Veterans and Agent Orange: Update 2012 (2014)

14


Conclusions and Recommendations

SYNOPSIS OF COMMITTEE CONCLUSIONS

The committee weighed the strengths and limitations of the epidemiologic evidence reviewed in its report and in previous Veterans and Agent Orange (VAO) reports. Although the studies published since Update 2010 are the subject of detailed evaluation here, the committee drew its conclusions in the context of the entire body of literature. The contribution of recent publications to the evidence database was considerable, but the committee did not weigh them more heavily merely because they were new. Epidemiologic methods and analytic capabilities have improved, but many of the recent studies were also particularly useful for the committee’s purpose because they produced results in terms of serum 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) concentrations or total toxic equivalents (TEQs), which take into account exposure to all dioxin-like chemicals, or because their findings consisted of observations on the aging population of primary concern, Vietnam veterans. The committee also notes that experimental data related to biologic plausibility of health conditions statistically associated with exposure to the components of Agent Orange have gradually emerged since the beginning of this series of VAO reports. The findings now do more to inform decisions about how to categorize the degree of association for individual conditions, so the committee for Update 2008 added a footnote to this effect to its summary tables. The committee for Update 2010 added a notation to Table 14-1 indicating the correspondence of the lymphohematopoietic cancers (LHCs) that have been found to have evidence of an association with herbicide exposure to the biologic understanding of the clonal derivation of LHCs that is the basis of the World Health Organization’s classification system for these neoplasms. The

TABLE 14-1 Summary of Ninth Biennial Update of Findings on Vietnam-Veterans, Occupational, and Environmental Studies Regarding Scientifically Relevant Association^aBetween Exposure to Herbicides and Specific Health Outcomes^b

^aThis change in wording was made to emphasize the scientific nature of the VAO task and procedures and reflects no change in the present committee’s criteria from those used in previous updates.

^bHerbicides indicates the following chemicals of interest: 2,4-dichlorophenoxyacetic acid (2,4-D), 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) and its contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD or dioxin), cacodylic acid, and picloram. The evidence regarding association was drawn from occupational, environmental, and veteran studies in which people were exposed to the herbicides used in Vietnam, to their components, or to their contaminants.

^cEvidence of an association is strengthened by experimental data supporting biologic plausibility, but its absence would not detract from the epidemiologic evidence.

^∗The committee notes the consistency of these findings with the biologic understanding of the clonal derivation of lymphohematopoietic cancers that is the basis of the World Health Organization classification system.

current committee did not modify the criteria used by previous VAO committees to assign categories of association to particular health outcomes, but decided to refer to the object of its evaluation as “scientifically relevant association” in the title of Table 14-1 to clarify that the strength of evidence evaluated, based on the quality of the scientific studies reviewed, was a fundamental component of the committee’s deliberations to address the imprecisely defined legislative target of “statistical association.”

On the basis of its evaluation of Vietnam veterans, occupational, and environmental studies, the committee assigned each health outcome to one of four categories of relative certainty of association with exposure to the herbicides that were used in Vietnam or to any of their components or contaminants (with no intention of specifying particular chemicals). This committee’s findings were the same as those of the committee for Update 2010 with a single exception: the committee voted unanimously to move stroke to the limited and suggestive category because of new evidence showing a statistically significant association of stroke with exposure to dioxin-like chemicals in the well-designed Prospective Investigation of the Vasculature in Uppsala Seniors; evidence of an overall increase in stroke or cerebrovascular disease associated with exposure to the chemicals of interest in Vietnam veteran, occupational, and environmental populations in the most relevant of previously considered studies; demonstrated biologic plausibil-

ity in human and animal studies; and the strong connection between stroke and hypertension, cardiovascular disease, and diabetes (three conditions already in the limited and suggestive category). The published data did not permit the committee to distinguish hemorrhagic from ischemic stroke, but given that only a small percentage of strokes are of the hemorrhagic type in Western populations, that was not seen to be an impediment. This change made by the current committee to the categorizations determined by the committee for Update 2010 (as presented in Table 1-1) is noted in boldface in Table 14-1.

Although the Department of Veterans Affairs (VA) did not find hypertension to be presumptively related to service in Vietnam (VA, 2010), on the basis of the total weight of available evidence the current committee reaffirmed the conclusion of the committees for Update 2006, for Update 2008, and for Update 2010 to categorize hypertension as having limited or suggestive evidence of association.

As mandated by Public Law (PL) 102-4, the distinctions among categories are based on statistical association, not on strict causality. The committee was directed to review the scientific data, not to recommend VA policy; therefore, conclusions reported in Table 14-1 are not intended to imply or suggest policy decisions. The conclusions are related to associations between exposure and outcomes in human populations, not to the likelihood that any individual’s health problem is associated with or caused by the chemicals in question.

COMMITTEE RECOMMENDATIONS

As part of its charge, the committee was asked to offer recommendations concerning the need, if any, for additional scientific studies to resolve uncertainties concerning the health effects of the chemicals of interest sprayed in Vietnam: 2,4-dichlorophenoxyacetic acid (2,4-D), 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) and its contaminant TCDD, picloram, and cacodylic acid. This chapter summarizes the committee’s recommendations.

Although progress continues to be made in understanding the health effects of exposure to the chemicals of interest and in elucidating the mechanisms underlying them, gaps in our knowledge remain. The scope of potential research on the chemicals is far-reaching, and what follows here is not an exhaustive list of future research that might have value. Many additional opportunities for progress in such fields as toxicology, the conduct of continuing or additional epidemiologic studies, and systematic and comprehensive integration of existing data have not been explicitly noted here. The committee for Update 2010 offered a high-priority recommendation that research be undertaken to address chronic obstructive pulmonary disease (COPD) from a morbidity perspective with appropriate functional diagnosis of COPD and collection of data to permit adjustment for smoking and other relevant confounders; the current committee is pleased to note that VA has started such a study on its Army Chemical Corps (ACC) cohort. Several of the recommendations detailed below were included in previous VAO

updates but have been included here to reiterate the committee’s conviction that more progress should be made in the research fields noted.

•  VA should evaluate possibilities for studying health outcomes in Vietnam-era veterans by using the existing administrative and health-services databases.

The original VAO committee recommended that the Department of Defense (DOD) and VA identify Vietnam service in the computerized index of records. Linking that information with the VA electronic medical-record and associated administrative databases, such as discharge-diagnosis and pharmacy-use records, should make it possible to assemble epidemiologic information on common health conditions for evaluation of possible associations with military service in Vietnam. Particular attention should be paid to the feasibility of conducting epidemiologic studies of conditions that have been noted to be of special interest but on which the current evidence is inadequate or insufficient to determine whether there is an association with herbicide exposure, such as COPD, brain cancer, tonsil cancer, melanoma (with particular attention given to ocular types), stroke, and Alzheimer disease. For very uncommon health outcomes, a case-control design would probably be most appropriate. Creative analysis of VA’s own data resources may well be the most effective way to address those outcomes and to gain a better understanding of the role of herbicide exposure in development of stroke, prostate cancer, and Parkinson disease in Vietnam veterans.

VA could possibly use its medical databases more effectively, particularly if there is concern about a perceived conflict of interest in surveying its own databases, by involving external analysts. For example, an independent panel could be commissioned to identify and assign priorities to database information that would aid future VAO committees in fulfilling their charge from Congress. Alternatively, or in addition, VA could establish an external advisory group that could recommend the most efficient mechanisms for mining the medical database information, which could include issuing requests for proposals for the conduct of analytic studies related to specific health outcomes of interest.

Finally, as noted in previous VAO updates, data related to the distribution of claims that have been filed by Vietnam veterans could be very informative. Although applications for compensation and appeals constitute a nonrepresentative, self-selected sample that is influenced by which conditions are already judged to be service-related, an effort to use existing VA information should include a more systematic review of the distribution of health outcomes in the databases. The information that had accumulated in VA’s records clearly generated a signal that motivated VA to ask prior VAO committees to make special evaluations of whether several specific malignancies were associated with herbicide exposure; ancillary information was adequate to enable the committees to conclude that chronic lymphocytic leukemia (CLL) and hairy-cell leukemia belong in the category of sufficient evidence of an association, but perhaps an answer for

Vietnam veterans concerning tonsil cancer will be found only by a case-control study that addresses deployment status and other emerging risk factors, such as viral infection.

In general, it is the committee’s conviction that improved data linkage and data-sharing between DOD and VA would greatly enhance the conduct of military epidemiology and the utility of its results. The committee endorses DOD’s efforts to improve collection of exposure data during current deployments; the impasses associated with missing exposure information will not impede investigations of health consequences in future veterans, as has been the case for Vietnam veterans. For optimal use, however, such DOD information on a veteran’s combat experience needs to be readily connected with future medical events, much of which resides with VA.

•  Available information should be gleaned from existing cohort studies.

In 2006, the Committee on the Disposition of the Air Force Health Study (AFHS) (IOM, 2006) recommended that all data from the AFHS be retained and suggested mechanisms by which the data could be made available to researchers. Since then, the Institute of Medicine (IOM) Medical Follow-up Agency (MFUA) has become the custodian of the data and biologic specimens (PL 109-364; 120 Stat. 2290); the specimens are held in storage at the Wright-Patterson Air Force Base under MFUA’s aegis, and funding has been provided for IOM to maintain and manage the materials and to make them available for research. A strong commitment by the federal government is required to provide sufficient funds to develop the infrastructure necessary to meet the goals of further research that uses these invaluable data and biospecimens. Moreover, dedicated funding is required so that focused analyses can be carried out by independent investigators, especially in relation to the research questions that concern the present committee. The investment would be a small fraction of the $143 million invested to date in the AFHS. Such research could clarify the various issues and would reap substantial benefits in the understanding of health issues of Vietnam veterans exposed to herbicides. Comprehensive longitudinal analyses of the data collected in the six intensive medical examinations—which include data on medical interventions (such as hospitalizations and emergency-department visits), cancer incidence, mortality, and exposures—should be conducted to investigate some or all of the health outcomes that may be associated with the exposures under consideration in the present report. Distillation of existing data could be enhanced by incorporating new results derived from assays of the biologic samples. For example, analysis of banked semen samples for epigenetic markers on sperm DNA and measurement of TCDD in seminal fluid, particularly in comparison with the subjects’ serum TCDD concentrations, could provide insight into the likelihood of male Vietnam veterans’ transmitting effects to their offspring, as well as supplementing general knowledge on paternal transmission.

In Spring 2012, the Committee on the Management of the Air Force Health

Study Data and Specimens issued a request for proposals to use the AFHS materials (data or specimens) for innovative research. Each submission was subject to an intensive review by the entire committee to ensure that it was scientifically sound and feasible. Seven proposals were approved. The investigations have begun only fairly recently, so no results are available. A second request for proposals was issued in May 2013. The committee enthusiastically supports these new and continuing research efforts.

Members of the ACC constitute the largest cohort of Vietnam veterans exposed directly to herbicides and TCDD. They were involved in the handling and distribution of the chemicals in Vietnam. ACC veterans who reported spraying herbicides as part of their duties have been shown to have increased serum TCDD concentrations; this highly exposed population has also been shown to be at increased risk for several diseases. Previous VAO committees recommended that VA conduct additional studies of ACC veterans because the population presents a unique opportunity to examine the association between health effects of exposure to TCDD and the herbicides used in Vietnam. Recently, VA launched the Army Chemical Corps Vietnam-Era Veterans Health Study to investigate the relationship between herbicide exposure during the Vietnam War and hypertension and COPD in ACC veterans. Information garnered from the study could benefit VA and future VAO committees as potential associations between exposure to the chemicals of interest and respiratory outcomes are evaluated.

Although about 250,000 US women served in the military during the Vietnam War and 5,000–7,000 women served in Vietnam, few data on the health of the deployed and nondeployed female veterans are available. More than a decade ago, Kang et al. (2000a,b) examined the prevalence of gynecologic cancers in female Vietnam veterans and of birth defects in their children. More recently, Cypel and Kang (2008) reported a mortality study of the population, but additional followup of the health status of the group has been lacking. In 2009, VA announced the start of the Health of the Vietnam Era Veteran Women’s Study, a 4-year undertaking to investigate the mental and physical health of deployed and nondeployed US women who served during the Vietnam War. Although findings from the study have yet to be published, the committee supports these VA efforts and hopes that the findings will help to elucidate how military service in Vietnam may have affected the health of female veterans who served.

At the direction of Congress, the National Vietnam Veterans Readjustment Study (1986–1988) investigated primarily psychiatric sequelae in a representative cohort of about 1,600 men and women. In 2000, Congress mandated (PL 106419) that VA assess the current physical and mental well-being of the members of that cohort. In 2001, VA contracted for the work, named the National Vietnam Veterans Longitudinal Study (NVVLS), but progress ceased within 2 years. The VA Office of Inspector General (VAOIG, 2005) ruled that “the Study was not properly planned, procured, or managed” but directed that it be completed and that provisions be made to avoid the previous problems. Because baseline

information is available on symptoms and chronic health problems in the original cohort, VAO committees have thought that completion of the NVVLS could generate useful information for future updates. On May 5, 2011, at a hearing of the House Veterans’ Affairs Committee, the chair of the VAO committee for Update 2008 had the privilege of testifying to that effect. In 2010, VA announced that a contractor had been engaged to conduct the study, with an expected completion date of 2013. Unfortunately from the perspective of VAO committees, the questionnaires and interviews are focusing on PTSD and psychologic issues rather than expanding the data-gathering to include the physical health status of this well-defined cohort. Particular attention should be paid to occurrence of conditions that have been noted to be of special interest but on which the current evidence is inadequate or insufficient to determine whether there is an association with herbicide exposure, such as COPD, brain cancer, tonsil cancer, melanoma (with particular attention given to ocular types), and Alzheimer disease. However, if a reliable, up-to-date estimate of overall mortality could be generated by following up this cohort, that presently elusive information would be quite helpful.

In 1978, the National Institute for Occupational Safety and Health (NIOSH) began to study US workers potentially exposed to TCDD. A total of 5,132 workers in 12 large manufacturing companies were included in the NIOSH cohort. The cohort has been a source of data very valuable in assessing the health effects associated with TCDD exposure, such as data from the recent study of PCP workers with and without TCDD exposure (Ruder and Yiin, 2011) that is referred to throughout the present report. The NIOSH studies have included high-quality exposure assessment, and evaluations of a wide array of health outcomes have been published. Given its value as an important source of epidemiologic data, the committee recommends that studies of the NIOSH cohort be extended.

The committee also notes that future analyses of health outcomes in those and other important study populations should be as specific as possible because generic findings, such as those for “all respiratory outcomes,” are not useful in addressing the committee’s charge to determine associations of herbicide exposures with specific health conditions.

•  Possible health effects in offspring of exposed men merit further investigation.

The rapidly expanding field of epigenetics is revealing the molecular mechanisms by which environmental agents can modify gene expression without changing DNA sequence long after exposure occurs, even in later generations—at least in the case of maternal exposure to some chemicals. There is a growing body of evidence that TCDD can induce epigenetic changes in animal models, but there are few data that support risk of adverse effects in offspring of men exposed to xenobiotics in general and the VAO chemicals of interest in particular.

VAO committees have been monitoring studies of morphologic birth defects and cancer in the offspring of exposed people. Two major information gaps

identified by the committee for Update 2010 concerning possible association between exposure of male Vietnam veterans to the chemicals of interest and the development of disease in their offspring remain unfilled: a paucity of studies of the endpoints that VAO committees have been monitoring related to paternal exposure in the absence of maternal exposure and a failure to review systematically defined clinical health conditions that are manifested later in the offspring’s lives. The few data on toxic contaminants in seminal fluid suggest that fetal exposure due to paternal transmission during later acts of intercourse is highly unlikely, but it now appears more physiologically possible that epigenetic modifications of sperm, including alterations in sperm mRNAs, microRNAs, and DNA after paternal exposure, might lead to changes in the offspring. The last of the few publications on birth defects in the offspring of male Vietnam veterans was published before the report on the children of female Vietnam veterans (Kang et al., 2000b), and none of the epidemiologic studies recently reviewed by the present committee assessed the role of paternal exposure in the occurrence of such effects. Thus, most of the available epidemiologic studies of effects in offspring are not relevant to the primary exposure group of concern: male Vietnam veterans. In addition, the epidemiologic studies of maternal exposure and adverse effects in offspring other than morphologic birth defects and cancer reviewed by this committee did not assess specific diseases in the offspring, but rather measured physiologic biomarkers that might or might not predict the potential for disease development later in life.

On the basis of those information gaps, the present committee favors renewed efforts to conduct epidemiologic studies on all the developmental effects in offspring that may be associated with paternal exposure. In addition, new studies should evaluate offspring for defined clinical health conditions that develop later in life, focusing on organ systems that have shown the greatest effects after maternal exposure, including neurologic, immune, and endocrine effects. Finally, although the committee recognizes that there is evidence that environmental exposures can affect later generations through fetal and germ line modifications, epidemiologic investigation designed to associate toxicant exposures with health effects manifested in later generations will be even more challenging to conduct than is research on adverse effects on the first generation. Thus, the committee recommends development of epidemiologic protocols to address the logistical challenge of determining whether adverse effects are being manifested in later generations as a result of paternal exposure: consideration must be given to the minimum sample size needed to detect changes if present, the most sensitive and reliable outcome measures that should be included, and the need for animal studies to provide mechanistic insight into documented epidemiologic associations. The best cohorts for revealing potential associations would be those with known, well-characterized exposure information. Another approach could be adopting a case-control structure and exploring whether information about Vietnam exposure or specific herbicide exposure could be ascertained in any of the many birth

cohorts that have been established in the last several decades. To home in on a paternal effect, however, it will be necessary to establish that the mothers did not have the opportunity for other than background exposure to the chemicals of interest. Epidemiologic studies of the potential for paternally mediated effects on health outcomes in offspring of Vietnam veterans (such as Operation Ranch Hand veterans or ACC) vs nondeployed Vietnam-era veterans are also strongly encouraged by this committee.

•  There is a need for epidemiologic studies of the incidence of COPD.

As discussed above, VA has initiated a renewed morbidity study of the ACC cohort and may publish findings in time for the next VAO update. The committee would welcome studies in additional populations of COPD diagnosed individuals using appropriate functional tests with adjustment for smoking status and consideration of comorbidities that could contribute to death from COPD.

•  There is a need for new animal models to elucidate mechanisms of diseases and disease progression.

The committee believes that experimental research in the mechanisms that underlie human health outcomes (particularly cardiovascular disease, B-cell cancers, and paternally mediated effects in offspring) could provide valuable information related to the risk of disease in Vietnam veterans and their children. The development of animal models of neurologic outcomes and various chronic health conditions and their progression would be useful for understanding the possible contributions of the chemicals of interest to compromise the health of aging Vietnam veterans. Specifically, determining the mechanism by which dioxin-like chemicals induce B-cell cancers and how exposure to dioxin-like chemicals alters susceptibility to obesity and components of metabolic syndrome would fill important knowledge gaps. Furthermore, animal models elucidating the effects of paternal exposure on the development of disease in offspring would be very informative, particularly in identifying the timing and duration of exposure that are most critical and the susceptibility of specific organ systems to disease development in offspring later in life. Animal studies of the mechanisms of inhibition of fetal growth, particularly in male offspring, after maternal exposure could help to elucidate findings seen in some epidemiologic studies that examined maternal exposure and birth weight.

The predecessors of this committee have offered similar recommendations of additional research to resolve outstanding questions. This committee remains particularly concerned about stroke, COPD, hypertension, melanoma, tonsil cancer, Alzheimer disease, and paternally transmitted effects on offspring.

REFERENCES¹

Alavanja MC, Sandler DP, Lynch CF, Knott C, Lubin JH, Tarone R, Thomas K, Dosemeci M, Barker J, Hoppin JA, Blair A. 2005. Cancer incidence in the Agricultural Health Study. Scandinavian Journal of Work, Environment and Health 31(Suppl 1):39–45.

Cypel Y, Kang H. 2008. Mortality patterns among women Vietnam-era veterans: Results of a retrospective cohort study. Annals of Epidemiology 18(3):244–252.

IOM (Institute of Medicine). 2006. Disposition of the Air Force Health Study. Washington, DC. National Academies Press.

Kang HK, Mahan CM, Lee KY, Magee CA, Selvin S. 2000a. Prevalence of gynecologic cancers among female Vietnam veterans. American Journal of Industrial Medicine 38(4):447–454.

Kang HK, Mahan CM, Lee KY, Magee CA, Mather SH, Matanoski G. 2000b. Pregnancy outcomes among US women Vietnam veterans. American Journal of Industrial Medicine 38(4):447–454.

Ruder AM, Yiin JH. 2011. Mortality of US pentachlorophenol production workers through 2005. Chemosphere 83(6):851–861.

VA (Department of Veterans Affairs). 2010. Health effects not associated with exposure to certain herbicide agents. Federal Register 75(109):32540–32553.

VAOIG (Department of Veterans Affairs Office of Inspector General). 2005. Audit of VA Acquisition Practices for the National Vietnam Veterans Longitudinal Study (Report No. 04-02330-212). Washington, DC.

__________________

¹Throughout this report, the same alphabetic indicator after year of publication is used consistently for a given reference when there are multiple citations by the same first author in a given year. The convention of assigning the alphabetic indicators in order of citation in a given chapter is not followed.