Prostate Cancer Control And Survival In Vietnam Veterans Exposed To Agent Orange

Methods and Material From May 1995 to January 2005, 81 Vietnam veterans (29 with Agent Orange exposure and 52 without) and 433 nonveterans of comparable age (mean age, 58 years) underwent prostate brachytherapy. The mean follow-up was 5.0 years. Biochemical progression-free survival (bPFS) was defined as a prostate-specific antigen (PSA) ≤ 0.40 ng/mL after nadir. Patients with metastatic prostate cancer or hormone refractory disease without obvious metastases who died of any cause were classified as died of prostate cancer. All other deaths were attributed to the immediate cause of death. Multiple parameters were evaluated for impact on survival. Results At 9 years, Agent Orange–exposed men were least likely to remain biochemically controlled (89.5%, 100%, and 97.2% in Agent Orange–exposed, nonexposed veterans, and nonveterans, respectively, p = 0.012). No significant differences in cause-specific (CSS) ( p = 0.832) or overall survival (OS) ( p = 0.363) were discerned. In multivariate analysis, CSS was best predicted by Gleason Score and day 0 D 90 , whereas Gleason Score, % positive biopsies, and D 90 predicted for bPFS. None of the evaluated parameters predicted for OS, however, a trend was identified for better OS in younger patients and those with a higher D 90 . In addition, Agent Orange exposure did not predict for any of the survival parameters. To date, 22 patients have died (metastatic prostate cancer two, second malignancies nine, cardiovascular disease eight, trauma two, and pulmonary one). Conclusions In this cohort of prostate brachytherapy patients, Agent Orange exposure did not statistically impact survival in multivariate analysis. Keywords Prostate cancer Agent Orange Vietnam veterans Brachytherapy Introduction Agent Orange, a defoliant widely used during the Vietnam War, contains the carcinogen dioxin. Conflicting reports have accumulated regarding Agent Orange exposure and the subsequent development of prostate cancer in Vietnam War veterans (1–8) . In both military personnel and occupational workers, an association between Agent Orange exposure and prostate cancer remains controversial. Concerns over potential effects of dioxin in the US and Australian veterans exposed to herbicides during the Vietnam War have resulted in several studies examining cancer mortality, including prostate cancer (1) . An excess of prostate cancer relative to the general population was reported in Australia's Vietnam veterans (6) and in a pilot case–control study in Michigan (8) , whereas no increase in prostate cancer was observed in other early studies of US veterans (2,3) . Initial results from the Air Force Health Study (AFHS) did not reveal an excess of prostate cancer. However, a followup study reported an excess of prostate cancer compared with the US white male population (4) . Pavuk et al reported the risk of prostate cancer increased with years of service in Southeast Asia but not with exposure to dioxin (9) . In addition, studies evaluating United States Air Force veterans of Operation Ranch Hand reported no increased risk of prostate cancer mortality or morbidity after a maximum 10-year follow-up (1,4,10) . The shortcomings of these veteran studies include the small numbers of prostate cancer cases and a lack of accurate individual exposure assessment. In a prostate biopsy series, Zafar and Terris reported the absence of a relationship between prostate cancer and Agent Orange exposure (5) . No statistically significant differences between Agent Orange and non–Agent Orange patients were noted about PSA, prostate cancer diagnosis, proportion of well-differentiated cancer or length of cancer in the biopsy cores (5) . After radical prostatectomy, Shah et al reported that Agent Orange–exposed patients were younger at diagnosis with earlier stage disease and lower preoperative PSA with similar biopsy Gleason scores (11) . Agent Orange exposure did not predict for Gleason 4 + 3 or higher histology, extracapsular extension, positive surgical margins, or seminal vesicle invasion. Despite these findings, Agent Orange–exposed patients had a greater risk of biochemical progression (11) . Contradicting results have also been reported after occupational exposures. One study reported an increased cancer risk in factory workers exposed to Agent Orange (12) . In contrast, a second quantitative study evaluating long-term mortality in 2,187 Dow Chemical Company workers exposed to substantial levels of dioxin revealed no increased risk of cancer (13) . As Vietnam Veterans age, prostate cancer will markedly increase with expectant apprehension regarding the impact of Agent Orange exposure on outcomes after therapeutic intervention including brachytherapy. In this study, we evaluated clinical, biochemical, and pathologic features along with biochemical control (bPFS), cause-specific survival (CSS) and overall survival (OS) in Vietnam veterans with and without Agent Orange exposure and in a nonveteran cohort of comparable age. Methods and materials From May 1995 to January 2005, 81 Vietnam veterans (29 with Agent Orange exposure determined by eligibility for Veterans Administration benefits and 52 without Agent Orange exposure) underwent permanent prostate brachytherapy by one of the authors (GSM) at the Schiffler Cancer Center. These 81 Vietnam veterans were compared with 433 consecutive nonveterans of comparable age undergoing brachytherapy during the same time interval. All patients were born between March 1937 and September 1953. All patients underwent brachytherapy more than 3 years before analysis. Before the formulation of a treatment plan, all biopsy slides were reviewed by a single pathologist (EA). Preplanning technique, intraoperative approach, and dosimetric evaluation have been described in detail (14–16) . Table 1 summarizes the clinical, treatment, and dosimetric parameters of the patient population stratified by veteran and Agent Orange status. Differences in presentation, treatment, and dosimetry were evaluated across the three cohorts for identification of statistically significant differences. No patient underwent seminal vesicle biopsy or pathologic lymph node staging. Risk group classification was as follows: low risk (PSA ≤ 10 ng/mL, Gleason score ≤ 6, and clinical stage ≤ T2a), intermediate risk (one adverse factor: PSA 10.1–19.9 ng/mL or Gleason score 7 or clinical stage T2b) and high risk (PSA ≥ 20 ng/mL or Gleason score ≥ 20 ng/mL or Gleason score ≥ 8 or clinical stage ≥ T2c). Of the 514 patients, 349 (67.9%) were hormone naïve and 165 (32.1%) received androgen deprivation therapy (ADT). Ninety-seven patients received short course (≤6 months ADT) and 68 received a prolonged course (>6 months ADT). ADT was initiated 3 months before implantation and consisted of a leutinizing hormone–releasing hormone agonist and an antiandrogen. Median ADT duration in short and extended course regimens were 4 and 12 months, respectively (range, 3–36 months). Of the 514 patients, 248 (48.2%) received supplemental external beam radiation therapy (XRT). In general, patients received 45 Gy in 1.8 Gy fractions using 15–18 MV photons delivered via a four-field conformal technique (opposed lateral and anteroposterior/posteroanterior). For patients with <10% risk of pelvic lymph node involvement, the target volume consisted of the prostate gland and seminal vesicles. For patients with >10% risk of pelvic lymph node involvement, the pelvic lymph nodes were also included in the target volume. The Partin tables were used to estimate pelvic lymph node risk (17) . In all cases, supplemental XRT was delivered before brachytherapy. The brachytherapy target volume consisted of the prostate gland with periprostatic treatment margins, including the proximal 1.0 cm of the seminal vesicles (14–16) . The minimal peripheral dose was delivered to the target area with the margin. Standard implant doses were used for monotherapy and boost regimens. At implantation, the prostate gland, periprostatic region, and base of the seminal vesicles were implanted (14–16) . Patients were monitored by physical examination, including digital rectal examination and serum PSA determinations at 3–6 months intervals. The endpoint of the analysis was cause-specific survival, biochemical progression-free survival (definition: PSA ≤ 0.40 ng/mL after nadir), and overall survival (18) . Multiple clinical, treatment, and dosimetric parameters were evaluated for impact on CSS, bPFS, and OS. Cause of death was determined for each deceased patient. Patients with metastatic prostate cancer or hormone refractory disease without obvious metastases who died of any cause were classified as died of prostate cancer. All other deaths were attributed to the immediate cause of death. Independent samples were tested by t -test and Fischer's exact test and Chi-squared test were applied to the clinical, treatment, and dosimetric parameters of the three patient cohorts to determine the significance of the differences. Selected parameters were included in a multivariate forward conditional Cox regression to identify predictors for CSS, bPFS, and OS. Kaplan–Meier survival analyses were applied to the data to generate survival curves and log-rank comparisons were used to determine significance between the curves of the three patient cohorts. All data were analyzed using Statistical Package for Social Sciences, Version 12.0, Software (SPSS, Chicago, IL). Statistical significance was set at p ≤ 0.05 for all analyses. Results Table 1 summarizes the clinical, treatment, and dosimetric parameters stratified by veteran and Agent Orange status. The median patient age at the time of implantation was 58.7 years with overall mean and median follow-up of 5.0 ± 2.4 and 4.5 years, respectively. Patients with Agent Orange exposure were statistically younger with a shorter follow-up and a statistically higher pretreatment PSA when compared with Vietnam veterans without Agent Orange exposure and nonveterans. There were no additional statistical differences between the three cohorts including Gleason score, percent positive biopsies, BMI, prostate volume, Day 0 postimplant dosimetry, clinical stage, isotope, use of XRT and/or ADT, perineural invasion, risk group or concomitant medical morbidities to include hypertension, diabetes, or tobacco use. For biochemically controlled patients, the median post-treatment PSA was <0.04 ng/mL for all three cohorts. Figures 1–3 illustrate the Kaplan–Meier CSS, bPFS, and OS stratified by veteran status and Agent Orange exposure. At 9 years, the CSS was 99.5%, 100%, and 100% for nonveterans, Agent Orange–exposed veterans and non–Agent Orange–exposed veterans, respectively ( p = 0.832). In contrast, 10 nonveterans, zero nonexposed veterans and three Agent Orange–exposed veterans developed biochemical progression with a resultant greater risk of biochemical progression in Agent Orange–exposed veterans (89.5%) when compared with nonveterans (97.2%) or nonexposed veterans (100%) ( p = 0.012). The difference in the biochemical progression-free curves was primarily a result of early failures in the Agent Orange–exposed cohort. Despite an increased risk in biochemical failure, to date, no statistical differences in overall survival have been noted between the three cohorts ( Fig. 3 , p = 0.363). In univariate and multivariate analysis, neither Agent Orange exposure nor veteran status predicted for CSS, bPFS, or OS ( Table 2 ). Cause-specific survival was best predicted by Gleason score and Day 0 D 90 , whereas biochemical progression-free survival was best predicted by Gleason score, percent positive biopsies, and D 90 . Biochemical progression-free survival for Gleason scores 5–6, 7, and 8–9 were 98.5%, 98.7%, and 80.7% ( p < 0.001). None of the evaluated parameters predicted for overall survival. However, a trend for better overall survival was noted in younger patients and those with a higher Day 0 D 90 . To date, 22 patients (4.3% of the study population) have died. Of the 22 patients, 20 were nonveterans and two were Agent Orange–exposed veterans. No non–Agent Orange–exposed veteran has died. The two Agent Orange–exposed veterans died of trauma and lung cancer, whereas the 20 nonveterans died of the following causes: metastatic prostate cancer two, cardiovascular disease eight, second malignancies eight, pulmonary one, and trauma one. Discussion Agent Orange is composed of multiple chemical compounds, including dioxin. Dioxins are synthetic, lipophilic, toxic chemicals with the most toxic of these being 2.3.7.8-tetrachlorodibenzo-p-dioxin (TCPD) (19) . The exact mechanism of any of the components of Agent Orange remains unknown. However, it is of great health concern that dioxin is detected in adipose tissue and blood samples of Vietnam veterans 18 years after exposure (20) . The potential causative role of Agent Orange in prostate cancer remains controversial with conflicting reports regarding Vietnam veterans and occupationally exposed workers (1–8,12,13) . In a prostate biopsy study, Zafar and Terris reported the absence of a significant relationship between prostate cancer and Agent Orange exposure without statistically significant differences in prebiopsy PSA, prostate cancer diagnosis, proportion of well-differentiated cancer or length of cancer in the biopsy cores (5) . In our study, although no statistically significant difference in biopsy Gleason scores were identified, patients with Agent Orange exposure presented with statistically higher pretreatment PSA ( Table 1 ). In the SEARCH database, Shah et al evaluated 1,400 patients undergoing radical prostatectomy in three Veterans Affairs Hospitals (11) . Ninety patients with self-reported Agent Orange exposure were identified. Patients with Agent Orange exposure were younger with earlier stage disease, lower preoperative PSA, and similar biopsy scores. In our series, Agent Orange–exposed patients were also younger than the other two cohorts ( Table 1 ). In the SEARCH database, Agent Orange–exposed patients were at greater risk of biochemical progression which is consistent with our brachytherapy findings using log-rank comparisons ( Fig. 2 ). However, in multivariate Cox regression analysis, Agent Orange did not affect bPFS, CSS, or OS ( Table 2 ). The shortcomings of our study include the small number of men with Agent Orange exposure and the inability to quantify Agent Orange exposure. An additional shortcoming of all Agent Orange studies is the number of prostate cancer cases in this age cohort will increase in future years because of the young age of these patients. A strength of our study, however, is that Agent Orange exposure was determined by eligibility for Veterans Administration benefits. Evaluation of large numbers of Agent Orange–exposed veterans will be mandatory to determine whether or not these men have a greater incidence of prostate cancer and/or a more virulent clinical course. However, in our series, a multivariate analysis failed to establish a relationship between Agent Orange exposure and poorer cause-specific, biochemical progression-free, and/or overall survival. Conclusion In this cohort of prostate brachytherapy patients, Agent Orange exposure did not have a statistically significant effect on survival in multivariate analysis. Most deaths were the result of second malignancies or cardiovascular disease. References [1] M. Pavuk J.E. Michalek N.S. Ketchum Prostate cancer in US Air Force veterans of the Vietnam War J Exposure Sci Environ Epidem 16 2006 184 190 [2] K.K. Watanabe H.K. Kang Mortality patterns among Vietnam veterans: a 24-year retrospective analysis J Occup Environ Med 38 1996 272 278 [3] R.W. Capp Update of cancer surveillance of veterans in Massachusetts, USA Int J Epidemiol 26 1997 679 681 [4] N.S. Ketchum J.E. Michalek E. Burton Serum dioxin and cancer in veterans of operation Ranch Hand Am J Epidemiol 143 1999 630 639 [5] M.B. Zafar M.K. Terris Prostate cancer detection in veterans with a history of Agent Orange exposure J Urol 166 2001 100 103 [6] P.J. Crane D.L. Bernard K.W. Horsley Mortality of Vietnam veterans: the veteran cohort study. A report of the 1996 cohort study of Australian Vietnam veterans 1997 Department of Veterans Affairs Canberra [7] F.Z. Akhtar H.D. Garabrant N.S. Ketchum Cancer in US Air Force veterans of the Vietnam War J Occup Environ Health 46 2004 123 136 [8] V.N. Giri A.E. Casidy J. Beebe-Dimmer Association between Agent Orange and prostate cancer: a pilot case-control study Urol 63 2004 757 760 [9] M. Pavuk J.E. Michalek A. Schecter Did TCDD exposure or service in Southeast Asia increase the risk of cancer in Air Force Vietnam veterans who did not spray Agent Orange? J Occup Environ Med 47 2005 335 342 [10] J.E. Michalek N.S. Ketchum F.Z. Akhtar Post-service mortality of US Air Force veterans occupationally exposed to herbicides in Vietnam: 15-year follow up Am J Epidemiol 148 1998 786 792 [11] S.R. Shah S.J. Freedland W.J. Aronson Agent Orange exposure and risk of advanced disease and biochemical progression after radical prostatectomy: results from SEARCH database J Urol 175 2006 42 [12] A. Manz J. Berger J.H. Dwyer Cancer mortality among workers in chemical plant contaminated with dioxin Lancet 338 1991 959 964 [13] K.M. Bodner J.J. Collins L.J. Bloemen Cancer risk for chemical workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin Occup Environ Med 60 2003 672 675 [14] G.S. Merrick W.M. Butler Modified uniform seed loading for prostate brachytherapy: rationale, design and evaluation Tech Urol 6 2000 78 84 [15] G.S. Merrick W.M. Butler A.T. Dorsey Seed fixity in the prostate/periprostatic region following brachytherapy Int J Radiat Oncol Biol Phys 46 2000 215 220 [16] G.S. Merrick W.M. Butler K.E. Wallner Dosimetry of an extracapsular annulus following permanent prostate brachytherapy Am J Clin Oncol 30 2007 228 233 [17] A.W. Partin L.A. Mangold D.M. Lamm Contemporary update of prostate cancer staging nomograms (Partin tables) for the new millennium Urol 58 2001 843 848 [18] G.S. Merrick W.M. Butler K.E. Wallner Androgen deprivation therapy does not impact cause-specific or overall survival in high-risk prostate cancer managed with brachytherapy and supplemental external beam Int J Radiat Oncol Biol Phys 68 2007 34 40 [19] A. Schecter H. McGee J.S. Stanley Dioxins and Dioxin-like chemicals in blood and semen of American Vietnam Veterans from the state of Michigan Am J Ind Med 30 1996 647 654 [20] A.J. Schecter J.J. Ryan J.D. Constable Partitioning of 2.3.7.8-chlorinated dibenzo-p-dixons and dibenzofurans between adipose tissue and plasma lipid of 20 Massachusetts Vietnam veterans Chemosphere 20 1990 951 958