Year of brachytherapy per-year increase ( ) ( ) No supplemental EBRT Supplemental EBRT < <0.01 ( ) ( ) Propensity score per 1% increase < ( ) ( ) PCa = prostate cancer; HR = hazard ratio; CI = confidence interval; HT = hormone therapy; CAD = coronary artery disease; PSA = prostate-specific antigen; AJCC = American Joint Committee on Cancer; EBRT = external-beam radiation therapy; ACM = all-cause mortality. * Using an interaction model for the low-risk PCa group, HT use was associated with a significantly increased risk of ACM in the subgroup of men with at least a single CAD risk factor (adjusted HR: 1.36; 95% CI, ; p = 0.01) but not in the subgroup of men with no CAD risk factors (adjusted HR: 1.19; 95% CI, ; p = 0.13).

5 181 Table 3 Unadjusted and adjusted hazard ratios and associated 95% confidence intervals describing the risk of all-cause mortality for patient, tumor, and treatment characteristics from the Cox regression univariable and multivariable analyses for the intermediate-risk prostate cancer risk group Clinical characteristic Intermediate-risk PCa No. of men No. of events Univariable analysis Multivariable analysis HR (95% CI) Adjusted HR (95% CI) No HT HT < ( ) ( ) Age (in years) on day of brachytherapy per-year increase < <0.01 ( ) ( ) No CAD risk factors One or more CAD risk factors ( ) ( ) Log (PSA in ng/ml) per-unit increase Gleason score ( ) ( ) ( ) ( ) AJCC [13] clinical stage T T ( ) ( ) T3 Year of brachytherapy per-year increase < ( ) ( ) No supplemental EBRT Supplemental EBRT < <0.01 ( ) ( ) Propensity score per 1% increase ( ) ( ) PCa = prostate cancer; HR = hazard ratio; CI = confidence interval; HT = hormone therapy; CAD = coronary artery disease; PSA = prostate-specific antigen; AJCC = American Joint Committee on Cancer; EBRT = external-beam radiation therapy. HR: 1.27; 95% CI, ; p < 0.01) but not in men with intermediate-risk (adjusted HR: 1.13; 95% CI, ; p = 0.15) or high-risk PCa (adjusted HR: 0.86; 95% CI, ; p = 0.28). Using an interaction model for the lowrisk group, NHT use was associated with a significantly increased risk of ACM in the subgroup of men with at least a single CAD risk factor (adjusted HR: 1.36; 95% CI, ; p = 0.01) but not in the subgroup of men with no CAD risk factors (adjusted HR: 1.19; 95% CI, ; p = 0.13) Estimates of all-cause mortality for men in each prostate cancer risk group stratified by neoadjuvant hormone therapy use As illustrated in Figure 1, in the setting of low-risk PCa, estimates of ACM (adjusted for age and use of supplemental EBRT) were significantly higher in men receiving NHT than in those who did not receive NHT (log-rank p = 0.005). However, as shown in Figures 2 and 3, there were no significant differences in the adjusted estimates of ACM when stratified by NHT use for men with intermediate-risk (log-rank p = 0.27) and high-risk PCa (log-rank p = 0.21). 4. Discussion It has recently been reported that NHT use is associated with an increased risk of mortality in men with preexisting CHF or MI [4]. However, this association has not been tested in men with no or at least a single CAD risk factor, as is commonly seen in the general population [21,22]. After adjusting for age, treatment year, supplemental EBRT use, number of CAD risk factors, treatment propensity score, and known PCa prognostic factors, our results demonstrate that for men with at least a single CAD risk factor, NHT use was associated with a significantly increased risk of ACM in the setting of low-risk but not intermediate- or high-risk PCa. This association was not present in men with no CAD risk factors. The clinical significance of these findings is that they further elucidate the potential risks of NHT use in men with PCa, now extending to those with at least a single CAD risk factor but not necessarily men with no CAD risk factors. These results, in conjunction with the well-established role of monotherapy with radical prostatectomy or RT for men

6 182 Table 4 Unadjusted and adjusted hazard ratios and associated 95% confidence intervals describing the risk of all-cause mortality for patient, tumor, and treatment characteristics from the Cox regression univariable and multivariable analyses for the high-risk prostate cancer risk group Clinical characteristic High-risk PCa No of men No. of events Univariable analysis Multivariable analysis HR (95% CI) Adjusted HR (95% CI) No HT HT ( ) ( ) Age (in years) on day of brachytherapy per-year increase < ( ) ( ) No CAD risk factors One or more CAD risk factors ( ) ( ) Log (PSA in ng/ml) per-unit increase ( ) ( ) Gleason score ( ) ( ) < ( ) ( ) 2002 AJCC [13] clinical stage T T ( ) ( ) T ( ) ( ) Year of brachytherapy per-year increase < ( ) ( ) No supplemental EBRT Supplemental EBRT ( ) ( ) Propensity score per 1% increase ( ) ( ) PCa = prostate cancer; HR = hazard ratio; CI = confidence interval; HT = hormone therapy; CAD = coronary artery disease; PSA = prostate-specific antigen; AJCC = American Joint Committee on Cancer; EBRT = external-beam radiation therapy. with low-risk PCa, suggest that multimodality regimens incorporating even short courses of NHT should be approached with caution in those men who also present with one or more CAD risk factors. However, for men with unfavorable-risk PCa and at least a single CAD risk factor requiring HT to maximize their PCa-specific outcomes based on the results of multiple randomized, controlled trials (RCTs) [9 12], the risks of HT use need to be balanced against their benefits in conjunction with the physician treating the underlying PCa, as recently asserted in a joint consensus panel statement by the American Heart Association, American Cancer Society, and the American Urological Association [23]. It is important to underscore that future RCTs using HT in men with PCa will need to be performed with upfront stratification by the presence or absence of CAD or CAD risk factors to provide level I evidence that could definitively demonstrate an interaction between HT use and cardiovascular comorbidity. Furthermore, additional studies will be required to determine whether interventions that address underlying occult CAD or aggressively manage preexisting as well as subsequent CAD risk factors can reduce the negative impact of HT use on mortality in these men. The results of this current retrospective analysis are hypothesis generating, and future prospective trials assessing both ACM and specific cardiovascular end points, including cardiovascular mortality, will be required for validation. Several points require further consideration. First, upon inspection of Figure 1, it becomes evident that the ageadjusted estimates of ACM stratified by NHT use do not diverge until about 4 yr, which is in contrast to what was shown for men with a history of CHF or MI, where these curves separate soon after NHT use [4]. Examples exist in the literature, where short courses of other routine medical therapy can be associated with late cardiovascular events, such as the use of oral contraceptives [24 27] in women with underlying CAD risk factors or the use of high-dose steroids in patients with illnesses such as asthma, chronic obstructive pulmonary disease, or systemic inflammatory conditions [28,29]. Second, it is reasonable to postulate that

7 [(Fig._1)TD$FIG] 183 [(Fig._3)TD$FIG] Fig. 1 Estimates of all-cause mortality (adjusted for age and use of supplemental external-beam radiation therapy) for the 5441 men in the study cohort with low-risk prostate cancer stratified by hormone therapy use. Log-rank = After applying the Bonferroni correction, s <0.017 are significant. HT = hormone therapy. Fig. 3 Estimates of all-cause mortality (adjusted for age and use of supplemental external-beam radiation therapy) for the 1360 men in the study cohort with high-risk prostate cancer stratified by hormone therapy use. Log-rank = After applying the Bonferroni correction, s <0.017 are significant. HT = hormone therapy. men with intermediate- or high-risk PCa did not have an increased risk of ACM, because the established efficacy of HT use on survival [9 12] in this setting negated the potential detrimental impact of HT use. This observation highlights the complex positive and negative interplay of HT use in men with PCa and can explain why in men not stratified by their PCa risk status NHT use was not associated with an increased risk of ACM in those with [(Fig._2)TD$FIG] Fig. 2 Estimates of all-cause mortality (adjusted for age and use of supplemental external-beam radiation therapy) for the 4365 men in the study cohort with intermediate-risk prostate cancer stratified by hormone therapy use. Log-rank = After applying the Bonferroni correction, s <0.017 are significant. HT = hormone therapy. only a single CAD risk factor [4]. The results of Radiation Therapy Oncology Group 9408 demonstrate a survival benefit from HT use in men with favorable-risk PCa [30]. However, the RT dose used is now considered suboptimal, and it is plausible that with contemporary doses of RT including those achieved through a brachytherapy-based approach such a survival advantage may disappear. Third, in all of the major randomized trials assessing RT with or without HT in men with unfavorable-risk PCa, there has been a benefit to the addition of HT [9 12]. It is important to keep in mind that only a minority of men in these studies had cardiovascular comorbidity, and no randomized trial to date has examined the impact of HT specifically on the subgroup of men with either known CAD or risk factors for CAD. In a recent postrandomization analysis of a trial comparing men with unfavorable-risk PCa treated with RT with or without 6 mo of HT, those men with moderate to severe comorbidity who were treated with HT had a nonsignificant trend toward an increased risk of ACM compared to men treated with RT alone [1]. Fourth, the ascertainment of CAD risk factors was limited to chart reviews and patient self-reporting, which are not always reliable. In addition, this study lacked information regarding family history for CAD and comorbidities unrelated to CAD and therefore was not able to adjust for such variables. Future studies should ensure the use of specific prospective guidelines in defining individual comorbidities and be inclusive of family history for CAD. Fifth, given that men with unfavorable-risk PCa are sometimes treated with longer durations of HT, it will be important in future studies to determine whether such extended durations have the potential to overwhelm the benefits of HT use in men who present with coexisting CAD risk factors. Finally, it is

8 184 important to consider that brachytherapy was administered over a 15-yr period across multiple centers by several physicians. This poses a major limitation to this retrospective analysis, even though year of brachytherapy was included as a covariate within the multivariable analysis. 5. Conclusions This study demonstrated that for men with no or at least a single risk factor for CAD, NHT use is associated with an increased risk of ACM in the setting of low-risk but not intermediate- or high-risk PCa. This effect is driven by the subgroup of men with at least a single risk factor for CAD. These results should heighten our awareness of the potential adverse effects of NHT in low-risk PCa and warrant prospective validation given the widespread use of NHT for prostate downsizing prior to brachytherapy. Author contributions: Akash Nanda had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Nanda, D Amico. Acquisition of data: Moran, Braccioforte, Dosoretz, Salenius, Katin, Ross. Analysis and interpretation of data: Nanda, Chen, D Amico. Drafting of the manuscript: Nanda, D Amico. Critical revision of the manuscript for important intellectual content: Nanda, Chen, D Amico. Statistical analysis: Nanda, Chen, D Amico. Obtaining funding: None. Administrative, technical, or material support: D Amico. Supervision: Nanda, D Amico. Other (specify): None. Financial disclosures: Akash Nanda certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: None. s [1] D Amico AV, Chen MH, Renshaw AA, Loffredo M, Kantoff PW. Androgen suppression and radiation vs radiation alone for prostate cancer: a randomized trial. JAMA 2008;299: [2] Pagliarulo V, Bracarda S, Eisenberger MA, et al. Contemporary role of androgen deprivation therapy for prostate cancer. Eur Urol 2012;61: [3] Heidenreich A, Bellmunt J, Bolla M, et al. EAU guidelines on prostate cancer. Part 1: screening, diagnosis, and treatment of clinically localised disease. Eur Urol 2011;59: [4] Nanda A, Chen MH, Braccioforte MH, Moran BJ, D Amico AV. Hormonal therapy use for prostate cancer and mortality in men with coronary artery disease-induced congestive heart failure or myocardial infarction. JAMA 2009;302: [5] Braga-Basaria M, Dobs AS, Muller DC, et al. Metabolic syndrome in men with prostate cancer undergoing long-term androgendeprivation therapy. J Clin Oncol 2006;24: [6] Keating NL, O Malley AJ, Freedland SJ, Smith MR. Does comorbidity influence the risk of myocardial infarction or diabetes during androgen-deprivation therapy for prostate cancer? Eur Urol 2013; 64: [7] Hu JC, Williams SB, O Malley AJ, Smith MR, Nguyen PL, Keating NL. Androgen-deprivation therapy for nonmetastatic prostate cancer is associated with an increased risk of peripheral arterial disease and venous thromboembolism. Eur Urol 2012;61: [8] Azoulay L, Yin H, Benayoun S, Renoux C, Boivin JF, Suissa S. Androgen-deprivation therapy and the risk of stroke in patients with prostate cancer. Eur Urol 2011;60: [9] Denham JW, Steigler A, Lamb DS, et al. Short-term androgen deprivation and radiotherapy for locally advanced prostate cancer: results from the Trans-Tasman Radiation Oncology Group randomised controlled trial. Lancet Oncol 2005;6: [10] Pilepich MV, Winter K, John MJ, et al. Phase III Radiation Therapy Oncology Group (RTOG) trial of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate. Int J Radiat Oncol Biol Phys 2001;50: [11] D Amico AV, Manola J, Loffredo M, Renshaw AA, DellaCroce A, Kantoff PW. 6-month androgen suppression plus radiation therapy vs radiation therapy alone for patients with clinically localized prostate cancer: a randomized controlled trial. JAMA 2004;292: [12] Bolla M, Collette L, Blank L, et al. Long-term results with immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC study): a phase III randomised trial. Lancet 2002;360: [13] Greene FL, Page DL, Fleming ID, et al. AJCC cancer staging manual. ed. 6. Philadelphia, PA: Lippincott; [14] Dosoretz AM, Chen MH, Salenius SA, et al. Mortality in men with localized prostate cancer treated with brachytherapy with or without neoadjuvant hormone therapy. Cancer 2010;116: [15] Newgard CD, Hedges JR, Arthur M, Mullins RJ. Advanced statistics: the propensity score a method for estimating treatment effect in observational research. Acad Emerg Med 2004;11: [16] Klein JP, Moeschberger ML. Survival analysis. ed. 2. New York, NY: Springer-Verlag; [17] Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53: [18] Cupples LA, Gagnon DR, Ramaswamy R, D Agostino RB. Ageadjusted survival curves with application in the Framingham Study. Stat Med 1995;14: [19] Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966;50: [20] Shaffer JP. Multiple hypothesis-testing. Ann Rev Psych 1995;46: [21] Greenland P, Knoll MD, Stamler J, et al. Major risk factors as antecedents of fatal and nonfatal coronary heart disease events. JAMA 2003;290: [22] Khot UN, Khot MB, Bajzer CT, et al. Prevalence of conventional risk factors in patients with coronary heart disease. JAMA 2003;290: [23] Levine GN, D Amico AV, Berger P, et al. Androgen-deprivation therapy in prostate cancer and cardiovascular risk: a science advisory from the American Heart Association, American Cancer Society, and American Urological Association: endorsed by the American Society for Radiation Oncology. Circulation 2010;121: [24] Tanis BC, van den Bosch MA, Kemmeren JM, et al. Oral contraceptives and the risk of myocardial infarction. N Engl J Med 2001;345: [25] Acute myocardial infarction and combined oral contraceptives: results of an international multicentre case-control study. WHO

9 Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Lancet 1997;349: [26] Beral V, Hermon C, Kay C, Hannaford P, Darby S, Reeves G. 185 [28] Wei L, MacDonald TM, Walker BR. Taking glucocorticoids by prescription is associated with subsequent cardiovascular disease. Ann Intern Med 2004;141: Mortality associated with oral contraceptive use: 25 year [29] Petri M, Lakatta C, Magder L, Goldman D. Effect of prednisone and follow up of cohort of women from Royal College of hydroxychloroquine on coronary artery disease risk factors in General Practitioners oral contraception study. BMJ 1999;318: systemic lupus erythematosus: a longitudinal data analysis. Am J Med 1994;96: [27] Rosenberg L, Palmer JR, Rao RS, Shapiro S. Low-dose oral contracep- [30] Jones CU, Hunt D, McGowan DG, et al. Radiotherapy and short-term tive use and the risk of myocardial infarction. Arch Intern Med androgen deprivation for localized prostate cancer. N Engl J Med 2001;161: ;365: