In their article, Chaime et al addressed an important question: the association between exposure to Agent Orange and 2,3,7,8-tetrachlorodibenzodioxin (TCDD) and the development of prostate cancer in American Vietnam veterans.1 TCDD is a known human carcinogen.2-4 The exposure history/questionnaire method used by Chamie et al has been shown repeatedly to be unreliable in categorizing dioxin exposure.5-9 The article by Chamie et al and the editorial by Shah and Terris10 fail to account for major weaknesses of their exposure assessment methodology, which undermines the validity of their conclusions.
Beginning in 1973, biomonitoring involving congener-specific measurement of TCDD and other dioxins in blood, milk, or adipose tissue became the “gold standard” for the assessment of dioxin exposure.4, 11, 12 In the early 1980s, the Centers for Disease Control and Prevention measured serum dioxin concentrations, using high-resolution gas chromatography/high-resolution mass spectrometry (HRGC/HRMS), in 646 enlisted ground troops who had served in areas that were heavily sprayed with Agent Orange and 97 controls who had not served in Vietnam. Dioxin concentrations were found to be similar in both groups. The CDC concluded that there was no measurable evidence of excess dioxin exposure in the ground troops.5
The Air Force studied veterans of Operation Ranch Hand, who were responsible for spraying most of the herbicides, including Agent Orange, over portions of South Vietnam. Unlike the ground troops, serum measurements of TCDD in these veterans, obtained in 1987, demonstrated a wide range of levels, from background to greater than 400 parts per trillion lipid, even decades after exposure.13 Continued analyses of the serum taken from the Operation Ranch Hand cohort clearly documented elevated TCDD exposure in enlisted personnel who had close contact with Agent Orange (eg, those who regularly cleaned spray tanks) versus exposure to such low levels if exposed at all that no elevation was noted in other occupational groups, such as officers.14, 15 When examining the morbidity of the Operation Ranch Hand cohort,16 associations between TCDD and all-site cancer and prostate cancer were reported after stratification based on years of service (before 1969, when the TCDD levels in Operation Ranch Hand veterans were highest, vs after 1969) and days of spraying (at least 30 days).15, 16
One argument against measuring serum dioxin is the expense of congener-specific dioxin analysis. To minimize costs, a National Institute for Occupational Safety and Health (NIOSH) worker study developed exposure indices and validated them with serum dioxin measurements in a carefully selected subsample of workers 35 years after exposure ended.17 Another argument against using biomonitoring in Vietnam veterans has been that although TCDD has a biologic half-life of 7 to 10 years, too many years have passed to accurately categorize exposure from the Vietnam conflict.1 This is not the case for highly exposed persons, as noted earlier, or when unique congener patterns document exposure. Serum dioxin measurements were recently used to categorize exposure to US Army Chemical Corps workers who sprayed herbicides around base camps during the war.18 Other studies have demonstrated elevated blood levels of TCDD (and sometimes other dioxins) as long as 30 to 40 years after exposure among workers19, 20 and some Vietnamese.21, 22
Assuming self-reported exposure resulted in serious misclassification of Agent Orange/TCDD exposure in the study by Chamie et al, that would lead to an attenuation of observed associations but does not explain the strong positive association observed. Chamie et al suggest the strong positive association they noted was because of the older age of their cohort compared with that of the Operation Ranch Hand cohort. However, the mean age of the Operation Ranch Hand cohort in 2003, at the time of data analysis for the prostate cancer study, was 63 years, which is slightly older than the mean age of 61 years of the California Vietnam veterans in 2006.16 Because of the availability of electronic records, Chamie et al chose to include only prostate cancer cases diagnosed after 1998. In the 17 years between 1982 and 1999, 101 prostate cancers were diagnosed among 2516 members of the Operation Ranch Hand study.16 If a similar rate of prostate cancer occurred in the cohort in the study by Chamie et al, approximately 500 prostate cancer cases may have not been included in their analysis because of the early cutoff date, thereby limiting their conclusions.
The study by Chamie et al is valuable in that it focuses attention on the issue of exposure to chemicals and prostate cancer in Vietnam veterans. The authors suggest that self-reported exposure to Agent Orange and its dioxin contaminant might be associated with a risk of prostate cancer that is twice than normal. This conclusion is biologically plausible based on studies in laboratory animals and is consistent with some epidemiologic studies in humans.23, 24 However, accurate methods for assessing exposure must be used. Application of biologic measurements in at least a representative sample would vastly enhance and validate the findings of the study by Chamie et al.1 Lastly, the findings of elevated levels of other toxic chemicals in the Vietnamese environment, including dibenzofurans and pesticides, as well as an association between time spent in Southeast Asia and cancer, suggest that focusing only on Agent Orange exposure with respect to health in veterans is no longer appropriate.22, 25
This letter has been reviewed by US Environmental Protection Agency but represents the personal views of the authors and is not Agency policy.